Diabetes Mellitus and Associated Hypertension, Vascular Disease, and Nephropathy

An Update

1. James R. Sowers , 2. Murray Epstein

+Author Affiliations1. From Wayne State University (Detroit, Mich) and Miami (Fla) University Schools of

Medicine and VA Medical Centers.1. Correspondence to James R. Sowers, MD, Wayne State University School of Medicine, UHC-4H,

4201 St Antoine, Detroit, MI 48201. Next Section

Abstract

Abstract Because considerable important information has been published since our previous review, this update concentrates on new findings with regard to cardiovascular and renal risk factors contributing to the striking morbidity and mortality of these coexisting diseases. For example, a large body of investigative data has recently emerged suggesting or delineating a pathogenic role for hyperglycemic-related glycosylation and oxidation of lipoproteins and vascular and renal tissues. Great strides have recently been made in the understanding of platelet, coagulation, lipoprotein, and endothelial abnormalities in the pathogenesis of cardiovascular and renal disease associated with diabetes mellitus and hypertension. Major progress has been made in clarifying the pathophysiology of glomerulosclerosis and other processes involved in the progression of diabetic nephropathy. Furthermore, accumulating data surveyed in this review address new and promising pharmacological interventions that specifically address these pathophysiological mechanisms.Key Words:

diabetes mellitus cardiovascular disease diabetic nephropathy

Diabetes mellitus and hypertension are interrelated diseases that strongly predispose an individual to atherosclerotic cardiovascular disease.1 2 An estimated 3 million Americans have both diabetes and hypertension.2 Hypertension is about twice as frequent in individuals with diabetes as in those without.2 Lifestyle and genetic factors are important factors contributing to both hypertension and diabetes mellitus. The prevalence of coexisting hypertension and diabetes appears to be increasing in industrialized nations because populations are aging and both hypertension and NIDDM incidence increases with age.1 2 Data obtained from death certificates show that hypertensive disease has been implicated in 4.4% of deaths coded to diabetes, and diabetes was involved in 10% of deaths coded to hypertensive disease.1 2 Indeed, an estimated 35% to 75% of diabetic cardiovascular and renal complications can be attributed to hypertension.1 2 Hypertension also contributes to diabetic retinopathy, which is the leading cause

of newly diagnosed blindness in the United States.2 For all these reasons, hypertension and diabetes should be recognized and treated early and aggressively.Essential hypertension accounts for the majority of hypertension in individuals with diabetes, particularly those with NIDDM (type II diabetes), who constitute more than 90% of people with a dual diagnosis of diabetes and hypertension.1 2 Hypertension often antedates and likely contributes to the development of nephropathy in many diabetic individuals.3 4 Diabetic nephropathy, which occurs after 15 years of diabetes in one third of people with IDDM (type I diabetes) and 20% of those with NIDDM, is an important contributing factor to the development of hypertension in the diabetic individual.1 2 The high BP associated with diabetic nephropathy is usually characterized by sodium and fluid retention and increased peripheral vascular resistance.1 2 Isolated systolic hypertension is considerably more common in diabetics, and supine hypertension with orthostatic hypotension is not uncommon in diabetic individuals with autonomic neuropathy.1 2

Increasing investigation has delineated an important role for several mechanisms acting together in mediating the pathogenesis of vascular disease in the diabetic hypertensive patient. We will review some of the more important mechanisms of cardiovascular and renal injury associated with diabetes mellitus and hypertension.

Previous Section Next Section

Mechanisms Contributing to Systemic Vascular Disease

Platelet Adhesion and Platelet AggregationPlatelet adhesion and platelet aggregation are often enhanced in both diabetes mellitus and

hypertension (Table 1⇓). The precise etiology of enhanced platelet reactivity in both disorders is complex, but it appears that abnormalities in platelet intracellular divalent cation metabolism may play an integral role. Both platelet intracellular calcium ([Ca2+]i) and magnesium ([Mg2+]i) have seminal roles in platelet activation.5 6 7 8 9 10 11 Platelet aggregation is associated with an elevation in [Ca2+]i, a necessary first event in the aggregation process. Enhanced platelet [Ca2+]i responses to LDL cholesterol have been observed in NIDDM patients with and without hypertension.6 Similarly, parallel hyperaggregation and platelet release reactions have been observed in NIDDM patients with and without hypertension.5 6In vitro, increased [Mg2+]i can exert an inhibitory effect on platelet aggregation.7There is considerable evidence that many patients with essential hypertension, as well as those with diabetes mellitus, have elevated platelet [Ca2+]i and decreased [Mg2+]i.7 8 9 Consequently, this altered balance between the relative intracellular concentrations of these divalent cations may contribute to the enhanced platelet aggregation in individuals with diabetes and hypertension.

View this table: In this window

  In a new window

Table 1.

