osteoporosis in chronic kidney disease
TRANSCRIPT
Oswdifriotoi
Kfpdd((tsilagtaafdp
HHU
Utd
5
Osteoporosis in Chronic Kidney Disease
John Cunningham, DM, FRCP, and Stuart M. Sprague, DO, on behalf of the Osteoporosis WorkGroup (Jorge Cannata-Andia, MD, Maria Coco, MD, Martine Cohen-Solal, MD,
Lorraine Fitzpatrick, MD, David Goltzmann, MD, Marie-Helene Lafage-Proust, MD,Mary Leonard, MD, Susan Ott, MD, Mariano Rodriguez, MD, PhD,
Catherine Stehman-Breen, MD, MS, Paula Stern, MD, and Jose Weisinger, MD)
pprams(oo
ptt(htsmtfttaoum
D
imwmtsTcpSaag
UR UNDERSTANDING of the pathogene-sis of bone disorders has progressed con-
iderably over the past 20 years. In individualsith normal kidney function, the focus has pre-ominantly been on osteoporosis, whereas inndividuals with impaired kidney function, theocus has been on the condition referred to asenal osteodystrophy. This pathologic understand-ng has led to significant advances in the therapyf bone disease as it affects these patient popula-ions. Although these advances have tended toccur in parallel, there has been relatively littlentegration between the two.
The work group on Osteoporosis in Chronicidney Disease attempted to deal with the inter-
ace between osteoporosis and renal osteodystro-hy. We recognized a major problematic issue ofefining osteoporosis in relationship to the boneisease observed in the chronic kidney diseaseCKD) patient. The National Institutes of HealthNIH) consensus statement on osteoporosis con-ains minimal reference to CKD as a cause ofecondary osteoporosis. This lack of integrations important for several reasons. First, both fieldsikely can learn from advances made in the other;nd second, assessment and management strate-ies may exhibit exploitable areas of overlap. Inhe clinical arena in particular, the increasingvailability of dual x-ray absorptiometry (DEXA)nd quantitative computed tomography (qCT)acilities for the measurement of bone mineralensity (BMD) in both the CKD and non-CKDopulations leads to the frequent identification of
From the Middlesex Hospital, University College Londonospitals, London, England; and Evanston Northwesternealthcare, Feinberg School of Medicine, Northwesternniversity, Evanston, IL.Address reprint requests to John Cunningham, DM, FRCP,
niversity College London Hospitals, The Middlesex Hospi-al, Mortimer St, London W1T 3AA, UK. E-mail:[email protected]© 2004 by the National Kidney Foundation, Inc.0272-6386/04/4303-0023$30.00/0
rdoi:10.1053/j.ajkd.2003.12.004
American Journal of66
atients with low BMD. In the case of the CKDatient with coexisting low BMD, the resultingeferral path may be confusing, leading either tonephrologist whose focus may not be on BMDeasurements and the management of osteoporo-
is, or conversely to an “osteoporosis” physicianrheumatologist, endocrinologist, gynecologist,r geriatrician) whose focus may not be on renalsteodystrophy and its management.The likelihood of the patient entering an appro-
riate pathway is further undermined by some ofhe terminology used. Thus, a measured BMDhat satisfies the World Health OrganizationWHO) definition of osteopenia or osteoporosisas a tendency to lead to the immediate applica-ion of that diagnostic label, even in circum-tances in which the underlying pathogenesisay be one of the specific forms of renal os-
eodystrophy. This, in turn, creates a tendencyor the management of such patients to focus onhe osteoporosis/osteopenia rather than on therue underlying disease. Hyperparathyroidism ordynamic bone disease may, therefore, go unrec-gnized with the potential for exacerbation ofnderlying disease by inappropriate manage-ent and treatment.
