Angiotensin II receptor blockade in diabetic nephropathy

Diabetic nephropathy has become the leading cause of end-stage renal disease in the United States.1 Approximately 35% of persons with type 1 diabetes will develop the triad of proteinuria, decreased glomerular filtration rate, and increased arterial pressure. For type 2 diabetes the incidence of nephropathy is approximately 20%. However, because 90% to 95% of diabetic patients are type 2, more than half of diabetic end-stage renal disease is currently associated with this form of the disorder.2 Diabetic nephropathy typically evolves through several phases. Microalbuminuria (urinary albumin excretion 30 to 300 mg/24 h) is the first clinical sign of diabetic damage to the kidney and is a harbinger of progressive kidney damage. In 20% to 50% of patients with sustained microalbuminuria, there is progression to more easily detectable proteinuria or macroalbuminuria (>300 mg/24 h) over the next 5 to 10 years.3 Once a patient with type 2 diabetes develops macroalbuminuria, there may be progression to frank nephrotic syndrome and further decline in renal function appears to be inevitable. Proteinuria accompanies a decline of glomerular filtration rate of approximately 4 to 12 mL/min/year.4 Although patients with type 1 diabetes are usually normotensive until overt renal disease develops, hypertension usually occurs before proteinuria and overt nephropathy in patients with type 2 diabetes. Both the onset of microalbuminuria and the progression of renal disease after the onset of macroalbuminuria are accelerated by hypertension.5 Sufficiently powered, placebo-controlled, randomized clinical outcomes trials on the influence of antihypertensive therapy in diabetics have not been performed and are unlikely to be performed in the future. Among both type I and type 2 diabetic individuals, the mortality rate has been found to be lowest in those with the most intensive antihypertensive therapy.6,7 The Collaborative Study Group compared the angiotensin-converting enzyme (ACE) inhibitor captopril with conventional antihypertensive therapy in patients with type 1 diabetes and nephropathy.8 Captopril treatment reduced the risk of a doubling of serum creatinine and the combined end points of death and end-stage renal failure (ESRD) by approximately one half. The renoprotective effect of captopril was observed to be greater than would be expected from blood pressure (BP) reduction alone. Two recently completed studies in type 2 diabetic patients with established proteinuria are the first randomized clinical trials to directly compare the cardiovascular and renal benefits of inhibitors of the renin-angiotensin system with conventional antihypertensive agents in this high-risk hypertensive population. The Irbesartan Diabetic Nephropathy Trial (IDNT) enrolled 1715 hypertensive patients with type 2 diabetes and nephropathy as defined by the presence of proteinuria and serum creatinine of 1.0 to 3.0 mg/dL.9 Patients were randomized to once-daily therapy with the angiotensin II–receptor blocker (ARB) irbesartan, the dihydropyridine calcium antagonist amlodipine, or placebo. Blood pressure was titrated toward a target of ≥135/85 mm Hg or a reduction of 10 mm Hg of the systolic BP if >145 mm Hg at baseline. Other antihypertensive agents (excluding ACE inhibitors, ARB, and calcium antagonists) could be added to all three arms to achieve target BP control. The primary end point of IDNT was the time to first occurrence of any of the following: doubling of baseline serum creatinine level, ESRD, or death from any cause. The primary composite outcome occurred in 39% of the patients assigned to the placebo group, 41.1% in the amlodipine group, and 32.6% in the irbesartan group. Doubling of serum creatinine occurred in 23.7% of the placebo group, 25.4% of the amlodipine group, and 16.9% of the irbesartan group. There were no significant differences in the total mortality rates of the three arms of the IDNT or in the secondary nonrenal cardiovascular composite outcomes that occurred in 25.3% of the placebo group, 22.6% of the amlodipine group, and 23.8% of the irbesartan group. Urinary protein excretion decreased on average by 33% in the irbesartan group as compared with 6% in the amlodipine group and 10% in the placebo group. Creatinine clearance declined most slowly in the irbesartan group (−5.5 mL/min/year) versus the amlodipine group (−6.8 mL/min/year) and the placebo group (−6.5 mL/min/year). The Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan Study (RENAAL) assessed the role of the angiotensin II-receptor antagonist losartan in patients with type 2 diabetes and proteinuria.10 A total of 1513 type 2 diabetic patients with proteinuria and serum creatinine of 1.3 to 3.0 mg/dL were randomized to losartan (50 to 100 mg once daily) or placebo. Conventional antihypertensive medications, excluding ACE inhibitors and ARB, were administered as needed to maintain a seated systolic BP level of <140/90 mm Hg for the duration of the study. The primary composite end point doubling of serum creatinine and ESRD and death was reached by 47.1% of the patients in the placebo group and by 43.5% in the losartan group. Doubling of serum creatinine occurred in 26% of the placebo group and in 21.6% of the losartan group. There were no differences between the two groups in the risk of death or the composite of morbidity and mortality from all cardiovascular causes. The level of proteinuria, as indexed by the urinary albumin-to-creatinine ratio in a first morning specimen, declined by 35% in the losartan group. The patients in both the IDNT and RENAAL studies were predominantly male (>60%) with a mean age of approximately 60 years. Ethnic diversity was greater in the RENAAL study (49% non-Hispanic white) than in the IDNT (72% non-Hispanic white). Blood glucose control at baseline as assessed by glycosylated hemoglobin levels was similar (8.2% to 8.5%) for patients in both studies. Baseline mean BPs in the RENAAL study (160/87 mm Hg) were higher than for those in