Abnormalities of Platelet Function in Diabetes Mellitus and Hypertension

In diabetes mellitus the balance between coagulative and fibrinolytic activities in the circulation

is affected in a number of ways10 11 12 13 14 (Table 2⇓). A procoagulant state in diabetes appears to be mediated in part by higher-than-normal levels of a number of coagulation factors. For example, an increase in the endothelium-derived von Willebrand factor occurs in diabetes mellitus, particularly in association with endothelial cell injury,10 11 12 13 14 microvascular and macrovascular damage,10 and poor diabetic control.12 13 14 High concentrations of factor VIII are related to hyperglycemia, accelerate the rate of thrombin formation, and contribute to occlusive vascular disease in diabetic patients. Levels of fibrinogen, factor VII, and thrombin-antithrombin complexes have also been reported to be elevated in diabetic patients.11 Elevated coagulation levels of these factors, particularly fibrinogen, are important for increasing the survival of the provisional clot matrix on transformation of fibrinogen to fibrin at the site of injured endothelium.11 Indeed, increased levels of thrombin-antithrombin complexes have been observed in diabetic patients in association with enhanced thrombin generation.11 High PAI-1 levels have been observed in patients with diabetes mellitus.10 11 12 13 Furthermore, elevated PAI-1 levels have been reported in untreated hypertensive individuals13 and in men with myocardial infarction at risk for reinfarction. Elevated levels of PAI-1 also appear to be associated with elevated serum levels of insulin and triglycerides.10 11 13 Indeed, insulin has been shown to stimulate PAI-1 synthesis in hepatocytes.13 Thus, it appears that hyperinsulinemia and associated insulin resistance, in addition to being integral to syndrome X, are independent risk factors, along with diabetes and hypertension, for an abnormal balance between coagulation and fibrinolysis.14

View this table: In this window

  In a new window

Table 2.

Coagulation and Lipoprotein Abnormalities Seen in Patients With Hypertension and Diabetes Mellitus

Lipoprotein AbnormalitiesAlthough hypertension and diabetes mellitus are both independent risk factors for ischemic heart disease, insulin resistance and hyperinsulinemia associated with hypertension and NIDDM also

likely contribute to accelerated atherogenesis15 16 1718 (Table 2⇑). A number of metabolic

abnormalities are often present in patients with diabetes and hypertension (Table 2⇑). Plasma levels of lipoprotein(a) have been noted to be elevated in diabetic individuals, particularly those with poor glycemic control.16 By inhibiting fibrinolysis, possibly via fibrin binding attributable to structural homology with apolipoprotein(a), lipoprotein(a) may delay thrombolysis and thus

contribute to plaque progression. Augmented lipoprotein oxidation has also been observed in diabetic states.17 18 Ox-LDL is not recognized by the classic LDL receptor but by the macrophage scavenger receptors.17 18 Foam cell uptake of LDL via the scavenger pathway is enhanced by oxidative modification of LDL. Once taken up by the foam cell, however, Ox-LDL degradation is impaired, which leads to further accumulation in the cell. Ox-LDL is toxic to endothelial cells, altering both structure and function.18 Ox-LDL increases adhesion of circulating monocytes to damaged endothelium, increasing their migration into the vascular intima.18

Glycosylation, the nonenzymatic linkage of glucose to proteins, has also been found to alter LDL particles in vivo.18 Importantly, apolipoprotein B, which regulates receptor-mediated uptake of LDL, can undergo glycosylation, thereby facilitating its atherogenicity in diabetic individuals. Similar to Ox-LDL, glycated LDL can enhance foam cell formation and is less well recognized by the native LDL receptor.18 Glycated LDL is also immunogenic, forming antibody-lipoprotein complexes that stimulate foam cell formation and enhance platelet aggregation.18Compared with normal LDL, glycated LDL sequestered in the arterial intima has a greater propensity to become bound by glucose-mediated cross-links to local matrix proteins. Once this occurs, the LDL particles may undergo even more extensive glycative and oxidative modification.18 Finally, evidence is accumulating that glycation in itself may render the particle more susceptible than normal to further oxidative damage.18

Endothelial Dysfunction in Diabetes and HypertensionA number of anatomic and functional abnormalities of the vascular endothelium are associated

with both diabetes mellitus and hypertension19 (Table 3⇓). In insulin-resistant states, endothelial cell lipoprotein lipase activity is decreased, as is the conversion of cholesterol ester–enriched very-low-density lipoprotein to LDL. The resulting large and abnormal cholesterol ester–enriched very-low-density lipoprotein is injurious to endothelial cells after receptor-mediated uptake.19Hyperglycemia appears to contribute to endothelial dysfunction as well.19 20 21Hyperglycemia activates protein kinase C in endothelial cells,20 which in turn may account for increased production of vasoconstrictor prostaglandins, endothelia and ACE, and platelet and vascular growth factors, which directly and indirectly enhance vasomotor reactivity and vascular remodeling and growth.17 18 19 20 Furthermore, hyperglycemia alters endothelial cell matrix production, which may contribute to basement membrane thickening. Hyperglycemia increases endothelial cell collagen IV and fibronectin synthesis and increases the activity of enzymes involved in collagen synthesis.21 Hyperglycemia also delays cell replication and increases endothelial cell death in part by enhancing oxidation and glycation.18

View this table: In this window

  In a new window

Table 3.

Alterations in Vascular Endothelium Associated With Diabetes Mellitus and Hypertension

Additional metabolic and hemodynamic factors may contribute to endothelial dysfunction in diabetes and hypertension. Hypercholesterolemia and perhaps hypertriglyceridemia impair endothelium-dependent relaxation.22 Both insulin and IGF have potentially important effects on endothelial cells. Insulin appears to have a modulating influence on glucose stimulation of protein kinase C and diacylglycerol in endothelial cells.23 One hypothesis for endothelial dysfunction in diabetes relates to elevated protein kinase C activity in diabetic vascular endothelium, which may enhance vascular tone, permeability, and atherosclerosis.23 Elevated diacylglycerol and protein kinase C levels are induced by hyperglycemia; insulin treatment that achieves euglycemia can prevent the increase in diacylglycerol levels and protein kinase C activity.23 24 These results suggest that impaired insulin action, as exists in NIDDM and hypertension, may contribute to the endothelial dysfunction seen in these disorders. However, a recent report suggests that hypertension is not associated with impaired carbohydrate metabolism or dyslipidemia, but endothelium-dependent vasodilation is preserved.25 This report provides additional credence to the concept that the metabolic abnormalities that often accompany hypertension and diabetes mellitus play a pivotal role in the development of endothelial dysfunction.VSMC Abnormalities: Role of Insulin and IGF-1The concept that decreased insulin action on VSMCs may explain the exaggerated vascular resistance associated with both NIDDM and IDDM was carefully reviewed in our previous updates on diabetes mellitus and hypertension.1 2 14 Accordingly, the current material represents an update of the understanding of this important topic. Research conducted over the past several years has shown that VSMC tissue, like skeletal muscle and adipocytes, is sensitive to the metabolic effects of both insulin and IGF-1.14 26 27 28 29 30 In this regard, these studies have