SUMMARY OF GROUP DISCUSSION
efinitions and Terminology
Definitions of osteoporosis vary considerablyn their applicability to the CKD patient. Theost widely used definition is that of the WHO,hich defines osteoporosis on the basis of BMDeasurements applied to the lumbar spine or to
he femoral neck. According to the WHO, a Tcore of �1.0 to �2.5 defines osteopenia, and a
score below �2.5 defines osteoporosis. Thislassification schema helps stratify appropriateatients according to fracture risk (Table 1).trictly speaking, however, these definitions arepplicable only to Caucasian women, and theirpplicability to men, children, and other ethnicroups and all cases of secondary osteoporosis
emains uncertain. The definition of osteoporosisKidney Diseases, Vol 43, No 3 (March), 2004: pp 566-571
dwedcitsddatmtsTfrd
vCtCpttpcimfttpiop
cfbqnFtpbdbsnpmf(fdrm
S
ovtvphdtrtcal
N
O
O
S
CONTROVERSIES IN MINERAL METABOLISM: OSTEOPOROSIS 567
eveloped by the NIH in its consensus statementas considered by the group to have more rel-
vance to the CKD patient. Here, osteoporosis isefined as “a skeletal disorder characterized byompromised bone strength predisposing to anncreased risk of fracture. Bone strength reflectshe integration of two main features—bone den-ity and bone quality.” The implication of thisefinition is that bone density is only one of theeterminants of bone strength and fracture risknd that other quality issues such as architecture,urnover, damage accumulation and repair, andineralization are all relevant. Unfortunately,
he latter are often exceedingly difficult to mea-ure, particularly in the routine clinical setting.his difficulty is likely to be one of the reasons
or inappropriate emphasis being placed on theelatively straightforward measurement of boneensity per se.Therefore, the term osteoporosis probably has
ery limited diagnostic meaning in the context ofKD. Furthermore, the WHO groupings are ex-
remely unlikely to be applicable in patients withKD, based as they are on different patientopulations. The work group recognized thathere are no independent data based upon BMDhat demonstrate clinical outcomes in the CKDopulation. Thus, the group considered that thelassification of bone disease should be based onntegration of bone quantity, which may be esti-ated from BMD, and bone quality, which is a
unction, in part, of bone turnover. Because “os-eoporosis” may be a misleading term, implyinghat BMD measurements are the only importantredictor of fractures, the work group consideredt more appropriate to describe this constellationf bone disorders in CKD as “renal osteodystro-
Table 1. WHO and B
Classification Definition
ormal Bone density is no morethe young adult norma
steopenia (low bone mass) Bone density is 1 SD to 2the young adult normato �2.5 SDs).
steoporosis Bone density �2.5 SD badult normal value (�2
evere osteoporosis Bone density �2.5 SD badult normal value and1 or more fragility fract
hy.” This takes account of the notion that the p
ompromised bone strength in CKD patients is aunction of bone turnover (assessed by boneiopsy), bone density (assessed by DEXA orCT), and bone architecture (for which there areo current in vivo measurement technologies).urthermore, we proposed that the categoriza-
ion of bone density be simplified by classifyingatients into 2 groups, namely those with “lowone density” and those with “normal/high boneensity.” Thus, a patient may have high-turnoverone disease with low or normal/high bone den-ity, or low-turnover bone disease with low orormal/high bone density. Furthermore, we pro-osed the standardization of BMD measure-ents be based on Z-score (bone mass adjusted
or race, age, and sex) as opposed to T-scoreadjusted to peak mass of a young adult adjustedor sex and race). A low bone density would beefined as a Z-score of �1 or less. These latterecommendations were controversial and did noteet with universal approval.
pecific Issues in Relation to CKD
The understanding of the pathology of renalsteodystrophy in patients with CKD has ad-anced greatly over the past 10 years. In parallel,here has been somewhat less spectacular ad-ance in our understanding of the associatedathophysiology, partly because of the enormouseterogeneity of bone and mineral disorders thatevelop in patients with CKD.1 In sharp contrasto the non-CKD patient with osteopenia/osteopo-osis, in the renal osteodystrophy patient there ishe potential for low bone mineral density tooexist with an enormous range of functionalbnormalities, from the very high-turnover boneesions seen in patients with uncontrolled hyper-
ensity Classification
Risk
SD below.
● Risk of fracture is very low.
below(�1 SD
● There is a 4� risk of fracture comparedto a normal BMD.
e young.
● There is an 8� risk of fracture comparedto a normal BMD.
e younghas been2.5 SD).