the IDNT study (152/82 mm Hg). The entry requirement and baseline levels of serum creatinine were virtually identical for the two studies. The entry criteria for IDNT included a 24-h urinary protein of ≥900 mg. For RENAAL the qualifying urinary protein level at study entry was ≥300 mg per milligram of creatinine in a first morning specimen on two occasions, equivalent to approximately 500 mg/day. The RENAAL mean duration of follow-up (3.4 years) was somewhat longer than that of IDNT (2.6 years). The primary outcome findings for both studies (ie, a composite of doubling of serum creatinine, occurrence of ESRD, or death) were qualitatively the same and quantitatively similar for the ARB-treated groups, with relative benefit of 20% observed in the IDNT and 16% reported for the RENAAL. In the IDNT, patients assigned to the amlodipine-based therapy had outcomes that were similar to those with other conventional antihypertensive agents. The relative risk reductions experienced by the ARB-treated groups in both studies were significant for the renal-specific outcomes of doubling of serum creatinine (33% for IDNT and 25% for RENAAL) and ESRD (28% for IDNT and 25% for RENAAL). Both studies found lower rates of hospitalization for heart failure in the ARB groups. However, total death and total cardiovascular event rates were not different among any of the groups in either the IDNT or the RENAAL study. The IDNT and RENAAL were well-designed trials that fill important gaps in the therapeutic approach to the nephropathy of type 2 diabetes. Until the IDNT and RENAAL results became known, it had been suspected but never proved that therapy based on inhibitors of the renin-angiotensin system can have a renoprotective effect in type 2 diabetic individuals with proteinuric renal disease. The pharmacologic actions of ARB including the renoprotection demonstrated by the RENAAL and IDNT derive solely from antagonism of angiotensin II activity. Thus, the IDNT and RENAAL show unequivocally that the renin-angiotensin system must contribute to the nephropathic process in a significant minority of patients with type 2 diabetes. Another important result of the IDNT study was that use of the long-acting dihydropyridine calcium antagonist, although not as renoprotective as irbesartan, was at least equivalent in safety and efficacy to conventional antihypertensive therapies. Microalbuminuria is a strong predictor of all-cause mortality and cardiovascular morbidity and mortality in type 2 diabetes. Although the mechanism underlying the association between microalbuminuria and mortality is not clear, the presence of microalbuminuria may reflect a generalized defect in vascular permeability leading to atherogenesis. Several recent studies have demonstrated that therapy with angiotensin-II–receptor antagonists is associated with a reduction of microalbuminuria and a slowing of progression to macroalbuminuria.11–13 Some have raised concern that physicians are slow to adopt consensus- and evidence-based recommendations despite unequivocally positive results in randomized clinical trials.2 Indeed, future recommendations that ARB therapy should be administered to all diabetic patients with proteinuria may not be greeted by clinicians and patients with the hoped-for level of enthusiasm. A fundamental process of preventive therapeutics is that of treating the many to help the few. This paradigm as applied to vaccine inoculation programs and preventive cardiology has been among the crowning achievements of health care during the last century. Why not draw the analogy to inoculation programs and just administer ACE inhibitors or ARB to all diabetic patients or diabetic patients with proteinuria? Some have suggested that it is laziness or ignorance that limits the widespread application of such therapies. Certainly there is a level of ignorance that we all share. It may be helpful to begin to examine the problem in the context of the polypharmaceutical therapies usually employed in the complicated type 2 diabetic patient with proteinuria. These patients also tend to have resistant hypertension and hyperlipidemia with at least one or more additional chronic symptomatic illnesses plus the not infrequent acute illnesses. In accordance with established clinical guidelines, such a patient under meticulous medical management will usually require two to three hypoglycemic agents, three to five antihypertensive agents, one to two lipid-lowering agents, and one or more additional agents for other disorders. For the IDNT trial the irbesartan group required an average of 3.8 antihypertensive medications (including the study drug) to maintain target BP levels that were less rigorous than currently recommended (≤130/80 mm Hg).2 In practice, it is unusual to encounter a well-managed diabetic patient with nephropathy on fewer than eight different prescription medications. It is a hurdle for anyone, regardless of personal resources, to be able to remember and afford to take all of these medications on a daily basis for years or even decades. Another reason for reluctance to prescribe additional medications is that there is very limited understanding about the potential for serious drug interactions of such complex regimens, particularly in the setting of renal impairment. Despite the major achievements of these well-designed studies, ARB are far from being a cure for diabetic nephropathy. The relative risk reductions for the primary composite end points of the RENAAL and IDNT trials were significant in the aggregate. However, the benefits were observed in a meaningful way by only a minority (16% to 20%) of those at risk. Individually, the two reports suggest a greater protective effect with time; although it is not reassuring that the trial with the longer duration was associated with slightly lesser relative risk reductions. The ARB groups for the trials had average rates of decline in creatinine clearance of approximately 0.8 to 1.0 mL/min/year slower than did the placebo groups. From a public health perspective the slightly