confirmed that insulin and IGF-1 regulate VSMC cation metabolism (Fig 1⇓) and that both hormones stimulate VSMC glucose uptake and metabolism.14 28 30 Insulin and IGF-1 are structurally related, share receptors, and have similar postreceptor actions. Unlike insulin, which must traverse the endothelium before acting on VSMCs in vivo, IGF-1 is synthesized by VSMCs and is more likely to act in an autocrine and paracrine fashion. Importantly, recent investigations indicate that IGF-1, like insulin, attenuates vasoconstrictive responses and increases blood flow in regional vascular beds and lowers BP in healthy individuals.26 27 IGF-1 is expressed, synthesized, and secreted by VSMCs and appears to exert its actions on VSMCs in an autocrine fashion.14 IGF-1, like insulin,29 increases VSMC Na+,K+-ATPase activity, suggesting one

mechanism by which IGF-1 modulates VSMC [Ca2+]i (Fig 1⇓) and attenuates vascular contractility.14 30

View larger version: In this page

  In a new window

Download as PowerPoint Slide

Figure 1.Diagram shows mechanisms regulating divalent cation metabolism and contraction in VSMCs and proposed targets of insulin and IGF-1 action. Pivotal steps in regulation of divalent cation metabolism are indicated by circled numbers. DAG indicates diacylglycerol; IP3, inositol 1,4,5-trisphosphate.

Roles of Insulin and IGF-1 in AtherosclerosisInsulin and IGF-1 may exert their atherogenic effects through influences on both vascular endothelial cells and VSMCs.14 30 Both insulin and IGF-1 increase mitogenic signaling pathways and thymidine incorporation into DNA in vascular endothelial cells and VSMCs.14 24 30 Insulin likely mediates most of its VSMC proliferative effects through IGF-1 rather than insulin receptors.31 Both IGF-1 and IGF-2 enhance proteoglycan synthesis by microvascular and large-vessel endothelial cells.21 Insulin also increases the uptake and esterification of lipoprotein cholesterol by VSMCs and decreases its deesterification and release by these cells.14 25 Insulin resistance and hyperinsulinemia may also promote atherosclerosis by retarding the fibrinolytic process.12 13 14 Insulin stimulates PAI-1 production,12 and there is a strong relation between plasma insulin levels and fibrinogen and PAI-1 levels.12 13 Both insulin and IGF-1 appear to function as progression factors or cofactors and promote the proliferative properties of several cytokines, including tumor necrosis factor.14 24 Thus, elevated circulating levels of insulin, as exist in type II diabetes in many patients with essential hypertension, may contribute directly or in conjunction with IGF to the accelerated atherosclerosis associated with these conditions.Role of Hyperglycemia in the Vascular Abnormalities Associated With Diabetes and HypertensionChronic hyperglycemia may exacerbate the vascular disease associated with diabetes mellitus and hypertension.30 At high concentrations glucose has a direct, toxic effect (independent of osmolality) on vascular endothelial cells.19 20 21 This toxic effect may lead to decreased endothelium-mediated vascular relaxation, increased vasoconstriction, promotion of VSMC hyperplasia, vascular remodeling, and atherosclerotic events.14 30 High glucose concentrations, as seen in the diabetic hyperglycemic state, have been shown to induce the overexpression of fibronectin and collagen IV in cultured human vascular endothelial cells.32Increased expression of fibronectin and collagen IV further contributes to endothelial cell dysfunction. Fibronectin is a glycoprotein with a critical role in cell matrix interactions,32 and its overexpression may contribute to both diabetic vascular disease30 and thickening of the glomerular basement membrane and mesangial hyperplasia.32

Hyperglycemia also accelerates the formation of nonenzymatic advanced glycosylation products,

which accumulate in vascular tissue33 (Table 4⇓). There are a number of sites where nonenzymatic protein glycosylation can affect key processes in atherogenesis and vascular remodeling. Indeed, a close association has been noted between the accumulation of increased levels of AGE and vascular disease.33 Glycosylation of collagen results in increased rigidity and decreased responsiveness to collagenase (proteinase digestion).33 There is also abnormal covalent cross-linking and an enhanced ability to bind nonglycosylated proteins, such as LDL, albumin, and immunoglobulins. Lipoprotein binding to glycosylated collagen may not only prolong its intimal residence time but also enhance its susceptibility to oxidation within the arterial intima.33

View this table: In this window

  In a new window

Table 4.