● There is a 20� risk of fracture comparedto a normal BMD.
one D
than 1l value.5 SDsl value
elow th.5 SD)
elow ththere
ures (�
arathyroidism to the profound reduction of bone
rnmt
oaatadstsdbpbtCimo
dmtmAfmbimaiusmbahrsafstsst
lBsbicbHastdpwssoaedbAhm
S
riamriwcudbtcmd
uatiwi
CUNNINGHAM AND SPRAGUE568
emodelling activity seen in patients with ady-amic bone disease. That the optimal manage-ent of low BMD is the same at these 2 ex-
remes of bone turnover is exceedingly unlikely.We considered available evidence in a range
f patient groups, namely CKD stages 2, 3, 4,nd 5 (although it is questionable whether predi-lysis and dialysis patients should be consideredogether), renal replacement therapy by hemodi-lysis, renal replacement therapy by peritonealialysis, and finally renal transplantation. Discus-ion also extended to the pediatric CKD popula-ion. These diverse treatment modalities for end-tage kidney disease likely have profoundlyiffering effects on the evolution of metabolicone disease. The hemodialysis population isrobably the best studied, with posttransplantone disease also receiving considerable atten-ion over the past 5 to 10 years. Patients withKD Stages 2, 3, and 4 remain relatively under-
nvestigated, particularly in regard to the skeletalanifestations of their disordered mineral metab-
lism.Imaging and bone measurements are currently
ominated by DEXA technology. Currently, thisethod is generally applied to spine and hip, but
he spine clearly presents problems in CKD, withisleading elevation of measured BMD in theP projection due to the aortic calcifications
requently observed in CKD patients. Measure-ent of BMD at the hip and radius may give
etter precision with fewer artifacts. Agreements needed on the optimum sites of DEXAmeasure-
ent in the CKD population. qCT has definitedvantages, particularly for spine measurementn CKD patients, as this technique can definencontaminated regions of interest (cortical ver-us trabecular), which will provide meaningfuleasurements of bone density. The distinction
etween cortical and trabecular bone is clearlydvantageous. The increased remodeling rate inigh-turnover disease results in thickened, scle-otic trabeculae, while low-turnover disease re-ults in abnormally thin trabeculae. Tradition-lly, assessment of renal osteodystrophy hasocused on trabecular morphology, to the exclu-ion of cortical bone—illogical in that hyperpara-hyroidism, for example, may cause increasedubperiosteal, endocortical, and intracortical re-orption resulting in cortical bone loss on mul-
2
iple surfaces. pWe considered at length the implications ofow BMD in a patient with CKD. Correlations ofMD with bone histology are poor. BMD mea-
urement per se has no role in the assessment ofone turnover. Although fracture risks are greatlyncreased in CKD patients, fracture risk does notorrelate well with DEXA measurements of tra-ecular BMD in patients with kidney disease.2-8
emodialysis patients with vertebral fracturesnd fragility fractures have comparable lumbarpine BMD, compared to those without frac-ures.9,10 Certainly a BMD measurement aloneoes not constitute a diagnosis, but rather formsart of the overall data set required in a patientith CKD and bone disease. In particular, BMD
hould be viewed in the context of the clinicalcenario, laboratory variables, and bone histol-gy, if available.11 The group also considered thepplication of well-established biochemical mark-rs and agreed that there were limitations, un-oubtedly due to the mixed nature of the boneiology as well as effect of renal clearance.12
ssessment of calcium, phosphorus, parathyroidormone (PTH), and bone alkaline phosphataseay have limited utility.
