revision sistematica riesgo de acidosis con tratamiento metformina

Riesgo de acidosis láctica fatal y no fatal con el uso de metforminapara la diabetes mellitus tipo 2

Salpeter S, Greyber E, Pasternak G, Salpeter E

Reproducción de una revisión Cochrane, traducida y publicada en La Biblioteca Cochrane Plus, 2008, Número 2

Producido por

Si desea suscribirse a "La Biblioteca Cochrane Plus", contacte con:

Update Software Ltd, Summertown Pavilion, Middle Way, Oxford OX2 7LG, UKTel: +44 (0)1865 513902 Fax: +44 (0)1865 516918E-mail: [email protected] web: http://www.update-software.com

Usado con permiso de John Wiley & Sons, Ltd. © John Wiley & Sons, Ltd.Ningún apartado de esta revisión puede ser reproducido o publicado sin la autorización de Update Software Ltd.Ni la Colaboración Cochrane, ni los autores, ni John Wiley & Sons, Ltd. son responsables de los errores generadosa partir de la traducción, ni de ninguna consecuencia derivada de la aplicación de la información de esta Revisión,ni dan grantía alguna, implícita o explícitamente, respecto al contenido de esta publicación.El copyright de las Revisiones Cochrane es de John Wiley & Sons, Ltd.El texto original de cada Revisión (en inglés) está disponible en www.thecochranelibrary.com.

http://www.update-software.com

ÍNDICE DE MATERIAS

RESUMEN...................................................................................................................................................................1

RESUMEN EN TÉRMINOS SENCILLOS....................................................................................................................2

ANTECEDENTES........................................................................................................................................................2

OBJETIVOS.................................................................................................................................................................4

CRITERIOS PARA LA VALORACIÓN DE LOS ESTUDIOS DE ESTA REVISIÓN......................................................4

ESTRATEGIA DE BÚSQUEDA PARA LA IDENTIFICACIÓN DE LOS ESTUDIOS....................................................4

MÉTODOS DE LA REVISIÓN.....................................................................................................................................5

DESCRIPCIÓN DE LOS ESTUDIOS..........................................................................................................................7

CALIDAD METODOLÓGICA.......................................................................................................................................8

RESULTADOS.............................................................................................................................................................8

DISCUSIÓN.................................................................................................................................................................9

CONCLUSIONES DE LOS AUTORES......................................................................................................................10

POTENCIAL CONFLICTO DE INTERÉS...................................................................................................................10

FUENTES DE FINANCIACIÓN..................................................................................................................................11

REFERENCIAS.........................................................................................................................................................11

TABLAS......................................................................................................................................................................23

Characteristics of included studies.....................................................................................................................23

Characteristics of excluded studies....................................................................................................................96

CARÁTULA................................................................................................................................................................98

RESUMEN DEL METANÁLISIS.................................................................................................................................99

GRÁFICOS Y OTRAS TABLAS................................................................................................................................100

01 Fatal/nonfatal lactic acidosis........................................................................................................................100

01 Incidencia de acidosis láctica por paciente por año (metformina menos sin metformina)....................100

02 Niveles del lactato sanguíneo......................................................................................................................101

01 Efecto neto del tratamiento, niveles de lactato (mmol/L)......................................................................101

02 Niveles medios de lactato con el tratamiento (mmol/L)........................................................................102

03 Niveles pico de lactato estimulado (mmol/L)........................................................................................103

Riesgo de acidosis láctica fatal y no fatal con el uso de metformina para la diabetes mellitus tipo 2 i

Copyright © John Wiley & Sons Ltd. Usado con permiso de John Wiley & Sons, Ltd.

Riesgo de acidosis láctica fatal y no fatal con el uso de metforminapara la diabetes mellitus tipo 2

Salpeter S, Greyber E, Pasternak G, Salpeter E

Esta revisión debería citarse como:Salpeter S, Greyber E, Pasternak G, Salpeter E. Riesgo de acidosis láctica fatal y no fatal con el uso de metformina para la diabetesmellitus tipo 2 (Revisión Cochrane traducida). En: La Biblioteca Cochrane Plus, 2008 Número 2. Oxford: Update Software Ltd.Disponible en: http://www.update-software.com. (Traducida de The Cochrane Library, 2008 Issue 2. Chichester, UK: John Wiley& Sons, Ltd.).Fecha de la modificación más reciente: 16 de noviembre de 2005Fecha de la modificación significativa más reciente: 16 de noviembre de 2005

RESUMEN

AntecedentesLa metformina es un agente oral hipoglucemiante que se utiliza en el tratamiento de la diabetes mellitus tipo 2. Los resultadosdel UK Prospective Diabetes Study (Estudio prospectivo de diabetes del Reino Unido) indican que el tratamiento con metforminase asocia a una reducción en la mortalidad total en comparación con otros tratamientos hipoglucemiantes. Sin embargo, se creeque la metformina aumenta el riesgo de acidosis láctica y se la considera contraindicada para muchas enfermedades hipoxémicascrónicas que se pueden asociar con la acidosis láctica, como la enfermedad cardiovascular, renal, hepática y pulmonar, y la edadavanzada.

ObjetivosEvaluar la incidencia de la acidosis láctica fatal y no fatal con el uso de la metformina en comparación con el placebo y otrostratamientos que reducen la glucosa en pacientes con diabetes mellitus tipo 2. Un objetivo secundario fue evaluar los niveles delactato sanguíneo para aquellas personas tratadas con metformina en comparación con placebo o tratamientos sin metformina.

Estrategia de búsquedaSe realizó una búsqueda en The Cochrane Library (hasta 8/2005), MEDLINE (hasta 8/2005), EMBASE (hasta 11/2000), OLDMEDLINE y REACTIONS (hasta 8/2005) para identificar todos los estudios de tratamiento con metformina desde 1966 hastaagosto de 2005. Se utilizó el Cumulated Index Medicus para buscar artículos pertinentes desde 1959 hasta 1965. La búsqueda seamplió mediante el examen de las referencias de los artículos identificados y el contacto con los principales investigadores. Fechade la última búsqueda: agosto 2005.

Criterios de selecciónLos ensayos prospectivos en pacientes con diabetes tipo 2 que duraron más de un mes se incluyeron si evaluaban la metformina,sola o en combinación con otros tratamientos, en comparación con el placebo o cualquier otro tratamiento que reduzca la glucosa.También se incluyeron estudios observacionales de cohortes del tratamiento con metformina que duraron más de un mes.

Recopilación y análisis de datosDos revisores seleccionaron de forma independiente los ensayos para ser incluidos, evaluaron la calidad de los estudios y extrajeronlos datos. La incidencia de la acidosis láctica fatal y no fatal se registró como casos por pacientes tratados por año, para eltratamiento con metformina y para el placebo u otros tratamientos. Se calculó el límite superior para la verdadera incidencia delos casos en los grupos con metformina y sin metformina mediante las estadísticas de Poisson. En un segundo análisis, se midieronlos niveles de lactato como un cambio neto desde el valor inicial o como valores medios del tratamiento (basales y estimuladosmediante alimentos o ejercicios) para los grupos de tratamiento y comparación. Los resultados agrupados se registraron comodiferencia de medias ponderada (DMP) en mmol/L, mediante el modelo de efectos fijos para los datos continuos.

Resultados principalesLos datos combinados de los 206 ensayos comparativos y los estudios de cohortes no revelaron casos de acidosis láctica fatal ono fatal en 47 846 pacientes tratados por año con metformina o en 38 221 pacientes tratados por año en el grupo sin metformina.

Página 1


Mediante el uso de las estadísticas de Poisson con intervalos de confianza del 95%, el límite superior para la verdadera incidenciade la acidosis láctica asociada a la metformina fue 6,3 casos por cada 100 000 pacientes tratados por año y el límite superior parala verdadera incidencia de la acidosis láctica en el grupo sin metformina fue de 7,8 casos por cada 100 000 pacientes tratados poraño. No hubo diferencia alguna en los niveles de lactato, ya sea en los niveles promedios del tratamiento o en un cambio netodesde el valor inicial, para la metformina en comparación con el placebo u otros tratamientos sin biguanidas. Los niveles promediode lactato fueron ligeramente inferiores en el tratamiento con metformina que en el de fenformina (DMP -0,75 mmol/L; IC del95%: -0,86; -0,15).

Conclusiones de los autoresNo hay pruebas en los ensayos comparativos prospectivos o en los estudios observacionales de cohortes de que la metforminase asocie con un mayor riesgo de acidosis láctica o un aumento en los niveles del lactato, en comparación con otros tratamientoshipoglucemiantes si se prescribe en las condiciones del estudio.

✦

RESUMEN EN TÉRMINOS SENCILLOS

El uso de la metformina probablemente no está vinculado al riesgo de la acidosis lácticaDurante mucho tiempo, se ha creído que la metformina, una medicación usada para reducir los niveles de glucosa en pacientescon diabetes mellitus, aumenta el riesgo de un trastorno metabólico conocido como acidosis láctica. Esta revisión resumió losdatos de todos los estudios comparativos y observacionales conocidos que duraron más de un mes y no encontró casos de acidosisláctica fatal o no fatal en 47 846 pacientes tratados por año con metformina o en 38 221 pacientes tratados por año sin metformina.Los niveles promedio de lactato medidos durante el tratamiento con metformina no fueron diferentes que para el placebo o paraotros medicamentos usados para tratar la diabetes. En resumen, no hay pruebas en la actualidad, de la asociación de la metforminaa un mayor riesgo de acidosis láctica si se prescribe bajo las condiciones del estudio.

✦

ANTECEDENTES

La diabetes mellitus es un trastorno metabólico que resulta deun defecto en la secreción de insulina, la acción de la insulinao en ambos (DeFronzo 1999). Una consecuencia de esto es lahiperglucemia crónica con trastornos en el metabolismo decarbohidratos, grasas y proteínas. Entre las complicaciones alargo plazo de la diabetes mellitus se encuentran la retinopatía,la nefropatía, la neuropatía y un mayor riesgo de enfermedadescardiovasculares. Para obtener un resumen detallado de ladiabetes mellitus, ver título "Información adicional" en el GrupoCochrane de Trastornos Metabólicos y Endocrinos (Metabolicand Endocrine Disorders Group) en la Cochrane Library (ver"Acerca de la Colaboración Cochrane", "Grupos Colaboradoresde Revisión [Collaborative Review Groups]). Para obtener unaexplicación de los términos metodológicos, ver el glosarioprincipal en The Cochrane Library.

La metformina es un agente hipoglucemiante que se ha usadocon mayor frecuencia durante los últimos años, en especial enpacientes obesos con diabetes tipo 2 en la que los niveles deglucemia no se pueden controlar de otra manera que a travésde medios farmacológicos. Sin embargo, todavía debenabordarse tres preguntas principales. Primero, ¿cómo afecta elfármaco a la mortalidad total y al desarrollo de complicacionesa largo plazo relacionadas con la diabetes? y ¿son estos efectossimilares en todos los grupos de pacientes con diabetes tipo 2?

En segundo lugar, ¿cuál es el efecto en los factores de riesgocardiovascular como la obesidad, dislipidemia y la hipertensión?y ¿se asocia este efecto con los cambios en la morbilidad y lamortalidad cardiovascular? La tercera pregunta aborda lasinquietudes por la seguridad del fármaco; específicamente ¿cuáles el riesgo de acidosis láctica fatal y no fatal asociada con eluso de metformina? Estas tres preguntas se tratarán en tresrevisiones separadas. Esta revisión evalúa el riesgo de laacidosis láctica atribuida al uso de metformina en comparacióncon el placebo y otros agentes usados para el control glucémicoen pacientes con diabetes tipo 2. Se actualizarán las revisionesde forma continua para incluir estudios relevantes nuevos.

MetforminaEl clorhidrato de metformina es una biguanida que ha sido deuso clínico durante más de 45 años (DeFronzo 1999; Sterne1959). A diferencia de las sulfonilureas, las biguanidas no tienenun efecto hipoglucemiante en las personas sanas y no estimulanla liberación de insulina (Cusi 1996). A través de su efectohipoglucemiante, la metformina reduce los niveles de glucemiaen ayunas y posprandiales en pacientes con diabetes tipo 2.Aunque no se ha establecido completamente el mecanismopreciso de este efecto, las pruebas indican que el fármaco mejorala sensibilidad tanto periférica como hepática a la insulina,reduce la producción de glucosa hepática basal y aumenta lacaptación estimulada por la insulina y utilización de la glucosapor los tejidos periféricos (AHFS 1999). La metformina, aun

Riesgo de acidosis láctica fatal y no fatal con el uso de metformina para la diabetes mellitus tipo 2

Página 2


en dosis excesivas, normalmente no reduce las concentracionesde glucosa por debajo de la euglucemia. La metformina seacumula en la pared del intestino pero no parece tener efectosclínicamente importantes en la absorción de la glucosa.

Además de su influencia en el metabolismo de carbohidratos,se cree que la metformina tiene otros efectos positivosrelacionados a la diabetes tipo 2 y su pronóstico a largo plazo.Es posible que haya una mejoría moderada en los lípidos séricos,especialmente una reducción de los triglicéridos séricos enayunas así como en las concentraciones de colesterol total yLDL (LDL = lipoproteína de baja densidad). Además, eltratamiento con metformina puede asociarse con la pérdida depeso o una estabilización en el aumento de peso. Entre losmecanismos sugeridos para este efecto se encuentran la ausenciade un efecto hiperinsulinémico (que de estar presente puedeaumentar el apetito o la lipogénesis) y la ingesta dietéticareducida causada por los efectos adversos gastrointestinales dela metformina. Actualmente, no hay pruebas concluyentes sobreel efecto de la metformina en el sistema fibrinolítico y laagregación plaquetaria que tienen una función en el desarrollode la trombosis en las arterias coronarias (Palumbo 1998).

Estudios de la metforminaSe han publicado diversos ensayos que usan la metformina solao en combinación con otros fármacos en pacientes con diabetesmellitus tipo 2. El UK Prospective Diabetes Study (UKPDS)fue el primer ensayo grande que evaluó los resultados clínicosa largo plazo en relación al tratamiento con metformina enpersonas que presentaban una diabetes tipo 2. El estudio incluyópacientes con sobrepeso con una diabetes tipo 2 diagnosticadarecientemente, con un promedio de 53 años de edad, que nopresentaban una coronariopatía o una contraindicación altratamiento. Los resultados indicaron que la monoterapia conmetformina producía una reducción en las variables deevaluación relacionadas con la diabetes y también en lamortalidad relacionada con la diabetes y en la mortalidad total,comparada con la insulina, el tratamiento con sulfonilureas osólo la dieta (UKPDS-34 1998). No hubo casos de acidosisláctica en ninguno de los grupos.

Recientemente se han publicado tres metanálisis que evaluaronel efecto de la metformina en la regulación de la glucosa y elpeso en comparación con el placebo o las sulfonilureas.(Campbell 1995; Johansen 1999; Guthrie 1997). El metanálisisde Campbell indicó que la metformina tuvo un efectocomparativo hipoglucemiante con relación a las sulfonilureas,pero hubo una mayor reducción de peso en el grupo demetformina. En los metanálisis similares de Johansen y Guthrie,la metformina y la sulfonilurea reducían la glucemia y lahemoglobina glucosilada de forma significativa en comparacióncon el placebo, pero el peso corporal fue significativamenteinferior en el grupo de metformina comparado con el desulfonilurea. Estos estudios no evaluaron la mortalidad, loscriterios de valoración relacionados con la diabetes o los efectosadversos del tratamiento.

Efectos adversos de la metforminaSe informa que los efectos adversos, principalmentegastrointestinales, ocurren en el 20% al 30% de los pacientesque reciben un tratamiento con metformina y se debeinterrumpir la administración del fármaco en menos del 5% delos pacientes (DeFronzo 1999). La diarrea, las náuseas, losvómitos, la distensión abdominal, los calambres o doloresabdominales, la flatulencia y la anorexia son los síntomasgastrointestinales más comunes asociados al tratamiento conmetformina. Otros efectos adversos que se informaron son lacefalea, la agitación, el mareo y el cansancio.

La acidosis láctica es una enfermedad rara, metabólica ypotencialmente fatal que puede ocurrir siempre que exista unahipoperfusión tisular substancial y una hipoxia (Kreisberg 1980;Olivia 1970). La acidosis láctica se caracteriza por la elevadaconcentración de lactato sanguíneo (que excede los 45 mg/dlo los 5,0 mEq/L), un pH sanguíneo reducido (menor que 7,35)y trastornos de electrólitos con una mayor insuficiencia deaniones. La mortalidad en los casos informados esaproximadamente del 42% al 50% (Bailey 1996; Misbin 1998).Se cree que las biguanidas reducen la gluconeogénesis de laalanina, el piruvato y el lactato y es posible que los niveles delácido láctico se acumulen en ciertas circunstancias (Stang 1999).Una biguanida anterior, la fenformina, se retiró del mercadoporque se asociaba con una tasa informada de acidosis lácticade 40 a 64 casos cada 100 000 pacientes tratados por año(DeFronzo 1999; Stang 1999). La metformina difiere de lafenformina en la estructura molecular y en la farmacocinética(Sulkin 1997). Se piensa que la metformina, a diferencia de lafenformina, mejora la oxidación de la glucosa sin afectar demanera significativa la producción de lactato en ayunas de lostejidos periféricos (Cusi 1996).

Se desconoce la verdadera incidencia de la acidosis lácticainducida por metformina. La Food and Drug Administrationha calculado que la tasa de acidosis láctica fatal o no fatal esde cinco casos cada 100 000 personas tratadas en el curso deun año (Misbin 1998). Los estudios basados en la poblaciónhan calculado una tasa de dos a nueve casos de acidosis lácticaen usuarios de metformina cada 100 000 personas por año(Campbell 1985; Stang 1999; Wilholm 1993). Sin embargo, lamayoría de los casos informados se presentaron en pacientescon enfermedades agudas graves, como la insuficiencia renal,que podrían haber sido la causa de la acidosis láctica (Brown1998; Misbin 1998). Para calcular el riesgo específicamenteatribuible a la metformina, se debe evaluar la tasa deantecedentes de la acidosis láctica en pacientes con diabetestipo 2 que no han recibido un tratamiento con metformina. Coneste fin, se utilizó una base de datos para medir las tasas deincidencia en los pacientes con diabetes tipo 2 en los EstadosUnidos antes de que se introdujera la metformina y se encontróuna tasa de nueve casos cada 100 000 personas por año (Brown1998). Esto plantea la cuestión de si los pacientes con diabetestipo 2 corren un mayor riesgo de desarrollar acidosis láctica


Página 3


mediante el uso de metformina en comparación con otrostratamientos que reducen la glucosa.

Los datos disponibles indican que el uso de la metformina enpacientes obesos con diabetes mellitus tipo 2 se asocia a unareducción en la morbilidad y mortalidad cardiovascularcomparado con la insulina, la sulfonilurea o sólo la dieta(UKPDS-34 1998). Sin embargo, actualmente el uso de lametformina se considera contraindicado en muchasenfermedades crónicas que puedan aumentar el riesgo de laanoxia tisular (falta de oxígeno) y el desarrollo de la acidosisláctica, como las enfermedades cardiovasculares, renales,pulmonares y hepáticas. Estas restricciones reducen de manerasignificativa el número de pacientes que podrían beneficiarsedel tratamiento con metformina. Esta revisión evalúa el riesgode la acidosis láctica fatal y no fatal asociada con la metformina.Otros efectos adversos asociados con el uso de la metforminase evalúan en otra revisión.

OBJETIVOS

Evaluar el riesgo de la acidosis láctica fatal y no fatal asociadaal uso de la metformina en personas con diabetes mellitus tipo2, en comparación con el placebo u otros tratamientos parareducir la glucosa. Un objetivo secundario fue evaluar losniveles del lactato sanguíneo, medidos al inicio y durante eltratamiento, para comparar la metformina con el placebo u otrostratamientos hipoglucemiantes.

CRITERIOS PARA LA VALORACIÓN DE LOSESTUDIOS DE ESTA REVISIÓN

Tipos de estudios

Se incluyeron los ensayos clínicos prospectivos en pacientescon diabetes mellitus tipo 2 si evaluaban la metformina, sola oen combinación con otros tratamientos, en comparación con elplacebo u otro tratamiento hipoglucemiante, y si duraban almenos un mes. Se incluyeron los ensayos clínicos aunque noestuvieran asignados al azar o cegados. Además, se incluyeronen el análisis todos los estudios observacionales de cohortesque evaluaron durante al menos un mes el uso de la metformina,siempre que proporcionaran el número de pacientes y laduración del tratamiento. Se evaluaron por separado los ensayosexcluidos que duraron menos de un mes para ver si había algúncaso de acidosis láctica.

Tipos de participantes

Los participantes estudiados fueron adultos con diabetes mellitustipo 2. Para ser coherentes con los cambios en los criterios declasificación y diagnóstico de la diabetes mellitus tipo 2 con elpaso de los años, se debería establecer el diagnóstico mediantelos criterios estándar en el momento del ensayo.

Tipos de intervención

Metformina, sola o en combinación con otros tratamientos,versus el placebo o una de las siguientes intervenciones usadascon la intención de disminuir los niveles de la glucosasanguínea:

1. Placebo2. Sulfonilurea (por ejemplo, glibenclamida)3. Tiazolidinediona (por ejemplo, rosiglitazona)4. Meglitinida (por ejemplo, repaglinida)5. Inhibidor de la alfa glucosidasa (por ejemplo, acarbosa ymiglitol)6. Insulina7. Intervención no farmacológica (por ejemplo, dietas)8. Cualquier combinación de las anteriores

Los datos sobre los participantes tratados con fenformina no seincluyeron en el análisis para la acidosis láctica, pero sí seincluyeron en las mediciones de los niveles de lactato.

Tipos de medidas de resultado

Medidas de resultados primarias1. Muerte descrita como consecuencia de acidosis láctica2. Casos informados de acidosis láctica no fatal, definidos porel investigador

Medidas de resultadosecundarias1. Niveles del lactato sanguíneo para la metformina encomparación con el placebo u otros tratamientos sin biguanidasy comparado con la fenformina.

Modificadores del efectoCasos informados de insuficiencia renal o del cambio encualquier enfermedad concomitante hipóxica (p.ej.,enfermedades pulmonares). Si se tenían que identificar casosde acidosis láctica, se evaluaba su asociación con la enfermedadconcomitante.

ESTRATEGIA DE BÚSQUEDA PARA LAIDENTIFICACIÓN DE LOS ESTUDIOS

Dos investigadores (SS, EG) desarrollaron de forma conjuntaestrategias de búsqueda con la ayuda de un bibliotecario delservicio de información y el coordinador de búsqueda deensayos del Grupo Cochrane de Trastornos Metabólicos yEndocrinos (Cochrane Metabolic and Endocrine DisordersGroup).

Búsquedas electrónicasSe realizó una búsqueda exhaustiva de las siguientes bases dedatos para identificar ensayos clínicos en humanos o metanálisisrelevantes: The Cochrane Library (incluyendo el RegistroCochrane de Ensayos Controlados [Cochrane Controlled TrialsRegister] (CENTRAL) y la Database of Abstracts of Reviewsof Effectiveness (DARE)) (8/2005), MEDLINE (1966-8/2005),OLD MEDLINE, Reactions (1983-8/2005), y EMBASE


Página 4


(1974-11/2000). La siguiente estrategia de búsqueda paraMedline se adaptó para buscar otras bases de datos médicas:

NOTAS: a menos que se mencione lo contrario, los términosde búsqueda son términos de texto libre; MeSH: Medical subjectheading (título de tema médico) (término índice médico deMEDLINE); un asterisco (*) significa cualquier carácter(es)"

Diabetes mellitus tipo 21. Diabetes mellitus, non-insulin-dependent [MeSH, allsubheadings and categories included]2. NIDDM3. (Non insulin* dep*) OR (Noninsulin* dep*) OR (Non insulindep*)4. (Typ* II diabet*) OR (Typ* 2 diabet*) OR (diabet* typ* 2)OR (diabet* typ* II)5. #1 OR #2 OR #3 OR #4

Metformin6. Biguanides [MeSH, all subheadings and categories included]7. Biguanid*8. Metformin [MeSH, all subheadings and categories included]9. Glucophag*10. Metformin*11. #6 or #7 or #8 or #9 or #10

TYPE 2 DIABETES AND METFORMIN12. #5 AND #11

Búsqueda manualLa búsqueda se amplió aún más después de la investigación delas referencias de los artículos o revisiones identificadas, y delos resúmenes de un simposio clínico, que se informó en larevista Diabetologia, volumen 43, suplemento 1, 2000. Se utilizóel Cumulated Index Medicus para buscar artículos pertinentesdesde 1959 hasta 1965.

Otras búsquedasAdemás, se intentó establecer contacto con los autores de losestudios identificados para obtener referencias adicionales,ensayos no publicados, ensayos en curso o datos que faltabanno informados en los ensayos originales. De igual manera, seestableció contacto con el fabricante de la metformina,Bristol-Myers Squibb Company, para obtener información sobrelos ensayos, publicados o no publicados, acerca de lametformina.

Se incluyeron estudios publicados en cualquier idioma. No seidentificaron palabras clave adicionales de relevancia duranteninguna de las búsquedas electrónicas o las demás búsquedas.Si, en búsquedas futuras, se encuentran palabras claveadicionales, se modificarán las estrategias de la búsquedaelectrónica para incorporar estos términos. La fecha de labúsqueda más reciente fue agosto 2005.

MÉTODOS DE LA REVISIÓN

Selección de ensayos

Dos revisores (GP, SS), de forma independiente, examinaronlos títulos, los resúmenes y las palabras clave de cada registroencontrado en la búsqueda. Se obtuvieron artículos completospara una evaluación posterior cuando la información dadasugería que el estudio evaluaba el uso de la metformina enpacientes con diabetes mellitus. En caso de dudas con respectoa estos criterios, derivadas de la información proporcionada enel título y el resumen, se recuperó el artículo completo paraaclararlas. Además, se obtuvo cualquier ensayo clínicopotencialmente relevante encontrado al explorar las referenciasde artículos identificados o revisiones.

Dos investigadores (SS, EG) evaluaron de forma independientelos estudios obtenidos para su inclusión, y se logró el consensoen los casos de controversias. El acuerdo del porcentajeobservado entre los evaluadores de la inclusión antes delconsenso se midió a través de la estadística kappa (Fleiss 1981).

Evaluación de la calidad de los ensayosLa calidad metodológica de cada estudio se evaluó según loscriterios de calidad modificados a partir de Schulz, Jadad yStroup (Jadad 1996; Schulz 1995; Stroup 2000). Se dividieronlos estudios en cinco categorías.

Para los ensayos controlados aleatorios, se estudiaron lossiguientes factores:1. Minimización del sesgo de selección - a) ¿fue elprocedimiento de asignación al azar adecuado? b) ¿fue elocultamiento de la asignación adecuado?2. Minimización del sesgo de ejecución: ¿estaban losparticipantes y las personas que administraban el tratamientocegados a la intervención?2. Reducción del sesgo de deserción - a) ¿fueron los retiros ylos abandonos: completamente descritos? b) ¿se realizó elanálisis por intención de tratar (intention-to-treat analysis)?4. Minimización del sesgo de detección - ¿estaban losevaluadores de resultado cegados a la intervención?

Según estos criterios, los ensayos se subdividieron ampliamenteen estas tres categorías:A - se cumplieron todos los criterios de calidad: bajo riesgo desesgo.B - uno o más de los criterios de calidad se cumplieron sóloparcialmente: riesgo moderado de sesgo.C - no cumple con uno o más criterios: alto riesgo de sesgo.

Para los ensayos no aleatorios, se utilizaron los siguientescriterios:D - Ensayos controlados abiertos no aleatoriosE - Estudios observacionales de cohortes

Dos revisores (SS, EG) evaluaron de forma independiente cadaensayo y se logró el consenso en los casos de desacuerdo. Secalculó el acuerdo entre evaluadores antes del consenso a travésde la estadística kappa. Sin embargo, debido a que no seencontraron eventos en los resultados, no se realizaron análisisde sensibilidad mediante las evaluaciones de calidad.

Extracción de los datos


Página 5


Dos revisores (SS, EG) extrajeron de forma independiente losdatos sobre detalles de la población del estudio, la intervencióny los resultados mediante un formulario de extracción de datosestándar. El formulario de extracción de datos incluía lossiguientes ítems:

1. Información general: publicado/ no publicado, título, autores,dirección de contacto, idioma de publicación, año depublicación, publicaciones duplicadas, patrocinador, contexto.2. Características del ensayo: diseño, duración, asignación alazar (y método), cegamiento (simple, doble y triple), métodoy verificación del cegamiento.3. Intervención(es): placebo incluido, intervenciones (dosis,vía, tiempo), intervenciones de comparación (dosis, vía, tiempo),fármacos (dosis, vía, tiempo).4. Pacientes: criterios de inclusión y exclusión, número total ynúmero de participantes en los grupos de comparación, sexo,edad, características seleccionadas al inicio del estudio, criteriosde diagnóstico, semejanza entre los grupos al inicio del estudio(con inclusión de cualquier comorbilidad), retiros o pérdidasdurante el seguimiento (descripción), subgrupos.5. Resultados: muertes atribuidas a la acidosis láctica, acidosisláctica no fatal, niveles de lactato, insuficiencia renal,empeoramiento de las enfermedades hipoxémicasconcomitantes, duración de seguimiento, calidad de lanotificación de los resultados.

Las diferencias en la extracción de datos se resolvieron medianteel consenso y la consulta del artículo original. Los casos deacidosis láctica se tabulaban según el informe del investigador.Además, se buscó la información de los autores de los estudiosprincipales.

Análisis de los datosPara expresar el efecto del tratamiento sobre la acidosis lácticafatal y no fatal como diferencia de riesgo, se consideró laincidencia de eventos del uso de metformina, sola o encombinación con otros tratamientos, y luego se restó laincidencia de eventos del uso del placebo o los tratamientosalternativos. Cuando se encontraban eventos no fatales, elprimer evento se consideraba para cualquier paciente. Se habíapropuesto combinar los resultados mediante el modelo deefectos fijos para datos dicotómicos. A continuación, ladiferencia de riesgo se convertía en el número necesario paradañar (NND). Además, los resultados se podrían expresar comoel riesgo relativo de la acidosis láctica asociada con el uso demetformina en comparación con el placebo o el tratamiento sinmetformina. Sin embargo, cuando no se encontraron casos deacidosis láctica en alguno de los grupos de tratamiento, el límitesuperior para la verdadera incidencia de la acidosis láctica enel grupo de metformina y en el grupo sin metformina se calculópor separado a través de las estadísticas de Poisson.

Se probó la heterogeneidad entre estudios al utilizar laestadística de Ji cuadrado para presuponer la homogeneidad,con una significación estadística establecida en p < 0,1. Lasposibles fuentes de heterogeneidad se evaluaron mediante

análisis de sensibilidad y de subgrupos, como se describe acontinuación. Esto no se realizó debido a que no se encontraroncasos de acidosis láctica. Se probó el sesgo del estudio pequeñomediante la técnica del gráfico de embudo (funnel plot). Elanálisis se realizó mediante MetaView 4.1 (Cochrane Software).