Potential Vascular Effects of Nonenzymatic Protein Glycosylation in Diabetes Associated With Hypertension

A membrane-associated macrophage receptor that specifically recognizes proteins to which AGE are bound has been identified.33 The binding of these proteins to macrophage receptors induces the synthesis and secretion of tumor necrosis factor and interleukin-1.33 These cytokines in turn stimulate other cells to increase protein synthesis and to proliferate.33 AGE proteins also increase platelet-derived growth factor secretion, enhance endothelial cell permeability, and are chemotactic for blood monocytes.33 Apolipoproteins and LDL are glycosylated, and this process leads to enhanced cholesterol ester synthesis and accumulation in macrophages and impaired degradation and release.33 Thus, prolonged hyperglycemia may result in enhanced production of extracellular matrix and in proliferation of VSMCs as a result of an increase in the number of highly cross-linked proteins with AGE, with resulting vascular hypertrophy and remodeling. The fact that chronic hyperglycemia is associated with decreased elasticity of vascular arterial wall connective tissue and elevated pulse pressures may be related in part to AGE.30 33

Several other mechanisms associated with chronic hyperglycemia may also contribute to the hypertension frequently accompanying diabetes mellitus. For example, elevated glucose levels have been reported to increase VSMC [Ca2+]i,34which could result in increased vascular tone.2 14 30 Glucose can be extracted by renal proximal tubule cells by an active process in which sodium and glucose cotransporters are involved. Hyperglycemia results in hyperfiltration of glucose, which in turn stimulates the proximal tubular sodium-glucose cotransporter.35This process is insulin independent and rapidly operative, as elevated proximal tubular cell sodium concentration and increased Na+,K+-ATPase activity occur within days of inducing hyperglycemia.35 The resultant sodium retention caused by hyperglycemia can help explain the increased total exchangeable sodium seen in diabetic hypertensive patients.1 14

Previous Section Next Section

Metabolic Abnormalities and Renal Glomerular Disease: Analogy to Vascular Atherosclerosis

Pathophysiological changes of diabetic nephropathy are histologically different from those of other types of renal disease.1 36 37 In humans, mesangial expansion leads to the earliest lesions seen in IDDM.38 In experimental studies the earliest pathological change is an increase in the thickness of the glomerular basement membrane; usually the volume of the glomerulus is increased compared with that of nondiabetic subjects. Subsequently, the amount of matrix in the mesangium increases, and in some patients this mesangial expansion can progress to increasingly more severe diffuse or nodular glomerulosclerosis.36 Basement membranes are specialized regions of extracellular matrix composed of type IV collagen, laminin, entactin/nidogen, and proteoglycans, which together form a complex meshlike structure.37 Sievelike permselectivity is one important function of basement membrane that may be gradually lost in diabetes mellitus, with associated progressively increasing proteinuria.37 Nonenzymatic glycosylation of long-lived proteins, such as the basement membrane components type IV collagen and laminin, and cross-link formation of these components appear to lead to modification of basement membrane ultrastructure and loss of permselectivity.37AGE can also have important influences on renal mesangium. AGE binding to mesangial cells can lead to a significant increase in fibronectin and several other mesangial structural proteins.37 AGE binding to mesangial AGE receptors can lead to an increase in basement membrane collagen as well.37 Thus, hyperglycemia clearly contributes to the development of diabetic nephropathy.36

Mesangial cells share many properties with VSMCs. They are known to contract in response to vasoactive agents such as angiotensin II, vasopressive platelet-activating factor, and endothelin-1.1 19 39 40 41 Contraction of mesangial cells is important because the mesangium binds together capillary loops; contraction of the mesangium thus can alter capillary flow, pressures, or both. The mesangial cell synthesizes several growth factors that may act in an autocrine/paracrine fashion. These include IGF-1, platelet-derived growth factor, platelet-activating factor, endothelin, prostanoids, and interleukin-1. Thus, the mesangial cell has supportive, filtrative, and synthetic functions. The glomerular endothelial cell is separated from the mesangial cell of the kidney only by its basement membrane; passage of substances between these two cells is easily accomplished. Substances elaborated by the endothelium affect mesangial cell growth contraction and protein synthesis. For example, endothelin-1 increases growth and extracellular matrix production by mesangial cells. Endothelium-derived relaxing factor and vasodilator prostaglandins inhibit mesangial cell growth and contraction.39 Furthermore, increased local production of growth factors by stimulated endothelial and inflammatory cells can result in mesangial expansion. For example, endothelial damage can result in platelet activation, with release of platelet-derived growth factor and other growth factors that can also enhance mesangial cell proliferation and matrix overproduction.19 39

Mesangial cell abnormalities typically appear after 5 to 15 years of clinical diabetes in individuals with type I disease.36 37 38 The most common abnormality observed by light microscopy is diffuse intercapillary sclerosis due to expansion of the mesangium. Nodular

intercapillary sclerosis is seen in approximately 25% of patients with diabetic nephropathy.36 37 38 The pathophysiological alterations that occur with focal and diffuse glomerulosclerosis are similar to those that occur in vascular glomerulosclerosis, with mesangial cell changes paralleling those in VSMCs, including proliferation, hypertrophy, foam cell accumulation, appearance of extracellular matrix material, deposits of amorphous debris, and evolving sclerosis. The observed mesangial expansion is primarily related to mesangial cell hypertrophy and an accumulation of mesangial cell matrix. A number of hormones have been demonstrated to alter mesangial cell growth in vitro, but only angiotensin II and vasopressin have been clearly demonstrated to promote hypertrophy,19 36 37 38 which is more prominent in glomerulosclerosis than is proliferation. This may help explain why ACE inhibitors have been shown to slow the progression of glomerulosclerosis in both type I and type II diabetic patients.4243