pecial Clinical Groups
Pediatrics. The pediatric CKD populationaises particular issues and challenges. Growthn children with kidney disease is poor and,lthough effectively treated with growth hor-one, the use of this therapy is sporadic. The
elationship between bone turnover and growthn children is unclear. There is a major problemith slowing of bone growth, in particular corti-
al thickness, at puberty, and after puberty, “catchp” growth is likely to be incomplete. Thisecrement has an inordinately large effect onone strength, because bone strength is related tohe 4th power of the periosteal radius. Furtheromplicating factors include hypogonadism andetabolic acidosis, both of which alter growth
ynamics.Diabetes. In patients with CKD, there is little
nderstanding of the distinction between type 1nd type 2 diabetes. Diabetic patients exhibit aendency for low-turnover bone lesions, but thiss not invariable. There is also a tendency to-ards relatively low PTH, but again, this is not
nvariable. Fracture risk is increased in posttrans-
lantation diabetic patients relative to nondia-
bdstdfi
op2nenoibopstarmacp
N
base
teeeaotitTweTaaTtaputm
pisottsshrepwitibaisb
sifyi
CONTROVERSIES IN MINERAL METABOLISM: OSTEOPOROSIS 569
etic graft recipients.13 Relative fracture risk ofiabetic versus nondiabetic patients in CKDtages 3 to 5 is unclear. Appropriate strategies forhe management of established adynamic boneisease in diabetic patients have not been de-ned.Therapies. The use of standard therapies for
steoporosis used in the general population inatients with CKD is highly controversial (Table). Agents considered included the bisphospho-ates (oral and intravenous), estrogen, selectivestrogen receptor modulators (SERMs), calcito-in, and androgens/anabolic steroids. The impactf these agents on BMD in CKD is uncertain, ass the effect of these treatments on bone turnover,one quality, and fracture risk. Safety issues aref concern with estrogen, SERMs, and bisphos-honates. Calcitonin is probably safe in thisetting, but this therapy has to be considered inhe light of uncertain efficacy. None of thesegents has licensed indications within the field ofenal osteodystrophy in the United States or inost European countries. At present it is safest
nd likely most efficacious to focus initially onorrecting bone turnover abnormalities in CKDatients.
ewer Approaches
PTH/calcimimetics/calcilytics. There haseen a resurgence of interest in the use of PTH asbone anabolic agent in patients with osteoporo-
is. Recent studies have added further support to
Table 2. Use of Traditional OsteoporosisTherapeutic Agents in CKD
EstrogenPotential use in hypogonadismSafety data lackingIncreased drug half-lifeEfficacy unknown—especially long-term use
Selective estrogen receptor modulators (SERMs)Safety data lackingEfficacy unknown
BisphosphonatesEfficacy unknown—especially long-term useSafety data lackingTheoretically dangerous in adynamic bone diseaseShould have a bone biopsy
CalcitoninEfficacy unknownProbably safe
arlier work pointing to the efficacy of intermit- o
ent parenteral PTH on improving BMD.14 How-ver, the scenario in renal osteodystrophy isntirely different. PTH is already substantiallylevated in many patients and is thought to exertnet catabolic effect on the skeleton. The role oftherwise normal PTH pulsatility is unclear inhe CKD environment. Nevertheless, the possibil-ty exists for creating pulsatile PTH secretion inhese patients by a number of different means.hus, low calcium dialysate, possibly alternatingith high calcium dialysate, is clearly capable of
ffecting abrupt changes to PTH concentration.he effects on bone are unknown. Calcimimeticgents, used cyclically, are likely to be capable ofchieving brisk and transient reduction of PTH.he use of calcilytic agents alone or in combina-
ion with calcimimetics might achieve greatermplitude of PTH pulsatility, but again this ap-roach is currently speculative. Therefore, these of exogenously administered PTH as aherapy for renal osteodystrophy is not recom-ended.Kidney transplantation. The postrenal trans-
lant environment is deleterious to the skeletonn a number of ways and is usually applied to thekeleton on a background of pre-existing renalsteodystrophy associated with CKD. The pat-ern of bone loss following kidney transplanta-ion is brisk in the early phase, with a tendency totabilize after 1 year. Evidence from a number oftudies indicates that fracture rates are evenigher in transplanted patients than in compa-able dialysis patients, and this increase is great-st in diabetic recipients of kidney/pancreas trans-lants.13,15-18 Compared with normal subjects,omen aged 25 to 44 experience an 18-fold
ncrease in fracture rate following renal transplan-ation, and in the age range of 45 to 64 years thencrease is 34-fold.18 Histological studies haveeen conducted in the posttransplant subgroupnd a variety of lesions have been identified,ncluding high turnover and low turnover le-ions. Neither of these lesions is well predictedy PTH concentration.19,20
In the phase of rapid bone loss, histologicaltudies suggest that bone resorption rates arencreased and, in relative terms at least, boneormation is decreased. Bone turnover after 1ear is variable and there is marked heterogene-ty between and within patient groups.20 Choice
f therapy is difficult, therefore, and suggests
ttbtc
cafnpbWrdgpiwcpchte
A
tc
ccgbtaifeammttfnaast
dcattiiaot
Cgaodntisbtntpa
CUNNINGHAM AND SPRAGUE570
hat strategies will need to be individualized ifhe variation in underlying bone pathology is toe matched with appropriate therapies. Of course,his presupposes that the underlying pathologyan be characterized adequately.