Una vez que los resultados agrupados no revelaron casos deacidosis láctica, se decidió informar sobre ensayos que midieranlos niveles del lactato sanguíneo para la metformina, comparadocon el placebo o los tratamientos sin biguanidas, y también encomparación con la fenformina. Se analizaron tres resultadospara el grupo de metformina en contraste con los grupos decomparación: (1) el cambio en los niveles de lactato desde elinicio hasta el tratamiento, (2) la media de los niveles de lactatoregistrados durante el tratamiento, y (3) el cambio en los nivelesde lactato del tratamiento desde un estado basal hasta laestimulación máxima, ya sea con alimentos o ejercicios. Losresultados se registraron como diferencia de medias ponderada(DMP), en mmol/L, y se agruparon mediante el uso del modelode efectos fijos para los datos continuos.

Análisis de subgruposSe planificó realizar análisis de subgrupos a fin de investigarla asociación de la acidosis láctica con los siguientes factores:

1. Pacientes con enfermedades concomitantes hipoxémicas,p.ej., insuficiencia renal crónica (creatinina > 1,5 mg/dl) oinsuficiencia renal, insuficiencia cardíaca congestiva,enfermedad hepática, enfermedades pulmonares y arteriopatíaperiférica.2. Edad superior a los 65 años.3. Uso de metformina administrada como monoterapia o encombinación con otros fármacos.4. Diferentes intervenciones de comparación.

Estos análisis no se realizaron debido a que no había casos paraestudiar. En cambio, se obtuvo información acerca de cuántospacientes eran mayores de 65 años o cuántos se considerabaque presentaban enfermedades hipoxémicas concomitantes.

Análisis de sensibilidadSe planificaron análisis de sensibilidad para investigar lainfluencia de los siguientes factores en el tamaño del efecto:

1. Repetición del análisis con exclusión de los estudios nopublicados, ensayos no aleatorios y ensayos sin cegamiento.2. Repetición del análisis según la calidad del estudio, como seespecificó anteriormente.3. Repetición del análisis con exclusión de estudios muy largoso de gran tamaño para establecer cuánto dominan los resultados.4. Repetición del análisis con exclusión de estudios financiadospor patrocinadores industriales.

La solidez de los resultados también se comprobó mediante larepetición de los análisis que utilizaban diferentes medidas detamaño del efecto (diferencia de riesgo, riesgo relativo, etc.) ydiferentes modelos estadísticos (modelos de efectos fijos yaleatorios). Debido a que no se encontraron casos de acidosisláctica, no se realizaron los análisis de sensibilidad.


Página 6


DESCRIPCIÓN DE LOS ESTUDIOS

Ensayos identificadosLa búsqueda en las bases de datos electrónicas identificó 1192artículos, y de éstos, 192 eran estudios potencialmenterelevantes acerca del uso de la metformina en pacientes condiabetes tipo 2 (181 de MEDLINE y 11 artículos adicionalesde EMBASE). Después de revisar los artículos y lasbibliografías, el Cumulated Index Medicus y los resúmenes deun simposio clínico, se identificaron 70 estudios adicionales.De estos 262 estudios, 205 cumplieron con los criterios deinclusión. No se encontraron otros artículos a través de lacorrespondencia con los autores, pero se recibió del Dr. EvertineAbbink información de un ensayo no publicado adicional.

Acuerdo entre evaluadoresLa puntuación kappa para el acuerdo entre evaluadores en laselección de ensayos fue 0,88 (IC: 0,78 a 0,98) lo que indicabuen acuerdo. Se alcanzó el consenso sobre los demás ensayosal verificar de forma conjunta el artículo.

Datos faltantesSe intentó establecer contacto con 102 de los autores para losensayos comparativos a través de la dirección decorrespondencia mencionada y se recibieron 30 respuestas. Los30 que respondieron declararon no conocer caso alguno deacidosis láctica en cualquiera de sus 34 ensayos. Además, elfabricante de la metformina, Bristol-Myers Squibb Company,respondió y declaró que no tenía ensayos no publicados parainformar. Proporcionaron una lista de ensayos relacionados conla metformina, pero todos se habían analizado a través de labúsqueda.

Estudios excluidosLos estudios fueron excluidos por las siguientes razones: doseran estudios de cohorte retrospectivos que no suministraroninformación sobre el número de pacientes o la duración detratamiento (Debry 1966a; Debry 1966b), 12 eran estudios decohorte prospectivos que no suministraron información sobreel número de pacientes o duración de tratamiento (Bernard1965; Carpentier 1975; Chow 1995; Clauson 1996; Debry 1964;Messens 1965; Messens 1966; Muntoni 1965; Nauck 1993;Rambert 1961; Sugawara 1962; Teitelbaum 1963), 33 ensayoscomparativos prospectivos tuvieron una duración inferior a unmes (Bonfigli 1999; Bruneder 1978; Cacciapuoti 1991; Fery1997; Galuska 1994; Gibson 1995; Gin 1982; Gin 1985; Gin1989; Giugliano 1979; Isnard 1991; Isnard 1996; Irsigler 1978;Ismail 1978; Jansson 1996; Leslie 1987; Lim 1970; Orlikowska1966; Panahloo 1995; Perriello 1994; Pilger 1978; Prager 1983;Rigas 1968; Rizkalla 1986; Sambol 1996; Scarpello 1998;Schaffalitzky 1979; Signore 1996; Slama 1984; Sum 1992;Trischitta 1983; Turner 1995; Zapecka-Dubno 1999), y diezfueron análisis o revisiones retrospectivos (Aguilar 1992b;Connolly 1996; Daniel 1997; Guthrie 1997; Johansen 1999;Lalau 1994; Lalau 1995; Nauck 1997; O'Connor 1998; Selby1999).

Estudios Incluidos

Estudios y participantesDe los 206 estudios analizados, 148 fueron ensayoscomparativos prospectivos, 46 fueron estudios de cohortesprospectivos y 12 fueron estudios de cohortes retrospectivos.Se realizó el seguimiento de un total de 47 096 participantesde 86 067 pacientes tratados por año, con 30 294 participantes(47 846 pacientes tratados por año) en el grupo de metforminay 16 782 participantes (38 221 pacientes tratados por año) enel grupo sin metformina. La edad media de los participantesdel grupo de metformina fue 57,1 (+/- 8,9) años y un 61% eranhombres. En el grupo sin metformina, la media de edad fue57,2 (+/ - 9,1) años y un 61% eran hombres. De los datosdisponibles se calculó que un 24% de los participantes eranmayores de 65 años, que tuvieron un seguimiento de hastaaproximadamente 20 650 pacientes tratados por año. Laduración media del ensayo fue 2,1 años, con un rango de 0,08a 10,7 años. El tamaño medio del estudio en el grupo demetformina fue de 147 participantes con un rango de 6 a 7,227.El tamaño promedio del estudio en el grupo sin metformina fuede 81 participantes con un rango de 8 a 2 796. La tasa deabandono se estimó en un 9,2%.

Criterios de exclusión de los estudiosDe los 194 estudios prospectivos, la insuficiencia renal semencionó como criterio de exclusión en 105 (54%), lasenfermedades cardiovasculares en 76 (39%), las enfermedadeshepáticas en 104 (54%), las enfermedades pulmonares en 33(17%) y la edad superó los 65 en 34 (18%).

IntervencionesSe administró la metformina en dosis diarias de 1 a 3 gramos,con ajustes clínicos de la dosis. Los tratamientos decomparación incluían placebo, dieta, insulina, gliburida,gliclazida, glipizida, glibenclamida, clorpropramida,tolbutamida, acarbosa, nateglinida, repaglinida, miglitol,troglitazona, pioglitazona y la goma de guar.

Medidas de resultadoLos resultados midieron el control glucémico incluido (glucosaurinaria y sanguínea, HbA1, HbA1c, insulina y nivelesc-péptido), la sensibilidad de la insulina mediante un bolo deglucosa, el peso, el consumo de energía, los lípidos, laslipoproteínas, la fructosamina, los ácidos grasos libres, elfibrinógeno, el inhibidor del activador de plasminógeno, lafrecuencia cardíaca, la presión arterial, los niveles de lactato,el bicarbonato, las cetonas, la microalbuminuria, las pruebasde la función renal y hepática, la tasa bruta de mortalidad, lamortalidad relacionada con el tiempo y la diabetes relacionadacon los criterios de valoración (muerte súbita, muerte por hipero hipoglucemia, infarto del miocárdico, accidentecerebrovascular, insuficiencia renal y la enfermedad oculardiabética).

Sólo 19 ensayos estaban diseñados específicamente para evaluarla incidencia de la acidosis láctica, aunque en casi todos losensayos se describieron los efectos secundarios o los eventos


Página 7


adversos. Se realizaron intentos para establecer contacto conlos autores de los ensayos y 30 investigadores contestaron, todosconfirmaron que no había casos de acidosis fatal o no fatalconocidos en ninguno de sus 34 ensayos. El bicarbonato séricoo los niveles de lactato se midieron en 99 de los estudiosincluidos (48%). En los ensayos comparativos, 25 midieron losniveles de lactato durante los tratamientos con o sin metformina(Bjorntorp 1978; Botha 1977; Campbell 1994; Cavallo-Perin1989; Cryer 2005; Cusi 1996; Damsbo 1998; De Silva 1979;DeFronzo 1995; Fritsche 2000; Gregorio 1989; Gregorio 1990;Hother-Nielsen 1989; Inzucchi 1998; Jackson 1987; Josephkutty1990; Klein 1991; McAlpine 1988; Nattrass 1977; Pedersen1989; Rachmani 2002; Raptis 1996; Teupe 1991; Velussi 1992;Wu 1990).

CALIDAD METODOLÓGICA

Se asignaron las puntuaciones de calidad metodológica de losestudios mediante los criterios descritos con anterioridad. Paradeterminar una puntuación de calidad se utilizó sólo lainformación publicada en los ensayos. La puntuación kappapara el acuerdo entre los evaluadores fue 0,84 (intervalo deconfianza [IC] del 95%: 0,77 a 0,91) lo que indica buen acuerdo.Después que se logró el consenso en los ensayos restantes, sieteensayos recibieron una puntuación de A; 58 ensayos recibieronuna puntuación de B; otros 58 una puntuación de C; 25 ensayosuna puntuación de D; y 58 estudios más una puntuación de E.De los estudios analizados, 65 eran ensayos controladosaleatorios a doble ciego (siete describieron el método deasignación al azar y el ocultamiento de la asignación). Otros83 fueron ensayos comparativos de simple ciego o abiertos (58aleatorios y 25 no aleatorios). Los 58 estudios de cohortesfueron todos abiertos y observacionales. La tasa de abandonopromedio fue del 9,2%.

RESULTADOS

Incidencia de la acidosis lácticaCuando se combinaron los datos de los estudios de cohortescon los ensayos comparativos prospectivos, no se informóningún caso de acidosis láctica fatal o no fatal en el grupo demetformina, con un total de 47 846 pacientes tratados por añoy ningún caso en el grupo sin metformina, que representaban38 221 pacientes tratados por año. Mediante el uso de lasestadísticas de Poisson con un intervalo de confianza del 95%,el límite superior para la verdadera incidencia de la acidosisláctica asociada a la metformina fue de 6,3 casos por cada 100000 pacientes tratados por año y el límite superior para laincidencia de la acidosis láctica en el grupo sin metformina fuede 7,8 casos por cada 100 000 pacientes tratados por año.Cuando se combinaron los datos de los grupos de metforminay sin metformina, el límite superior para la verdadera incidenciade la acidosis láctica en todos los pacientes con diabetes tipo 2fue 3,5 casos cada 100 000 pacientes tratados por año.

Asociación con enfermedades hipoxémicas concomitantesOtro resultado que se evaluó fue el número de participantes quemostraron un empeoramiento en sus enfermedadesconcomitantes hipoxémicas durante el ensayo. No fue posiblerealizar una evaluación exacta de la incidencia de lainsuficiencia renal o el empeoramiento de otras enfermedadesporque dos de los ensayos grandes no proporcionaron datosadecuados (Fisman 2001; UKPDS-34 1998). No se pudoproporcionar esta información a través de la correspondenciacon los autores de estos ensayos.

No hubo información suficiente para calcular el número departicipantes estudiados que presentaban enfermedadeshipoxémicas concomitantes como la insuficiencia renal, lasenfermedades cardiovasculares, hepáticas o pulmonares. Encambio, cada uno de los ensayos incluidos en este análisis secaracterizó en cuanto a si alguna de estas enfermedades semencionaba como criterio de exclusión. Si los pacientes seenumeraron como saludables o con el uso de contraindicacionesestándar, se supuso que todas estas condiciones fueronexcluidas. La insuficiencia renal se definió normalmente comoun nivel de creatinina de más de 1,5 mg/dl. De los 194 estudiosprospectivos, 89 (46%) permitieron la inclusión de lainsuficiencia renal, según 28 244 pacientes por año de uso dela metformina y 186 (96%) permitieron la inclusión de al menosuna de las contraindicaciones mencionadas anteriormente. Delos datos disponibles, se calculó que el 24% de los participantesen los estudios tenían una edad superior a los 65 años, yrecibieron un seguimiento de hasta aproximadamente 11 483pacientes por año de uso de metformina.

Niveles del lactato sanguíneoPara aquellos ensayos que proporcionaron los datos, el nivelde lactato inicial medido antes del tratamiento con metforminafue de 1,13 +/-0, 25 mmol/L. No hubo diferencias en el cambioneto de los niveles de lactato desde el inicio del estudio para lametformina en comparación con el placebo o con lostratamientos sin biguanidas, con una diferencia de mediasponderada (DMP) de 0,12 mmol/L (intervalo de confianza [IC]del 95% -0,01; 0,25). El nivel de lactato promedio durante eltratamiento con metformina fue de 1,24 +/-0,31 mmol/L, elcual no fue significativamente diferente de las comparacionessin biguanidas (DMP 0,04 mmol/L; IC del 95%: 0,00 a 0,13,P = 0,07) y fue de 0,75 mmol/L más bajo que con la fenformina(IC del 95%: -0,86 a -0,65). El nivel de lactato durante eltratamiento con metformina, medido antes y después de laestimulación (a través de una comida o ejercicios enérgicos)fue 2,3 +/- 1,7 mmol/L. Este resultado no fue significativamentediferente para la metformina en comparación con el grupo sinbiguanidas (DMP 0,09 mmol/L; IC del 95%: -0,03; 0,22) o conla fenformina (DMP -0,37 mmol/L; IC del 95%: -1,06; 0,32).Cuatro ensayos que midieron los niveles de lactato noproporcionaron datos para ser analizados, pero informaronniveles normales durante el tratamiento con metformina y sinmetformina (DeFronzo 1995; Fritsche 2000; Gregorio 1989;Raptis 1996).


Página 8


Se observó una posible heterogeneidad en los tres ensayos quemidieron los niveles de lactato después de la estimulación conalimentos o ejercicios durante el tratamiento con metforminao los tratamientos sin biguanidas, probablemente debido a quecada uno se realizó bajo distintas condiciones. Los resultadosno fueron significativamente diferentes cuando se utilizó elmodelo de efectos aleatorios (DMP 0,04 mmol/L; IC del 95%:-0,45; 0,53). Además, se observó cierta heterogeneidad en lostres ensayos que medían los niveles promedio de lactato parael tratamiento con metformina en comparación con el defenformina. Cuando se utilizó el modelo de efectos aleatorios,la diferencia ya no era estadísticamente significativa (-0,64mmol/L; IC del 95%: -1,63; 0,35).

Análisis del sesgo de publicaciónLos gráficos de embudo (funnel plots) del tamaño del efectoversus el error estándar se evaluaron para los ensayos incluidosen el análisis. El gráfico de embudo (funnel plot) utilizado parala incidencia de la acidosis láctica no pudo aportar pruebas afavor o en contra de la posibilidad del sesgo del estudiopequeño, debido a que en ninguno de los ensayos se encontraroncasos de acidosis láctica. Los gráficos de embudo (funnel plots)para medir los niveles de lactato no mostraron pruebas de unsesgo significativo del estudio pequeño.

DISCUSIÓN

ResumenPara evaluar el riesgo de la acidosis láctica atribuida al uso demetformina, se analizaron los datos agrupados de todos losensayos comparativos prospectivos y los estudiosobservacionales de cohortes conocidos que duraron más de unmes. No se encontraron casos en 206 ensayos con 47 846pacientes tratados por año con metformina. En realidad, en larevisión de 47 ensayos adicionales que se excluyeron del análisis(aquellos que duraron menos de un mes o tuvieron una duraciónincierta) no se encontraron casos de acidosis láctica. Ladiferencia de riesgo para el tratamiento con metformina encomparación con el tratamiento sin metformina, calculadomediante las estadísticas de Poisson fue de 0,00 cada 100 000pacientes tratados por año (IC del 95%: -7,8; + 6,3). Esto indicaque el límite superior para la verdadera incidencia de la acidosisláctica asociada con la metformina es de 6 casos cada 100 000pacientes tratados por año y el límite superior para la incidenciacon otros tratamientos sin biguanidas es de 8 cada 100 000pacientes tratados por año. En los ensayos que midieron losniveles del lactato sanguíneo, no hubo diferencias significativaspara la metformina en comparación con el placebo o con lostratamientos sin biguanidas; y fueron inferiores para lametformina que para la fenformina.

La duración media de los estudios incluidos en esta revisiónfue de 2,1 años con un amplio rango de un mes hasta 10,7 años.Debido a que no se encontraron casos de acidosis láctica enninguno de los ensayos, no se pudo evaluar la asociación de laacidosis láctica a la duración del tratamiento. Además, se

evaluaron los ensayos excluidos con menos de un mes deduración para ver si la acidosis láctica ocurría poco después delinicio del tratamiento, pero no se encontraron casos.

Actualmente, se considera que la metformina estácontraindicada en pacientes con insuficiencia renal crónica,anormalidades de la función hepática, insuficiencia cardíacacongestiva, vasculopatía periférica, enfermedades pulmonareso en pacientes mayores de 65 años de edad, debido a que estasenfermedades pueden aumentar el riesgo de la anoxia tisular y,por consiguiente, el desarrollo de la acidosis láctica. En estarevisión, 187 (96%) de los 194 estudios prospectivospermitieron la inclusión de pacientes con al menos una de estascontraindicaciones y se estimó que el 24% de todos losparticipantes eran mayores de 65 años, sin efectos adversosdetectados. Sin embargo, es incierta la cantidad de participantescon estas enfermedades que se incluyeron en los ensayos, demanera que no se puede evaluar la seguridad de la metforminaen presencia de estas contraindicaciones estándar. Un ensayo(Rachmani 2002) cuestionó las contraindicaciones estándar alestudiar a 393 pacientes, con al menos una contraindicaciónpara el uso de metformina, y no encontró ningún caso deacidosis láctica en más de cuatro años de duración del ensayo.Todos los pacientes de este ensayo tenían insuficiencia renal,con niveles de creatinina plasmática medios de 1,5 a 2,5 mg/dl(nivel medio 1,8 mg/dl).

Limitaciones de la revisiónEsta revisión tiene algunas limitaciones. Esencialmente todoslos datos incluidos en este análisis provenían de ensayospublicados y esto puede haber producido resultados sesgados.Un gráfico de embudo (funnel plot) del tamaño del efecto versusel error estándar no pudo aportar pruebas para el sesgo depublicación significativo, ya que no se encontraron casos enninguno de los ensayos. Es interesante observar que muchosde los ensayos comparativos incluidos en el análisis recibieronel patrocinio de compañías farmacéuticas que producíanfármacos hipoglucemiantes distintos a la metformina, en cuyocaso un sesgo sería publicar los efectos adversos para lametformina.

Otra de las dificultades es que para evaluar el riesgo de unaaparición rara como la acidosis láctica, es posible que seanecesario evaluar a más de 48 000 pacientes tratados por añocon metformina. Es especialmente difícil evaluar el riesgo dela acidosis láctica en presencia de contraindicaciones estándarcomo la insuficiencia renal o la hepática porque es incierto elnúmero exacto de participantes que presentaban estasenfermedades. Por ese motivo, no se pueden establecerconclusiones acerca de la seguridad del uso de la metforminaen presencia de estas enfermedades. A pesar de estaslimitaciones, la conclusión más importante de esta revisión esque, en la actualidad, no hay pruebas en los ensayoscomparativos prospectivos o los estudios observacionales decohortes que apoyen la hipótesis de la asociación de lametformina a un mayor riesgo de acidosis láctica.


Página 9


Generalización y aplicabilidad de los resultadosLa metformina es un medicamento hipoglucemiante de lafamilia de las biguanidas que se ha utilizado durante más de 45años (Sterne 1959). Se ha demostrado que el tratamiento conmetformina en pacientes con sobrepeso que presentan diabetestipo 2 reduce la mortalidad cardiovascular y total encomparación con la insulina, la sulfonilurea o sólo la dieta(UKPDS-34 1998). La preocupación respecto al riesgo deacidosis láctica ha dado lugar a recomendaciones acerca de lasuspensión de la metformina en personas con enfermedadescrónicas que en sí mismas puedan causar la acidosis láctica. Sise siguen estas recomendaciones, se reducirá el número depacientes aptos para recibir metformina a aproximadamente lamitad (Brown 1998). Se ha encontrado que en la práctica clínicaestas contraindicaciones estándar se desatienden en gran manera,ya que del 54% al 73% de los pacientes tratados con metforminapresentan al menos una contraindicación al tratamiento (Holstein1999; Sulkin 1997). En un estudio transversal, el 19% de lospacientes tratados con metformina que ingresaron a un hospitalpresentaron una insuficiencia renal concomitante (Holstein1999).

Se ha implicado a la metformina como causa de acidosis lácticadebido a que se había asociado una biguanida relacionada, lafenformina, a diversos casos de acidosis láctica, y se retiró delmercado de los EE.UU. en 1977 (Aguilar 1992b). A pesar desus semejanzas, la fenformina tiene una estructura químicasignificativamente diferente a la metformina. A diferencia dela metformina, la fenformina puede deteriorar la fosforilaciónoxidativa del hígado, y aumentar de esta manera la producciónde lactato a través de vías anaeróbicas (Cavallo-Perin 1989;Irsigler 1978; Pilger 1978; Sirtori 1994; Velussi 1992). Encambio, la metformina inhibe la gluconeogénesis hepática sinalterar el recambio de lactato o la oxidación de lactato (Cusi1996; Scheen 1996; Stacpoole 1998). Además de los ensayosanalizados en esta revisión, otros ensayos han confirmado queel tratamiento con metformina no eleva de forma significativalos niveles del lactato sanguíneo, aun en presencia de unadeficiencia renal o edad avanzada (Connolly 1996; Debry 1964;Giugliano 1993; Irsigler 1978; Lalau 1990; Menzies 1989;Pagano 1983; Pilger 1978; Trischitta 1983).

En la actualidad, las únicas pruebas que indican la asociacióndel uso de metformina con la acidosis láctica provienen deinformes de aproximadamente 330 casos que ocurrieron enpacientes durante el tratamiento con metformina (Bergman1978; Gan 1992; Lalau 1994; Luft 1978). La incidencia deacidosis láctica en pacientes tratados con metformina se hacalculado en estudios de población y va de 2 a 9 casos cada 100000 pacientes tratados por año (Misbin 1998; Stang 1999;Wilholm 1993). La mayoría de los casos informados se encontróen pacientes con enfermedades subyacentes graves que en símismas podrían haber causado la acidosis láctica.

También se ha encontrado acidosis láctica en pacientesdiabéticos no tratados con metformina, que habitualmentepresentan enfermedades con una hipoperfusión tisular

significativa o hipoxia (Aguilar 1992b). Para evaluar la tasa deacidosis láctica en pacientes diabéticos que recibierontratamientos distintos a la metformina, un estudio de poblaciónrealizó un seguimiento de pacientes con diabetes tipo 2 que setrataron en los EE.UU. antes de la aparición de la metforminay después del retiro de la fenformina (Brown 1998). Este estudiohalló que la tasa de acidosis láctica confirmada eraaproximadamente de 10 cada 100 000 pacientes tratados poraño, lo que equivale a la encontrada en el tratamiento conmetformina. Otro estudio evaluó todos los casos de acidosismetabólica no cetónica en pacientes con diabetes tipo 2 queocurrieron en 609 ingresos de emergencia en un hospitaluniversitario (Aguilar 1992b). Las tasas de acidosis no cetónicacada 1000 ingresos en urgencias fueron 29 para la sulfonilurea,32 para la sulfonilurea más la fenformina, 48 para la insulinay ningún caso para aquellos tratados con metformina. Todoslos casos de acidosis metabólica no cetónica encontrados seasociaron con enfermedades precipitantes graves que podríanhaber causado la acidosis láctica. Los investigadores llegarona la conclusión de que el determinante principal para laaparición de la acidosis láctica es la disfunción sistémicasubyacente y no el tratamiento en sí. Para apoyar esa conclusión,los resultados de esta revisión revelan que no hay pruebas deun mayor riesgo de acidosis láctica asociada con el uso demetformina si se prescribe en las enfermedades del estudio yse tienen en cuenta las contraindicaciones.

CONCLUSIONES DE LOS AUTORES

Implicaciones para la práctica

No hay pruebas en los ensayos comparativos prospectivos o enlos estudios observacionales de cohortes de que el tratamientocon metformina aumente la incidencia de la acidosis láctica sise prescribe en las enfermedades del estudio, y se tienen encuenta las contraindicaciones. Esta revisión no fue capaz deevaluar cuantitativamente la seguridad del tratamiento conmetformina en presencia de cada una de las enfermedadesconcomitantes hipóxicas.

Implicaciones para la investigación

Será necesario realizar grandes ensayos prospectivos ycomparativos en pacientes con diabetes mellitus tipo 2 quepresenten enfermedades que se consideran, en la actualidad,contraindicaciones para su uso. Por ejemplo, se podría realizarun ensayo de gran tamaño en pacientes que se sabe presentanuna insuficiencia renal crónica. Los resultados a seguir incluiríanla incidencia de la acidosis láctica así como las complicacionesrelacionadas con la diabetes y la mortalidad total.

POTENCIAL CONFLICTO DE INTERÉS

Ninguno conocido.


Página 10


FUENTES DE FINANCIACIÓN

Recursos externos

• La información sobre los recursos de apoyo no está

disponible

Recursos internos

• Santa Clara Valley Medical Center USA

✦

REFERENCIAS

Referencias de los estudios incluidos en esta revisión

Aarsand 1998 {published data only}Aarsand AK, Carlsen SM. Folate administration reduces circulatinghomocysteine levels in NIDDM patients on long-term metformin treatment.Journal of Internal Medicine 1998;244:169-74.

Abbasi 1997 {published data only}Abbasi F, Kamath V, Rizvi AA, Carantoni M, Chen YD, Reaven GM.Results of a placebo-controlled study of the metabolic effects of the additionof metformin to sulfonylurea-treated patients. Evidence for a central roleof adipose tissue. Diabetes Care 1997;20(12):1863-9.

Abbasi 1998 {published data only}Abbasi F, Carantoni M, Chen YD, Reaven GM. Further evidence for acentral role of adipose tissue in the antihyperglycemic effect of metformin[see comments]. Diabetes Care 1998;21(8):1301-5.

Abbink 2001 {unpublished data only}Abbink A. (no title). .

Abbink 2000 {published data only}Abbink EJ, Tack CJ, Pickkers P Jansen van Rosendaal A, Smits P. Vasculareffects of glibenclamide, glimepiride, metformin and acarbose in Type 2DM patients. Diabetologia 2000;43(Suppl 1):A39.

Aguilar 1992a {published data only}Aguilar CA, Wong B, Gomez-Perez FJ, Rull JA. Combination daytimechlorpropamide-metformin/bedtime insulin in the treatment of secondaryfailures in non insulin dependent diabetes. Revista de Investigacion Clinica1992;44(1):71-6.

Allen 1961 {published data only}Allen GE, Montgomery DAD, Weaver JA. [Dimethylbiguanide in thetreatment of diabetes mellitus]. La Revue Francaise d'EndocrinologieClinique 1961;2:347-52.

Andras 1962 {published data only}Andras K, Tamas H, Gyorgy K. [Experience with biguanide treatment fordiabetes mellitus]. Orvosi hetilap 1962;103:1029-32.

Aviles-Santa 1999 {published data only}Aviles-Santa L, Sinding J, Raskin P. Effects of metformin in patients withpoorly controlled, insulin-treated type 2 diabetes mellitus. A randomized,double-blind, placebo-controlled trial. Annals of Internal Medicine1999;131(3):182-8.

Azerad 1960 {published data only}Azerad E, Lasry M. [Treatment of diabetes with NN-Dimethyl Biguanide].Gazette Medicale de France 1960;1(7):1555-8.

Bacci 1961 {published data only}Bacci L, Cassarini F. [N.N. dimethyl-biguanide (DMGG) in the treatmentof Diabetes meliitus]. Minerva Medica 1961;52:2770-3.

Bayraktar 1996 {published data only}Bayraktar M, Van Thiel DH, Adalar N. A comparison of acarbose versusmetformin as an adjuvant therapy in sulfonylurea-treated NIDDM patients.Diabetes Care 1996;19(3):252-4.

Beisswenger 1999 {published data only}Beisswenger PJ, Howell SK, Touchette AD, Lal S, Szwergold BS.Metformin reduces systemic methylglyoxal levels in type 2 diabetes.Diabetes 1999;48(1):198-202.

Belcher 2005 {published data only}Blecher G, Schernthaner G. Changes in liver tests during 1-year treatmentof patients with type 2 diabetes with pioglitazone, metformin or glicazinde.Diabetic Medicine 2005;22:973-9.

Bell 1997 {published data only}Bell DSH, Mayo MS. Outcome of metformin-facilitated reinitiation of oraldiabetic therapy in insulin-treated patients with non-insulin-dependentdiabetes mellitus. Endocrinology Practice 1997;3:73-6.

Beyer 1975 {published data only}Beyer G. [Results in the treatment of diabetes with glucophage retard].Therapie der Gegenwert 1975;114(9):1426-37.

Bingle 1964 {published data only}Bingle JP, James JI. Metformin combined with chlorpropamide in thetreatment of Diabetes mellitus. Practitioner 1964;192:567-72.

Bjorntorp 1978 {published data only}Bjorntorp P, Carlstrom S, Fagerberg SE, Hermann LS, Holm AG, ScherstenB, et al. Influence of phenformin and metformin on exercise inducedlactataemia in patients with diabetes mellitus. Diabetologia 1978;15(2):95-8.

Blonde 2002 {published data only}Blonde L, Rosenstock J, Mooradian AD, Piper B-A, Henry D.Glyburide/Metformin combination product is safe and efficacious in patientswith type 2 diabetes failing sulphonylurea therapy. Diabetes Obesity &Metabolism 2002;4:368-75.

Boronat 2000 {published data only}Boronat CM, Marrero AD, La Roche BF, Ojeda PA, Carillo DA, NovoaMFJ. [Effectiveness of treatment with metformin in patients with type 2diabetes mellitus poorly controlled with insulin treatment]. Revista ClinicaEspanola 2000;200(2):74-6.

Botha 1977 {published data only}Botha JL, Vinik AI, Jackson WP. Association of lactic acidosis withbiguanide therapy [letter]. South African Medical Journal 1977;52(8):301-2.