Diabetic nephropathy has become the leading cause of end-stage renal disease in the United States.2 36 37 38 Hypertension is acknowledged to be a major risk factor in the progression of diabetic renal disease.2 Diabetic nephropathy, defined as the appearance of proteinuria, elevated arterial BP, and diminished GFR, will develop in as many as 40% of IDDM patients.2 36 37 38 In patients with the onset of diabetes at an early age, renal disease is an important contributor to mortality, accounting for up to 31% of all deaths.2 36 37 38 Renal disease also complicates NIDDM in adults and contributes significantly to morbidity and mortality in this group.2 36 A smaller percentage of patients with NIDDM develop renal disease, and the incidence of end-stage renal disease is strongly related to the duration of diabetes. However, more than 50% of diabetic end-stage renal disease is associated with NIDDM because this form of diabetes constitutes more than 90% of diabetic patients.2 36

In both types of diabetes mellitus the appearance of clinically detectable proteinuria (>200 μg/min of urinary albumin excretion) signals the onset of the relentless progression of diabetic nephropathy, which is typically followed by deterioration to end-stage renal disease.2 36 37 38 Currently, it is believed that diabetic nephropathy develops as a result of the interplay of the metabolic abnormalities inherent to diabetes (eg, hyperglycemia) and hemodynamic abnormalities of the renal microcirculation that result in progressive structural and functional glomerular abnormalities.36 37 38

Previous Section Next Section

Treatment of Hypertension in Patients with Diabetes Mellitus

General Goals of TherapyNo large, population-based, randomized trials of hypertension treatment in diabetic patients have been conducted. Nevertheless, as proposed by a recent consensus statement,2 the goal of treating hypertension in diabetic patients should be to prevent death and disability associated with high BP. In addition, other reversible risk factors for cardiovascular disease need to be addressed. For example, target-organ involvement should be considered when a treatment plan is being formulated. The major focus of clinical and investigative efforts has been on retarding the

progression of diabetic nephropathy (see below) and to a lesser extent reducing cardiovascular morbidity and mortality. The diagnosis of hypertension should be based on multiple BP measurements obtained in a standardized fashion on at least three occasions. Supine, sitting, and standing BPs should be measured in all diabetic patients. Automated ambulatory BP monitoring may be especially helpful in the diabetic patient for evaluating BP control over a 24-hour period to document the absence of the usual nocturnal fall in BP in diabetes (especially with autonomic dysfunction or nephropathy). Ambulatory BP monitoring may also be useful in documenting episodic hypertension, orthostatic hypotension, or resistant hypertension, which are relatively common in diabetic individuals with accompanying hypertension.2

What Is the BP Goal in Diabetics With Hypertension?Although the optimal BP level during antihypertensive treatment in patients with diabetic nephropathy has not been defined, a review of the relationship between the rate of fall in GFR and the BP level during antihypertensive treatments suggests that we should strive for lower goal BP than recommended by current guidelines.244 First, the benefit of reducing BP has been demonstrated most clearly when treatment is instituted before GFR is markedly reduced.44 This emphasizes the concept that efforts to reduce BP should begin before serum creatinine is elevated. Second, the best results apparently have been achieved by reducing BP below conventionally accepted levels.2 44 Indeed, the smallest decline in renal function is found in patients with BP levels around 130/85 mm Hg, a level that is readily attainable during treatment in incipient diabetic nephropathy before the decline in GFR has started. This is also the goal of

BP attainment recommended in a recent consensus2 44 (Fig 2⇓).

Figure 2.Chart shows suggested approach to hypertension therapy in diabetic individuals. Treatment goal is to maintain BP at less than 130/85 mm Hg. Diabetic renal disease, autonomic dysfunction, and adverse effects on glucose and lipid metabolism must be considered before the course of therapeutic intervention is chosen. (Adapted from Reference 22 with modification.)

Nonpharmacological Therapy in Diabetic Hypertensive PatientsLifestyle modifications may serve as definitive therapy for mild hypertension in diabetic patients or as an adjunct to pharmacological therapy to lower the number and dose of antihypertensive drugs.2 The diet recommended by the American Diabetes Association, which is low in calories and fat, high in carbohydrate and soluble fiber, and moderately low in protein, has been reported to lower BP in diabetic patients.2 Moderate salt restriction reduces systolic BP, which is often inordinately elevated in diabetic patients.2 Weight reduction is important, particularly in type II diabetics, and improves glucose tolerance as well as reducing BP.2 8 For example, for each 10-lb weight reduction, systolic and diastolic pressures can be expected to decrease by 10 and 5 mm Hg, respectively.2 8Moderate but regular aerobic exercise improves glycemic and lipemic control and helps with weight reduction.Results of the Diabetes Control and Complications Trial (DCCT),45 a 7-year study of more than 1440 patients, demonstrated that intensive insulin therapy reduced the occurrence of