Therapies assessed in this patient group in-lude bisphosphonates, vitamin D, calcitonin,nd estrogen (Table 3). The strongest data existor bisphosphonate therapy. Pamidronate, ibandro-ate, or zoledronate given at the time of trans-lantation effectively retard and largely preventone loss in the early posttransplant period.21-23
hether this approach also decreases fractureisk is unclear, nor is it clear whether potentiallyeleterious ultrastructural effects on bone areenerated. Bisphosphonate use in the renal trans-lant recipient has become quite commonplacen a number of countries, despite the relativelyeak evidence base. Safety issues remain a con-
ern, and in most countries, the use of bisphos-honates in renal transplant recipients is unli-ensed. Data for calcitriol, calcitonin, and gonadalormone use in posttransplant patients are rela-ively sparse and no clear recommendations havemerged.
reas Requiring Further Investigation
The consensus of the work group was thathere is an urgent need for more studies, espe-
Table 3. Therapeutic Approaches AfterRenal Transplant
Minimize glucocorticoid useCalcineurin inhibitor
Role controversialProbably minimize use
CalciumProbably benignMay be helpfulEffectiveness unproven
Vitamin D or CalcitriolMay be helpfulEffectiveness unprovenRisk of hypercalcemia
BisphosphonatesProtect BMD early onTheoretically reasonable for high turnover onlyFracture outcomes/long-term safety both unknown
CalcitoninFew data
Gonadal hormonesFew data
ially those with bone biopsy data. There is a i
ritical yet unknown linkage between BMD andlinical end points. Thus, studies are also ur-ently required to investigate the relationshipetween BMD and fracture, BMD and boneurnover, BMD and growth, and, finally, BMDnd vascular calcification. Clinical outcome stud-es in CKD and transplantation are relativelyew, and future studies should be powered toxamine the impact of therapies on fracture ratend the diagnostic utility of noninvasive assess-ents of bone metabolism such as biochemicalarkers. The role of recurrent CKD in the failing
ransplant setting is unknown; it is likely thathese patients often receive inadequate treatmentor their recurring renal osteodystrophy. Hypogo-adism is highly prevalent in both transplantednd dialyzed patients and is underinvestigatednd undertreated. Opportunity and need exist fortudies to evaluate the role of sex steroid interven-ion on preventing bone loss in CKD patients.
There is a great opportunity to advance theatabase in the posttransplant patient group be-ause many new immunosuppressive protocolsre being evaluated, in most cases sponsored byhe pharmaceutical industry. Transplanted pa-ients are extremely vulnerable to skeletal morbid-ty, and these studies should, wherever possible,nclude measures to assess skeletal health. Thispproach would rapidly increase the size of theverall database and also draw out specific advan-ages and drawbacks of the new regimens.
CONCLUSION
The management of reduced BMD in theKD and posttransplant populations provides aood example of clinical practice tending to runhead of our understanding of mechanisms andf the clinical evidence base. The potential foroing harm is considerable and there is an urgenteed to strengthen the underpinning of the some-imes empirical management of these patients. Its critical to understand that BMD is only aingle assessment tool for the understanding ofone mass in renal osteodystrophy and the ex-rapolation of diagnostic categories of BMD ab-ormalities (eg, osteoporosis and osteopenia) inhe general population to the CKD population isroblematic. More data exist on the diagnosisnd treatment of abnormalities of bone turnover
n CKD patients. Clinicians are encouraged to
fr
N
bf
iA
vt
fD
rd
hhf3
fJ
a
tdf1
hdK
m5
fJ
pdJ
oJ
ri
bR
It
pk2
Bo1
tt
it
a
CONTROVERSIES IN MINERAL METABOLISM: OSTEOPOROSIS 571
ocus first on correction of this component ofenal osteodystrophy.