Boyd 1992 {published data only}Boyd K, Rogers C, Boreham C, Andrews WJ, Hadden DR. Insulin,glibenclamide or metformin treatment for non insulin dependent diabetes:heterogenous responses of standard measures of insulin action and insulinsecretion before and after differing hypoglycaemic therapy. DiabetesResearch 1992;19(2):69-76.

Brown 1999 {published data only}Brown JB, Pedula KL. Metformin as secondary therapy in a definedpopulation with Type 2 Diabetes. Clinical Therapeutics1999;21(10):1678-87.

Cairns 1977 {published data only}Cairns SA, Shalet S, Marshall AJ, Hartog M. A comparison of phenforminand metformin in the treatment of maturity onset diabetes. Diabete &Metabolisme 1977;3(3):183-8.


Página 11


Calle-Pascual 1995 {published data only}Calle-Pascual AL, Garcia-Honduvilla J, Martin-Alvarez PJ, Vara E, CalleJR, Munguira ME, et al. Comparison between acarbose, metformin, andinsulin treatment in type 2 diabetic patients with secondary failure tosulfonylurea treatment. Diabete & Metabolisme 1995;21(4):256-60.

Campbell 1988 {published data only}Campbell IW, Menzies DG, McBain AM, Brown IRF. Effects of metforminon blood pressure and microalbuminuria in diabetes mellitus. Diabete &Metabolisme 1988;14:613-7.

Campbell 1994 {published data only}Campbell IW, Menzies DG, Chalmers J, McBain AM, Brown IR. One yearcomparative trial of metformin and glipizide in type 2 diabetes mellitus.Diabete & Metabolisme 1994;20(4):394-400.

Canivet 1962 {published data only}Canivet J, Roux P. [The treatment of Diabetes mellitus: Statistical resultsafter a review of 1190 cases]. La Presse Medicale 1962;70:2038-40.

Carpentier 1975 {published data only}Carpentier J, Luyckx AS, Lefebvre PJ. Influence of metformin onarginine-induced glucagon secretion in human diabetes. Diabete &Metabolisme 1975;1:23-8.

Carter 2005 {published data only}Carter AM, Bennett CE, Bostock JA, Grant PJ. Metformin reducesC-reactive protein but not complement factor C3 in overweight patientswith type 2 diabetes mellitus. Diabetic Medicine 2005;22:1282-4.

Cavallo-Perin 1989 {published data only}Cavallo-Perin P, Aluffi E, Estivi P, Bruno A, Carta Q, Pagano G, et al. Thehyperlactatemic effect of biguanides: a comparison between phenforminand metformin during a 6-month treatment. European Review for MedicalPharmacological Science 1989;11(1):45-9.

Cefalu 2002 {published data only}Cefalu WT, Schneider DJ, Carlson HE, Gan Lim L, Izon MP, Kappor A,Bell-Farrow A, Terry JG, Sobel BE. Effect of combination glipizideGITS/Metformin on fibrinolytic and metabolic parameters in poorlycontrolled type 2 diabetic subjects. Diabetes Care 2002;25:2123-8.

Chakrabarti 1965 {published data only}Chakrabarti R, Hocking ED, Fearnley GR. Fibrinolyitic effect of metforminin coronary-artety disease. Lancet 1965;2:256-9.

Chan 1993 {published data only}Chan JC, Tomlinson B, Critchley JA, Cockram CS, Walden RJ. Metabolicand hemodynamic effects of metformin and glibenclamide in normotensiveNIDDM patients. Diabetes Care 1993;16(7):1035-8.

Chiasson 1994 {published data only}Chiasson JL, Josse RG, Hunt JA, Palmason C, Rodger NW, Ross SA, etal. The efficacy of acarbose in the treatment of patients withnon-insulin-dependent diabetes mellitus. A multicenter controlled clinicaltrial [see comments]. Annals of Internal Medicine 1994;121(12):928-35.

Wolever TM, Chiasson JL, Josse RG, Hunt JA, Palmason C, Rodger NW,et al. Small weight loss on long-term acarbose therapy with no change indietary pattern or nutrient intake of individuals with non-insulin-dependentdiabetes. International Journal of Obesity and Related Metabolic Disorders1997;21(9):756-63.

Wolever TM, Chiasson JL, Josse RG, Hunt JA, Palmason C, Rodger NW,et al. No relationship between carbohydrate intake and effect of acarboseon HbA1c or gastrointestinal symptoms in type 2 diabetic subjectsconsuming 30-60% of energy from carbohydrate [see comments]. DiabetesCare 1998;21:1612-8.

Chiasson 2001 {published data only}Chiasson J-L, Naditch L. The synergistic effect of miglitol plus metformincombination therapy in the treatment of type 2 diabetes. Diabetes Care2001;24(6):989-94.

Cho 1992 {published data only}Cho YW, Yang DH, Oh DY, Baick SH, Kim SK, Kim SJ, et al. Plasmat-PA and PAl-1 antigen concentrations in non-insulin dependent diabeticpatients: effects of treatment modality on fibrinolysis. Korean Journal ofInternal Medicine 1992;7(2):81-6.

Clarke 1965 {published data only}Clarke BF, Duncan LJP. Combined Metformin-Chlorpropamide therapyin 108 diabetic sulphonylurea-failures. Lancet 1965;1:1248-51.

Clarke 1968 {published data only}Clarke BF, Duncan LJ. Comparison of chlorpropamide and metformintreatment on weight and blood-glucose response of uncontrolled obesediabetics. Lancet 1968;1(7534):123-6.

Clarke 1977 {published data only}Clarke BF, Campbell IW. Comparison of metformin and chlorpropamidein non-obese, maturity-onset diabetics uncontrolled by diet. British MedicalJournal 1977;2(6102):1576-8.

Collier 1989 {published data only}Collier A, Watson HH, Patrick AW, Ludlam CA, Clarke BF. Effect ofglycaemic control, metformin and gliclazide on platelet density andaggregability in recently diagnosed type 2 (non-insulin-dependent) diabeticpatients. Diabete & Metabolism 1989;15(6):420-5.

Cryer 2005 {published data only}Cryer DR, Nicholas SP, Henry DH, Mills DJ, Stadel BV. Comparativeoutcomes study of metformin intervention versus conventional approach.Diabetes Care 2005;28:539-43.

Cusi 1996 {published data only}Cusi K, Consoli A, DeFronzo RA. Metabolic effects of metformin onglucose and lactate metabolism in noninsulin-dependent diabetes mellitus.Journal of Clinical Endocrinology and Metabolism 1996;81:4059-67.

D'Argenzio 1996 {published data only}D'Argenzio R, Cavallo P, Merante D, Morelli A. [Comparison of twotreatment models in type-II diabetic patients with poor metabolic control:Preformed combination of glibenclamide 2,5 mg + metformin 400 mg ormono-therapy with sulfonylurea at maximal doses? An evaluation at sixmonths]. Minerva Endocrinologica 1996;21(3):101-10.

Damsbo 1998 {published data only}Damsbo P, Hermann LS, Vaag A, Hother-Nielsen O, Beck-Nielsen H.Irreversibility of the defect in glycogen synthase activity in skeletal musclefrom obese patients with NIDDM treated with diet and metformin. DiabetesCare 1998;21(9):1489-94.

Davidson 2000 {published data only}Davidson J, Mooradian A, Piper B. Effect of metformin/glyburide tabletson HBA1c in first-line treatment of Type 2 Diabetes. Diabetologia2000;43(Suppl 1):A184.

De Silva 1979 {published data only}De Silva SR, Betteridge DJ, Shawe JEH, Cudworth AG, Alberti KGMM.Metformin and clofibrate in maturity onset diabetes mellitus: advantagesof combined treatment. Diabete & Metabolisme 1979;5:223-9.

DeFronzo 1991 {published data only}DeFronzo RA, Barzilai N, Simonson D. Mechanism of metformin actionin obese and lean noninsulin-dependent diabetic subjects. Journal of ClinicalEndocrinology and Metabolism 1991;73:1294-1301.

DeFronzo 1995 {published data only}DeFronzo RA, Goodman A and the Metformin Investigator Group.Combined metformin/glyburide treatment in NIDDM patients not optimallyresponding to maximum dose sulfonyluria: Results of a multicenter trial.Diabetes 1993;42(Suppl 1):146A (abs 455).

*DeFronzo RA, Goodman AM. Efficacy of metformin in patients withnon-insulin-dependent diabetes mellitus. The Multicenter Metformin StudyGroup [see comments]. New England Journal of Medicine1995;333(9):541-9.


Página 12


Goodman AM and the Metformin Investigator Group. Efficacy and safetyof metformin in NIDDM: Results of a muticenter trial. Diabetes1993;42(Suppl 1):Abs 178.

Derosa 2003 {published data only}Derosa G, Mugellini Am, Ciccarelli L, Crescenzi G, Fogari R. Comparisonof glycaemic control and cardiovascular risk profile in patients with type2 diabetes during treatment with either repaglinide or metformine. DiabetesResearch and Clinical Practice 2003;60:161-9.

Dies 1978 {published data only}Dies H, Lozano Castaneda O, Garcia Viveros M, Schulte J, Figueroa J,Rull JA. [Long range results (5 years) in the treatment of stable diabeteswith low doses of clorpropamide-metformin (author's transl)]. La Revistade Inverstigation Clinica 1978;30(2):131-4.

Donnelly 1960 {published data only}Donnelly P. Dimethyldiguanide in diabetes mellitus: Report of a small trial.Journal of the Irish Medical Association 1960;47:142-5.

Dornan 1991 {published data only}Dornan TL, Heller SR, Peck G, Gregory R, Tattersall RB. Metforminmonotherapy for obese non-insulin-dependent diabetes. A double-blindevaluation. Diabetic Medicine 1988;5(Suppl 2):5 (Abs A17).

Dornan TL, Heller SR, Peck GM, Tattersall RB. Double-blind evaluationof efficacy and tolerability of metformin in NIDDM. Diabetes Care1991;14(4):342-4.

Elkeles 1991 {published data only}Elkeles RS. The effects of oral hypoglycaemic drugs on serum lipids andlipoproteins in non-insulin-dependent diabetes (NIDDM). Diabete &Metabolisme 1991;17(1 Pt 2):197-200.

Erle 1999 {published data only}Erle G, Lovise S, Stocchiero C, Lora L, Coppini A, Marchetti P, et al. Acomparison of preconstituted, fixed combinations of low-dose glyburideplus metformin versus high-dose glyburide alone in the treatment of type2 diabetic patients. Acta Diabetologia 1999;36(1-2):61-5.

Fanghanel 1996 {published data only}Fanghanel G, Sanchez-Reyes L, Trujillo C, Sotres D, Espinosa-Campos J.Metformin's effects on glucose and lipid metabolism in patients withsecondary failure to sulfonylureas. Diabetes Care 1996;19(11):1185-9.

Fanghanel 1998 {published data only}Fanghanel G, Silva U, Sanchez-Reyes L, Sisson D, Sotres D, Torres EM.Effects of metformin on fibrinogen levels in obese patients with type 2diabetes. La Revista de Investigacion Clinica 1998;50(5):389-94.

Ferner 1988 {published data only}Ferner RE, Rawlins MD, Alberti KG. Impaired beta-cell responses improvewhen fasting blood glucose concentration is reduced innon-insulin-dependent diabetes. Quarterly Journal of Medicine1988;66(250):137-46.

Fisman 2001 {published data only}Fisman EZ, Tenenbaum A, Benderly M, Goldbourt U, Behar S, Motro M.Antihyperglycemic treatment in diabetics with coronary disease: Increasedmetformin-associated mortality over a 5-year follow-up. Cardiology1999;91:195-202.

Fisman EZ, Tennenbaum A, Boyko V, Benderly M, Adler Y, FriedensohnM, et al. Oral antidiabetic treatment in patients with coronary disease:Time-related increased mortality on combined glyburide/metformin therapyover a 7.7-year follow-up. Clinical Cardiology 2001;24:151-8.

Fonseca 2000 {published data only}Fonseca V, Rosenstock J, Patwardhan R, Salzman A. Effect of metforminand rosiglitazone combination therapy in patients with type 2 diabetesmellitus: a randomized controlled trial. Journal of the American MedicalAssociation 2000;283(13):1695-702.

Fritsche 2000 {published data only}Fritsche A, Schmulling R-M, Haring H-U, Stumvoll M. Intensive insulintherapy combined with metformin in obese type 2 diabetic patients. ActaDiabetologia 2000;37(1):13-8.

Fujioka 2005 {published data only}Fujioka K, Brazg RL, Raz, I, Bruce S, Joyal S, Sanink R, Pans M. Efficacy,dose-response relationship and safety of once-daily extended-releasemetformin (Glucophage XR) in type 2 diabetic patients with inadequateglycaemic control despite prior treatment with diet and exercise: resultsfrom two double-blind, placebo-controlled studies. Diabetes Obesity &Metabolism 2005;7:28-39.

Galeone 1998 {published data only}Galeone F, Fiore G, Arcangeli A, Mannucci E. [Gliclazide and metformincombination in patients with type 2 diabetes. Preliminary data]. MinervaEndocrinologica 1998;23(3):71-5.

Garber 1997 {published data only}Garber AJ, Duncan TG, Goodman AM, Mills DJ, Rohlf JL. Efficacy ofmetformin in type II diabetes: results of a double-blind, placebo-controlled,dose-response trial. American Journal of Medicine 1997;103(6):491-7.

Garber 2002 {published data only}Garber AJ, Larsen J, Schneider SH, Piper BA, Henry D and theGlyburide/Metformin Initial Therapy Study Group. Simultaenousglyburide/metformin therapy is superior to component monotherapy as aninitial pharmacological treatment for type 2 diabetes. Diabetes Obesity &Metabolism 2002;4:201-8.

Garcia 1971 {published data only}Garcia Viveros M, Prado Vega R, Rull JA, Lozano-Castaneda O.[Chlorpropamide-metformin in obese diabetics or diabetics with side effectsfrom sulfonylureas. 2-year results]. La Prensa Medica Mexicana1971;36(7):347-9.

Giugliano 1993 {published data only}Giugliano D, Quatraro A, Consoli G, Minei A, Ceriello A, De Rosa N, etal. Metformin for obese, insulin-treated diabetic patients: improvement inglycaemic control and reduction of metabolic risk factors. European Journalof Clinical Pharmacology 1993;44(2):107-12.

Goldstein 2003 {published data only}Goldstein BJ, Pans M, Rubin CJ. Multicenter, randomized, double-mased,parallel-group assessment of simultaneous glipizide/metformin assecond-line pharmacologic treatment for patients with type 2 diabetesmellitus that is inadequately controlled by a sulfonylurea. ClinicalTherapeutics 2003:890-903.

Gonzalez-Ortiz 2004 {published data only}Gonzalez-Ortiz M, Martinez-Abundis E y el Grupo para el Tratamiento dela Diabetes Mellitus con Combinaciones. Efficacy and safety of glimepirideplus metformin in a single presentation, as combined therapy, in patientswith type 2 diabetes mellitus and secondary failure to glibenclamide, asmonotherapy [Eficacia y seguridad de la terapia hypoglycemiente oralcombinada de glimepirida mas metformina en una sola forma farmaceuticaen pacientes con diabetes mellitus type 2 y falla secundaria a monoterapiacon glibenclamida]. Revista de Investigacion Clinica 2004;56(3):327-333.

Gottlieb 1962 {published data only}Gottlieb B. Metformin in treatment of Diabetes mellitus. British MedicalJournal 1962;5279:680-2.

Grant 1991 {published data only}Grant PJ, Stickland MH, Booth NA, Prentice CR. Metformin causes areduction in basal and post-venous occlusion plasminogen activatorinhibitor-1 in type 2 diabetic patients. Diabetic Medicine 1991;8(4):361-5.

Grant 1996 {published data only}Grant PJ. The effects of high- and medium-dose metformin therapy oncardiovascular risk factors in patients with type II diabetes. Diabetes Care1996;19(1):64-6.


Página 13


Grant 1998 {published data only}Grant PJ. Metformin reduces circulating Factor VII concentrations inpatients with Type II Diabetes mellitus. Thrombosis and Haemostasis1998;80:209-10.

Gregorio 1989 {published data only}Gregorio F, Ambrosi F, Angelici F, Cristallini S, Dini FL, Vespasiani G,et al. [Body mass index, blood lactate and therapeutic effectiveness ofmetformin in type II diabetes mellitus]. Medicina (Firenze) 1989;9(2):200-4.

Gregorio 1990 {published data only}Gregorio F, Ambrosi F, Marchetti P, Cristallini S, Navalesi R, Brunetti P,et al. Low dose metformin in the treatment of type II non-insulin-dependentdiabetes: clinical and metabolic evaluations. Acta Diabetologica Latina1990;27(2):139-55.

Gregorio 1997 {published data only}Gregorio F, Ambrosi F, Manfrini S, Santucci A, Filipponi P. Meformin,plasma glucose and free fatty acids in type II diabetic out-patients: resultsof a clinical study. Diabetes Research and Clinical Practice1997;37(1):21-33.

Groop 1989 {published data only}Groop L, Widen E, Franssila-Kallunki A, Ekstrand A, Saloranta C, SchalinC, et al. Different effects of insulin and oral antidiabetic agents on glucoseand energy metabolism in type 2 (non-insulin-dependent) diabetes mellitus.Diabetologia 1989;32(8):599-605.

Groop 1991 {published data only}Groop L, Widen E. Treatment strategies for secondary sulfonylurea failure.Should we start insulin or add metformin? Is there a place for intermittentinsulin therapy?. Diabete & Metabolisme 1991;17(1 Pt 2):218-23.

Guillausseau 1997 {published data only}Guillausseau PJ. Monitoring of metabolic control in patients withnon-insulin-dependent diabetes mellitus on oral hypoglycaemic agents:value of evening blood glucose determination. Diabete & Metabolisme1997;14(9):798-802.

Gursoy 2000 {published data only}Gursoy N, Ilcol Y, Tuncel E, Imamoglu S, Erturk E. The effect of metforminon insulin resistance and serum lipid profiles. Diabetologia 2000;43(Suppl1):A150.

Haupt 1991 {published data only}Haupt E, Knick B, Koschinsky T, Liebermeister H, Schneider J, Hirche H.Oral antidiabetic combination therapy with sulphonylureas and metformin.Diabete & Metabolisme 1991;17(1 Pt 2):224-31.

Herman 1961 {published data only}Herman JB, Jackson WPU. Dimethyldiguanide (Glucophage, LA 6023) inDiabetes mellitus. South African Medical Journal 1961;35:286-8.

Hermann 1991a {published data only}Hermann LS, Bitzen PO, Kjellstrom T, Lindgarde F, Schersten B.Comparative efficacy of metformin and glibenclamide in patients withnon-insulin-dependent diabetes mellitus. Diabete & Metabolisme 1991;17(1Pt 2):201-8.

Hermann LS, Kjellstrom T, Nilsson-Ehle P. Effects of metformin andglibenclamide alone and in combination on serum lipids and lipoproteinsin patients with non-insulin-dependent diabetes mellitus. Diabete &Metabolisme 1991;17(1 Pt 2):174-9.

Hermann 1991b {published data only}Hermann LS, Karlsson JE, Sjostrand A. Prospective comparative study inNIDDM patients of metformin and glibenclamide with special referenceto lipid profiles. European Journal of Clinical Pharmacology1991;41(3):263-5.

Hermann 1994a {published data only}Hermann LS, Schersten B, Melander A. Antihyperglycaemic efficacy,response prediction and dose-response relations of treatment with metforminand sulphonylurea, alone and in primary combination. Diabete &Metabolisme 1994;11(10):953-60.

Hermann 1994b {published data only}Hermann LS, Schersten B, Bitzen PO, Kjellstrom T, Lindgarde F, MelanderA. Therapeutic comparison of metformin and sulfonylurea, alone and invarious combinations. A double-blind controlled study [see comments].Diabetes Care 1994;17(10):1100-9.

Higginbotham 1979 {published data only}Higginbotham L, Martin FIR. Double-blind trial of metformin in the therapyof non-ketotic diabetics. Medical Journal of Australia 1979;2(154-6).

Hirsch 1999 {published data only}Hirsch IB. Metformin added to insulin therapy in poorly controlled Type-IIDiabetes. Diabetes Care 1999;22(5):854.

Hoffmann 1997 {published data only}Hoffmann J, Spengler M. Efficacy of 24-week monotherapy with acarbose,metformin, or placebo in dietary-treated NIDDM patients: the Essen-IIStudy. American Journal of Medicien 1997;103(6):483-90.

Hollenbeck 1991 {published data only}Hollenbeck CB, Johnston P, Varasteh BB, Chen YD, Reaven GM. Effectsof metformin on glucose, insulin and lipid metabolism in patients with mildhypertriglyceridaemia and non-insulin dependent diabetes by glucosetolerance test criteria. Diabete & Metabolisme 1991;17(5):483-9.

Holman 1987 {published data only}Holman RR, Steemson J, Turner RC. Sulphonylurea failure in Type 2Diabetes: Treatment with a basal insulin supplement. Diabete &Metabolisme 1987;4:457-62.

Horton 2000 {published data only}Horton ES, Clinkenbeard C, Gatlin M, Foley J, Mallows S, Shen S.Nateglinide alone and in combination with metformin improves glycemiccontrol by reducing mealtime glucose levels in type 2 diabetes. DiabetesCare 2000;23(11):1660-5.

Horton 2004 {published data only}Horton ES, Doley JE, Shen SG, Baron MA. Efficacy and tolerability ofinitial combination therapy with nateglinide and metformin intreatment-naive patients with thpe 2 diabetes. Current Medical Researchand Opinion 2004;20(6):883-9.

Hother-Nielsen 1989 {published data only}Hother-Nielsen O, Schmitz O, Andersen PH, Beck-Nielsen H, PedersenO. Metformin improves peripheral but not hepatic insulin action in obesepatients with type II diabetes. Acta Endocrinologica (Copenhagen)1989;120(3):257-65.

Imano 1998 {published data only}Imano E, Kanda T, Nakatani Y, Nishida T, Arai K, Motomura M, et al.Effect of troglitazone on microalbuminuria in patients with incipient diabeticnephropathy. Diabetes Care 1998;21(12):2135-9.

Inzucchi 1998 {published data only}Inzucchi SE, Maggs DG, Spollett GR, Page SL, Rife FS, Walton V, et al.Efficacy and metabolic effects of metformin and troglitazone in type IIdiabetes mellitus [see comments]. New England Journal of Medicine1998;338(13):867-72.

Jackson 1962 {published data only}Jackson WPU. Combined oral therapy in Diabetes. South African MedicalJournal 1962;36:727-9.

Jackson 1987 {published data only}Jackson RA, Hawa MI, Jaspan JB, Sim BM, Disilvio L, Featherbe D, et al.Mechanism of metformin action in non-insulin-dependent diabetes. Diabetes1987;36(5):632-40.


Página 14


Jeppesen 1994 {published data only}Jeppesen J, Zhou M-Y, Chen Y-D I, Reaven GM. Effect of metformin onpostprandial lipemia in patients with fairly to poorly controlled NIDDM.Diabetes Care 1994;17(10):1093-9.

Johansen 1984 {published data only}Johansen K. Acarbose treatment of sulfonylurea-treated non-insulindependent diabetics. A double-blind cross-over comparison of analpha-glucosidase inhibitor with metformin. Diabete & Metabolisme1984;10(4):219-23.

Johnson 1993 {published data only}Johnson AB, Webster JM, Sum CF, Heseltine L, Argyraki M, Cooper BG,et al. The impact of metformin therapy on hepatic glucose production andskeletal muscle glycogen synthase activity in overweight type II diabeticpatients. Metabolism 1993;42(9):1217-22.

Johnson 1998 {published data only}Johnson M, Krosnick A, Carson P, McDade AM, Laraway K. Aretrospective chart review of uncontrolled use of metformin as an add-ontherapy in type 2 diabetes. Clinical Therapeutics 1998;20(4):691-8.

Jones 2000 b {published data only}Jones NP, Mather R, Owen S, Porter LE, Patwardhan R. Long-term efficacyof rosiglitazone as monotherapy or in combination with metformin.Diabetologia 2000;43(Suppl 1):A192.

Jones 2000a {published data only}Jones T, Jones NP, Sautter M. Addition of rosiglitazone to metformin iseffective in obese, insulin-resistant patients with Type 2 Diabetes.Diabetologia 2000;43(Suppl 1):A191.

Jones 2002 {published data only}Jones KL, Arslanian S, Peterokova VA, Park J-S, Tomlinson MJ. Effectof metformin in pediatric patients with type 2 diabetes: a randomizedcontrolled trial. Diabetes Care 2002;25:89-94.

Josephkutty 1990 {published data only}Josephkutty S, Potter JM. Comparison of tolbutamide and metformin inelderly diabetic patients. Diabete & Metabolisme 1990;7(6):510-4.

Josse 1995 {published data only}Josse RG. Acarbose for the treatment of type II diabetes: the results of aCanadian multi-centre trial [published erratum appears in Diabetes ResClin Pract 1995 Sep;29(3):215]. Diabetes Research and Clinical Practice1995;28(Suppl):S167-72.

Jung 2005 {published data only}Jung HS, Youhn B-S, Cho YM, Yu K-Y, Park HJ, Shin CS, Kim SY, LeeHK, Park KS. The effects of rosiglitazone and metformin on the plasmaconcentrations of resistin in patients with type 2 diabetes mellitus.Metabolism Clinical and Experimental 2005;54:314-20.

Karlsson 2005 {published data only}Karlsson HKR, Hallsten K, Bjornholm M, Tsuchida H, Chibalin AV,Virtanen KA, Heinonen OJ, Lonnqvist F, Nuutila P, Zierath JR. Effects ofmetformin anad rosiglitazone treatment on insulin signaling and glucoseuptake in patients with newly diagnosed type 2 diabetes. Diabetes2005;54:1459-67.

Kiayias 1999 {published data only}Kiayias JA, Vlachou ED, Papadodima EL. Metformin and Lipoprotein (a)levels. Diabetes Care 1999;22(4):859.

Kim 2002 {published data only}Kim Y-B, Ciaraldi TP, Kong A, Kim D, Chu N, Mohideen P, Mudaliar Su,Henry RR, Kahn BB. Troglitazone but not metformin restoresinsulin-stimulated phosphoinositide 3-kinase activity and increases p110betaprotein levels in skeletal muscle of type 2 diabetic subjects. Diabetes2002;51:443-8.

Kirk 1999 {published data only}Kirk JK, Pearce KA, Michielutte R, Summerson JH. Troglitazone ormetformin in combination with sulfonylureas for patients with type 2diabetes?. Journal of Family Practice 1999;48(11):879-82.

Klein 1975 {published data only}Klein W, Herrmann A. [Therapy of diabetes mellitus using metformin.Clinical study on 60 patients]. Medizinische Welt 1975;26(11):516-9.

Klein 1991 {published data only}Klein W. Sulfonylurea-metformin-combination versussulfonylurea-insulin-combination in secondary failures of sulfonylureamonotherapy. Results of a prospective randomized study in 50 patients.Diabete & Metabolisme 1991;17(1 Pt 2):235-40.

Lalau 1990 {published data only}Lalau JD, Vermersch A, Hary L, Andrejak M, Isnard F, Quichaud J. Type2 diabetes in the elderly: an assessment of metformin (metformin in theelderly). International Journal Clinical Pharmacocology Therapeutics andToxicology 1990;28(8):329-32.

Lalor 1990 {published data only}Lalor BC, Bhatnagar D, Winocour PH, Ishola M, Arrol S, Brading M, etal. Placebo-controlled trial of the effects of guar gum and metformin onfasting blood glucose and serum lipids in obese, type 2 diabetic patients.Diabete & Metabolisme 1990;7(3):242-5.

Lam 1998 {published data only}Lam KSL, Tiu SC, Tsang MW, Ip TP, Tam SCF. Acarbose in NIDDMpatients with poor control on conventional oral agents. Diabetes Care1998;21(7):1154-8.

Laurenti 1992 {published data only}Laurenti O, Bravi MC, Faldetta MC, De Mattia G. [Evaluation of theefficacy of metformin-glibenclamide treatment in overweight non-insulindependent diabetics]. La Clinica Terapeutica 1992;140(3):259-63.

Lawrence 2004 {published data only}Lawrence JM, Reid J, Taylor GJ, Stirling C, Reckless J. Favorable effectsof pioglitazone and metformin compared with glicazide on lipoproteinsubgractions in overweight patients with early type 2 diabetes. DiabetesCare 2004;27:41-6.

Lean 1983 {published data only}Lean ME, Borthwick LJ. Ciclazindol: an oral agent with weight reducingproperties and hypoglycaemic activity. European Journal of ClinicalPharmacology 1983;25(1):41-5.

Lee 1998 {published data only}Lee A, Morley JE. Metformin decreases food consumption and inducesweight loss in subjects with obesity with type II non-insulin-dependentdiabetes. Obesity Research 1998;6(1):47-53.

Lord 1983 {published data only}Lord JM, White SI, Bailey CJ, Atkins TW, Fletcher RF, Taylor KG. Effectof metformin on insulin receptor binding and glycaemic control in type IIdiabetes. British Medical Journal (Clinical Research Edition)1983;286(6368):830-1.

Lunetta 1996 {published data only}Lunetta M, DiMauro M. Different effect of acute and chronic oral metforminadministration on glucose and insulin response to bread and to pasta innon-insulin dependent diabetic patients. Diabetes Research and ClinicalPractice 1996;33(1):53-8.

Makimattila 1999 {published data only}Makimattila S, Nikkila K, Yki-Jarvinen H. Causes of weight gain duringinsulin therapy with and without metformin in patients with Type II diabetesmellitus. Diabetologia 1999;42(4):406-12.


Página 15


Manzella 2004 {published data only}Manzella D, Grella R, Esposito K, Giugliano D, Barbagallo M, PaolissoG. Blood pressure and cardiac auonomic nervous system in obese type 2diabetic patients: effect of metformin administration. American Journal ofHypertension 2004;17:223-7.

Marena 1994 {published data only}Marena S, Tagliaferro V, Montegrosso G, Pagano A, Scaglione L, PaganoG. Metabolic effects of metformin addition to chronic glibenclamidetreatment in type 2 diabetes [see comments]. Diabete & Metabolisme1994;20(1):15-9.

Marfella 1996 {published data only}Marfella R, Acampora R, Verrazzo G, Ziccardi P, De Rosa N, Giunta R,et al. Metformin improves hemodynamic and rheological responses toL-arginine in NIDDM patients. Diabetes Care 1996;19(9):934-9.

Marre 2002 {published data only}Marre M, Howlett H, Lehertt P, Allavoine T. Improved glycemic controlwith metformin-glibenclamide combined tabled therapy (Glucovance) intype 2 diabetic patients inadequately controlled on metformin. DiabeticMedicine 2002;19:673-80.

McAlpine 1988 {published data only}McAlpine LG, McAlpine CH, Waclawski ER, Storer AM, Kay JW, FrierBM. A comparison of treatment with metformin and gliclazide in patientswith non-insulin-dependent diabetes. European Journal of ClinicalPharmacology 1988;34(2):129-32.