microalbuminuria by 39% and that of albuminuria by 54%. In addition, intensified therapy patients had lower rates of serious retinopathy requiring photocoagulation, lower rates of decreased visual acuity, and fewer cases of nephropathy. Similar results were reported from a study designed to test the hypothesis that optimized glycemic control in type I diabetic recipients of renal allografts will prevent or delay diabetic renal lesions in the allograft. 46 This study was a prospective, controlled, and randomized trial of glycemic control in an inception cohort (ie, all patients were at stage 0 for diabetic renal lesions in the graft when randomized to the trial) of type I diabetic renal allograft recipients. The experimental group had maximized glycemic control, and the standard group received standard clinical diabetic care. Patients underwent baseline (before transplant) and 5-year posttransplant allograft biopsies. More than a twofold increase in the volume fraction of mesangial matrix per glomerulus occurred, as well as a threefold increase in arteriolar hyalinosis, greater widening of the glomerular basement membrane, and increase of volume fraction of the total mesangium in the patients receiving standard treatment compared with those with maximized glycemic control. This trial indicates a causal relationship between hyperglycemia and an important lesion of diabetic nephropathy, mesangial matrix expansion, in renal allografts transplanted into diabetic recipients. In summary, newly available data suggest that aggressive control of blood sugar in the very early stages of this disease process can provide significant protection against its development.Prospective clinical trials in type I diabetic patients with overt nephropathy47 48showed that even with moderate protein restriction, renal function is stabilized in diabetic nephropathy. Although both studies were limited to diabetic patients with overt nephropathy, it has been suggested that dietary protein restriction will produce even better results if implemented during the early stages of diabetic nephropathy. In recognition of these data, the American Diabetes Association has recommended that dietary protein be restricted to 0.8 g/kg body wt per day in all patients with diabetes other than children and pregnant or lactating women. It is still unclear to what extent dietary protein intake needs to be restricted to obtain maximal effects on delaying the progression of renal disease, but without producing metabolic side effects or malnutrition. For most diabetic patients, restricting dietary protein to 0.8 g/kg body wt per day would constitute a significant but practical reduction of their usual protein intake and would be likely to have a beneficial effect. Patients are also more likely to adhere to moderately protein-restricted diets than to more drastic restrictions.BP Control and Progression of Diabetic NephropathyHypertension invariably complicates the course of patients with diabetic nephropathy. Of the 35% to 40% of either IDDM or NIDDM patients who ultimately develop nephropathy, all at some time in their natural history will be hypertensive.2Numerous clinical trials of both IDDM and NIDDM hypertensive patients with nephropathy have assessed diverse forms of BP-lowering therapy on the progression of renal disease.1 2 44 These trials have all demonstrated that aggressive reduction of an elevated BP to levels below 140/90 mm Hg will retard the progression of diabetic renal disease. Furthermore, recent data suggest that some antihypertensive drugs may confer unique beneficial effects in attenuating the progression of this disease independent of their BP-lowering effects.42 43 44

Pharmacological Treatment of Hypertension in the Diabetic PatientPharmacological therapy should be initiated when lifestyle modifications are unsuccessful in

controlling hypertension in the diabetic individual2 (Fig 2⇑). The National Institutes of Health Consensus Panel recommended four classes of drugs that are effective as first-line single-agent therapy.2 Each drug class has potential advantages and disadvantages. Recent data from several large-scale hypertension treatment trials suggest that some classes may be preferred in the diabetic patient with nephropathy, and these considerations will be addressed. The five major classes of antihypertensive drugs currently being used in the United States for the diabetic hypertensive patient are discussed below.ACE InhibitorsACE inhibitors have no adverse effects on lipid levels or glycemic control.2Experimental studies have provided a theoretic framework for anticipating that ACE inhibition may preferentially retard the progression of diabetic renal disease. Studies over the past decade have demonstrated that the sustained increase in glomerular capillary pressure evoked in response to loss of renal mass produces a destructive sclerosing reaction.49 Administration of ACE inhibitors decreases glomerular capillary pressure, with a resultant reduction of glomerulosclerosis, suggesting that ACE inhibitor therapy may protect the injured kidney from hemodynamically mediated glomerular damage.49

The results of the first major attempt to compare patients randomized to an ACE inhibitor or alternative therapy were reported in 1992 from a study of 40 patients with IDDM and diabetic nephropathy randomized to treatment with either enalapril or metoprolol, generally combined with furosemide.50 Treatment with enalapril compared with metoprolol resulted in a highly statistically significant reduction in the rate of decline of GFR and in the level of proteinuria. Overall, there was no statistical difference in the BP reduction or BP achieved with the two treatments. Recently, the Diabetes Collaborative Study Group reported the results of a trial designed to determine whether the ACE inhibitor captopril is more effective in slowing the progression of diabetic nephropathy than are agents that act primarily by reducing BP.42 This was a randomized, controlled trial comparing captopril with placebo in patients with IDDM in whom urinary protein excretion was greater than or equal to 500 mg/d and serum creatinine concentration was less than or equal to 2.5 mg/dL. The BP goal was to achieve BP control during a median follow-up of 3 years. The primary end point was a doubling of the baseline serum creatinine concentration. Serum creatinine concentrations doubled in 25 patients in the captopril group compared with 43 patients in the conventional therapy group. The reduction in the risk of a doubling of serum creatinine concentration was 48% in the captopril group as a whole. Captopril therapy was associated with a 50% reduction in the risk of the combined end points of death, dialysis, and transplantation that was independent of the small disparity in BP between the groups. In a recent subgroup analysis of the data, remission of nephrotic-range proteinuria was observed in 7 of 42 patients assigned to captopril (16.7%; mean follow-up, 3.4±0.8 years) but in only 1 of 66 patients assigned to placebo. The findings were interpreted as suggesting that both