REFERENCES1. Hruska KA, Teitelbaum SL: Renal osteodystrophy.Engl J Med 333:166-174, 19952. Schober HC, Han ZH, Foldes AJ, et al: Mineralized
one loss at different sites in dialysis patients: Implicationsor prevention. J Am Soc Nephrol 9:1225-1233, 1998
3. Coco M, Rush H: Increased incidence of hip fracturesn dialysis patients with low serum parathyroid hormone.m J Kidney Dis 36:1115-1121, 20004. Piraino B, Chen T, Cooperstein L, et al: Fractures and
ertebral bone mineral density in patients with renal osteodys-rophy. Clin Nephrol 30:57-62, 1988
5. Atsumi K, Kushida K, Yamazaki K, et al: Risk factorsor vertebral fractures in renal osteodystrophy. Am J Kidneyis 33:287-293, 19996. Alem AM, Sherrard DJ, Gillen DL, et al: Increased
isk of hip fracture among patients with end-stage renalisease. Kidney Int 58:396-399, 20007. Bruce DG, St John A, Nicklason F, et al: Secondary
yperparathyroidism in patients from Western Australia withip fracture: Relationship to type of hip fracture, renalunction, and vitamin D deficiency. J Am Geriatr Soc 47:354-59, 19998. Ball AM, Gillen DL, Sherrard D, et al: Risk of hip
racture among dialysis and renal transplant recepients.AMA 288:3014-3018, 2002
9. Stehman-Breen C: Risk of hip fracture among dialysisnd renal transplant recipients. JAMA 288:3014-3018, 2002
10. Yamaguchi T, Kanno E, Tsubota J, et al: Retrospec-ive study on the usefulness of radius and lumbar boneensity in the separation of hemodialysis patients withractures from those without fractures. Bone 19:549-555,99611. Jamal SA, Chase C, Goh YI, et al: Bone density and
eel ultrasound testing do not identify patients with dialysis-ependent renal failure who have had fractures. Am J
idney Dis 39:843-849, 2002 t12. Urena P, De Vernejoul MC: Circulating biochemicalarkers of bone remodeling in uremic patients. Kidney Int
5:2141-2156, 199913. Chiu MY, Sprague SM, Bruce DS, et al: Analysis of
racture prevalence in kidney-pancreas allograft recipients.Am Soc Nephrol 9:677-683, 199814. Neer RM, Arnaud CD, Sanchetta JR, et al: Effect of
arathyroid hormone (1-34) on fractures and bone mineralensity in postmenopausal women with osteoporosis. N EnglMed 344:1434-1441, 200115. Julian BA, Laskow DA, Dubovsky J, et al: Rapid loss
f vertebral mineral density after renal transplantation. N EnglMed 325:544-550, 199116. Almond NK, Kwan JTC, Evans K, et al: Loss of
egional bone mineral density in the first 12 months follow-ng renal transplantation. Nephron 66:52-57, 1994
17. Horber FF, Casez JP, Steiger U, et al: Changes inone mass early after kidney transplantation. J Bone Mineres 9:1-9, 199418. Ramsey-Goldman R, Dunn JE, Dunlop DD, et al:
ncreased risk of fracture in patients receiving solid organransplants. J Bone Min Res 14:456-463, 1999
19. Monier-Faugere MC, Mawad H, Qi Q, et al: Highrevalence of low bone and occurrence of osteomalacia afteridney transplantation. J Am Soc Nephrol 11:1093-1099,00020. Cueto-Manzano A, Konel S, Hutchinson AJ, et al:
one loss in long-term renal transplantation: Histopathol-gy and densitometry analysis. Kidney Int 55:2021-2029,99921. Fan SLS, Almond MK, Ball E, et al: Pamidronate
herapy as prevention of bone loss following renal transplan-ation. Kidney Int 57:684-690, 2000
22. Grotz W, Nagel C, Poeschel D, et al: Effect ofbandronate on bone loss and renal function after kidneyransplantation. J Am Soc Nephrol 12:1530-1537, 2001
23. Haas M, Leko-Mohr Z, Roschger P, et al: Zoledroniccid to prevent bone loss in the first 6 months after renal
ransplantation. Kidney Int 63:1130-1136, 2003