McBain 1988 {published data only}McBain AM, Brown IR, Menzies DG, Campbell IW. Effects of improvedglycaemic control on calcium and magnesium homeostasis in type IIdiabetes. Journal of Clinical Pathology 1988;41(9):933-5.

McIntyre 1991 {published data only}McIntyre HD, Ma A, Bird DM, Paterson CA, Ravenscroft PJ, CameronDP. Metformin increases insulin sensitivity and basal glucose clearance intype 2 (non-insulin dependent) diabetes mellitus. Australian and NewZealand Journal of Medicine 1991;21(5):714-9.

Mehta 1963 {published data only}Mehta BJ, Vakil BJ, Narula DV, Vakil PR. Utility of Metformin as anon-hormonal agent in the treatment of Diabetes mellitus. Journal of theIndian Medical Association 1963;40:151-5.

Menzies 1989 {published data only}Menzies DG, Campbell IW, McBain A, Brown IR. Metformin efficacy andtolerance in obese non-insulin dependent diabetics: a comparison of twodosage schedules. Current Medical Research and Opinion 1989;11(5):273-8.

Moses 1999a {published data only}Moses R. Repaglinide in combination therapy with metformin in Type 2diabetes. Exp Clin Endocrinol Diabetes 1999;107(Suppl 4):S136-9.

Moses R, Slobodniuk R, Boyages S, Colagiuri S, Kidson W, Carter J, etal. Effect of repaglinide addition to metformin monotherapy on glycemiccontrol in patients with type 2 diabetes. Diabetes Care 1999;22(1):119-24.

Munk 1975 {published data only}Munk W. [Treatment of obese diabetic patients using glucophage retard].Zeltschrift fur Allgemeinmedzin 1975;51(14):681-3.

Nagi 1993 {published data only}Nagi DK, Ali VM, Yudkin JS. Effect of metformin on intact proinsulin anddes 31,32 proinsulin concentrations in subjects with non-insulin-dependent(type 2) diabetes mellitus. Diabete & Metabolisme 1996;13(8):753-7.

Nagi DK, Yudkin JS. Effects of metformin on insulin resistance, risk factorsfor cardiovascular disease, and plasminogen activator inhibitor in NIDDMsubjects. A study of two ethnic groups [see comments]. Diabetes Care1993;16(4):621-9.

Natali 2004 {published data only}Natali A, Baldeweg S, Toschi E, Capaldo B, Barbaro D, Gastaldelli AmYudkin JS, Ferrannini E. Vascular effects of improving metabolic controlwith metformin or rosiglitazone in type 2 diabetes. Diabetes Care2004;27:1349-57.

Nattrass 1977 {published data only}Nattrass M, Todd PG, Hinks L, Lloyd B, Alberti KG. Comparative effectsof phenformin, metformin and glibenclamide on metabolic rhythms inmaturity-onset diabetics. Diabetologia 1977;13(2):145-52.

Niazi 1998 {published data only}Niazi R, Muzaffar Z. Comparison of bedtime NPH insulin or metformincombined with glibenclamide in secondary sulphonylurea failure in obesetype II (NIDDM) patients. Journal of the Pakistani Medical Association1998;48(11):336-8.

Nosadini 1987 {published data only}Nosadini R, Avogaro A, Trevisan R, Tessari P, Duner E, Tiengo A, et al.Effect of metformin on insulin-stimulated glucose turnover and insulinbinding to receptors in Type 2 diabetes. Diabetes Care 1987;10:62-7.

Noury 1991 {published data only}Noury J, Nandeuil A. Comparative three-month study of the efficacies ofmetformin and gliclazide in the treatment of NIDD. Diabete & Metabolisme1991;17(1 Pt 2):209-12.

Ohnhaus 1983 {published data only}Ohnhaus EE, Berger W, Duckert F, Oesch F. The influence ofDimethylbiguanide on Phenprocoumon elimination and its mode of action.Klinische Wochenschrift 1983;61:851-8.

Pavo 2003 {published data only}Pavo I, Jermendy G, Varkonyi TT, Kerenyl Z, Gyimesi A, Shoustov S, etal. Effect of pioglitazone compared with metformin on glycemic controland indicators of insulin sensitivity in recently diagnosed patients with type2 diabetes. Journal of Clinical Endocrinology and Metabolism2003;88:1637-45.

Peacock 1984 {published data only}Peacock I, Tattersall RB. The difficult choice of treatment for poorlycontrolled maturity onset diabetes: tablets or insulin?. British MedicialJournal (Clinical Research Edition) 1984;288(6435):1956-9.

Peacock 1986 {published data only}Peacock I, Hawkins M, Heptinstall S. Platelet behaviour innon-insulin-dependent Diabetes - Influence of vascular complications,treatment and metabolic control. Thrombosis and heamostasis1986;55:361-5.

Pedersen 1965 {published data only}Pedersen J. The effect of Metformin on weight loss in obesity. ActaEndocrinologica 1965;49:479-86.

Pedersen 1989 {published data only}Pedersen O, Nielsen O, Bak J, Richelsen B, Beck-Nielsen H, Sorensen N.The effects of metformin on adipocyte insulin action and metabolic controlin obese subjects with type 2 diabetes. Diabete & Metabolisme1989;6(3):249-56.

Pirart 1961 {published data only}Pirart J, Rutman S. [A new oral antidiabetic medicine: N.N.Dimethylbiguanide. Clinical trial alternating a placebo with a sulfamide].Acta Clinica Belgica 1961;16:575-89.

Ponssen 2000 {published data only}Ponssen HH, Elte JW, Lehert P, Schouten JP, Bets D. Combined metforminand insulin therapy for patients with type 2 diabetes mellitus [In ProcessCitation]. Clinical Therapeutics 2000;22(6):709-18.

Prager 1986 {published data only}Prager R, Schernthaner G, Graf H. Effect of metformin on peripheral insulinsensitivity in non insulin dependent diabetes mellitus. Diabete &Metabolisme 1986;12(6):346-50.


Página 16


Puchegger 1964 {published data only}Puchegger R. [Experience with the biguanide Glucophage]. WienerKlinische Wechenschrift 1964;76:335-7.

Rachmani 2002 {published data only}Rachmani R, Slavachevski I, Levi Z, Zadok B-S, Kedar Y, Ravid M.Metformin in patients with type 2 diabetes mellitus: reconsideration oftraditional contraindications. Eur J Int Med 2002;13:428-433.

Rains 1988 {published data only}Rains SG, Wilson GA, Richmond W, Elkeles RS. The effect ofglibenclamide and metformin on serum lipoproteins in type 2 diabetes.Diabete & Metabolisme 1988;5(7):653-8.

Rains 1989 {published data only}Rains SG, Wilson GA, Richmond W, Elkeles RS. The reduction of lowdensity lipoprotein cholesterol by metformin is maintained with long-termtherapy. Journal of the Royal Society of Medicine 1989;82(2):93-4.

Raptis 1996 {published data only}Raptis AE, Tountas NB, Yalouris AG, Halvatsiotis PG, Raptis SA.Therapeutic effect of glibenclamide in a fixed combination with metforminor phenformin in NIDDM patients. Hormone and Metabolic Research1996;28(2):89-94.

Reaven 1992 {published data only}Reaven GM, Johnston P, Hollenbeck CB, Skowronski R, Zhang JC,Goldfine ID, et al. Combined metformin-sulfonylurea treatment of patientswith noninsulin-dependent diabetes in fair to poor glycemic control. Journalof Clinical Endocrinology and Metabolism 1992;74(5):1020-6.

Relimpio 1998 {published data only}Relimpio F, Pumar A, Losada F, Mangas MA, Acosta D, Astorga R. Addingmetformin versus insulin dose increase in insulin-treated but poorlycontrolled Type 2 diabetes mellitus: an open-label randomized trial. Diabete& Metabolisme 1998;15(12):997-1002.

Reyes 1969 {published data only}Reyes Larrasilla JM. [Chlorpropamide-dimethylbiguanide combination inthe control of diabetes mellitus with complications]. La Prensa MedicaMexicana 1969;34(9):378-81.

Riccio 1991 {published data only}Riccio A, Del Prato S, Vigili de Kretzenberg S, Tiengo A. Glucose andlipid metabolism in non-insulin-dependent diabetes, effect of metformin.Diabete & Metabolisme 1991;17(1 pt 2):180-4.

Robinson 1998 {published data only}Robinson AC, Burke J, Robinson S, Johnston DG, Elkeles RS. The effectsof metformin on glycemic control and serum lipids in insulin-treatedNIDDM patients with suboptimal metabolic control [see comments].Diabetes Care 1998;21(5):701-5.

Roden 2005 {published data only}Roden M, Laaksot M, Johns D, Widel M, Urquhart R, Richardson C, et al.Long-term effects of pioglitazone and metformin on insulin sensitivity inpatients with type 2 diabetes mellitus. Diabetic Medicine 2005;22:1101-6.

Rodger 1995 {published data only}Rodger NW, Chiasson JL, Josse RG, Hunt JA, Palmason C, Ross SA, etal. Clinical experience with acarbose: results of a Canadian multicentrestudy. Clinical Investigation in Medicine 1995;18(4):318-24.

Roger 1999 {published data only}Roger P, Auclair J, Drain P. Addition of benflouorex to biguanide improvesglycemic control in obese non-insulin-dependent diabets: a double-blindstudy versus placebo. Journal of Diabetes and its complications1999;13(2):62-7.

Rosenstock 1998 {published data only}Rosenstock J, Brown A, Fischer J, Jain A, Littlejohn T, Nadeau D, et al.Efficacy and safety of Acarbose in Metformin-treated patients with Type2 Diabetes. Diabetes Care 1998;21:2050-5.

Sanchez-Barba 1999 {published data only}Sanchez-Barba Izquierdo MI, Ibarra Rueda JM, Ruiz de Adana Perez T.[The combination of insulin and metformin in obese patients with type-2diabetes mellitus]. Atencion Primaria 1999;24(8):462-7.

Santos 1995 {published data only}Santos RF, Nomizo R, Wajhenberg BL, Reaven GM, Azhar S. Changes ininsulin receptor tyrosine kinase activity associated with metformin treatmentof type 2 diabetes. Diabete & Metabolisme 1995;21(4):274-80.

Schernthaner 2004 {published data only}Schernthaner G, Matthews DR, Charbonnel B, Hanefeld M, Brunetti P andthe Quarter Study Group. Efficacy and safety of pioglitazone versusmetformin in patients with type 2 diabetes mellitus: a double-blind,randomized trial. The Journal of Clinical Endocrinology & Metabolism2004;89:6-68-76.

Schneider 1990 {published data only}Schneider J, Erren T, Zofel P, Kaffarnik H. Metformin-induced changes inserum lipids, lipoproteins, and apoproteins in non-insulin-dependent diabetesmellitus. Atherosclerosis 1990;82(1-2):97-103.

Schulte 1973 {published data only}Schulte J, Garcia Viveros M, Rull J, Lozano-Castaneda O. [Long-term (3year) results of the treatment of stable diabetes with low doses ofchlopropamide-metformin]. La Prensa Medica Mexicana 1973;38(7):281-2.

Sieradzki 1999 {published data only}Siedradzki J, Soszynski P. [Assessment of efficacy and safety of acarbosein the treatment of diabetes mellitus. Observation study in the conditionsof general health care]. Przeglad Lekarski 1999;56(5):335-41.

Stades 2000 {published data only}Stades AM, Heikens JT, Holleman F, Hoekstra JB. Effect of metformin onglycaemic control in type 2 diabetes in daily practice: a retrospective study.Netherlands Journal of Medicine 2000;56(3):86-90.

Stalhammar 1991 {published data only}Stalhammer J, Bergman U, Boman K, Dahlen M. Metabolic control indiabetic subjects in three Swedish areas with high, medium and low salesof antidiabetic drugs. Diabetes Care 1991;14:12-19.

Sterne 1963 {published data only}Sterne J. [Report on five years experience with Metformin]. WienerMedizinische Wochenschrift 1963;113:599-602.

Stratmann 1965 {published data only}Stratmann FW. [Experience with Dimethylbiguanide in late failrues of oraldiabetes therapy]. Medizinische Welt 1965;49:2743-6.

Strowig 2002 {published data only}Strowig SM, Aviles-Santa ML, Raskin P. Comparison of insulinmonotherapy and combination therapy with insulin and metformin or insulinand troglitazone in type 2 diabetes. Diabetes Care 2002;25:1691-8.

Stumvoll 1995 {published data only}Stumvoll M, Nurjhan N, Perriello G, Dailey G, Gerich J. Metabolic effectsof metformin in non-insulin-dependent diabetes mellitus. New EnglandJournal of Medicine 1995;333:550-4.

Sundaresan 1997 {published data only}Sundaresan P, Lykos D, Daher A, Diamond T, Morris R, Howes LG.Comparative effects of glibenclamide and metformin on ambulatory bloodpressure and cardiovascular reactivity in NIDDM. Diabetes Care1997;20(5):692-7.

Swislocki 1999 {published data only}Swislocki ALM, Knuu Q, Liao E, Wu E, Beza F, Lopez J, et al. Safety andefficacy of metformin in a restricted formulary. American Journal ofManaged Care 1999;5:62-8.

Szanto 1964 {published data only}Szanto S. Combined trial of Acetohexamide and two Diguanide preparations.Irish Journal of Medical Science 1964:3-11.


Página 17


Taylor 1982 {published data only}Taylor KG, John WG, Matthews KA, Wright AD. A prospective study ofthe effect of 12 months treatment on serum lipids and apolipoproteins A-Iand B in Type 2 (non-insulin-dependent) diabetes. Diabetologia1982;23(6):507-10.

Tessari 1994 {published data only}Tessari P, Biolo G, Bruttomesso D, Inchiostro S, Panebianco G, VedovatoM, et al. Effects of metformin treatment on whole-body and splanchnicamino acid turnover in mild type 2 diabetes. Journal of ClinicalEndocrinology and Metabolism 1994;79(6):1553-60.

Tessier 1999 {published data only}Tessier D, Maheux P, Khalil A, Fulop T. Effects of gliclazide versusmetformin on the clinical profile and lipid peroxidation markers in type 2diabetes. Metabolism 1999;48(7):897-903.

Testa 1996 {published data only}Testa R, Bonfigli AR, Piantanelli L, Manfrini S, Testa I, Gregorio F.Relationship between plasminogen activator inhibitor type-1 plasma levelsand the lipoprotein(a) concentrations in non-insulin-dependent diabetesmellitus. Diabetes Research and Clinical Practice 1996;33(2):111-8.

Teupe 1991 {published data only}Teupe B, Bergis K. Prospective randomized two-years clinical studycomparing additional metformin treatment with reducing diet in type 2diabetes. Diabete & Metabolisme 1991;17(1 Pt 2):213-7.

Tikkainen 2004 {published data only}Tikkainen M, Hakkinen A, Korsheninnikova E, Nyman T, Makimattila S,Yki-Jarvinen H. Effects of rosiglitazone and metformin on liver fat content,hepatic insulin resistance, insulin clearance, and fene expression in adiposetissue in patients with type 2 diabetes. Diabetes 2004;53:2169-76.

Tosi 2003 {published data only}Tosi F, Muggeo M, Brun E, Spiazzi G, Perbelli L, Zanolin E, et al.Combination treatment with metformin and glibenclamide versus single-drugtherapies in type 2 diabetes mellitus: a randomized, double-blind,comparative study. Metabolism 2003;52:862-7.

Trischitta 1992 {published data only}Trischitta V, Italia S, Mazzarino S, Buscema M, Rabuazzo AM, SangiorgioL, et al. Comparison of combined therapies in treatment of secondary failureto glyburide. Diabetes Care 1992;15(4):539-42.

Trischitta 1998 {published data only}Trischitta V, Italia S, Raimondo M, Guardabasso V, Licciardello C, RunelloF, et al. Efficacy of combined treatments in NIDDM patients with secondaryfailure to sulphonylureas. Is it predictable?. Journal of EndocrinologicalInvestigation 1998;21(11):744-7.

UKPDS-34 1998 {published data only}Holman RR, Cull CA, Turner RC and UKPDS Study Group. A randomizeddouble-blind trial of Acarbose in Type II Diabetes shows improved glycemiccontrol over 3 years (UK Prospective Diabetes Study 44). Diabetes Care1999;22:960-4.

Turner R, Cull C, Holman R, and the UK Prospective Diabetes Study Group.United Kingdom Prospective Diabetes Study 17: a 9-year update of arandomized, controlled trial on the effect of improved metabolic controlon complications in non-insulin-dependent diabetes mellitus. Annals ofInternal Medicine 1996;124(1 Pt 2):136-45.

Turner RC, Cull CA, Frighi V, Holman RR, and the UK ProspectiveDiabetes Study Group. Glycemic control with diet, sulfonylurea, metformin,or insulin in patients with type 2 diabetes mellitus: progressive requirementfor multiple therapies (UKPDS 49). UK Prospective Diabetes Study(UKPDS) Group. Journal of the American Medical Association1999;281(21):2005-12.

U. K Prospective Diabetes Study (UKPDS) Group. U. K. prospectivediabetes study 16. Overview of 6 years' therapy of type II diabetes: aprogressive disease. U.K. Prospective Diabetes Study Group [published

erratum appears in Diabetes 1996 Nov;45(11):1655]. Diabetes1995;44(11):1249-58.

U.K. Prospective Diabetes Study Group. U.K. prospective diabetes study.II. Reduction in HbA1c with basal insulin supplement, sulfonylurea, orbiguanide therapy in maturity-onset diabetes. A multicenter study. Diabetes1985;34(8):793-8.

U.K. Prospective Diabetes Study Group. United Kingdom ProspectiveDiabetes Study (UKPDS) 13: Relative efficacy of randomly allocated diet,sulphonylurea, insulin, or metformin in patients with newly diagnosednon-insulin dependent diabetes followed for three years [see comments].BMJ 1995;310(6972):83-8.

U.K. Prospective Diabetes Study Group. United Kingdom ProspectiveDiabetes Study 24: a 6-year, randomized, controlled trial comparingsulfonylurea, insulin, and metformin therapy in patients with newlydiagnosed type 2 diabetes that could not be controlled with diet therapy[see comments]. Annals of Internal Medicine 1998;128(3):165-75.

U.K. Prospective Diabetes Study Group. UKPDS 28: a randomized trial ofefficacy of early addition of metformin in sulfonylurea-treated type 2diabetes. Diabetes Care 1998;21(1):87-92.

UK Prospective Diabetes Study (UKPDS) Group. Effect of intensiveblood-glucose control with metformin on complications in overweightpatients with type 2 diabetes (UKPDS 34). Lancet 1998;352:854-65.

Vannasaeng 1995 {published data only}Vannasaeng S, Ploybutr S, Nitiyanant W, Peerapatdit T, Vichayanrat A.Effects of alpha-glucosidase inhibitor (acarbose) combined with sulfonylureaor sulfonylurea and metformin in treatment of non-insulin-dependentdiabetes mellitus. Journal of the Medical Association of Thailand1995;78(11):578-85.

Velussi 1992 {published data only}Velussi M, Cernigoi AM, Viezzoli L, Caffau C. [Median-term (4 months)treatment with glibenclamide + metformin substituting for glibenclamide+ fenformin lowers the lacticemia levels in type-2 diabetics (NIDDM)]. LaClinical Terapeutica 1992;141(12):483-92.

Vigneri 1991 {published data only}Vigneri R, Trischitta V, Italia S, Mazzarino S, Rabuazzo MA, Squatrito S.Treatment of NIDDM patients with secondary failure to glyburide:comparison of the addition of either metformin or bed-time NPH insulinto glyburide. Diabete & Metabolisme 1991;17(1 Pt 2):232-4.

Willey 1992 {published data only}Willey KA, Moyneaux JE, Overland JE, Yue DK. The effects ofdexfenluramine on blood glucose control in patients with Type 2 Diabetes.Diabetic Medicine 1992;9:341-3.

Willey 1994 {published data only}Willey KA, Molyneaux LM, Yue DK. Obese patients with type 2 Diabetespoorly controlled by insulin and metformin: Effects of adjunctiveDexfenfluramine therapy on glycaemic control. Diabetic Medicine1994;11:701-4.

Willms 1999 {published data only}Willms B, Ruge D. Comparison of acarbose and metformin in patients withType 2 diabetes mellitus insufficiently controlled with diet andsulphonylureas: a randomized, placebo-controlled study. Diabetic Medicine1999;16(9):755-61.

Wilson 1989 {published data only}Wilson JA, Scott MM, Gray RS. A comparison of metformin versus guarin combination with sulphonylureas in the treatment of non insulindependent diabetes. Hormone and Metabolic Research 1989;21(6):317-9.

Wolever 1995 {published data only}Wolever TMS, Radmard R, Chiasson J-L, Hunt JA, Josse RG, PalmasonC, et al. One-year Acarbose treatment raises fasting serum acetate inDiabetic patients. Diabetic Medicine 1995;12:164-72.


Página 18


Wolever 2000 {published data only}Wolever TMS, Assiff L, Basu T, Chiasson J-L, Boctor M, Gerstein HC, etal. Miglitol, an alpha-glucosidase inhibitor, prevents the metformin-inducedfall in serum folate and vitamin B12 in subjects with type 2 diabetes.Nutrition Research 2000;20(10):1447-56.

Wu 1990 {published data only}Wu MS, Johnston P, Sheu WH, Hollenbeck CB, Jeng CY, Goldfine ID, etal. Effect of metformin on carbohydrate and lipoprotein metabolism inNIDDM patients. Diabetes Care 1990;13(1):1-8.

Wulffele 2000 {published data only}Wulffele MG, Kooy A, Ogterop C, Borger vd Burg B, Stehouwer CDA,Donker AJM. Metformin and insulin therapy decreases glycosylatedhemoglobin and insulin requirement in Type 2 Diabetes. Diabetologia2000;43(Suppl 1):A184.

Wulffele 2002 {published data only}Wulffele MG, Kooy A, Lehert P, Bets D, Ogterop JC, van der Burg BB,et al. Combination of insulin and metformin in the treatment of type 2diabetes. Diabetes Care 2002;25:2133-40.

Wulffele 2003 {published data only}Wulffele MG, Kooy A, Lehert P, Bets D, Ogterop JC, Can der Burg BB,et al. Effects of short-term treatment with metformin on serumconcentrations of homocysteine, folate and vitamin B12 in type 2 diabetesmellitus: a randomized, placebo-controlled trial. Journal of InternalMedicine 2003;254:455-63.

Wulffele 2005 {published data only}Wulffele MG, Kooy A, Lehert P, Bets D, Donker AHM, Stehouwer CDA.Does metformin decrease blood pressure in patients with type 2 diabetesintensively treated with insulin?. Diabetic Medicine 2005;22:907-13.

Yamanouchi 2005 {published data only}Yamanouchi T, Sakai T, Igarashi K, Ichiyanagi K, Watanabe H, KawasakiT. Comparison of metabolic effects of pioglitazone, metformin, andglimepiride over 1 year in Japanese patients with newly diagnosed type 2diabetes. Diabetic Medicine 2005;22:980-5.

Yki-Jarvinen 1999 {published data only}Yki-Jarvinen H, Ryysy L, Nikkila K, Tulokas T, Vanamo R, Heikkila M.Comparison of bedtime insulin regimens in patients with type 2 diabetesmellitus. A randomized, controlled trial [see comments]. Annals of InternalMedicine 1999;130(5):389-96.

Yu 1999 {published data only}Yu JG, Kruszynska YT, Mulford MI, Olefsky JM. A comparison oftroglitazone and metformin on insulin requirements in euglycemicintensively insulin-treated type 2 diabetic patients. Diabetes1999;48(12):2414-21.

Referencias de los estudios excluidos de esta revisión

Aguilar 1992bAguilar C, Reza A, Garcia JE, Rull JA. Biguanide related lactic acidosis:incidence and risk factors. Archives of Medical Research 1992;23(1):19-24.

Bernard 1965Bernard C, Bernard R. [Tolbutamide-dimethyl-biguanide association in thetreatment of diabetes mellitus]. Gazette Medicale de France1965;72(14):2763-5.

Bonfigli 1999Bonfigli AR, Manfrini S, Gregorio F, Testa R, De Sio G, Coppa G.Determination of plasma Metformin by a new cation-exchange HPLCtechnique. Therapeutic Drug Monitoring 1999;21:330-4.

Bruneder 1978Bruneder H, Klein HJ. [Blood lactate and biguanide therapy: comparisonbetween phenformin, metformin and buformin in 408 senile diabetics].Acta Medica Austriaca 1978;5(3):88-90.

Cacciapuoti 1991Cacciapuoti F, Spiezia R, Bianchi U, Lama D, D'Avino M, Varricchio M.Effectiveness of Glibenclamide on myocardial ischemic ventriculararrhythmias in non-insulin-dependent Diabetes mellitus. American Journalof Cardiology 1991;67:843-7.

Chow 1995Chow C-C, Tsand LWW, Sorensen JP, Cockram CS. Comparison of insulinwith or without continuation of oral hypoglycemic agents in the treatmentof secondary failure in NIDDM patients. Diabetes Care 1995;18(3):307-14.

Clauson 1996Clauson P, Karlander S, Steen L, Efendic S. Saytime Glibenclamide andbedtime NPH insulin compared to intensive insulin treatment in secondarySulphonylurea failure: a 1-year follow-up. Diabetic Medicine 1995;13:471-7.

Connolly 1996Connolly V, Kesson CM. Metformin treatment in NIDDM patients withmild renal impairment. Postgraduate Medical Journal 1996;72(848):352-4.

Daniel 1997Daniel JR, Hagmeyer KO. Metformin and insulin: is there a role forcombination therapy?. Annals of Pharmacotherapy 1997;31(4):474-80.

Debry 1964Debry G, Anziani, Cherrier, Laurent J. [Study of lactic acid levels on anempty stomach in diabetic patients treated with N-N-dimethylbiguanide].Diabete 1964;12:239-45.

Debry 1966aDebry G, Laurent J. [Results of the treatment of Diabetes mellitus in theadult with the association of Metformin and sulfamides (concerning 197cases)]. Journal de Medicine de Lyon 1966;47:407-13.

Debry 1966bDebry G, Laurent J. [Results of the treatment of Diabetes mellitus in theadult with metformin (concerning 202 cases0]. Journal de Medicine deLyon 1966;47:395-404.

Fery 1997Fery F, Plat L, Balasse EO. Effects of metformin on the pathways of glucoseutilization after oral glucose in non-insulin-dependent diabetes mellituspatients. Metabolism 1997;46(2):227-33.

Galuska 1994Galuska D, Nolte LA, Zierath JR, Wallberg-Henriksson H. Effect ofmetformin on insulin-stimulated glucose transport in isolated skeletal muscleobtained from patients with NIDDM. Diabetologia 1994;37:826-32.

Gibson 1995Gibson JM, Westwood M, Crosby SR, Gordon C, Holly JMP, Fraser W,et al. Choice of treatment affects plasma levels of insulin-like growthfactor-binding protein-1 in noninsulin-dependent Diabetes mellitus. Journalof Clinical Endocrinology and Metabolism 1995;80:1369-75.

Gin 1982Gin H, Slama G, Weissbrodt P, Poynard T, Vexiau P, Klein JC, et al.Metformin reduces post-prandial insulin needs in Type 1 (Insulin-dependent)diabetic patients: Assessment by the artificial pancreas. Diabetologia1982;23:34-6.

Gin 1985Gin H, Messerchmitt C, Brottier E, Aubertin J. Metformin improved insulinresistance in type I, insulin-dependent, diabetic patients. Metabolism1985;34(10):923-5.

Gin 1989Gin H, Freyburger G, Boisseau M, Aubertin J. Study of the effect ofmetformin on platelet aggregation in insulin-dependent diabetics. DiabetesResearch and Clinical Practice 1989;6(1):61-7.


Página 19


Giugliano 1979Giugliano D, Torella R, Improta L, Scognamiglio G. Effects of metforminon plasma glucose, insulin, FFA, glucagon, growth hormone and cortisolresponses to oral glucose in subjects with chemical diabetes. Farmacoedizione Pratica 1979;34(1):32-41.

Guthrie 1997Guthrie R. Treatment of non-insulin-dependent diabetes mellitus withmetformin. Journal of the American Board of Family Practice1997;10(3):213-21.

Irsigler 1978Irsigler K, Kritz H, Regal H, Foltin E. [The risk of lacticate acidosis: acomparison of the 3 biguanides in treatment of diabetics (authors' transl)].Wiener Klinische Wochenschrift 1978;90(10):332-7.

Ismail 1978Ismail AA, Badrawi H, Nasr S, Khater R, Hafez AA, El-Kirdassy ZH, etal. The effect of metformin on individual amino acid absorption. Journalof the Egyptian Medical Association 1978;61(9-10):673-92.

Isnard 1991Isnard F, Andre P. [Glycemic equilibrium and weight evolution in diabeticstaking metformin]. Journees Annuelles Diabetology Hotel-Dieu1991:297-314.

Isnard 1996Isnard F. [Long term treatment with biguanides in the natural evolution ofnon-insulin-dependent diabetes]. Journ Annu Diabetol Hotel Dieu1996:273-88.

Jansson 1996Jansson PA, Gudbjornsdottir HS, Andersson OK, Lonnroth PN. The effectof metformin on adipose tissue metabolism and peripheral blood flow insubjects with NIDDM. Diabetes Care 1996;19(2):160-4.

Johansen 1999Johansen K. Efficacy of metformin in the treatment of NIDDM:Meta-analysis. Diabetes Care 1999;22:33-7.

Lalau 1994Lalau JD, Lacroix C, De Cagny B, Fournier A. Metformin-associated lacticacidosis in diabetic patients with acute renal failure. A critical analysis ofits pathogenesis and prognosis. Nephrology Dialysis and Transplantation1994;9(Suppl 4):126-9.

Lalau 1995Lalau JD, Lacroix C, Compagnon P, de Cagny B, Rigaud JP, Bleichner G,et al. Role of metformin accumulation in metformin-associated lacticacidosis. Diabetes Care 1995;18(6):779-84.

Leslie 1987Leslie P, Jung RT, Isles TE, Baty J. Energy expenditure in non-insulindependent diabetic subjects on metformin or sulphonylurea therapy. ClinicalScience 1987;73(1):41-5.

Lim 1970Lim P, Khoo OT. Metformin compared with tolbutamide in the treatmentof maturity-onset diabetes mellitus. Medical Journal of Australia1970;1(6):271-3.

Messens 1965Messens Y, Margoulies M. [Treatment of 81 cases of diabetes mellitus withN.N. dimethyl-biguanide]. Revue Medicale de Liege 1965;20(22):607-13.

Messens 1966Messens Y, Margoulies M. [Treatment of diabetes mellitus with N.N.dimethy-biguanide]. Diabete 1966;14(2):74-9.

Muntoni 1965Muntoni S, Boero A, Corona M, Flores M. [Dimethyl-biguanide in thetreatment of diabetes mellitus]. La Clinica terapeutica 1965;35:227-51.

Nauck 1993Nauck MA, Kleine N, Orskov C, Holst JJ, Willms B, Creutzfeldt W.Normalization of fasting hyperglycaemia by exogenous glucagon-likepeptide 1 (17-36 amide) in Type 2 (non-insulin-dependent) diabetic patients.Diabetologia 1993;36:741-4.