BP control and reduced proteinuria contribute to the reduced rate of GFR loss in the remission group.Whereas most available clinical trials have assessed the effects of ACE inhibition in patients with IDDM, recent clinical trials in NIDDM patients have been reported. A long-term 5-year study evaluating the effects of ACE inhibition on proteinuria and on the rate of decline in renal function in NIDDM patients with microalbuminuria was recently conducted in Israel.43 In a randomized, double-blind, placebo-controlled trial of 94 patients ACE inhibition during the early stages of diabetic nephropathy resulted in long-term stabilization of plasma creatinine levels and of the degree of urinary loss of albumin. Lebovitz et al51 recently reported the results of a 3-year prospective, double-blind, placebo-controlled trial in NIDDM patients and demonstrated that an antihypertensive regimen that included the ACE inhibitor enalapril preserves renal function to a greater extent than does therapy with antihypertensive agents excluding ACE inhibitors. The rate of loss of GFR was significantly greater in patients with overt proteinuria at baseline (urinary albumin excretion >300 mg/24 h) compared with patients with baseline subclinical proteinuria (urinary albumin excretion ≤300 mg/24 h). Antihypertensive treatment with enalapril preserved GFR better in the patients with subclinical proteinuria at baseline than the other antihypertensive treatments that excluded the ACE inhibitor. Furthermore, only 7% of the enalapril-treated group progressed to clinical albuminuria compared with 21% of control patients. On the basis of these findings these investigators suggested that ACE inhibitors should be used as initial treatment for hypertensive NIDDM patients with or without microalbuminuria and not held in reserve until clinical albuminuria or proteinuria develops.An important meta-regression analysis of the relative effects of different antihypertensive agents on proteinuria and renal function in patients with diabetes was recently reported.52 This analysis assessed 100 controlled and uncontrolled studies that provided data on renal function, proteinuria, or both before and after treatment with an antihypertensive agent. Multiple linear regression analysis indicated that ACE inhibitors decreased proteinuria independent of changes in BP, treatment duration, and type of diabetes or stage of nephropathy. It was concluded that ACE inhibitors had a unique ability to decrease proteinuria independent of the reduction in proteinuria caused by changes in systemic BP.Despite these promising results, several caveats are in order. Lowering of BP per se may be the major factor contributing to the salutary renal effects of ACE inhibitors as well as other antihypertensive agents in patients with diabetes and hypertension.53 ACE inhibitors are not free of side effects. An infrequent but important risk of ACE inhibitors is an acceleration of renal insufficiency, particularly in patients with bilateral renal artery stenosis and possibly more commonly in patients with diabetes. Under conditions in which filtration pressure depends on angiotensin II, the converting enzyme inhibitors may cause a precipitous fall in GFR. This complication is most likely to occur in the presence of bilateral renal artery stenosis due to atheromatous plaques or severe congestive cardiac failure. ACE inhibitors may provoke hyperkalemia, particularly in those individuals with decrements in GFR or hyporeninemic hypoaldosteronism.2 Finally, care must be exercised in initiating ACE inhibitor therapy in patients receiving diuretics because BP may drop and renal function decline profoundly.2

Calcium AntagonistsThe published studies regarding calcium antagonists and diabetic renal disease have been widely divergent in their design and findings.53 54 The Melbourne Diabetic Nephropathy Study Group has reported the 24-month results of their prospective, randomized study comparing the effects of the ACE inhibitor perindopril with those of the calcium antagonist nifedipine on BP and microalbuminuria in 43 diabetic patients with persistent microalbuminuria.55 After 12 months of therapy the investigators observed that both drug regimens were equally efficacious in reducing BP and albumin excretion in hypertensive patients. However, the 2-year follow-up data demonstrated that proteinuria had returned to baseline in the nifedipine-treated group but remained decreased in the perindopril-treated group.55 Thus, more long-term prospective studies need to be conducted to determine the benefits of both dihydropyridine and nondihydropyridine calcium antagonists in patients with hypertension and associated diabetic nephropathy.In addition, it has been suggested that the combination of a calcium antagonist with a converting enzyme inhibitor should result in a greater reduction in urinary protein excretion and slowed morphological progression of nephropathy.56 One study compared the renal hemodynamic and antiproteinuric effects of a nondihydropyridine calcium antagonist and an ACE inhibitor alone and in combination in NIDDM patients with documented nephrotic range proteinuria, hypertension, and renal insufficiency.56 Patients treated with the combination of a calcium antagonist and ACE inhibitor manifested the greatest reduction in albuminuria. In addition, the decline in GFR was the lowest in this group. Although such an approach is extremely attractive, additional studies will be required to extend these initial observations.Thiazide DiureticsThiazide diuretics in small doses are used frequently and successfully to treat hypertension in

individual diabetic patients2 (Fig 2⇑). These drugs have been shown to reduce cardiovascular morbidity and mortality in large population-based randomized trials (Systolic Hypertension in the Elderly Program [SHEP]). If the dose is low (ie, 25 mg or less hydrochlorothiazide or chlorthalidone daily), adverse effects on carbohydrate metabolism, hypokalemia, and hypomagnesemia are uncommon. Because the diabetic hypertensive patient is generally volume expanded, diuretics are often necessary for adequate control of BP. The disadvantages of thiazides are that they cause short-term dyslipidemia, altered carbohydrate metabolism, hypokalemia, hypomagnesemia, and hyperuricemia in some patients, but these adverse effects are minimized at the low doses recommended.2 Diuretics are also very useful antihypertensive agents when used in conjunction with ACE inhibitors; this combination is often synergistic in lowering BP and minimizes the metabolic side effects of diuretics.β-BlockersSeveral concerns limit the usefulness of β-blockers in treating people with diabetes: (1) These agents may have adverse effects on glucose and lipid metabolism. (2) Most troublesome for the insulin-treated diabetic subject is the observation that β-blockers can interfere with awareness of hypoglycemia in patients with diabetes and perhaps also prolong the recovery from hypoglycemia.2The catecholamine-mediated symptoms of hypoglycemia-induced symptoms can