slower rate of decline of GFR in the ARB groups is important, but may be considered less impressive by the clinician managing individual patients. Patients in all treatment groups in both studies continued to lose GFR at a clinically significant rate with a decline rate of creatinine clearance in the range of 4.4 to 6.8 mL/min/year. Both the RENAAL and IDNT reports indicated that ARB therapy was associated with a significant reduction in the mean urinary protein levels. Disappointingly, neither study may have been powered to determine whether individuals with greater or lesser changes in urinary protein excretion rates differ in renal outcomes. If changes in urinary protein levels could be found to predict renal outcomes, then it is possible that patients for whom there is neither a favorable BP nor proteinuric response to an ARB might be better served with antihypertensive and other medical therapy not targeted toward the renin-angiotensin system. Results of the IDNT and RENAAL studies suggest that the renoprotective effects of ARB in type 2 diabetic individuals may not depend on the BP lowering action of these agents. It has been shown that in the aggregate all classes of antihypertensive drugs, when used as monotherapy, are equally effective (50% to 60%) in lowering BP. It is also recognized that individuals vary widely in BP responses to particular antihypertensive agents. Correspondingly, a significant proportion of patients receiving ARB will experience little or no BP response while possibly maintaining the potential for renal benefit. A core process of medical care is based on the ability to apply clinical and laboratory data to monitor the biologic effects of therapy in each patient. Some physicians may wisely hesitate at the prospect of adding another medication to an already expensive and complex regimen without an established way to interpret clinical (eg, BP) or laboratory parameters (eg, proteinuria) to predict, detect, or monitor the desired biologic response. The biologic mechanisms involved in the genesis and progression of diabetic nephropathy are exceedingly complex. The IDNT and RENAAL confirm that, for some patients, the renin-angiotensin system is an active and treatable participant in the process. The renin-angiotensin system consists of a cascade of enzymatic reactions that act on angiotensinogen and angiotensin I, leading to the formation of angiotensin II.14 Renin cleaves the biologically inert decapeptide angiotensin I from its substrate molecule, angiotensinogen. Angiotensin I is hydrolyzed to the biologically active octapeptide angiotensin II through the action of ACE, an enzyme in relative abundance that has substrate affinity for a broad range of small peptides. The first-order kinetics of this reaction sequence is such that under physiologic conditions it is the level of renin that is the prime determinant of the rate of angiotensin II formation.14 The intrarenal actions of angiotensin II in the initiation and progression of diabetic nephropathy are incompletely understood. Angiotensin II plays a critical role in the development of glomerulosclerosis by increasing glomerular capillary pressure via preferential constriction of efferent arterioles.15 There are also highly complex actions of angiotensin II on the activation of cytokines, such as transforming growth factor–β, that mediate the hypertrophic and proliferative responses of fibroblast and mesangial cells and promote secretion of collagen and extracellular matrix proteins within the diabetic kidney.16 In the intact human kidney virtually all intrarenal angiotensin II generation is renin dependent.17 The kidney is the only known source of active renin in the circulation, and renal synthesis and release of renin are under tight and elaborate physiologic control.14 Experimental infusions of angiotensin II have been shown to produce necrotizing arterial lesions, multifocal myocardial lesions and renal tubular necrosis.18 These findings provide support to the notion that angiotensin II is vasculotoxic. Over the past three decades, retrospective and prospective observational studies have obtained differing results on the relationship of circulating plasma renin activity (PRA) and cardiovascular end organ injury.19 The larger body of evidence suggests a positive relationship between PRA and coronary and all-cause cardiovascular mortality.20–22 The relationship between intrarenal and plasma renin levels in diabetic patients is unknown. Whether circulating levels of renin or its inactive precursor, prorenin, predict or affect renal or nonrenal cardiovascular injury in diabetic patients remains to be investigated. Indeed we know almost nothing about how to determine which diabetic patients are most and least likely to benefit from pharmacologic interference with the renin-angiotensin system. To what extent could the baseline or on-treatment PRA or other measures of the renin-angiotensin system add predictive value? What is the probability of renal protection for patients who experience neither an antihypertensive nor antiproteinuric effect of such treatment? Do we indiscriminately administer ARB or ACE inhibitors on a lifetime basis to the millions of diabetic patients as suggested by cost-benefit analysis?23 Perhaps it is time to raise the bar for clinical trials in cardiovascular medicine. We should strive to do better than settle for aggregate results in which much fewer than half of the treated subjects benefit. Future clinical investigations including outcomes trials need to be powered to test hypotheses that facilitate greater understanding of how to individualize therapy based on sound pathophysiologic concepts. There is no doubt that the design and conduct of such studies will be more cumbersome and more expensive. However, learning how to apply therapeutic interventions to those who are most likely to benefit could produce huge long-term cost-savings in the decades to come. More importantly, providing the correct treatment to the right patient is the essence of good medicine.