Nauck 1997Nauck MA, Holst JJ, Wilms B. Glucagon-like peptide 1 and its potentialin the treatment of non-insulin-dependent Diabetes mellitus. Hormone andMetabolic Research 1997;29:411-6.

O'Connor 1998O'Connor PJ, Spann SJ, Woolf SH. Care of adults with type 2 diabetesmellitus. A review of the evidence [see comments]. Journal of FamilyPractice 1998;47(5 Suppl):S13-22.

Orlikowska 1966Orlikowska W. [Influence of N.N. dimethyl biguandie on certain elementsof lipid metabolism in diabetic patients]. Le Diabete 1966;14:183-9.

Panahloo 1995Panahloo A, Mohamed-Ali V, Lane A, Green F, Humphries SE, YudkinJS. Determinants of plasminogin activatory inhibitor 1 activiry in treatedNIDDM and its relation to a polymorphosm in the Plasminogen ActivatorInhibitor 1 gene. Diabetes 1995;44(37-42).

Perriello 1994Perriello G, Misericordia P, Volpi E, Santucci A, Santucci C, Ferannini E,et al. Acute antihyperglycemic mechanisms of Metformin in NIDDM:Evidence for suppression of lipid oxidation and hepatic glucose production.Diabetes 1994;43:920-8.

Pilger 1978Pilger E, Schmid P, Goebel R. [Effect of biguanide therapy on lactatemetabolism during graded submaximal ergometric testing]. Acta MedicaAustriaca 1978;5(3):91-5.

Prager 1983Prager R, Schernthaner G. Insulin receptor binding to monocytes, insulinsecretion, and glucose tolerance following metformin treatment. Resultsof a double-blind cross-over study in type II diabetics. Diabetes1983;32(12):1083-6.

Rambert 1961Rambert P, Canivet J, Quichaud J, Spitz B. [Treatment of diabetes mellituswith NN-dimethyl-diguanide. Experience of 177 cases]. Semaine desHospitaux de Paris 1961;37:247-54.

Rigas 1968Rigas AN, Bittles AH, Hadden DR, Montgomery DA. Circadian variationof glucose, insulin, and free fatty acids during long-term use of oralhypoglycaemic agents in diabetes mellitus. British Medical Journal1968;3(622):25-8.

Rizkalla 1986Rizkalla SW, Elgrably F, Tchobroutsky G, Slama G. Effects of metformintreatment on erythrocyte insulin binding in normal weight subjects, in obesenon diabetic subjects, in type 1 and type 2 diabetic patients. Diabete &Metabolisme 1986;12(4):219-24.

Sambol 1996Sambol NC, Chiang J, O'Conner M, Liu CY, Lin ET, Goodman AM, et al.Pharmacokinetics and pharmacodynamics of metformin in healthy subjectsand patients with noninsulin-dependent diabetes mellitus. Journal of ClinicalPharmacology 1996;36(11):1012-21.

Scarpello 1998Scarpello JH, Hodgson E, Howlett HC. Effect of metformin on bile saltcirculation and intestinal motility in type 2 diabetes mellitus. DiabeticMedicine 1998;15(8):651-6.


Página 20


Schaffalitzky 1979Schaffalitzky de Muckadell OB, Mortensen H, Lyngsoe J. Metabolic effectsof glucocorticoid and ethanol administration in phenformin- andmetformin-treated obese diabetics. Acta Medica Scandinavia1979;206(4):269-73.

Selby 1999Selby JV, Ettinger B, Swain BE, Brown JB. First 20 months' experiencewith use of metformin for type 2 diabetes in a large health maintenanceorganization. Diabetes Care 1999;22(1):38-44.

Signore 1996Signore A, Fiore V, Chianelli M, Procaccini E, Barone R, Ronga G, et al.The effect of metformin on liver blood flow in vivo in normal subjects andpatients with non insulin dependent diabetes. Diabetes Research andClinical Practice 1996;33(2):83-7.

Slama 1984Slama G, Jean-Joseph P, Goicolea I, Elgrably F, Haardt MJ, CostagliolaD, et al. Sucrose taken during mixed meal has no additional hypergycemicaction over isocaloric amounts of starch in well-controlled diabetics. Lancet1984;July 21:122-5.

Stefanovic 1999Stefanovic V, Antic S, Mitic-Zlatkovic M, Vlahovic P. Reversal of increasedlymphocyte PC-1 activity in patients with type 2 diabetes treated withmetformin. Diabetes/Metabolism Research Reviews 1999;15(6):400-4.

Sugawara 1962Sugawara Y, Nitmi T, Sato T, Nakai Y, Mivata K, Kano H, et al. [Clinicaltrial with DMBG (Melbin) in diabetes mellitus]. Naika hokan, Japanesearchives of Internal Medicine 1962;9:405-7.

Sum 1992Sum C-F, Webster JM, Johnson AB, Catalano C, Cooper BG, Taylor R.The effect of intravenous Metformin on glucose metabolism duringhyperglycemia in Type 2 Diabetes. Diabetic Medicine 1992;9:61-6.

Teitelbaum 1963Teitelbaum M, Le Marchant J-M. [Biguanides and sulfamides in smalldoses in the initial treatment of diabetes in obese patients]. Diabete1963;11:342-3.

Trischitta 1983Trischitta V, Gullo D, Pezzino V, Vigneri R. Metformin normalizes insulinbinding to monocytes from obese nondiabetic subjects and obese type IIdiabetic patients. Journal of Clinical Endocrinology and Metabolism1983;57(4):713-8.

Turner 1995Turner RC, Holman RR. Lessons from IK prospective diabetes study.Diabetes Research and Clinical Practice 1995;28(Suppl):S151-S157.

Zapecka-Dubno 1999Zapecka-Dubno B, Czyzyk A, Dworak A, Bak MI. Effect of oral antidiabeticagents on plasma amylin level in patients with non-insulin-dependentDiabetes mellitus (Type 2). Arzliche-Forschung/ Drug Research1999;49(1):330-4.

Referencias adicionales

AHFS 1999Metformin hydrochloride. American Hospital Formulary Service DrugInformation. Bethesda: American Society of Health-System Pharmacists,Inc, 1999:2755-63.

Bailey 1996Bailey CJ, Turner RC. Metformin. New England Journal of Medicine1996;334:574-9.

Bergman 1978Bergman U, Boman G, Wiholm B-E. Epidemiology of adverse drugreactions to phenformin and metformin. British Medicial Journal1978;2:464-6.

Berlin 1989Berlin JA, Laird NM, Sacks HS, Chalmers TC. A comparison of statisticalmethods for combining event rates from clinical trials. Statistics in Medicine1989;8:141-51.

Brown 1998Brown JB, Pedula K, Barzilay J, Herson MK, Latare P. Lactic acidosisrates in type 2 diabetes. Diabetes Care 1998;21:1659-63.

Campbell 1985Campbell IW. Metformin and the sulphonylureas: The comparative risk.Hormone and Metababolism Research Supplement 1985;15:105-11.

Campbell 1995Campbell IW, Howlett HCS. Worldwide experience of metformin as aneffective glucose-lowering agent: A meta-analysis. Diabetes MetabolismReviews 1995;11 (Suppl 1):S57-62.

DeFronzo 1993DeFronzo RA, Goodman A, and the Metformin Investigator Group.Combined metformin/glyburide treatment in NIDDM patients not optimallyresponding to maximum dose sulfonylurea: Results of a mutlicenter trial.Diabetes 1993;42(Suppl 1):146A (Abs 455).

DeFronzo 1999DeFronzo RA. Pharmacologic therapy for type 2 diabetes mellitus. Annalsof Internal Medicine 1999;131:281-303.

Fleiss 1981Fleiss JL. Statistical methods for rates and proportions. 2nd Edition. NewYork: Wiley, 1981:217-34.

Gan 1992Gan SC, Barr J, Arieff AI, Pearl RG. Biguanide-assoicated lactic acidosis:case report and review of the literature. Archives of Internal Medicine1992;152:2333-6.

Holstein 1999Holstein A, Nahrwold D, Hinze S, Egbert E-H. Contra-indications tometformin therapy are largely disregarded. Diabetic Medicine1999;16:692-6.

Jadad 1996Jadad AR, Moore A, Carroll D, Jenkinson C, Reynolds DJM, GavaghanDJ, et al. Assessing the quality of reports of randomized clinical trials: Isblinding necessary?. Controlled Clinical Trials 1996;17:1-12.

Kreisberg 1980Kreisberg RA. Lactate homerostasis and lactic acidosis. Annals of InternalMedicine 1980;92(Part 1):227-37.

Luft 1978Luft D, Schmulling, Eggstein M. Lactic acidosis in biguanide-treateddiabetics: a review of 330 cases. Diabetologia 1978;14:75-87.

Misbin 1998Misbin RI, Green L, Stadel BV, Gueriguian JL, Gubbi A, Fleming GA.Lacic acidosis in patients with diabetes treated with metformin. NewEngland Journal of Medicine 1998;338:265-6.

Olivia 1970Olivia PB. Lactic acidosis. American Journal of Medicine 1970;48:208-25.

Palumbo 1998Palumbo PJ. Metformin: Effects of cardiovascular risk factors in patientswith non-insulin-dependent diabetes mellitus. Journal of Diabetes and itsComplications 1998;12:110-9.

Scheen 1996Scheen AJ. Clinical pharmacokinetics of metformin. ClinicalPharmacokinetics 1996;30(5):359-71.


Página 21


Schulz 1995Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias:Dimensions of methodological quality associated with estimates of treatmenteffects in controlled trials. Journal of the American Medical Association1995;273:408-12.

Sirtori 1994Sirtori CR, Pasik C. Re-evaluation of a biguanide, metformin: mechanismof action and tolerability. Pharmacology Research 1994;30(3):187-228.

Stacpoole 1998Stacpoole PW. Matformin and lactic acidosis: guilt by association?. DiabetesCare 1998;21(10):1587-8.

Stang 1999Stang MR, Wysowski DK, Butler-Jones D. Incidence of lactic acidosis inmetformin users. Diabetes Care 1999;22:925-7.

Sterne 1959Sterne J. [Treatment of Diabetes mellitus with N.N. Dimethyl GuanilGuanidine (LA 6023, glucophage)]. Therapie 1959;14:625-30.

Stroup 2000Stroup DF,Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie , et al.Meta-analysis of observational studies in epidemiology. A proposal forreporting. Journal of the American Medical Association2000;283(15):2008-12.

Sulkin 1997Sulkin TV, Bosman D, Krentz AJ. Contraindications to metformin therapyin patients with NIDDM. Diabetes Care 1997;20:925-8.

Wilholm 1993Wilholm BE, Myrhed M. Metformin-associated lactic acidosis in Sweden1977-1991. European Journal of Clinical Pharmacology 1993;44:589-91.

* El asterisco señala los documentos más importantes para este estudio


Página 22


TABLAS

Characteristics of included studies

Aarsand 1998Study

TRIAL DESIGN: Retrospective cohort studyDURATION: at least one year, then 12 weeks.

Methods

COUNTRY: NorwaySETTING: Endocrinology center Treatment N: 28, with 14 on folate and 14 on placebo.Metformin + placebo AGE: 57+/-2.8. Metformin + folate AGE: 62+/-2.5. Metformin +placebo SEX: 79% men. Metformin + folate SEX: 71% men. INCLUSION: patients withtype 2 DM, treated with metformin for a minimum of 1 yearEXCLUSIONS: vitamin use that would interfere with the study.

Participants

TREATMENT: metformin, at least 1g/day. One-half of patients on folate 0.25 mg/day+ iron 60mg/day, and one-half on iron 60mg/day.COMPARISON: none.

Interventions

Fasting homocysteine, cysteine, cysteinylglycine, vitamin B12, and folate.Outcomes

Notes

DAllocation concealment

Abbasi 1997Study

TRIAL DESIGN: Randomised controlled trialDURATION: 3 months

Methods

COUNTRY: United StatesSETTING: research laboratory Treatment N: 15Control N: 8.Treatment AGE: 53 +/-3Control AGE: 51 +/-4Treatment SEX: 64% menControl SEX: 87% malesINCLUSION: Type 2 DMEXCLUSIONS: abnormal laboratory values, vascular disease

Participants

TREATMENT: metformin-blind versus open-label metformin, dosage adjused clinically.COMPARISON: placebo

Interventions

Fasting and postprandial glucose, insulin, and free fatty acids.Outcomes

Notes

CAllocation concealment

Abbasi 1998Study

TRIAL DESIGN: Prospective cohort studyDURATION: 6 months

Methods

COUNTRY: United StatesSETTING: outpatient and research centerTreatment N: 11Control N: 0AGE: not listedSEX: not listedINCLUSION: diet-treated type 2 DM EXCLUSIONS: laboratory abnormalities, diabeticvascular complications, or abnormal electrocardiogram

Participants

Página 23




INTERVENTION: metformin 1-2.5 g/day COMPARISON: noneInterventions

Plasma glucose, insulin, and free fatty acids.Outcomes

Notes


Abbink 2001Study

TRIAL DESIGN: Double-blind randomised controlled trial - unpublishedDURATION: 2 months

Methods

COUNTRY: NetherlandsSETTING: outpatientTreatment N: 12Control N: 12AGE: unclearSEX: not listed.INCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 500 mg TID COMPARISON: GlibenclamideInterventions

Glucose, HbA1.Outcomes

Notes

BAllocation concealment

Abbink 2000Study

TRIAL DESIGN: Abstract of a double-blind randomised controlled trialDURATION: 2 months

Methods

COUNTRY: NetherlandsSETTING: outpatientTreatment N: 12Control N: 60AGE: unclearSEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dosage adjusted clinically COMPARISON: glibenclamindeor glimerperide or acarbose

Interventions

Vasodilator responses to diazoxide.Outcomes

Notes


Aguilar 1992aStudy


Methods

Country: Mexico. Setting: diabetes institute. Treatment N: 9. Control N: 0. Age: unclear.Sex: 26% men. Inclusion: type 2 DM with secondary failure to oral agents. Exclusions:insulin dependence.

Participants

TREATMENT: metformin 1200 mg/day, chlorpropamide 375 mg/day, and bedtimeinsulin 0.1 U/kg/day COMPARISON: none

Interventions

Página 24




Fasting glucose, HbA1c, insulin dose, and glucose tolerance.Outcomes

Notes


Allen 1961Study


Methods

COUNTRY: FranceSETTING: outpatientTreatment N: 57Control N: 0AGE: >40SEX: not listedINCLUSION: poorly controlled DM EXCLUSION: none listed

Participants

TREATMENT: metformin, dosage unclear COMPARISON: noneInterventions

GlycemiaOutcomes

Notes


Andras 1962Study

TRIAL DESIGN: Prospective cohort studyDURATION: approximately 1 month

Methods

COUNTRY: unclearSETTING: outpatientTreatment N: 20Control N: 0AGE: not listedSEX: not listedINCLUSION: maturity-onset DM EXCLUSIONS: none listed

Participants

TREATMENT: metformin, dosage unclearCOMPARISON: none

Interventions

GlycemiaOutcomes

Notes


Aviles-Santa 1999Study

TRIAL DESIGN: Randomised controlled trialDURATION: 6 months

Methods

Página 25




COUNTRY: United StatesSETTING: University clinicTreatment N: 21Control N: 22Treatment AGE: 53 +/-4Control AGE: 54 +/-8Treatment SEX: 28% menControl SEX: 45% menINCLUSION: Poorly controlled Type 2 DM on insulinEXCLUSIONS: pregnancy, creatinine > 1.5, hepatic enzymes double normal, medicalconditions that could promote lactic acidosis.

Participants

TREATMENT: Metformin + insulinCOMPARISON: placebo + insulin

Interventions

Weight, HbA1, and lipids.Outcomes

Notes


Azerad 1960Study

TRIAL DESIGN: Prospective cohort studyDURATION: average 24 months

Methods

COUNTRY: FranceSETTING: outpatientTreatment N: 200Control N: 0AGE: not listedSEX: not listedINCLUSION: DMEXCLUSIONS: none listed

Participants

TREATMENT: metformin, with goal of 3 g/day, maximum 5 g/day.COMPARISON: none

Interventions

Glycemia, and glucosuria.Outcomes

Notes


Bacci 1961Study

TRIAL DESIGN: Retrospective cohort studyDURATION: 3-6 months, average 4.5 months

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 42Control N: 0AGE: not listedSEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: metformin, dosage adjusted clinically COMPARISON: noneInterventions

Glycemia and glucosuria.Outcomes

Página 26




Notes


Bayraktar 1996Study

TRIAL DESIGN: Crossover randomised controlled trialDURATION: 2 months

Methods

COUNTRY: TurkeySETTING: University clinic Treatment N: 36Control N: 36AGE: 30-63SEX: 100% menINCLUSION : Type 2 DM with poor controlEXCLUSIONS: microvascular or macrovascular complIcations, liver functionabnormalities.

Participants

TREATMENT: Metformin 500mg TIDCOMPARISON: acarbose

Interventions

Insulin , c-peptide, fibrinogen, lipids, HbA1.Outcomes

Notes


Beisswenger 1999Study

TRIAL DESIGN: Retrospective cohort studyDURATION: 3 months

Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 30Control N: 0AGE: 62+/-8SEX: 56% menINCLUSION: Type 2 DM, some on metformin treatment and some not EXCLUSIONS:renal or hepatic impairment or cardiac disease

Participants

TREATMENT: metformin 500-2500 mg/dayCOMPARISON: none

Interventions

HbA1c, methylglyoxal levels, D-lactate, and glucose.Outcomes

Notes


Belcher 2005Study

TRIAL DESIGN: Double-blind randomised controlled trialDURATION: 12 months

Methods

COUTNRY: United Kingdom SETTING: outpatientTreatment N: 917Control N: 2796Age: 57 +/- 9Sex: 55% menInclusion: type 2 DMExlcusions: ALT levels greater than 2.5 times upper limit of normal

Participants

Página 27




TREATMENT: metformin 2500 mg/dayCOMPARISON: pioglitazone 45 mg/day

Interventions

liver enzyme levelsOutcomes

Notes


Bell 1997Study

TRIAL DESIGN: Prospective cohort studyDURATION: 6 weeks

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 55Control N: 0AGE: 60.2+/-13SEX: 45% menINCLUSION: patients with non-insulin-dependent DM treated with insulinEXCLUSIONS: creatinine > 1.5 mg/dl, or c-peptide < 0.8 ng/ml

Participants

TREATMENT: metformin, 1-3 g/day, some with glyburide or insulin, dosage titratedclinically COMPARISON: none

Interventions

Insulin requirement, HbA1, BMI, and % successfully changed to oral therapy.Outcomes

Notes


Beyer 1975Study


Methods

COUNTRY: GermanySETTING: outpatientTreatment N: 24Control N: 0AGE: not listedSEX: 36% menINCLUSION: adult-onset DM EXCLUSION: none listed

Participants

TREATMENT: metformin, dosage titrated clinicallyCOMPARISON: none

Interventions

Glucose and weight.Outcomes

Notes


Bingle 1964Study

TRIAL DESIGN: Blinded randomised controlled trial (unclear if double-blind)DURATION: 2 months

Methods

Página 28




COUNTRY: United KingdomSETTING: outpatientTreatment N: 22Control N: 22AGE: unclearSEX: not listedINCLUSION: Type 2 DM not controlled on sulfonylureas EXCLUSIONS: none listed

Participants

TREATMENT: Metformin 1-2 g/day + chlorpropamideCOMPARISON: placebo + chlorpropamide

Interventions

Plasma glucose and weight.Outcomes

Notes


Bjorntorp 1978Study

TRIAL DESIGN: Prospective, cross-over comparative trial; not randomisedDURATION: 8 weeks

Methods

COUNTRY: SwedenSETTING: outpatientTreatment N: 21Control N: 21AGE: 58SEX: 52% menINCLUSION: Type 2 DM on long-term biguanide treatment EXCLUSIONS: abnormalrenal function or liver function

Participants

TREATMENT: Metformin, 1.5-3.0 g/dayCOMPARISON: phenformin, 50-100 mg/day (not analysed)

Interventions

Fasting glucose and fasting lactate levels.Outcomes

Notes


Blonde 2002Study


Methods

COUNTRY: United States SETTING: outpatientTreatment N: 476Control N: 164Age: 55.6 +/- 9.4Sex: 57% menInclusion: type DM uncontrolled on sulfnylurea treatmentExclusions: hepatic or renal dysfunction, congestive heart failure

Participants

TREATMENT: metformin 1 g/day, with and without glyburideCOMPARISON: glybruide 20 mg/day

Interventions

HbA1, fasting glucoseOutcomes

Notes


Página 29




Boronat 2000Study

TRIAL DESIGN: Retrospective cohort studyDURATION: average of 12 months

Methods

COUNTRY: SpainSETTING: Endocrine centerTreatment N: 21Control N: 0AGE: unclearSEX: 5% menINCLUSION: obese insulin-treated patients with type 2 DM, also on metforminEXCLUSIONS: none listed

Participants

TREATMENT: Insulin and metformin, dose adjusted clinicallyCOMPARISON: none

Interventions

HbA1c, weight, and insulin requirement.Outcomes

Notes


Botha 1977Study

TRIAL DESIGN: Open-label cross-over trial; not randomisedDURATION: 1 month

Methods

COUNTRY: South AfricaSETTING: general practiceTreatment N: 21Control N: 21AGE: unclearSEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dose adjusted clinicallyCOMPARISON: phenformin, buformin (not analysed), and untreated controls.

Interventions

Heart rate, blood lactate, and lactate/pyruvate ratios, at baseline and with exercise.Outcomes

Notes


Boyd 1992Study

TRIAL DESIGN: Randomised controlled trialDURATION: 6 weeks

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 8Control N: 19Treatment AGE: 64+/-6.2Control AGE: 63.5+/-7.6Treatment SEX: 37% menControl SEX: 68% menINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

Página 30




TREATMENT: Metformin, dosage adjusted clinicallyCOMPARISON: glibenclamide or insulin

Interventions

Insulin sensitivity, HbA1, weight.Outcomes

Notes


Brown 1999Study

TRIAL DESIGN: Retrospective cohort studyDURATION: average 11.6 months

Methods

COUNTRY: United StatesSETTING: patients in an HMO registryTreatment N: 3402Control N: 0AGE: > 30SEX: 53% menINCLUSION: Type 2 DM on metformin treatmentEXCLUSIONS: none listed

Participants

TREATMENT: metformin, 1000-2550 mg/day COMPARISON: noneInterventions

HbA1c, and fructosamine.Outcomes

Notes


Cairns 1977Study

TRIAL DESIGN: Open-label randomised controlled trialDURATION: 4 weeks

Methods

COUNTRY: United Kingdom SETTING: outpatientTreatment N: 39Control N: 67AGE: 57SEX: 21% menINCLUSION: Type 2 DMEXCLUSIONS: renal failure, congestive heart failure

Participants

TREATMENT: Metformin 850 mg BIDCOMPARISON: phenformin (not analysed)

Interventions

Fasting glucose, body weight, and lipidsOutcomes

Notes


Calle-Pascual 1995Study

TRIAL DESIGN: Open-label comparative trial; not randomisedDURATION: 4 months

Methods

Página 31




COUNTRY: SpainSETTING: outpatientTreatment N: 12Control N: 24AGE: unclearSEX: 50% menINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 850 mg TIDCOMPARISON: insulin or acarbose

Interventions

Lipids, blood pressure, HbA1, body weight, insulin sensitivity.Outcomes

Notes


Campbell 1988Study

TRIAL DESIGN: Open-label randomised controlled trialDURATION: 12 months

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 38Control N: 24AGE: 54+/-6.1SEX: 64% menINCLUSION: Type 2 DM, diet failedEXCLUSIONS: congestive heart failure, renal failure, liver function abnormalities

Participants

TREATMENT: Metformin, dosage adjusted clinically. COMPARISON: glipizideInterventions

Blood pressure, heart rate, microalbuminuria.Outcomes

Notes


Campbell 1994Study


Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 24Control N: 24Treatment AGE: 57+/-10Control AGE: 57+/-9Treatment SEX: 33% menControl SEX: 33% menINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, 500 mg BID to 3,000 mg/day maximum.COMPARISON: glipizide, 5 mg/day to 39 mg/day maximum BID

Interventions

Glucose, HbA1, lipids, lactate levelsOutcomes

Página 32




Notes


Canivet 1962Study

TRIAL DESIGN: Retrospective cohort studyDURATION: average 66 months

Methods

COUNTRY: FranceSETTING: outpatientTreatment N: 180Control N: 0AGE: not listedSEX: not listedINCLUSION: DM, 180 treated with metforminEXCLUSIONS: none listed

Participants

TREATMENT: metformin, dose unclearCOMPARISON: some patients treated with other agents, not analysed

Interventions

Plasma glucoseOutcomes

Notes


Carpentier 1975Study


Methods

COUNTRY: BelgiumSETTING: outpatientTreatment N: 11Control N: 0AGE: 58.8SEX: 45% menINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: metformin 1.5 g/day + arginine infusion 11.7 mg/kg/minCOMPARISON: none

Interventions

Blood glucose, free fatty acids, and glycagon.Outcomes

Notes


Carter 2005Study

TRIAL DESIGN: Double-blind randomised controlled trial DURATION: 6 monthsMethods

COUNTRY: United Kingdom SETTING: outpatientTreatment N: 26Control N: 16Age: not statedSex: not statedInclusion: poorly controlled overweight patients with type 2 DMExlcusions: not stated

Participants

TREATMENT: metformin 1.5 to 3 g/day COMPARISON: placeboInterventions

Página 33




C-reactive proteitn, complement factor C3Outcomes

Notes


Cavallo-Perin 1989Study

TRIAL DESIGN: Double-blind crossover randomised controlled trialDURATION: 6 months

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 10Control N: 10AGE: 51+/-2.1SEX: 60% menINCLUSION: Type 2 DMEXCLUSIONS: liver or kidney disease, heart failure, other drugs, or chronic infection

Participants

TREATMENT: Metformin 850 mg BIDCONTROL: phenformin 50 mg BID (not analysed)

Interventions

Weight, glucose, HbA1, and blood lactate levels at different times of day.Outcomes

Notes

AAllocation concealment

Cefalu 2002Study

TRIAL DESIGN: Open-label randomised controlled trial DURATION: 4.5 monthsMethods

COUNTRY: United States SETTING: outpatientTreatment N: 91Control N: 91Age: 35-70Sex: not statedInclusion: type 2 DMExclusion: not stated

Participants

TREATMENT: metformin 850 mg TID with and without glipizideCONTROL: glipizide 20 mg/day

Interventions

Glycemic control, body weight, abdominal fat distribution, PAI-1 levelsOutcomes

Notes


Chakrabarti 1965Study

TRIAL DESIGN: Single-blind crossover comparative trial; not randomisedDURATION: 2 months placebo, 4 months treatment

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 27Control N: 27AGE: 56.3SEX: 95% menINCLUSION: Type 2 DM with coronary artey disease, claudicationEXCLUSIONS: none listed

Participants

Página 34




TREATMENT: metformin 500 mg TIDCOMPARISON: placebo

Interventions

Cholesterol, plasma fibrinogen.Outcomes

Notes


Chan 1993Study

TRIAL DESIGN: Crossover randomised controlled trialDURATION: 4 weeks

Methods

COUNTRY: Hong Kong and United KingdomSETTING: outpatientTreatment N: 24Control N: 24AGE: 48.5+/-2.4SEX: 50% menINCLUSION: Type 2 DMEXCLUSIONS: renal insufficiency, hypertension

Participants

TREATMENT: Metformin, dosage adjusted clinically COMPARISON: glybenclanideInterventions

Weight, body mass index (BMI), lipids, blood pressure, systemic vascular resistanceindex.

Outcomes

Notes


Chiasson 1994Study

TRIAL DESIGN: Randomised controlled trial of acarbose versus placebo. Metforminin non-randomised treatment strata.DURATION: 1 year

Methods

COUNTRY: CanadaSETTING: multicenterTreatment N: 83Control N: 271Treatment AGE: 57.4+/-1.1Control AGE: 57+/-1.1Treatment SEX: 51% menControl SEX: 58% menINCLUSION: Type 2 DMEXCLUSIONS: gastrointestinal disease, various medications

Participants

TREATMENT: Main: acarbose versus placebo Treatment strata: metformin (dosageadjusted clinically), diet, sulfonylurea, insulin

Interventions

Postprandaial glucose, HbA1, lipds, c-peptide levels.Outcomes

Notes


Chiasson 2001Study

TRIAL DESIGN: Double-blind randomised controlled trialDURATION: 36 weeks

Methods

Página 35




COUNTRY: CanadaSETTING: multicenterTreatment N: 156Control N: 162Treatment AGE: 57.9+/-8.6Control AGE: 57.3+/-9Treatment SEX: 77% menControl SEX: 74% menINCLUSION: Type 2 DMEXCLUSIONS: cardiovascular events, gastrointestinal disease, history of lactic acidosis,major debilitating disease

Participants

TREATMENT: Metformin, dosage adjusted clinically, or metformin + miglitol.COMPARISON: miglitol or placebo

Interventions

Fasting and postprandial glucose, HbA1, insulin, weight.Outcomes

Notes


Cho 1992Study

TRIAL DESIGN: Open-label comparative trial; not randomisedDURATION: 36 days

Methods

COUNTRY: KoreaSETTING: University centerTreatment N: 22Control N: 27AGE: unclearSEX: 47% menINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 0.5-1.5 g/dayCOMPARISON: insulin or sulfonylurea

Interventions

Plasma t-PA and PAI-1 antigenOutcomes

Notes


Clarke 1965Study

TRIAL DESIGN: Prospective cohort studyDURATION: Average 21 months

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 108Control N: 0AGE: > 30 to < 60SEX: 38% menINCLUSION: DM, treatment failures with sulfonyureasEXCLUSIONS: ketonuria, bicarbonate < 17 mEq/L, or serious organic disease

Participants

TREATMENT: metformin, 1 g/dayCOMPARISON: none

Interventions

Página 36




Glycemica, glycosuria, and weight.Outcomes

Notes


Clarke 1968Study

TRIAL DESIGN: Crossover randomised controlled trialDURATION: 1 year

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 139Control N: 139Treatment AGE: 59Control AGE: 57SEX: not listedINCLUSION: Obese patients with Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 1-3 g/dayCOMPARISON: chlorpropamide

Interventions

Weight, blood glucose.Outcomes

Notes


Clarke 1977Study

TRIAL DESIGN: randomised controlled trialDURATION: 1 year

Methods

COUNTRY: United Kingdom SETTING: outpatientTreatment N:131Control N: 146Treatment AGE: 60Control AGE: 60Treatment SEX: 31% menControl SEX: 31% menINCLUSION: Newly diagnosed Type 2 DMEXCLUSIONS: malignancy, congestive heart failure, obesity, other hypoglycemicmedications.

Participants

TREATMENT: Metformin, 1-3 g/dayCOMPARISON: chlorpropamide

Interventions

Blood glucose, weight.Outcomes

Notes


Collier 1989Study


Methods

Página 37




COUNTRY: ScotlandSETTING: outpatientTreatment N: 12Control N: 12Treatment AGE: 53.3Control AGE: 55.5SEX: 50% menINCLUSION: Type 2 DMEXCLUSIONS: abnormal renal function, smokers, aspirin.