be blunted if not abolished. The reflex tachycardia that serves to warn the patient of hypoglycemia may be blocked, putting the patient at greater risk of progressing to central nervous system symptoms. (3) β-Blockers can reduce peripheral blood flow and worsen claudication and vasospasm in patients who already have a compromised peripheral vascular system. (4) Finally, when β-blockers are added to diuretics, an aggravation of the hyperglycemic effect of the latter may occur.2Results from a recent investigation suggest that obese elderly patients treated with β-blockers and diuretics were at greater risk of developing NIDDM compared with obese elderly normotensive individuals.57 Thus, except under special circumstances (eg, in the presence of angina pectoris and after myocardial infarction), β-blockers should no longer be used as first-line antihypertensive medications in these patients.α1-Blockersα1-Blockers have been recommended for the treatment of diabetic hypertension on the basis of their efficacy, lack of adverse effects on glucose or insulin metabolism, and neutral or perhaps beneficial effect on the lipid profile.2 These agents infrequently produce or exacerbate sexual dysfunction and may permit improvement of sexual function when substituted for a central sympatholytic agent.2 Currently, there is no reported clinical evidence that peripheral α-blockers aggravate diabetic orthostatic hypotension, as is said to be the case with centrally acting sympatholytics.2

There are both advantages and disadvantages to the use of central sympatholytic antihypertensive agents (eg, clonidine, guanabenz, methyldopa, guanfacine) in the diabetic patient. Advantages include their lipid-neutral2 and minimal-to-absent hyperglycemic effects.2 However, centrally acting sympatholytic medications may worsen or unmask both orthostatic hypotension and sexual dysfunction in diabetic patients.2 In summary, although central sympatholytic agents have few if any metabolic side effects, some of their putative other adverse effects render them less than ideal antihypertensive agents for the management of diabetic patients.Summary of Pharmacological TherapyThe available data at present indicate that overall, ACE inhibitors produce more beneficial effects on diabetic nephropathy (reductions in proteinuria) than other antihypertensive agents studied. It has recently been reported that predictions in proteinuria are related to reduced rates of decline in renal function.58 It should be pointed out, however, that not all ACE inhibitors have been studied in diabetic nephropathy. Considering their efficacy in lowering BP, low incidence of side effects, and renal protective properties, ACE inhibitors appear to be a logical first-choice class of drugs for use in hypertensive diabetic patients with albuminuria (>30 mg/24 h) when economically feasible and not contraindicated. ACE inhibitors may also be beneficial in normotensive patients with albuminuria, but the evidence is less clear. When ACE inhibitors are contraindicated or ineffective, other antihypertensive agents should be used.Concerns Regarding Sexual DysfunctionClinically relevant side effects associated with antihypertensive pharmacological therapy appear to be more frequently observed in diabetic individuals than nondiabetic people with hypertension.2 59 Diabetic hypertensive people often manifest a side-effect profile very similar to that seen in elderly patients treated with antihypertensive drugs. For example, as with the elderly,

many diabetic individuals have reduced baroreceptor sensitivity and the tendency toward orthostatic hypotension in response to antihypertensive drug therapy.2 59 Diabetes mellitus, hypertension, and advancing age are independently associated with an increased prevalence of

sexual dysfunction in both women and men2 59 (Fig 3⇓). Hypertension, neuropathy, vascular insufficiency, and physiological problems all have been implicated in impotence, impaired ejaculation, and decreased libido in men and decreased vaginal lubrication, orgasmic

dysfunction, and decreased libido in women2 59 (Fig 3⇓). The absence of standards for evaluation of sexual dysfunction in women has led to underreporting and inadequate comprehension of the extent of this problem in women with diabetes and hypertension.

View larger version: In this page

  In a new window

Download as PowerPoint Slide

Figure 3.Chart shows interrelation of factors in sexual dysfunction in hypertensive diabetic patients (see Reference 22 ).

Any antihypertensive medication can contribute to sexual dysfunction, perhaps some more than others, and this should be a consideration in evaluating people with both hypertension and diabetes who have this problem.2 59 For example, several carefully conducted trials have indicated that men taking thiazide diuretics have a relatively high prevalence of sexual dysfunction, including a decrease in libido, difficulty in gaining and maintaining an erection, and difficulty with ejaculation.2 59 Thus, patients with both diabetes and hypertension should be evaluated for sexual dysfunction; appropriate therapy, including changes in medication or referral for sex counseling, should become routine in their clinical care.Previous Section Next Section

Future Considerations

Increasing investigation should also focus on identifying appropriate antihypertensive agents that not only lower BP but also reduce cardiovascular risk and retard the rate of progression of diabetic renal disease. In light of the recent report of the Diabetes Collaborative Study Group demonstrating that ACE inhibition retards the progression of diabetic nephropathy in IDDM patients, it would be important to ascertain whether this intervention is also applicable to patients with NIDDM. In light of conflicting evidence that calcium antagonists may also confer renal protection in diabetic nephropathy, prospective long-term studies will be needed to further delineate and corroborate these actions. Furthermore, in light of the attractive theoretic framework commending their use, clinical trials should be initiated to assess the long-term

effects of interventions including 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors, aminoguanidine, acarbose, and aldose reductase inhibitors on the natural course of decline of GFR and, if possible, on the progression of anatomic abnormalities. Finally, in light of divergent results on the effects of protein restriction, additional studies are warranted to delineate the precise role of this intervention in both IDDM and NIDDM patients.

Previous Section Next Section

Selected Abbreviations and Acronyms

ACE =angiotensin-converting enzymeAGE =advanced glycosylation end product(s)BP =blood pressureGFR =glomerular filtration rateIDDM =insulin-dependent diabetes mellitusIGF =insulin-like growth factor(s)LDL =low-density lipoproteinNIDDM=non–insulin-dependent diabetes mellitusOx-LDL=oxidized low-density lipoproteinPAI =plasminogen activator inhibitorVSMC =vascular smooth muscle cell

Previous Section Next Section

Acknowledgments