Participants

TREATMENT: Metformin, dosage adjusted clinicallyCOMPARISON: gliclazide

Interventions

Platelet density profiles and aggregability studies.Outcomes

Notes


Cryer 2005Study

TRIAL DESIGN: Open-label randomised controlled trial DURATION: 12 monthsMethods

COUNTRY: United StatesSETTING: outpatientTreatment N: 7227Control N: 1505Age: 58.5 +/- 13Sex: 37% menInclusion: type 2 DM suboptimally controlled on diet or sulfonylureaExclusions: standard

Participants

TREATMENT: metformin 2.6 g/dayCONTROL: usual care

Interventions

Serious adverse effects such as lactic acidosisOutcomes

Notes


Cusi 1996Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 10Control N: 10.Treatment AGE: 51+/-3Control AGE: 54+/-3Treatment SEX: 40% menControl SEX: 60% menINCLUSION: Type 2 DM, with body weight stableEXCLUSION: sedentary or strenuous activities, renal disease, hepatic disease or othersignificant organ system disease

Participants

TREATMENT: Metformin 500 mg BID to 2500 mg/day maximum + glibenclamide, doseon clinical groundsCOMPARISON: glibenclamide + placebo

Interventions

Página 38




Glucose, HbA1, lipids, blood lactate.Outcomes

Notes


D'Argenzio 1996Study

TRIAL DESIGN: Open-label, nonrandomised comparative trialDURATION: 6 months

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 23Control N: 57AGE: 56SEX: 40% menINCLUSION: Poorly controlled Type 2 DMEXCLUSIONS: cardiac, liver or renal disease, contraindication to oral hypoglycemicmedications

Participants

TREATMENT: Metformin, dosage adjusted clinically + glibenclamideCOMPARISON: glibenclamide or diet

Interventions

Basal glucose, HbA1, renal and liver functions, lipids.Outcomes

Notes


Damsbo 1998Study


Methods

COUNTRY: SwedenSETTING: outpatientTreatment N: 9Control N: 9Treatment AGE: 51Control AGE: 53Treatment SEX: 78% menControl SEX: 66% menINCLUSION: Obese patients with Type 2 DMEXCLUSIONS: abnormal renal, liver fucntion, or cardiac function

Participants

TREATMENT: Metformin 1-3 g/dayCOMPARISON: placebo

Interventions

Insulin sensitivity, plasma glucose, insulin, c-peptide, free fatty acids, lactate levels.Outcomes

Notes


Davidson 2000Study

TRIAL DESIGN: Abstract; randomised controlled trial, placebo-controlled; unclear ifsingle-blind

Methods

Página 39




COUNTRY: United StatesSETTING: outpatientTreatment N: 484Control N: 161AGE: not listedSEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dosage adjusted clinically, versus metformin + glyburideCOMPARISON: glyburide or placebo

Interventions

HbA1Outcomes

Notes


De Silva 1979Study


Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 21Control N: 20AGE: 55SEX: 32% menINCLUSION: Type 2 DMEXCLUSIONS: renal or liver abnormalities

Participants

TREATMENT: Metformin 1.5 g/day + placeboCOMPARISON: clofibrate + placebo

Interventions

Fasting glucose, urinary glucose, lipids and fibrinogen.Outcomes

Notes


DeFronzo 1991Study

TRIAL DESIGN: Open-label cross-over trialDURATION: 3 months

Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 14Control N: 14AGE: 60+/-3SEX: 71% menINCLUSION: Obese and lean type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 1-2.5 g/dayCOMPARISON: no metformin

Interventions

Insulin sensitivity, glucose tolerance test, continuous indirect calorimetry, and lipids.Outcomes

Página 40




Notes


DeFronzo 1995Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 566Control N: 355Treatment AGE: 53+/-1Control AGE: 55+/-1Treatment SEX: 43% menControl SEX: 49% menINCLUSION: Obese patients with Type 2 DMEXCLUSIONS: creatinine > 1.4, abnormal liver functions, cardiovascular disease

Participants

TREATMENT: Metformin 850 mg TID or metformin + glyburideCOMPARISON: glyburide or placebo

Interventions

HbA1c, fasting and postprandial glucoseOutcomes

Notes


Derosa 2003Study

TRIAL DESIGN: Open-label randomised trialDURATION: 12 months

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 56Control N: 56Age: 54 +/- 9Sex: 50% menInclusion: type 2 DMExclusion: hypertension, heart disease, abnormal renal function, or drugs that interactwith treatment

Participants

TREATMENT: metformin 2.5 g/dayCONTROL: repaglinide 4 mg/day

Interventions

Fasting plasma insulin, postprandial plasma insulin, lipid profile, homocysteineOutcomes

Notes


Dies 1978Study

TRIAL DESIGN: Prospective cohort studyDURATION: at least 5 years

Methods

Página 41




COUNTRY: MexicoSETTING: outpatientTreatment N: 25Control N: 0AGE: 56SEX: 35% menINCLUSION: adult-onset DMEXCLUSIONS: none listed

Participants

TREATMENT: metformin 560 mg/day + chlorpropamide 175 mg/dayCOMPARISON: none

Interventions

Fasting and postprandial glucose, glycosuria, and weight.Outcomes

Notes


Donnelly 1960Study

TRIAL DESIGN: Prospective cohort studyDURATIONS: average 6 months

Methods

COUNTRY: IrelandSETTING: outpatientTreatment N: 25Control N: 0AGE: 21-77Sex: 22% menINCLUSION: type 2 DMEXCLUSION: ketonuria or infection

Participants

TREATMENT: metformin, dosage adjusted clinicallyCOMPARISON: none

Interventions

GlycosuriaOutcomes

Notes


Dornan 1991Study


Methods

COUNTRY: United KingdomSETTING: university clinicTreatment N: 30Control N: 30Treatment AGE: 55+/-1Control AGE: 55+/-1Treatment SEX: 53% menControl SEX: 30% menINCLUSION: Diet-treated Type 2 DMEXCLUSIONS: ketonuria, renal or liver dysfunction, congestive heart failure

Participants

TREATMENT: Metformin 500 mg QD-TID COMPARISON: placeboInterventions

Glucose, BMI, c-peptide, blood pressure, lipids.Outcomes

Página 42




Notes


Elkeles 1991Study

TRIAL DESIGN: 1) Open-label cross-over randomised controlled trial2) Single-blind crossover trialDURATION: 3 months, then 6 weeks

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 63Control N: 49AGE: < 70 yearsSEX: 64% menINCLUSION: Type 2 DMEXCLUSIONS: renal or liver disease

Participants

TREATMENT: 1) Metformin, dosage adjusted clinically 2) MetforminCOMPARISONS: 1) glibenclamide2) placebo

Interventions

Serum lipids, lipoproteins, glucose, HbA1.Outcomes

Notes


Erle 1999Study

TRIAL DESIGN: Double-blind crossover randomised controlled trialDURATION:

Methods

COUNTRY:SETTING:Treatment N:Control N:AGE:SEX:INCLUSION: Type 2 DMEXCLUSIONS:

Participants

TREATMENT: Metformin, dosage adjusted clinically, + glyburideCOMPARISON: placebo + glyburide

Interventions

Glycemic controlOutcomes

Notes


Fanghanel 1996Study


Methods

Página 43




COUNTRY: MexicoSETTING: outpatientTreatment N: 30Control N: 30Treatment AGE: 52.1+/- 8.8Control AGE: 51.2+/-8.5Treatment SEX: 40% menControl SEX: 30% menINCLUSION: Type 2 DM, obeseEXCLUSIONS: abnormal liver functions, cardiomyopathy, lung disease

Participants

TREATMENT: Metformin 850 mg BID-TIDCOMPARISON: insulin BID

Interventions

Lipids, HbA1, blood pressure, BMI.Outcomes

Notes


Fanghanel 1998Study


Methods

COUNTRY: MexicoSETTING: outpatientTreatment N: 30Control N: 30AGE: 49+/-9.6SEX: 38% menINCLUSION: Type 2 DM with sulfonylurea failureEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 0.85-2.5 g/dayCOMPARISON: insulin

Interventions

Plasma glucose, fibrinogen, body mass index.Outcomes

Notes


Ferner 1988Study

TRIAL DESIGN: Open-label nonrandomised comparative trialDURATION: 3 months

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 6Control N: 12Treatment AGE: 56Control AGE: 56Treatment SEX: 67% menControl SEX: 50% menINCLUSION: Type 2 DMEXCLUSIONS: other medication, ketosis, ketonuria

Participants

TREATMENT: Metformin, dose adjusted clilnicallyCOMPARISON: tolbutamide or diet

Interventions

Página 44




Insulin sensitivity under euglycemic insulin clampOutcomes

Notes


Fisman 2001Study

TRIAL DESIGN: Open-label nonrandomised comparative trialDURATION: 7.7 years

Methods

COUNTRY: IsraelSETTING: research instituteTreatment N: 332Control N: 1943Treatment AGE: 60.1+/-6.5Control AGE: 59.9+/-6.6Treatment SEX: 66% menControl SEX: 76% menINCLUSION: Type 2 DM with coronary artery diseaseEXCLUSIONS: pacemaker, cerebrovascular disease, malignant disease, estrogenreplacement, and insulin treatment

Participants

TREATMENT: Metformin or metformin + sulfonylurea, dose adjusted clinicallyCOMPARISON: sulfonylurea or diet

Interventions

Crude mortality rate, time-related mortality, and cause of deathOutcomes

Notes


Fonseca 2000Study

TRIAL DESIGN: Prospective cohort study of metformin in a randomised controlled trialof rosiglitazoneDURATION: 6.5 months

Methods

COUNTRY: United StatesSETTING: mulitcenter outpatientTreatment N: 348Control N: 0AGE: 58+/-9SEX: 68% menINCLUSION: type 2 DMEXCLUSIONS: renal or hepatic disease, angina, congestive heart failure, abnormallaboratory result, or chronic use of insulin

Participants

TREATMENT: metformin 2.5 g/day + placebo, metformin + rosiglitazone 4 mg/day, ormetformin + rosiglitazone 8 mg/day.

Interventions

HbA1c, fasting glucose, insulin sensitivity, weight, and lipids.Outcomes

Notes


Fritsche 2000Study

TRIAL DESIGN: Double-blind cross-over randomised controlled trialDURATION: 10 weeks

Methods

Página 45




COUNTRY: GermanySETTING: outpatientTreatment N: 26Control N: 26AGE: 51+/-9SEX: not listedINCLUSION: Severely obese type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dosage adjusted clinically, + insulinCOMPARISON: placebo + insulin

Interventions

Glucose, insulin, c-peptide, HbA1c, lipids, weight, venous lactic acid.Outcomes

Notes


Fujioka 2005Study

TRIAL DESIGN:2 double-blind randomised controlled trialsDURATION: 3 months and 4 months

Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 663Control N: 202Age: 56 +/- 10Sex: 50% menInclusion: type 2 DM inadequately controlled on diet and exerciseExclusions: standard

Participants

TREATMENT: metformin XR 500 -2000 mg/dayCOMPARISON: placebo

Interventions

HbA1c, fasting glucose and insulin, lipid profilesOutcomes

Notes


Galeone 1998Study


Methods

COUNTRY: ItalySETTING: diabetes referral centerTreatment N: 57Control N: 0AGE: 61+/-3.4SEX: 54% menINCLUSION: type 2 DM for at least 5 yearsEXCLUSIONS: hepatic or liver abnormalities, neurological, psychological or cardiacdisease

Participants

TREATMENT: metformin 1500 mg/day and glicazide 120mg/dayCOMPARISON: none

Interventions

HbA1c, 24-hour glycosuria, and fasting and postprandial glucose.Outcomes

Página 46




Notes


Garber 1997Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 222Control N: 229Treatment AGE: 57+/-10Control AGE: 55+/-11Treatment SEX: 62% menControl SEX: 56% menINCLUSION: Type 2 DM, not controlledEXCLUSIONS: significant disease or contraindication likely to affect diabetes

Participants

TREATMENT: Metformin, dosage adjusted clinicallyCOMPARISON: placebo

Interventions

Fasting glucose and HbA1.Outcomes

Notes


Garber 2002Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 317Control N; 321Age: 56 +/- 10Sex: 53% menInclusion: type 2 DM that failed diet and exerciseExclusions: polyurea, weight loss, acidosis, insulin treatment

Participants

TREATMENT: metformin 500 mg BID with and without glyburideCONTROL: gluburide 2.5 mg BID or placebo

Interventions

HbA1, fasting and postprandial glucoseOutcomes

Notes


Garcia 1971Study

TRIAL DESIGN: Prospective cohort studyDURATION: 2 years

Methods

Página 47




COUNTRY: MexicoSETTING: outpatientTreatment N: 23Control N: 0AGE: 53.6SEX: 26% menINCLUSION: DM, treated with sulfonylureasEXCLUSIONS: none listed

Participants

TREATMENT: metformin + chlorpropamide in combination, dosage titrated clinicallyCOMPARISON: none

Interventions

Fasting and postprandial glucose, glucosuria.Outcomes

Notes


Giugliano 1993Study


Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 27Control N:AGE: not listedSex: 23% menINCLUSION: Obese patients with Type 2 DMEXCLUSIONS: intercurrent illness, age > 70, creatinine > 1.2 mg/dl, ischemic or wastingdisease

Participants

TREATMENT: Metformin, dosage adjusted clinicallyCOMPARISON: placebo

Interventions

HbA1, lipids, c-peptide, blood pressure, and BMI.Outcomes

Notes


Goldstein 2003Study

TRIAL DESIGN: Double-blind randomised controlled trialDURATION: 4.5 months

Methods

COUNTRY: United StatesSETTING: outpatient Treatment N: 163Control N: 84Inclusions: type 2 DM inadequately controlled on sulfonylureaExclusions: renal and hepatic dysfunction, cardiovascular diseaese, acidosis or long-terminsulin treatment

Participants

TREATMENT: metformin 2 g/day with or without glipizideCONTROL: glipizide 30 mg/day

Interventions

BMI, HbA1, fasting glucoseOutcomes

Notes


Página 48




Gonzalez-Ortiz 2004Study


Methods

COUNTRY: MexicoSETTING: outpatientTreatment N: 67Control N: 37Age: 53 +/- 7Sex: 52% menInclusion: type 2 DM with secondary failure to monotherapy with glibenclamideExlcusions: cardiovascular, renal or hepatic dysfunction, insulin treatment, pregnancy

Participants

TREATMENT: metformin 1-2 gm/day with or without glimepiride 2-4 mg/dayCOMPARISON: Glimipiride 2-4 mg/day

Interventions

HbA1c, adverse eventsOutcomes

Notes


Gottlieb 1962Study


Methods

COUNTRY: United Kingdom SETTING: inpatient then outpatientTreatment N: 39Control N: 0AGE: 21 - >80SEX: 58% menINCLUSION: patients with DM, poorly controlled on previous regimenEXCLUSIONS: none listed

Participants

TREATMENT: metformin 1-3 g/dayCOMPARISON: none

Interventions

Weight, and glycemiaOutcomes

Notes


Grant 1991Study


Methods

COUNTRY: United KingdomSETTING:Treatment N: 21Control N: 17Treatment AGE: 59.5+/-9Control AGE: 63.2+/-9.6SEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, low and high doseCOMPARISON: placebo

Interventions

Página 49




Plasminogen activator inhibitor, BMI, glucose, HbA1, insulin, lipids.Outcomes

Notes


Grant 1996Study


Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 52Control N: 23AGE: not listedSEX: not listedINCLUSION: Obese patients with Type 2 DMEXCLUSIONS: insulin therapy, BMI < 25, fasting glucose < 6 mmol/L

Participants

TREATMENT: Metformin 3 g/dayCOMPARISON: placebo

Interventions

Lipids, HbA1, insulin, BMI, plasminogen activator inhibitor.Outcomes

Notes


Grant 1998Study

TRIAL DESIGN: randomised controlled trialDURATION: 6 months

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 27Control N: 17AGE: not listedSEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 1.5 g/day or metformin 3 g/dayCOMPARISON: placebo

Interventions

Plasma insulin, glucose, lipids, and factor VII levels.Outcomes

Notes


Gregorio 1989Study

TRIAL DESIGN: Double-blind crossover randomised controlled trialDURATION: 5 weeks

Methods

Página 50




COUNTRY: ItalySETTING: outpatientTreatment N: 53Control N: 53AGE: not listedSEX: not listedINCLUSION: Type 2 DM, poor controlEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dosage adjusted clinically, + sulfonylureaCOMPARISON: placebo + sulfonylurea

Interventions

Weight, lipids, insulin, HbA1, and lactate levels.Outcomes

Notes


Gregorio 1990Study

TRIAL DESIGN: Single-blind comparative trial. Patients were their own controlsDURATION: 5 weeks

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 20Control N: 10AGE: 50-63Treatment SEX: 45% menControl SEX: 40% menINCLUSION: Type 2 DM with poor controlEXCLUSIONS: heptic, renal or vascular disease

Participants

TREATMENT: Metformin, dosage adjusted clinically, + sulfonylureaCOMPARISON: placebo + sulfonylurea

Interventions

Glucose, insulin, c-peptide, fructosamine, lipids, lactate, pyruvate, alanine, and glycerol.Outcomes

Notes


Gregorio 1997Study


Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 68Control N: 0AGE: 67+/-1.2SEX: 43% menINCLUSION: type 2 DMEXCLUSIONS: liver or renal abnormality, respiratory insufficiency or congestive heartfailure

Participants

TREATMENT: metformin 2350 mg/dayCOMPARISON: none

Interventions

Lactate, free fatty acids, lipids, insulin, c-peptide, plasma metformin, and glucose.Outcomes

Página 51




Notes


Groop 1989Study


Methods

COUNTRY: FinlandSETTING: outpatientTreatment N: 12Control N: 12AGE: not listedSEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: cardiac, renal, hepatic or endocrine disease, intercurrent illness

Participants

TREATMENT: Metformin 500 mg TID + glibenclamideCOMPARISON: insulin

Interventions

Glucose, lipids, weight, BMI, basal hepatic glucose productionOutcomes

Notes


Groop 1991Study


Methods

COUNTRY: FinlandSETTING: outpatientTreatment N: 12Control N: 24Treatment AGE: 60+/-2Control AGE: 59+/-2Treatment SEX: 50% menControl SEX: 50% menINCLUSION: Type 2 DM with sulfonylurea failureEXCLUSIONS: intercurrent illness, hepatic, renal or cardiac disease

Participants

TREATMENT: Metformin 1.5 g/day + glibenclamide.COMPARISON: insulin

Interventions

Blood glucose, HbA1, lipids, energy expenditure, glucose and fat oxidation.Outcomes

Notes


Guillausseau 1997Study

TRIAL DESIGN: Open-label, nonrandomised comparative trialDURATION: at least 3 months

Methods

Página 52




COUNTRY: FranceSETTING: outpatientTreatment N: 26Control N: 36Treatment AGE: 60+/-10Control AGE: 60+/-12.9Treatment SEX: 73% menControl SEX: 63% menINCLUSION: Type 2 DM on sulfonylureaEXCLUSIONS: none listed

Participants

TREATMENT: Metfomin, dosage adjusted clinically + gliclazideCOMPARISON: gliclazide

Interventions

Fasting and postprandial glucose, and HbA1.Outcomes

Notes


Gursoy 2000Study

TRIAL DESIGN: Abstract of a prospective cohort studyDURATION: 3 months

Methods

COUNTRY: TurkeySETTING: outpatientTreatment N: 20Control N: 0AGE: 49+/-8SEX: 80% menINCLUSION: obese and nonobese patients with type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 2.5 g/dayCOMPARISON: none

Interventions

Insulin sensitivity, lipid profiles, lactate, and BMI.Outcomes

Notes


Haupt 1991Study


Methods

COUNTRY: GermanySETTING: multicenter outpatientTreatment N: 1823Control N: 0AGE: 64.8SEX: 39% menINCLUSION: type 2 DM, poorly controlledEXCLUSIONS: nephropathy, previous treatment with metformin, and insulin-dependence

Participants

TREATMENT: metformin 850-2550 mg/day + sulfonylurea, dosage titrated clinicallyCOMPARISON: none

Interventions

Postprandial glucose, HbA1, weight, blood pressure, and lipids.Outcomes

Página 53




Notes


Herman 1961Study

TRIAL DESIGN: Prospective cohort studyDURATION: approximately 1 month

Methods

COUNTRY: South AfricaSETTING: outpatientTreatment N: 47Control N: 0AGE: not listedSEX: not listedINCLUSION: maturity-onset DM, juvenile-onset patients were studied but not analysedEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 1.5-3 g/dayCOMPARISON: none

Interventions

Fasting glucose and glucose tolerance.Outcomes

Notes


Hermann 1991aStudy


Methods

COUNTRY: SwedenSETTING: outpatientTreatment N: 122Control N: 45AGE: 60SEX: 64% menINCLUSION: Type 2 DMEXCLUSIONS: cardiac, renal or hepatic disease, alcohol abuse, severe chronic disease

Participants

TREATMENT: Metformin 1 g BID or metformin + glibenclamideCOMPARISON: glibenclamide

Interventions

Fasting glucose, HbA1, weight.Outcomes

Notes


Hermann 1991bStudy

TRIAL DESIGN: Open-label crossover randomised controlled trialDURATION: 1 year

Methods

Página 54




COUNTRY: SwedenSETTING: outpatientTreatment N: 22Control N: 22AGE: 59SEX: 72% menINCLUSION: Type 2 DMEXCLUSIONS: renal or liver dysfunction

Participants

TREATMENT: Metformin 0.5-3 g/dayCOMPARISON: glibenclamide

Interventions

Fasting glucose, lipds, c-peptide, HbA1.Outcomes

Notes


Hermann 1994aStudy

TRIAL DESIGN: randomised controlled trialDURATION: 3 months.

Methods

COUNTRY: SwedenSETTING: regional health centersTreatment N: 110Control N: 34AGE: 34-74SEX: 64% menINCLUSION: Type 2 DMEXCLUSIONS: contraindications to the medications, or insulin requirements

Participants

TREATMENT: Metformin or metformin, dosage adjusted clinically, + glibenclamideCOMPARISON: glibenclamide

Interventions

Fasting glucose, body weight, and c-peptide levels.Outcomes

Notes


Hermann 1994bStudy


Methods

COUNTRY: SwedenSETTING: regional health centersTreatment N: 108Control N: 36AGE: 60SEX: 63% menINCLUSION: Type 2 DMEXCLUSIONS: insulin treatment, contraindications to the medications

Participants

TREATMENT: Metformin, dosage adjusted clinicallyCOMPARISON: glibenclamide

Interventions

Fasting glucose, c-peptide levels, HbA1, blood pressure.Outcomes

Página 55




Notes


Higginbotham 1979Study

TRIAL DESIGN: Double-blind cross-over randomised controlled trialDURATION: 2 months

Methods

COUNTRY: AustraliaSETTING: outpatientTreatment N: 17Control N: 17AGE: 31-79SEX: 29% menINCLUSION: Type 2 DMEXCLUSIONS: renal or liver insufficiency, retinopathy

Participants

TREATMENT: Metformin, dosage unclearCOMPARISON: glibenclamide

Interventions

Fasting and postprandial glucose, weight, insulin and lactate levels.Outcomes

Notes


Hirsch 1999Study


Methods

COUNTRY: United StatesSETTING: research centerTreatment N: 25Control N: 25AGE: not listedSEX: not listedINCLUSION: Type 2 DM with poor control on insulinEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 2.5 g/dayCOMPARISON: placebo

Interventions

Weight, HbA1, insulin, c-peptide, or insulin dose.Outcomes

Notes


Hoffmann 1997Study

TRIAL DESIGN: randomised controlled trial. Single blind with respect to metformintreatmentDURATION: 6 months

Methods

Página 56




COUNTRY:SETTING: multicenterTreatment N: 31Control N: 63Treatment AGE: 55.9Control AGE: 59.2Treatment SEX: 45% menControl SEX: 28.5% menINCLUSION: Type 2 DM, previously on dietEXCLUSIONS: renal, liver or cardiovascular disease, malignancy, pregnancy, infection

Participants

TREATMENT: Metformin 850 mg BIDCOMPARISON: acarbose or placebo.

Interventions

Fasting and postprandial glucose, insulin, lipids, HbA1.Outcomes

Notes


Hollenbeck 1991Study


Methods

COUNTRY: United StatesSETTING: Veteran's Administration outpatientTreatment N: 9Control N: 0AGE: 63+/-3SEX: 89% menINCLUSION: Type 2 DM, with elevated triglyceridesEXCLUSIONS: significant diseases or medication that could interfere with carbohydratemetabolism

Participants

TREATMENT: Metformin 2.5 g/dayCOMPARISON: none

Interventions

HbA1c, plasma insulin, free fatty acids, triglyceride, and lipids.Outcomes

Notes


Holman 1987Study

TRIAL DESIGN: crossover randomised controlled trialDURATION: 2 months

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 18Control N: 45AGE: 57+/-11SEX: 33% menINCLUSION: Type 2 DMEXCLUSIONS: cardiovascular disease

Participants

TREATMENT: Metformin, dosage adjusted clinically, or metformin + sulfonylureaCOMPARISON: sulfonylurea or sulfonylurea + insulin versus insulin

Interventions

Página 57




Fasting glucose, c-peptide, HbA1.Outcomes

Notes


Horton 2000Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 350Control N: 351AGE: 56-59Treatment SEX: 58% menControl SEX: 61% menINCLUSION: Type 2 DMEXCLUSIONS: renal impairment, significant diabetic complications

Participants

TREATMENT: Metformin 500 mg TID or metformin + nateglinindeCOMPARISON: nateglinide or placebo

Interventions

Fasting glucose, HbA1.Outcomes

Notes


Horton 2004Study

TRIAL DESIGN: Double-blind randomised controlled trial DURATION: 6 monthsMethods

COUNTRY: United States SETTING: outpatientTreatment N: 193Control N: 297Age: 57 +/- 1.1Sex: 60% menInclusion: type 2 DM, treatment naiveExclusions: renal dysfunction, diabetic complications

Participants

TREATMENT: metformin 500 mg TID with and without nateglinide COMPARISON:nateglinide 120 mg before meals

Interventions

HbA1, fasting and postprandial glucose, post-load insulinOutcomes

Notes


Hother-Nielsen 1989Study


Methods

Página 58




COUNTRY: DenmarkSETTTING: outpatientTreatment N: 9Control N: 9AGE: not listedSEX: not listedINCLUSION: Obese patients with Type 2 DMEXCLUSIONS: renal or liver dysfunction

Participants

TREATMENT: Metformin 500 mg TIDCOMPARISON: placebo

Interventions

Insulin requirements, glucose, insulin, lactate levels.Outcomes

Notes


Imano 1998Study


Methods

COUNTRY: JapanSETTING: outpatientTreatment N: 13Control N: 17Treatment AGE: 66+/-8Control AGE: 62+/-13Treatment SEX: 23% menControl SEX: 29% menINCLUSION: Type 2 DM with microalbuminuriaEXCLUSIONS: abnormal liver function

Participants

TREATMENT: Metformin 500 mg TIDCOMPARISON: troglitazone

Interventions

Lipids, blood pressure, BMI, fasting and postprandial glucose, albumin-to-creatinineratio.

Outcomes

Notes


Inzucchi 1998Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 29Control N: 24Treatment AGE: 51+/-13Control AGE: 56+/-12Control SEX: 43% menTreatment SEX: 47% menINCLUSION: Type 2 DMEXCLUSIONS: abnormal renal or hepatic function, recent atherosclerotic event

Participants

Página 59




TREATMENT: Metformin 1g BIDCOMPARISON: troglitazone

Interventions

Postprandial glucose, HbA1, glucose tolerance, insulin, c-peptide.Outcomes

Notes


Jackson 1962Study

TRIAL DESIGN: Retrospective cohort studyDURATION: approximately 1 month

Methods

COUNTRY: South AfricaSETTING: outpatientTreatment N: 26Control N: 0AGE: not listedSEX: not listedINCLUSION: mild, not-ketosis-prone DMEXCLUSIONS: ketosis

Participants

TREATMENT: Metformin 1-3 g/dayCOMPARISON: none.

Interventions

Glycemia, and dose of sulfonylurea.Outcomes

Notes


Jackson 1987Study

TRIAL DESIGN: Single-blind cross-over trialDURATION: 4.9 months average

Methods

COUNTRY: United KingdomSETTING: general practiceTreatment N: 10Control N: 10AGE: 56.6+/-1.9SEX: 100% menINCLUSION: Type 2 DM, nonobeseEXCLUSIONS: excessive physical activity or a metabolic disorder

Participants

TREATMENT: Metformin, dose adjusted clinically COMPARISON: placeboInterventions

Plasma glucose, hepatic glucose output, forearm glucose uptake, and blood lactatelevels.

Outcomes

Notes


Jeppesen 1994Study

TRIAL DESIGN: Open-label cross-over trialDURATION: 12 weeks glipizide and 8 weeks metformin added

Methods

Página 60




COUNTRY: United StatesSETTING: research centerTreatment N: 16Control N: 16AGE: 57+/-3SEX: 63% menINCLUSION: Type 2 DM, poorly controlledEXCLUSIONS: patients not "in good health".

Participants

TREATMENT: Metformin, dosage adjusted clinically + glipizideCOMPARISON: glipizide

Interventions

Postprandial and steady-state glucose, lipids, free fatty acids.Outcomes

Notes


Johansen 1984Study


Methods

COUNTRY: DenmarkSETTING: outpatientTreatment N: 10Control N: 10AGE: 59SEX: 30% menINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin 500 mg/day + placebo/day. COMPARISON: acarbose +placebo

Interventions

Postprandial glucose, HbA1, urinary glucose.Outcomes

Notes


Johnson 1993Study


Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 8Control N: 12AGE: 58+/-8SEX: 62% menINCLUSION: Newly diagnosed obese untreated Type 2 DMEXCLUSIONS: renal or hepatic abnormalities

Participants

TREATMENT: Metformin 0.85-2.5 g/dayCOMPARISON: placebo

Interventions

Insulin sensitivity, HbA1, insulin, c-peptide, skeletal muscle biopsy, glucose synthetaseactivity.

Outcomes

Página 61




Notes


Johnson 1998Study

TRIAL DESIGN: Retrospective cohort studyDURATION: average 9 months

Methods

COUNTRY: United StatesSETTING: Diabetes center chart reviewTreatment N: 124Control N: 0AGE: not listedSEX: not listedINCLUSION: patients with type 2 DM treated with metforminEXCLUSIONS: none listed

Participants

TREATMENT: metformin 500-2500 mg/day, with other medications as neededCOMPARISON: none.

Interventions

Insulin dose, BMI, and HbA1c.Outcomes

Notes


Jones 2000 bStudy

TRIAL DESISN: Abstract: open-label extension study of a randomised controlled trial.DURATION: 30 months

Methods

COUNTRY: United States. SETTING: outpatient. Treatment N: Control N: Age: notlisted. Sex: not listed. Inclusion: Type 2 DM. Exclusions: none listed.

Participants

TREATMENT: Metformin, dosage adjusted clinically, + rosiglitazone. COMPARISON:rosiglitazone

Interventions

Lipds, HbA1c, beta-cell function.Outcomes

Notes


Jones 2000aStudy

TRIAL DESIGN: Abstract of a prospective cohort trial. Some data reported in Fonseca2000. Remaining data analysedDURATION: 6 months

Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 102Control N: 0AGE: not listedSEX: not listedINCLUSION: type 2 DM, poorly controlled on metforminEXCLUSIONS: none listed

Participants

TREATMENT: metformin, dosage adjusted clinically + placebo, or metformin +rositglitazone 4 mg/day, or metformin + rosiglitazone 8 mg/day

Interventions

Fasting glucose and BMI.Outcomes

Página 62




Notes


Jones 2002Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 42Control N: 40Age: 14 +/- 1.8Sex: 30% menInclusion: pediatric patients age 10-16 with type 2 DMExclusions: creatinine > 76 mcmole/L, hepatic dysfunction

Participants

TREATMENT: metformin up to 2 g/dayCOMPARISON: placebo

Interventions

Fasting glucose, HbA1cOutcomes

Notes


Josephkutty 1990Study


Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 20Control N: 20Treatment AGE: 76.5Control AGE: 80.5Treatment SEX: 30% menControl SEX: 30% menINCLUSION: Type 2 DM patients, aged 65 or olderEXCLUSIONS: renal or liver function abnormalities, recent congestive heart failure

Participants

TREATMENT: Metformin 1g BIDCOMPARISON: tolbutamide

Interventions

Fasting insulin, glucose, lactate levels, lipids and weight.Outcomes

Notes


Josse 1995Study

TRIAL DESIGN: randomised controlled trial of acarbose versus placebo. Metformin innonrandomised treatment strataDURATION: 12 months

Methods

Página 63




COUNTRY: CanadaSETTING: outpatientTreatment N: 83Control N: 271AGE: 57.4+/-1.1SEX: 64% menINCLUSION: Type 2 DMEXCLUSION: debilitating disease, gastrointestinal disease

Participants

TREATMENT: Main: acarbose versus placebo. Treatment strata: Metformin (dosageadjusted clinically), diet, sulfonylureas, insulin

Interventions

Postprandial glucose, HbA1.Outcomes

Notes


Jung 2005Study


Methods

COUNTRY: South KoreaSETTING: outpatientTreatment N: 13Control N: 14Age: 57 +/- 10Sex: 45% menInclusion: type 2 DM on sulfonylueraExlcusions: standard

Participants

TREATMENT: metformin 1 g/dayCOMPARISON: rosiglitazone 4 mg/day

Interventions

Anthropometric parameters, fasting plasma glucose, HbA1, lipid profile, adiponectin,resistin

Outcomes

Notes


Karlsson 2005Study


Methods

COUNTRY: FinlandSETTING: outpatientTreatment N: 9Control N: 21Age: 58 +/- 2.1Sex: 80% menInclusion: newly diagnosed type 2 DMExclusions: cardiovascular, renal or hepatic dysfunction, anemia

Participants

TREATMENT: 2 g/dayCOMPARISON: rosiglitazone 4 mg BID or placebo

Interventions

Euglycemic clamp measurements, skeletal muscle biopsies, insulin receptor substrateOutcomes

Página 64




Notes


Kiayias 1999Study

TRIAL DESIGN: Comparative trial; not randomisedDURATION: 3 months

Methods

COUNTRY: GreeceSETTING: outpatientTreatment N: 33Control N: 16AGE: 64.6+/-9.5SEX: 51% menINCLUSION: Poorly controlled type 2 DMEXCLUSIONS: proteinuria, smokers, various medications

Participants

TREATMENT: Metformin, dosage adjusted clinically, or metformin + sulfonylureaCOMPARISON: sulfonylurea

Interventions

Lipoprotein (a) levels, lipids, HbA1.Outcomes

Notes


Kim 2002Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 7Control N: 7Age: 56 +/- 1Sex: 79% menInclusion: type 2 DMExlcusions: standard

Participants

TREATMENT: metformin 2.5 g/dayCOMPARISON: troglitazone 600 mg/day

Interventions

Glucose disposal rate, HbA1, fasting glucoseOutcomes

Notes


Kirk 1999Study


Methods

Página 65




COUNTRY: United StatesSETTING: outpatientTreatment N: 15Control N: 16Treatment AGE: 50.5Control AGE: 54.5Treatment SEX: 64% menControl SEX: 31% menINCLUSION: Type 2 DMEXCLUSIONS: women of childbearing potential, renal or hepatic disease, alcoholabuse, various medications

Participants

TREATMENT: Metformin 0.5-1 g BIDCOMPARISON: troglitazone 200-400 mg/day.

Interventions

HbA1, fasting glucose and C-peptide.Outcomes

Notes


Klein 1975Study


Methods

COUNTRY: GermanySETTING: outpatientTreatment N: 60Control N: 0AGE: not listedSEX: 48% menINCLUSION: maturity-onset DM EXCLUSION: none listed

Participants

TREATMENT: Metformin, dosage titrated clinically, some with chlorpropamideCOMPARISON: none

Interventions

Glucose, weight, and lipids.Outcomes

Notes


Klein 1991Study

TRIAL DESIGN: Open-label randomised controlled trialDURATION: 1 year

Methods

COUNTRY: GermanySETTING: outpatientTreatment N: 16Control N: 19Treatment AGE: 68+/-10Control AGE: 66+/-11Treatment SEX: 27% malesControl SEX: 20% malesINCLUSION: Type 2 DM with failure with sulfonylureaEXCLUSIONS: renal insufficiency with creatinine > 1.2, acute or severe disease, variousmedications

Participants

Página 66




TREATMENT: Metformin, dosage adjusted clinically, + sulfonylureaCOMPARISON: insulin + sulfonylurea

Interventions

Weight, blood pressure, insulin, c-peptide, HbA1, lipids, liver and renal function, andlactate levels.

Outcomes

Notes


Lalau 1990Study

TRIAL DESIGN: Prospective cohort studyDURATION: 2 months.

Methods

COUNTRY: FranceSETTING: outpatientTreatment N: 24Control N: 0AGE: 74+/-1.5SEX: 67% menINCLUSION: patients over the age of 70 with type 2 DMEXCLUSIONS: creatinine clearance < 30 ml/min

Participants

TREATMENT: metformin, 1770-2550 mg/day COMPARISON: noneInterventions

Creatinine clearance, lactate levels.Outcomes

Notes


Lalor 1990Study


Methods

COUNTRY: United KingdomSETTING: hospital clinicTreatment N: 38Control N: 38AGE: 58SEX: 46% menINCLUSION: Obese patients with type 2 DMEXCLUSIONS: previous treatment with metformin or guar

Participants

TREATMENT: Metformin, dosage adjusted clinically, + placeboCOMPARISON: Guar + placebo

Interventions

Fasting glucose, weight, and lipids.Outcomes

Notes


Lam 1998Study

TRIAL DESIGN: Prospective cohort trial with 91% on metforminDURATION: 6 months

Methods

Página 67




COUNTRY: ChinaSETTING: three-center outpatientTreatment N: 90Comparison N: 0AGE: 35-70SEX: 45% menINCLUSION: Type 2 DM with poor control on oral hypoglycemicsEXCLUSIONS: abnormal liver and renal function, significant diseases or conditions,ketonuria,abnormal gutmotility, lactose intolerance, pregnancy and lactation

Participants

TREATMENT: 91% on metformin, dosage adjusted clinically, + acarbose, 150-300mg/day, or metformin + placeboCOMPARISON: 9% on other oral agents + acarbose or placebo. These patients notanalysed.

Interventions

Fasting and postprandial glucose, HbA1c, insulin levels, and lipids.Outcomes

Notes


Laurenti 1992Study

TRIAL DESIGN: Open-label comparative trialDURATION: 6 months

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 30Control N: 30AGE: 38-63SEX: not listedINCLUSION: Type 2 DM with poor control on sulfonylureaEXCLUSIONS: congestive heart failure, nephropathy, liver function abnormalities

Participants

TREATMENT: Metformin, dosage adjusted clinically, + glibenclamideCOMPARISON: sulfonylurea alone

Interventions

Fasting and postprandial glucose, insulin, fructosamine, and BMI.Outcomes

Notes


Lawrence 2004Study

TRIAL DESIGN: Open-label randomised controlled trial DURATION: 3 monthsMethods

COUNTRY: United KingdomSETTING: outpatient Treatment N: 20Control N: 10Age: 60 +/- 9Sex: 60% menInclusion: overweight type 2 DMExclusions: Creatinine > 150 mcmole/L, congestive heart failure, hepatic dysfunction

Participants

TREATMENT: metformin 500 mg BIDCOMPARISON: pioglitazone 30 mg/day or glicazine 80 mg.day

Interventions

HbA1, lipid profile, glucose, BMIOutcomes

Página 68




Notes


Lean 1983Study

TRIAL DESIGN: Prospective cohort study or metformin in a randomised controlled trialof ciclazindolDURATION: 2 months

Methods

COUNTRY: United Kingdom SETTING: outpatientTreatment N: 10Control N: 0AGE: 42-68SEX: 30% menINCLUSION: obese patients with type 2 DM, treated with metforminEXCLUSIONS: hepatic or renal impairment, heart disease, psychiatric or alcoholproblems

Participants

TREATMENT: metformin 500 mg BID + placebo or metformin + ciclazindol 25-75mg/dayCOMPARISON: none

Interventions

Plasma insulin, triglycerides, lactate pyruvate, and weight.Outcomes

Notes


Lee 1998Study


Methods

COUNTRY: United StatesSETTING: University centerTreatment N: 24Control N: 120Treatment AGE: 59+/-3Control AGE: 61+/-2SEX: 0 menINCLUSION: Obese type 2 DMEXCLUSIONS: major illnes, cardiac, renal or hepatic disorder, medicine known to affectbody weight or cholesterol metabolism

Participants

TREATMENT: Metformin 850 mg BIDCOMPARISON: placebo

Interventions

Food consumption and weight lossOutcomes

Notes


Lord 1983Study

TRIAL DESIGN: Open-label cross-over trial with untreated controlsDURATION: 4 weeks

Methods

Página 69




COUNTRY: United KingdomSETTING: outpatientTreatment N: 8Control N: 8AGE: 61+/-5SEX: 38% menINCLUSION: Obese, type 2 DMEXCLUSIONS: abnormal renal or liver function

Participants

TREATMENT: Metformin 500 mg TIDCOMPARISON: no metformin

Interventions

Glucose tolerance test, urinary glucose, and HbA1.Outcomes

Notes


Lunetta 1996Study

TRIAL DESIGN: Prospective cohort studyDURATION: 1 month

Methods

COUNTRY: ItalySETTING: outpatient clinicTreatment N: 12Control N: 0AGE: 55+/-5SEX: 50% menINCLUSION: Type 2 DM for at least one year, with good glycemic controlEXCLUSIONS: diabetic neuropathy, gastroparesis or diarrhea

Participants

TREATMENT: metformin 850 mg BID, then a single dose of metformin 850 mg orplaceboCOMPARISON: none

Interventions

Postprandial glucose.Outcomes

Notes


Makimattila 1999Study


Methods

COUNTRY: FinlandSETTING: outpatientTreatment N: 13Control N: 39Treatment AGE: 54+/-2Control AGE: 58+/-3SEX: not listedINCLUSION: Type 2 DMEXCLUSIONS: congestive heart failure, cardiovascular disase, seizure, liver diseaseunrelated to DM

Participants

TREATMENT: Metformin 2 g/day + insulin NPH QHS COMPARISON: insulin BIDInterventions

Weight gain, urinary glucose, and HbA1.Outcomes

Página 70




Notes


Manzella 2004Study

TRIAL DESIGN: Blinded randomised controlled trial DURATION: 4 monthsMethods

COUNTRY: ItalySETTING: outpatientTreatment N: 60Control N: 60Age: 57 +/- 11Sex: 55% menInclusion: obese type 2 DMExclusions: coronary artery disease

Participants

TREATMENT: metformin 850 mg BIDCOMPARISON: placebo

Interventions

Fasting glucose, insulin, triglyceride, free fatty acids, insulin resistance by HOMAmethod

Outcomes

Notes


Marena 1994Study


Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 10Control N: 10AGE: 60.8+/-10.7SEX: 60% menINCLUSION: Type 2 DM with poor controlEXCLUSIONS: hepatic, renal, pulmonary or cardiac dysfunctions

Participants

TREATMENT: Metformin, dosage adjusted clinically, + glibenclamideCOMPARISON: placebo + glibenclamide

Interventions

Fasting glucose, HbA1, weight, insulin sensitivity.Outcomes

Notes


Marfella 1996Study


Methods

Página 71




COUNTRY: ItalySETTING: outpatientTreatment N: 10Control N: 0AGE: 47+/-0.8SEX: 50% menINCLUSION: newly diagnosed patients with type 2 DM, with mild hyperglycemiaEXCLUSIONS: evidence of microvascular or macrovascular complications

Participants

TREATMENT: Metformin 1700 mg/dayCOMPARISON: none

Interventions

Weight, glucose, HbA1, insulin, lipids, blood pressure, heart rate, platelet aggregation,blood viscosity, blood filterability, epinephrine, and norepinephrine.

Outcomes

Notes


Marre 2002Study


Methods

COUNTRY: France, Belgium, Netherlands, Denmark, PortugalSETTING: outpatientTreatment N: 308Control N: 103Age: 58 +/- 11Sex: 60% menInclusion: type 2 DM inadequately controlled on metforminExclusions: creatinine 127 mcmole/L, hypoxic states, hepatic dysfunction

Participants

TREATMENT: metformin 2.5 g/day with and without glibenclamideCOMPARISON: glimenclamide 20 mg/day

Interventions

HbA1, fasting glucose, fructosamine, lipid profileOutcomes

Notes


McAlpine 1988Study

TRIAL DESIGN: Open-label crossover trialDURATION: 3 months

Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 27Control N: 27AGE: 58SEX: 57% menINCLUSION: Type 2 DMEXCLUSIONS: significant renal or hepatic impairment, various medications

Participants

TREATMENT: Metformin, dosage adjusted clinicallyCOMPARISON: glicazide

Interventions

Weight, fasting and postprandial glucose.Outcomes

Página 72




Notes


McBain 1988Study


Methods

COUNTRY: Scotland. SETTING: outpatient. Treatment N: 14. Control N: 20. Treatmentage: 56.5. Control age:56.3.Treatment sex: 36% men. Control sex: 35% men. Inclusion:Type 2 DM. Exclusions: low weight, Abnormal renal function, liver function.

Participants

TREATMENT: Metformin 500mgBID. COMPARISON: glipizide 5mg/day.Interventions

Calcium and magnesium levelsOutcomes

Notes


McIntyre 1991Study

TRIAL DESIGN: Open-label cross-over trialDURATION: 6 weeks

Methods

COUNTRY: AustraliaSETTING: outpatientTreatment N: 9Control N: 9AGE: 48-75SEX: 44% menINCLUSION: Type 2 DMEXCLUSIONS: renal or liver abnormalities

Participants

TREATMENT: metformin 1.5-3 g/dayCOMPARISON: diet

Interventions

Postprandial glucose, total insulin, and c-peptide levels.Outcomes

Notes


Mehta 1963Study

TRIAL DESIGN: Prospective cohort studyDURATION: approximately 1 month.

Methods

COUNTRY: IndiaSETTING: outpatientTreatment N: 41Control N: 0AGE: not listedSEX: not listedINCLUSIONS: patients with DM on medications other than metforminEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dosage unclearCOMPARISON: none.

Interventions

Glycemia, glucosuria.Outcomes

Página 73




Notes


Menzies 1989Study


Methods

COUNTRY: United KingdomSETTING: outpatientTreatment N: 64Control N: 0AGE: 64+/-9SEX: 41% menINCLUSION: obese patients with type 2 DMEXCLUSIONS: ketosis, or abnormal electrolytes or renal function

Participants

TREATMENT: Metformin 1.5-2 g/day or 2.5-3 g/day COMPARISON: noneInterventions

Plasma glucose, HbA1, and lactate.Outcomes

Notes


Moses 1999aStudy


Methods

COUNTRY: AustraliaSETTING: outpatientTreatment N: 54Control N: 28Treatment AGE: 57.8Control AGE: 60.3Treatment SEX: 63% menControl SEX: 54% menINCLUSION: Type 2 DM with poor control on metforminEXCLUSIONS: clincally significant renal insufficiency, abnormal liver functions, cardiacdiasease, history of lactic acidosis

Participants

TREATMENT: Metformin, dosage adjusted clinically, + placebo; or metformin +repaglinideCOMPARISON: repaglinide + placebo

Interventions

Fasting glucose, and HbA1.Outcomes

Notes


Munk 1975Study

TRIAL DESIGN: Open-label comparative trialTRIAL DURATION: 6 months

Methods

Página 74




COUNTRY: GermanySETTING: outpatientTreatment N: 40Control N: 20AGE: unclearSEX: 55% malesINCLUSION: Type 2 DMEXCLUSIONS: none listed

Participants

TREATMENT: Metformin, dosage unclear, or metformin + insulin COMPARISON:Sulfonylurea

Interventions

Lipids, liver function studies, and glucose.Outcomes

Notes


Nagi 1993Study


Methods

COUNTRY: United States.SETTING: outpatient. Treatment N: 27. Control N: 27. Age: 56.8 +/-8.9. Sex: not listed.Inclusion: Type 2 DM. Exclusions: cardiovascular disease, thromboembolic disease,renal or hepatic disease, retinopathy.

Participants

TREATMENT: Metformin, dosage adjusted clinically. COMPARISON: placeboInterventions

Fasting glucose, lipids, BMI, insulin, c-peptide, blood pressure, plasminogen activatorinhibitor, and other factors

Outcomes

Notes


Natali 2004Study


Methods

COUNTRY: ItalySETTING: outpatient:Treatment N: 28Control N: 46Age: 58 +/- 9Sex: 70% menInclusion: type 2 DMExclusions: renal or hepatic dysfrunction, congestive heart failure

Participants

TREATMENT: Metformin 500 mg TIDCOMPARISON: placebo

Interventions

Insulin sensitivity by euglycemic clamp, fat-free mass, response to acetycholineOutcomes

Notes


Página 75




Nattrass 1977Study

TRIAL DESIGN: Open-label, cross-over comparative trialDURATION: 1 month

Methods

Country: United Kingdom. Setting: outpatient.Treatment N: 6. Control N: 6. Age: 50-57.Sex: 67% men. Inclusion: Type 2 DM longer than 3 years. Exclusions: hepatic or renaldisease.

Participants

TREATMENTt: Metformin 500mg TID. COMPARISON: intervention: phenformin 50mgBID (not analysed) or glibenclamide, 2.5-5mg/day.

Interventions

Blood glucose, lactate, pyruvate, 3-hydroxybutyrate, acetoacetate,ketones, lactatopyruvate ratios, and cyclic AMP.

Outcomes

Notes


Niazi 1998Study


Methods

COUNTRY: Pakistan. SETTING: outpatient. Treatment N: 18. Control N: 36. Treatmentage: 50 +/-11. Control age: 48 +/-11. Treatment sex: 61% men. Control sex: 56% men.Inclusion: Type 2 DM with sulfonylurea failure. Exclusions: cardiomegaly, lung disease,malnutrition, infection, various medications.

Participants

TREATMENT: Metformin 0.5-3g/day. COMPARISON: insulinInterventions

Lipids, blood pressure, weight, and BMI.Outcomes

Notes


Nosadini 1987Study

TRIAL DESIGN: Open-label trial with patients as own controlsDURATION: 1 month

Methods

COUNTRY: Italy. SETTING: outpatient. Treatment N: 7. Control N: 7. Age: 46 +/-5.Sex: 57% men. Inclusion: Type 2 DM. Exclusions: age > 65.

Participants

TREATMENT: Metformin 850mg TID. COMPARISON: dietInterventions

Glucose turnover and insulin bindingOutcomes

Notes


Noury 1991Study


Methods

COUNTRY: France. SETTING: outpatient. Treatment N: 30. Control N: 27. Age: 55+/-9.1. Treatment sex: 53% men. Control sex: 44% men. Inclusion: Type 2 DM.Exclusions: renal or hepatic disease.

Participants

TREATMENT: Metformin 1700mg/day. COMPARISON: glicazideInterventions

Blood glucose, insulin levels, and weight lossOutcomes

Página 76




Notes


Ohnhaus 1983Study

TRIAL DESIGN: Open-label randomised controlled trialDURATION: 1.5 months

Methods

COUNTRY: Switzerland. SETTING: outpatient. Treatment N: 12. Control N: 12. Age:not listed. Sex: not listed. Inclusion: Type 2 DM pts on phenprocoumon. Exclusions:none listed.

Participants


Phenprocoumon pharmacokineticOutcomes

Notes


Pavo 2003Study


Methods

COUNTRY: RussiaSETTING: outpatientTreatment N: 100Control N: 105Age: 55 +/- 9Sex: 50% menInclusion: recently diagnosed type 2 DM naive to oral medicationsExclusions: hepatic and renal dysfunction, congestive heart failure

Participants

TREATMENT: metformin 2.5 g/dayCOMPARISON: pioglitazone 45 mg/day or placebo

Interventions

HbA1c, fasting glucose, insuliln resistance by HOMA methodOutcomes

Notes


Peacock 1984Study

TRIAL DESIGN: Prospective cohort studyDURATION: at least 3 months

Methods

Country: United Kingdom. Setting: outpatient setting. Treatment N: 33. Control N: 0.Age: 58. Sex: 60% men. Inclusion: patients with type 2 DM, treated with high doses ororal hypoglycemics. Exclusions: history or ketosis or good control on oral agents.

Participants

TREATMENT: metformin, dosage unclear, + glibenclamide, dosage adjusted clinically.After 3 months, some were treated additionally with insulin. COMPARISON: none.

Interventions

Fasting glucose, HbA1, and fasting c-peptide.Outcomes

Notes


Página 77




Peacock 1986Study

TRIAL DESIGN: Comparative trialDURATION: 6 months

Methods

Country: United Kingdom. Setting: outpatient.Treatment N: 27. Control N: 20.Treatmentage: 59.9 +/-2.1. Control age: 56.7 +/-2.1. Treatment sex: 59% men. Control sex: 66%men. Inclusion: Type 2 DM. Exclusions: none listed.

Participants

TREATMENT: Metformin, dosage unclear, + glibenclamide. COMPARISON: insulinInterventions

Platelet reactivity (ADP release, adrenaline release and NaAA threshold), and fastingglucose, HgA1.

Outcomes

Notes


Pedersen 1965Study


Methods

Country: Denmark. Setting: inpatient and outpatient. Treatment N: 20. Control N: 0.Age: not listed. Sex: not listed. Inclusion: maturity-onset DM. Exclusions: none listed.

Participants

TREATMENT: metformin, dose titrated up clinically, 1-4g/day. COMPARISON: none.Interventions

Plasma glucoseOutcomes

Notes


Pedersen 1989Study

TRIAL DESIGN: Double-blind crossover randomised controlled trialDURATION: 1 month

Methods

Country: Denmark. Setting: outpatient. Treatment N: 10. Control N: 10. Age: 53 +/-9.Sex: 20% men. Inclusion: Obese pts with Type 2 DM. Exclusions: renal or liverdysfunction.

Participants

TREATMENT: Metformin 500mg TID. COMPARISON: placeboInterventions

Fasting and postprandial glucose, fructosamine, insulin, c-peptide, and adipocite insulinreceptor binding.

Outcomes

Notes


Pirart 1961Study

TRIAL DESIGN: Retrospective cohort studyDURATION: 3 motnhs

Methods

Country: Belgium. Setting: outpatient. Treatment N: 107. Control N: 0. Age: not listed.Sex: not listed. Inclusion: type 2 DM, poorly controlled on a single agent. Exclusions:obesity.

Participants

TREATMENT: metformin, unclear dose. COMPARISON: some patients treated withother agents, not analysed.

Interventions


Página 78




Notes


Ponssen 2000Study


Methods

Country: Netherlands. Setting: outpatient. Treatment N: 31. Control N: 62. Age: 62.+/-10. Sex: 77% men. Inclusion: Type 2 DM. Exclusions: renal insufficency withCreatinine clearance < 50 ml.min, hepatic disease, cardiovascular disease, alcoholabuse, various medications.

Participants

TREATMENT: Metformin, dosage unclear, + insulin COMPARISON: placebo + insulinInterventions

Glucose, fructosamine, insulin requirements, lipds, BMI, and HbA1.Outcomes

Notes


Prager 1986Study

TRIAL DESIGN: Open-label trial, cross-over, with patients as their own controlsDURATION: 3 months control then 1 month metformin

Methods

Country: Austria. Setting: outpatient. Treatment N: 12. Control N: 12. Age: 35-62. Sex:16% men. Inclusion: Type 2 DM. Exclusions: vascular disease, renal failure, liverfunction abnormalities.

Participants


Insulin sensitivity, fasting glucose, and HbA1.Outcomes

Notes


Puchegger 1964Study


Methods

COUNTRY: Germany. SETTING: outpatient. Treatment N: 43. Control N: 0. Age:notlisted. Sex: 28% men. Inclusion: patients with DM. Exclusions: none listed.

Participants

TREATMENT: metformin, alone or in combination with insulin, dosage adjusted clinically.COMPARISON: none.

Interventions

Plasma glucose.Outcomes

Notes


Rachmani 2002Study


Methods

Página 79




COUNTRY: IsraelSETTING: outpatientTreatment N: 195Control N: 198Age: 64.5 +/- 4Sex: 60% nebInclusion: type 2 DM with at least one traditional contraindicationExclusions: liver cirrhosis, actue myocardial infarction or pulmonary edema withinprevious 30 days, CO2 narcosis, malginancy

Participants

TREATMENT: metformin, dose adjusted clinicallyCOMPARISON: no metformin

Interventions

Lactic acid levels, lactic acidosis, cardiovascular events, complicationsOutcomes

Notes


Rains 1988Study

TRIAL DESIGN: Single-blind randomised controlled trialDURATION: 3 months

Methods

COUNTRY: United Kingdom/ SETTING: hospital clinic. Treatment N: 35. Control N:70. Age: not listed. Sex: not listed. Inclusion: Type 2 DM. Exclusions: age > 70, BUN> 6 mmol/L, abnormal liver functions.

Participants

TREATMENT: Metformin 1-3g/day. COMPARISON: placeboInterventions

Plasma glucose, lipoproteins, and HbA1.Outcomes

Notes


Rains 1989Study

TRIAL DESIGN: Open-label cross-over randomised controlled trialDURATION: 1.5 months

Methods

Country: United Kingdom. Setting: diabetes clinic. Treatment N: 28. Control N: 14. Age:unclear. Sex: 64% men. Inclusion: Type 2 DM. Exclusions: none listed.

Participants

TREATMENT: Metformin, dosage unclear. COMPARISON: glibenclamideInterventions

Weight, lipds, glucose, and HbA1.Outcomes

Notes


Raptis 1996Study

TRIAL DESIGN: Open-label crossover randomised controlled trialDURATION: 3 months

Methods

Country: Greece. Setting: University center. Treatment N: 30. Control N: 30. Age: 60+/-7.5. Sex: 57% men. Inclusion: Type 2 DM. Exclusions: cardiac, renal, hepatic failure,autoimmune disease.

Participants

TREATMENT: Metformin, dosage adjusted clinically, + glibenclanide. COMPARISON:phenformin + glibenclanide

Interventions

Página 80




Postprandial glucose, HgA1c, lipids, and blood lactate levels.Outcomes

Notes


Reaven 1992Study

TRIAL DESIGN: Nonrandomised open-label trialDURATION: 3 months

Methods

Country: United States. Setting: research center. Treatment N: 13. Control N: 13. Age:57 +/-2. Sex: 77% men. Inclusion: Type 2 DM with poor control on sulfonylureas.Exclusions: Other drugs that effect lipids.

Participants

TREATMENT: Metformin 0.5-2.5g/day. COMPARISON: glipizide.Interventions

Insulin sensitivity, glucose, and HbA1.Outcomes

Notes


Relimpio 1998Study


Methods

Country: Spain. Setting: outpatient. Treatment N: 31. Control N: 29. Treatment age:65 +/-8. Control age: 66 +/-6.Treatment sex: 21% men. Control sex: 40% men. Inclusion:Poorly controlled insulin-treated Type 2 DM. Exclusions: life-threatening condition,common contraindication to treatment, renal insufficiency.

Participants

TREATMENT: Metformin, dosage adjusted clilnically, + insulin. COMPARISON: insulinincrease.

Interventions

Lipids, HbA1, and fasting glucose.Outcomes

Notes


Reyes 1969Study

TRIAL DESIGN: Prospective cohort studyDURATION: 1 month

Methods

Country: Mexico. Setting: outpatient. Treatment N: 53. Control N: 0. Age: not listed.Sex: 28% men. Inclusion: DM, poorly controlled on sulonylureas. Exclusions: nonelisted

Participants

TREATMENT: metformin, 1600-2400mg/day + chlorpropamide 500-750mg/day.COMPARISON: none

Interventions


Notes


Riccio 1991Study

TRIAL DESIGN: Prospective comparative trial, with control group for less than 1 month.Metformin data analysedDURATION: 4 weeks

Methods

Página 81




Country: Italy. Setting: medical center. Treatment N: 6. Control N: 0. Treatment age:48+/-2. Sex: not listed. Inclusion: non-insulin-dependent type DM. Exclusion: nonelisted.

Participants

TREATMENT: metformin 850mg BID. COMPARISON: none.Interventions

Basal and insulin-mediated glucose, free-fatty acid metabolism, and lipds.Outcomes

Notes


Robinson 1998Study


Methods

Country: United Kingdom. Setting: teaching hospital clinic. Treatment N: 35. ControlN: 35. Treatment age: 61.3. Control age: 56.1. Treatment sex: 37% men. Control sex:21% men. Inclusion: Insulin-treated Type 2 DM. Exclusions: childbearing age, anotheranihyperglycemic medication, renal insufficiency with creatinine > 125.

Participants

TREATMENT: Metformin 1-2 g/day. COMPARISON: placeboInterventions

Fasting glucose, HbA1, lipids, weight, and blood pressure.Outcomes

Notes


Roden 2005Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 917Control N: 916Age: 57 +/- 8.5Sex: 55% menInclusion: type 2 DM naive to metformin pioglitazoneExlcusions: not stated

Participants

TREATMENT: metformin 2.5 g/dayCOMPARISON: pioglitazone 45 mg/day

Interventions

Insulin sensitivity, fasting serum glucose and insulinOutcomes

Notes


Rodger 1995Study

TRIAL DESIGN: randomised controlled trial of acarbose vs placebo. Metformin innon-randomised treatment strataDURATION: 12 months

Methods

COUNTRY: Canada. SETTING: outpatient. Treatment N: 74. Control N: 242. Age:unclear. Sex: not listed. Inclusion: Type 2 DM. Exclusions: lactose intolerance,debilitating disease, gastrointestinal disease, various medications.

Participants

Página 82




TREATMENT: Main: acarbose vs placebo.Treatment strata: metformin (dosage adjustedclinically), diet, sulfonylurea, insulin.

Interventions

Postprandial glucose, HbA1, insulin, and c-peptide.Outcomes

Notes


Roger 1999Study


Methods

COUNTRY: France. SETTING: community-based multi-center study. Treatment N:127, with 63 on metformin + benflurex and 64 on metformin + placebo. Control N: 0.Age: not listed. Sex: not listed. Inclusion: obese patients with uncontrolled type 2 DM,treated with metformin. Exclusions: young patients, severe inervurrent illnes, kidneyor liver failure, severehypertension, chronic pancreatitis,and alcoholism.

Participants

TREATMENT: metformin 850mg BID + benflurex 150mg TID or metformin + placebo.COMPARISON: none.

Interventions

Basal and stimulated insulin, HgA1, and body weight.Outcomes

Notes


Rosenstock 1998Study

TRIAL DESIGN: Prospective cohort study of metformin in a randomised controlled trialof acarboseDURATION: 6 months

Methods

COUNTRY: United States. SETTING: multicenter outpatient.Treatment N: 148. ControlN: 0. Age: 56.7. Sex: 74% men. Inclusion: metformin-treated patients with type 2 DM.Exclusions: acute or chronic acidosis, persistent ketonuria, or a history of ketoacidosis.

Participants

TREATMENT: metformin 2-2.5g/day + placebo or metformin +acarbosis 75-300mg/day.COMPARISON: none.

Interventions

HbA1c, glucose, insulin, triglycerides, and plasma metformin levels.Outcomes

Notes


Sanchez-Barba 1999Study


Methods

COUNTRY: Spain. SETIING: outpatient. Treatment N: 30. Control N: 0. Age: not listed.Sex: not listed. Inclusion: type 2 DM. Exclusions: none listed

Participants

TREATMENT: metformin, dosage adjusted clinically + insulin, dosage adjusted clinically.COMPARISON: none

Interventions

HgA1c, and plasma glucose.Outcomes

Notes


Página 83




Santos 1995Study

TRIAL DESIGN: Prospective cohort studyDURATION: 2.5 months

Methods

COUNTRY: Brazil. SETTING: metabolic laboratory. Treatment N: 14. Control N: 0.Age: 44+/-2. Sex: 36% men. Inclusions: type 2 DM, on no medications. Exclusions:prior insulin treatment.

Participants

TREATMENT: metformin 850mg BID. COMPARISON: none.Interventions

Fasting glucose, HbA1, fasting insuling, lipids, and insuling receptor tyrosine kinaseactivity.

Outcomes

Notes


Schernthaner 2004Study


Methods

COUNTRY: 12 European countriesSETTING: outpatientTreatment N: 597Control N; 597Age: 57 +/- 9Sex: 60% menInclusion: poorly controlled type 2 DMExlcusions: standard

Participants

TREATMENT: metformin850 mg TIDCOMPARISON: pioglitazone 45 mg/day

Interventions

HbA1c, fasting glucose and insulin, lipid profilesOutcomes

Notes


Schneider 1990Study


Methods

COUNTRY: Germany. SETTING: outpatient. Treatment N: 18. Control N: 18. Treatmentage: 60.4, Control age: 61.5.Treatment sex: 44% men. Control sex: 56% men. Inclusion:Patients with Type 2 DM and hyperlipoproteinemia. Exclusions: cardiovascular disease,pulmonary disease, hepatic or gastrointestinal diseaes, malignancy or psychiatricdisorder.

Participants

TREATMEN: Metformin, dosage adjusted clinically. COMPARISON: placeboInterventions

Lipids, and lipoproteins.Outcomes

Notes


Schulte 1973Study


Methods

Página 84




COUNTRY: Mexico. SETTING: outpatient. Treatment N: 53. Control N: 0. Age: 57.Sex: 33% men. Inclusion: adult-onset DM. Exclusions: none listed.

Participants

TREATMENT: metformin + chlorpropamide, dose adjusted clinically. COMPARISON:none.

Interventions

Fasting and postprandial glucose, weight, and glycosuria.Outcomes

Notes


Sieradzki 1999Study

TRIAL DESIGN: Acarbose trial. Metformin in nonrandomised treatment strataDURATION: 2 motnhs

Methods

COUNTRY: Poland. SETTING: outpatient. Treatment N: 106. Control N: 374. Age:31-88. Sex: 44% men. Inclusion: Type 2 DM. Exclusions: none listed.

Participants

TREATMENT: Metformin, dosage adjusted clinically, +/- sulfonylurea + acarbose.COMPARISON: sulfonyurea + acarbose or acarbose

Interventions

Fasting and posprandial glucose, urinary glucose, and lipids.Outcomes

Notes


Stades 2000Study

TRIAL DESIGN: Retrospective cohort studyMethods

Country: Netherlands. Setting: outpatient clinic. Treatment N: 65. Control N: 0. Age:64.5. Sex: not listed. Inclusion: patients with type 2 DM on metformin treatment for atleast 6 months. Exclusions: insufficient follow-up time, or no HgA1c on record.

Participants

Study duration: median 32 months. Treatment: metformin, dosage adjusted clinically.Comparison: none.

Interventions

HbA1c and body weight.Outcomes

Notes


Stalhammar 1991Study

TRIAL DESIGN: Retrospective cohort studyDURATION: 35 months

Methods

COUNTRY: Sweden. SETTING: Swedish population study. Treatment N: 81. ControlN: 0. Age: 50-74 years. Sex: 51% men. Inclusion: Patients with type 2 DM receivingmetformin. Exclusions: none listed

Participants

TREATMENT: metformin, dosage adjusted clinically. COMPARISON: none.Interventions

HbA1c and BMI.Outcomes

Notes


Página 85




Sterne 1963Study

TRIAL DESIGN: Prospective cohort studyDURATION: 60 motnhs

Methods

COUNTRY: Germany. SETTING: outpatient. Age: not listed. Sex: not listed. Inclusions:maturity-onset DM. Exclusions: none listed.

Participants

TREATMENT: metformin, dosage titrated clinically, alone or in combination with insulinor sulfonyrureas. COMPARISON: none

Interventions

Glycemia, side effects.Outcomes

Notes


Stratmann 1965Study


Methods

COUNTRY: Germany. SETTING: outpatient. Treatment N: 92. Control N: 0. Age: notlisted. Sex: not listed. Inclusion: patients with DM, who have failed oral sulfonylureas.Exclusions: none listed.

Participants

TREATMENT: metformin, dosage adjusted clinically. COMPARISON: none.Interventions

Level of glycemic control.mOutcomes

Notes


Strowig 2002Study


Methods

COUNTRY: United StatesSETTING: outpatientTreatment N: 27Control N: 61Age: 52 +/- 9Sex: 50% menInclusion: type 2 DM inadequately treated on insulinExclusions: renal or hepatic dysfunction

Participants

TREATMENT: metformin 2 g/day + insulinCOMPARISON: insulin with or without troglitazone 600 mg/day

Interventions

HbA1c, body weight, lipid profileOutcomes

Notes


Stumvoll 1995Study

TRIAL DESIGN: Prospective comparative trial, with controlDURATION: 4 monthsgroup studied for less than 1 month. Metformin data analysed

Methods

Página 86




Country: United States. Setting: outpatient. Treatment N: 10. Control N: 0. Age: 58+/-9.Sex: 60% men. Inclusion: healthy obese type 2 DM. Exclusions: none listed, but allwere described as healthy.

Participants

Study duration: 4 months. Treatment intervention: metfomin 800-2550mg/day.Comparison intervention: none.

Interventions

HbA1, fasting glucose, weight, plasma glucose turnover, and lactate conversion toglucose.

Outcomes

Notes


Sundaresan 1997Study

TRIAL DESIGN: Double-blind randomised controlled trialMethods

COUNTRY: Australia. SETTING: outpatient. Treatment N: 14. Control N: 14. Age:40-73. Sex: 64% men. Inclusion: Type 2 DM. Exclusions: BMI > 40 different from idealbody weight, vascular disease, microvscular disease.

Participants

TREATMENT: Metformin 1-2g/day. COMPARISON: glibenclamicdeInterventions

Norepinephrine levels, blood pressure, and forearm vascular resistance.Outcomes

Notes


Swislocki 1999Study

TRIAL DESIGN: Retrospective cohort studyDURATION: 5 motnhs

Methods

COUNTRY: United States. SETTING: Veteran's Administration Health Care system.Treatment N: 251. Comparison: 0. Age: mot listed. Sex: not listed. Inclusion: patientswith type 2 DM receiving metformin. Exclusions: none listed.

Participants

TREATMENT: metformin, doses adjusted clinically. COMPARISON: none.Interventions

HbA1c, weight and blood pressure.Outcomes

Notes


Szanto 1964Study

TRIAL DESIGN: Open-label comparative trialDURATION: 9 months

Methods

COUNTRY: Ireland. SETTING: diabetes clinic. Treatment N: 10. Control N: 9. Age:51-76. Sex: 45% men. Inclusion:Type 2 DM not controlled on sulfonylueas. Exclusions:hypoglycemia.

Participants

TREATMENT: Metformin, dosage unclear. COMPARISON: phenformin (not analyses).Then acetohexamide-biguanide combination was given.

Interventions

Weight, blood glucose, and insulin dose.Outcomes

Notes


Página 87




Taylor 1982Study

TRIAL DESIGN: Nonrandomised open-label trialDURATION: 12 months

Methods

COUNTRY: United Kingdom. SETTING: outpatient. Treatment N: 23. Control N: 71.Age: 51-52 years. Treatment sex: 43% men. Control sex: 77% male. Inclusion: Type2 DM, obese and nonobese. Exclusions: renal or hepatic disease.

Participants

TREATMENT: Metformin (obese) 500mg TID. COMPARISON: glibenclamice (nonobese)2.5-15mg/day.

Interventions

Lipids and apolipoproteinsOutcomes

Notes


Tessari 1994Study

TRIAL DESIGN: Double-blind randomised controlled trialDURATION: 1 month

Methods

COUNTRY: Italy. SETTING: outpatient. Treatment N: 11. Control N: 6. Treatment age:53 +/-3. Control age: 60 +/-3> Treatment sex: 55% men. Control sex: 33% men.Inclusion: Diet-treated Type 2 DM. ExclusionsL cardiovascular, gastrointestinalpulmonary or renal disease.

Participants

TREATMENT: Metformin, dosage adjusted clinically. COMPARISON: placeboInterventions

Postprandial phenylalanine kinetics, weight, free fatty acids, BMI, and HbA1.Outcomes

Notes


Tessier 1999Study


Methods

COUNTRY: Canada. SETTING: outpatient. Treatment N: 18 Control N: 18. Treatmentage: 59.1 +/- 7.1. Control age: 59.3 +/-7.3. Treatment sex: 16% men. Control sex: 44%men. Inclusion: Type 2 DM. Exclusions: acute cardiovascular or neurological events,malignancy, various medications.

Participants

TREATMENT: Metformin 0.75-2.5g/day. COMPARISON: gliclazideInterventions

HbA1, fructosamine, glucose tolerance test.Outcomes

Notes


Testa 1996Study


Methods

COUNTRY: Italy. SETTING: outpatient clinic and metabolic laboratory. Treatment N:80. Control N: 0. Age: 63+/-9.7. Sex: 63% men. Inclusion: type 2 DM with good glycemiccontrol on sulfonylureas for at least 2 years.. Exclusions: previous insulin treatment

Participants

TREATMENT: 1.2-1.7g/day. COMPARISON: none.Interventions

Página 88




Plasminogen activator inhibitor, lipoprotein(a), and BMI.Outcomes

Notes


Teupe 1991Study


Methods

COUNTRY: Germany. SETTING: outpatient. Treatment N: 50. Control N: 50. Treatmentage: 51.5 +/-10. Control age: 56 +/-8. Treatment sex: 40% males. . Control sex: 40%males. Inclusion: Type 2 DM, poor control. Exclusions: age > 70, creatinine > 1.2, ivercirhosis, ischemia or wasting disease, sever acute disease.

Participants

TREATMENT: Metformin, dosage adjusted clinically, + diet. COMPARISON: dietInterventions

Weight, lipids, HbA1, c-peptide, and lactate levels.Outcomes

Notes


Tikkainen 2004Study


Methods

COUNTRY: FinlandSETTING: outpatientTreatment N: 11Control N: 9Age: 30.6 +/- 3.5Sex: 35% menInclusion: type 2 DM treated with dietExclusions: cardiovascular or renal disease

Participants

TREATMENT: metformin 1 g BID + placeboCOMARISON: rosiglitazone 4 mg BID + placebo

Interventions

HbA1c, insulin, free fatty acid, body weight, adiponectinOutcomes

Notes


Tosi 2003Study

TRIAL DESIGN: Double-blind randomised controlled cross-over trialDURATION: 6 months for each treatment armDURATION:

Methods

COUNTRY: ItalySETTING: outpatientTreatment N: 88Control N: 88Age: 57.3 +/- 7Sex: 70% menInclusion: type 2 DMExclusions: severe cardiovascular, renal or hepatic disease, insulin treatment,

Participants

Página 89




TREATMENT: metformin 3 g/day with or without glibenclamideCOMPARISON: glibenclamide 15 mg/day

Interventions

HbA1c, fasting glucoseOutcomes

Notes


Trischitta 1992Study

TRIAL DESIGN: randomised controlled trial cross-overDURATIPN: 2 months for each arm

Methods

COUNTRY: Italy. SETTING: outpatient. Treatment N: 20. Control N: 20. Age: 53.6+/-2.1. Sex: not listed. Inclusion: Type 2 DM with sulfonylurea. Exclusions: renal, liver,cardiovascular or systemic disese.

Participants

TREATMENT: Metformin 500mg TID. COMPARISON: insulinInterventions

Fasting and postprandial glucose, c-peptide, HbA1, weight, and lipids.Outcomes

Notes


Trischitta 1998Study

TRIAL DESIGN: randomised controlled trial cross-overDURATION: 2 months

Methods

COUNTRY: Italy. SETTING: outpatient. Treatment N: 50. Control N: 50. Age: 55.7+/-1.2. Sex: 24% men. Inclusion: Type 2 DM. Exclusions: none listed.

Participants

TREATMENT: Metformin 850mg TID + glibenclamide. COMPARISON: insulin +glibenclamide

Interventions

Fasting glucose, HbA1, c-peptide, and weight.Outcomes

Notes


UKPDS-34 1998Study

TRIAL DESIGN: Open-label randomised controlled trial.Methods

Country: United Kingdom Setting: large multicenter.Treatment N: 683. Control N: 1631.Treatment age: 53 +/-8. Control age: 53 +/-8. Treatment sex: 46% men. Control sex:46% men. Inclusion: Type 2 DM. Exclusions: severe vascular disease, acceleratedhypertension, renalfailure with creatinine > 175 mmol/L, life thretening disease, severeasthma, myocardial infarction in past year, current angina, congestive heart failure.n = 1704

Participants

Trial duration: 6.6 - 10.7 years. Treatment: Metformin 850mg QD-TID. Comparison:diet, sulfonylurea, or insulin

Interventions

DM-related endpoint (sudden death, death for hyper- or hypoglycemia, myocardialinfarction, stroke, renal failure, amputation, eye problems), diabetes-related death,all-cause mortality, HgA1, microalbuminuria.

Outcomes

Notes


Página 90




Vannasaeng 1995Study

TRIAL DESIGN: Open-label trial of acarbose. Metformin in nonrandomised treatmentstrataDURATION: 6 months

Methods

COUNTRY: Thailand. SETTING: Outpatient. Treatment N: 24. Control N: 12. Age: 50.4+/-1.5, Sex: 19% men. Inclusion: Type 2 DM. Exclusions: pregnancy, liver disorder,renal insufficiency with Creatinine > 2 mg/dl.

Participants

TREATMENT: Metformin, dosage adjusted clinically, + sulfonylurea + acarbose.COMPARISON: sulfonylurea + acarbose

Interventions

Fasting glucose, HbA1, lipids, insulin and c-peptide.Outcomes

Notes


Velussi 1992Study

TRIAL DESIGN: Open-label cross-over nonrandomised comparative trialDURATION: 4 months

Methods

COUNTRY: Italy. SETTING: general practive. Treatment N: 60. Control N: 60. Age: 68+/- 3 Sex: 53% men. Inclusion: Type 2 DM with hypertension. Exclusions: none listed.

Participants

TREATMENT: Metformin, dosage adjusted clinically, + glibenclamide, doses on clinicalgrounds. COMPARISON: Phenformin + glibenclamide (not analysed).

Interventions

Fasting glucose, HbA1c, basal C-peptide, glucosuria, and lactate levels.Outcomes

Notes


Vigneri 1991Study


Methods

COUNTRY: Italy. SETTING: outpatient. Treatment N: 12. Control N: 12. Age: 52.3+/-2.1 Sex: not listed. Inclusion: Type 2 DM with failure to sulfonylureas. Exclusions:none listed.

Participants

TREATMENT: Metformin, dosage adjusted clinically, + glyburide. COMPARISON:insulin + glyburide

Interventions

Fasting and postprandial glucose, and HbA1.Outcomes

Notes


Willey 1992Study

TRIAL DESIGN: Prospective cohort studyMethods

Country: Australia. Setting: outpatient. Treatment N: 38. Control N: 0. Age: 54+/-1.7.Sex: 44% men. Inclusion: overweight patients with type 2 DM, and HbA1c >normal.Exclusions: none listed.

Participants

Trial duration: 3 months.Treatment intervention: metformin, 1-3g/day, + dexfenfluramineor metformin + placebo. Comparison: none.

Interventions

Página 91




Body weight, HbA1c, blood pressure, and fructosamine.Outcomes

Notes


Willey 1994Study


Methods

COUNTRY: Australia. SETTING: diabetes center. Metformin treatment N: 20 (10 ondexfenfluramine, 10 on placebo). Age: 55+/-1.9. Sex: 30% men. Inclusion: Type 2 DM,overweight with poor control, on maximum dose metformin. Exclusions: none listed.

Participants

TREATMENT: Metformin, 1-3g/day, + dexfluramine or metformin + placebo.COMPARISON: none.

Interventions

Weight, BMI, and HbA1c.Outcomes

Notes


Willms 1999Study

TRIAL DESIGN: randomised controlled trial. Single-blind for metformin versus other.DURATION: 3 months

Methods

COUNTRY: Germany. SETTING: outpatient. Treatment N: 29. Control N: 60. Treatmentate: 53.4. Control age: 59.2. Treatment sex: 48% males. Control sex: 48% males.Inclusion: Type 2 DM. Exclusions: Severe hepatic or renal abnormalities, respiratoryinsufficiency, conditions that predispose to tissue anoxia.

Participants

TREATMENT: MetfORmin 850mg BID. COMPARISON: acarbose or placeboInterventions

Body weight, and HbA1.Outcomes

Notes


Wilson 1989Study


Methods

COUNTRY: United Kingdom. SETTING: outpatient. Treatment N: 15. Control N: 45.Age: 65 +/-2. Sex: 80% men. Inclusion: Type 2 DM on sulfonylureas. Exclusion: Nonelisted.

Participants

TREATMENT: Metformin 500mg TID. COMPARISON: guar 5gmTID.Interventions

Glucose, HbA1, and lipidsOutcomes

Notes


Wolever 1995Study

TRIAL DESIGN: Double-blind randomised controlled trial of acarbose versus placebo.Metformin in 1 of 4 non-randomized treatment strata.DURATION: 11 years

Methods

Página 92




COUNTRY: Canada. SETTING: outpatient. Treatment N: 83. Control N: 271. Treatmentage: 55.8. Control age 57.6. Treatment sex: 44% men. Control sex: 57% males.Inclusion: Type 2 DM. Exclusions: renal or liver abnormalities.

Participants

TREATMENT: acarbos vs placebo. Treatment strata: Metformin (dosage adjustedclinically), diet, sulfonylurea, insulin

Interventions

Lipids, HbA1, and serum acetate levelsOutcomes

Notes


Wolever 2000Study


Methods

COUNTRY: Canada. SETTING: outpatient. Treatment N: 109. Control N: 90. Treatmentage: 58.7 +/-1.1. Control age: 59.5 +/-1.1. Treatment sex: 80% men. Control sex: 69%men. Inclusion: Type 2 DM. Exclusions: insulin treatment, major debilitating disease,recent cardiovascular event or surgery, various medication, renal or liver idsease,emotional disorder.

Participants

TREATMENT: Metformin 500 mgTID or metformin + miglitol. COMPARISON: miglitolor placebo

Interventions

Serum folate and B12 levels, and HbA1.Outcomes

Notes


Wu 1990Study


Methods

COUNTRY: United States. SETTING: inpatient and outpatient.Treatment N: 12. ControlN: 0. Age: 56+/-3. Sex: 58% men. Inclusion: type 2 DM. Exclusion: significant illness,or medication that could affect carbohydrate metabolism.

Participants

TREATMENT: metformin 2.5g/day. COMPARISON: none.Interventions

Fasting and postprandial glucose, HbA1c, insulin binding, lactate and lipids.Outcomes

Notes


Wulffele 2000Study

TRIAL DESIGN: Abstract of randomised controlled trial, placebo-controlledDURATION: 4 months

Methods

COUNTRY: NetherlandsSETTING: outpatientTreatment N: 95Control N: 95AGE: not listedSEX: not listedINCLUSION: Type 2 DM treated with insulinEXCLUSIONS: none listed

Participants

Página 93




TREATMENT: Metformin, dosage unclear, + insulinCOMPARISON: placebo + insulin

Interventions

Daily dose insulin, and HbA1.Outcomes

Notes


Wulffele 2002Study


Methods

COUNTRY: The NetherlandsSETTING: outpatientTreatment N: 171Control N: 182Age: 60 +/- 10Sex: 45% m4nInclusion: type 2 DM controlled with insulinExlcusions: renal insufficiency with GFR < 50, congestive heart failure

Participants

TREATMENT: metformin, dose adjusted clinically, + insulinCOMPARISON: placebo + insulin

Interventions

Insulin requirements, lipid profile, glycemic controlOutcomes

Notes


Wulffele 2003Study


Methods

COUNTRY: The NetherlandsSETTING: outpatientTreatment N: 196Control N; 194Inclusion: type 2 DMExlcusions: renal insufficiency with GFR < 50, congestive heart failure, pregnancy

Participants

TREATMENT: metformin, dose adjusted clinicallyCOMPARISON: placebo

Interventions

Homocystein, folate, vitamin B12, body weight, glycemic controlOutcomes

Notes


Wulffele 2005Study


Methods

Página 94




COUNTRY: THe NetherlandsSETTING: outpatientTreatment N: 89Control N: 93Age: 58 +/- 11Sex: 48% menInclusion: type 2 DMExclusions: congestive heart failure, serious illness, renal insufficiency with GFR < 50

Participants

TREATMENT: metformin 2.5 g/dayCOMPARISON: placebo

Interventions

Systolic, diastolic, mean blood pressure, 24-hour blood pressureOutcomes

Notes


Yamanouchi 2005Study


Methods

COUNTRY: JapanSETTING: outpatientTreatment N: 39Control N: 75Age: 55.4 +/- 9Sex: 55% menInclusion: newly diagnosed type 2 DMExclusions: standard

Participants

TREATMENT: metformin750 mg/dayCOMPARISON: pioglitazone 30-45 mg/day or glimepiride 1-2 mg/day

Interventions

Fasting glucose, free fatty acid, HbA1c, blood pressure, lipid profileOutcomes

Notes


Yki-Jarvinen 1999Study

TRIAL DESIGN: randomised controlled trialDURATION: 1 year

Methods

COUNTRY: FinlandSETTING: multicenterTreatment N: 48Control N: 48AGE: 58+/-1SEX: not listedINCLUSION: Poorly controlled type 2 DMEXCLUSIONS: congestive heart failure, liver diseae, creatinine > 120

Participants

TREATMENT: Metformin, dosage adjusted clinically, + placebo or metformin + glyburideCOMPARISON: insulin + glyburide + placebo or BID insulin

Interventions

Weight, HbA1, plasma glucose, insulin, lipids.Outcomes

Página 95




Notes


Yu 1999Study

TRIAL DESIGN: randomised controlled trialDURATION: 4 weeks

Methods

COUNTRY: United StatesSETTING: research laboratoryTreatment N: 10Control N: 10Treatment AGE: 49+/-9Control AGE: 51+/-9Treatment SEX: 70% menControl SEX: 80% menINCLUSION: Type 2 DM with suboptimal contolEXCLUSIONS: renal or liver abnormalities

Participants

TREATMENT: Metformin 1-2.5 g/dayCOMPARISON: troglitazone

Interventions

Fasting glucose, insulin sensitivity.Outcomes

Notes


Notas:BID= two times a day; BMI=body mass index; DM=diabetes mellitus; TID=three times a day

Characteristics of excluded studies

Reason for exclusionStudy

TRIAL DESIGN: Retrospective analysis studyAguilar 1992b

TRIAL DESIGN: Prospective cohort, with varying durations of treatmentBernard 1965

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthBonfigli 1999

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthBruneder 1978

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthCacciapuoti 1991

TRIAL DESIGN: Prospective cohort trial, that did not give length of treatmentChow 1995

TRIAL DESIGN: Prospective cohort trial, that did not give lenght of treatmentClauson 1996

TRIAL DESIGN: Retrospective analysis studyConnolly 1996

TRIAL DESIGN: Retrospective meta-analysisDaniel 1997

TRIAL DESIGN: Prospective cohort study, of varying durationsDebry 1964

TRIAL DESIGN: Retrospective cohort study, with no durations givenDebry 1966a

TRIAL DESIGN: Retrospective cohort study, with no durations givenDebry 1966b

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthFery 1997

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthGaluska 1994

Página 96




TRIAL DESIGN: Prospective comparative trial, as part of another UKPDS trial, with patientsstudied less than 1 month

Gibson 1995

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthGin 1982

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthGin 1985

TRIAL DESIGN: Prospective cohort study, lasting less than 1 monthGin 1989

TRIAL DESIGN: Prospective cohort study, lasting less than 1 monthGiugliano 1979

TRIAL DESIGN: Retrospective meta-analysisGuthrie 1997

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthIrsigler 1978

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthIsmail 1978

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthIsnard 1991

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthIsnard 1996

TRIAL DESIGN: Prospective cohort study, lasting less than 1 monthJansson 1996

TRIAL DESIGN: Retrospective meta-analysisJohansen 1999

TRIAL DESIGN: Retrospective analysisLalau 1994

TRIAL DESIGN: Retrospective analysisLalau 1995

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthLeslie 1987

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthLim 1970

TRIAL DESIGN: Prospective cohort study of varying durationsMessens 1965

TRIAL DESIGN: Prospective cohort study of varying durationsMessens 1966

TRIAL DESIGN: Prospective cohort study, of varying durationsMuntoni 1965

TRIAL DESIGN: Prospective cohort study, of unclear durationNauck 1993

TRIAL DESIGN: Retrospective review of 4 trialsNauck 1997

TRIAL DESIGN: Retrospective meta-analysisO'Connor 1998

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthOrlikowska 1966

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthPanahloo 1995

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthPerriello 1994

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthPilger 1978

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthPrager 1983

TRIAL DESIGN: Prospective cohort study of varying durations.Rambert 1961

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthRigas 1968

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthRizkalla 1986

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthSambol 1996

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthScarpello 1998

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthSchaffalitzky 1979

TRIAL DESIGN: Retrospective analysisSelby 1999

TRIAL DESIGN: Prospective cohort study, lasting less than 1 monthSignore 1996

Página 97




TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthSlama 1984

TRIAL DESIGN: Prospective cohort study, of unclear durationStefanovic 1999

TRIAL DESIGN: Prospective cohort study, of varying durationSugawara 1962

TRIAL DESIGN: Prospective cohort study, lasting less than 1 monthSum 1992

TRIAL DESIGN: Prospective cohort study, of unclear durationTeitelbaum 1963

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthTrischitta 1983

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthTurner 1995

TRIAL DESIGN: Prospective comparative trial, lasting less than 1 monthZapecka-Dubno 1999

CARÁTULA

Riesgo de acidosis láctica fatal y no fatal con el uso de metformina parala diabetes mellitus tipo 2

Titulo

Salpeter S, Greyber E, Pasternak G, Salpeter EAutor(es)

SHELLEY SALPETER : Desarrollo del protocolo, selección de ensayos,evaluación de la calidad de los ensayos, extracción de datos, análisis de datos,preparación del texto, manejo de las referencias.

Contribución de los autores

ELIZABETH GREYBER: Estrategia de búsqueda, evaluación de la calidad delos ensayos, extracción de los datos, preparación del manuscrito.

GARY PASTERNAK: Selección de los ensayos.

EDWIN SALPETER: Análisis de los datos, evaluación estadística.

2001/1Número de protocolo publicadoinicialmente

2002/2Número de revisión publicadainicialmente

16 noviembre 2005Fecha de la modificación másreciente"

16 noviembre 2005"Fecha de la modificaciónSIGNIFICATIVA más reciente

El autor no facilitó la informaciónCambios más recientes

El autor no facilitó la informaciónFecha de búsqueda de nuevosestudios no localizados

El autor no facilitó la informaciónFecha de localización de nuevosestudios aún noincluidos/excluidos

Página 98



01 agosto 2005Fecha de localización de nuevosestudios incluidos/excluidos

El autor no facilitó la informaciónFecha de modificación de lasección conclusiones de losautores

Prof Edwin SalpeterCenter for Radiophysics and Space ResearchCornell University612 Space Sciences BuildingIthaca14853NYUSATélefono: +1 607 2554937E-mail: [email protected]: +1 607 2553433

Dirección de contacto

CD002967Número de la Cochrane Library

Cochrane Metabolic and Endocrine Disorders GroupGrupo editorial

HM-ENDOCCódigo del grupo editorial

RESUMEN DEL METANÁLISIS

01 Fatal/nonfatal lactic acidosis

Tamaño del efectoMétodo estadísticoNº departicipantes

Nº deestudios

Resultado

0.00 [0.00, 0.00]Diferencia de riesgo(efectos fijos) IC del 95%

6296014801 Incidencia de acidosis lácticapor paciente por año (metforminamenos sin metformina)

02 Niveles del lactato sanguíneo

Tamaño del efectoMétodo estadísticoNº departicipantes

Nº deestudios

Resultado

0.12 [-0.01, 0.25]Diferencia de mediasponderada (efectos fijos) ICdel 95%

222701 Efecto neto del tratamiento,niveles de lactato (mmol/L)

-0.09 [-0.13, -0.05]Diferencia de mediasponderada (efectos fijos) ICdel 95%

15471902 Niveles medios de lactato conel tratamiento (mmol/L)

0.08 [-0.05, 0.20]Diferencia de mediasponderada (efectos fijos) ICdel 95%

92403 Niveles pico de lactatoestimulado (mmol/L)

Página 99



GRÁFICOS Y OTRAS TABLAS

Fig. 01 Fatal/nonfatal lactic acidosis

01.01 Incidencia de acidosis láctica por paciente por año (metformina menos sin metformina)

Página 100



Fig. 02 Niveles del lactato sanguíneo

02.01 Efecto neto del tratamiento, niveles de lactato (mmol/L)

Página 101



02.02 Niveles medios de lactato con el tratamiento (mmol/L)

Página 102



02.03 Niveles pico de lactato estimulado (mmol/L)

Página 103



revision sistematica riesgo de acidosis con tratamiento metformina

Documents