Renoprotection by blocking the RAAS in diabetic nephropathy—fact or fiction?

Diabetic nephropathy is characterized by proteinuria, blood pressure elevation, a relentless decline in renal function and a high risk of cardiovascular disease. In 1992, based on clinical studies, it was suggested that angiotensin-converting enzyme inhibitors (ACEIs) offer renoprotection in diabetic nephropathy—i.e. an effect protecting the kidney function above and beyond what was offered by similar blood pressure reduction with other antihypertensive agents [ 1 ]. This study was conducted in type 1 diabetic patients with moderately impaired renal function. The finding was confirmed and extended by the Collaborative Study Group [ 2 ] in a study with a median follow-up, 3 years; range, 1.8–4.8, demonstrating that in type 1 diabetic patients, ACEIs were associated with a 50% [95% confidence intervals (CI) 18–70] reduced risk of dialysis/transplantation or death. Subsequently blockade of the renin–angiotensin system (RAS) was investigated in type 2 diabetic patients using angiotensin II receptor blockers (ARB), and a renoprotective effect was again demonstrated, compared with placebo (standard antihypertensive therapy) in hypertensive microalbuminuric type 2 diabetic patients in the IRMA 2 trial lasting 2 years, where a 70% reduction in the progression to overt nephropathy was observed in patients receiving the maximal dose of irbesartan 300 mg daily [ 3 ]. Again, this effect was shown to be present beyond the blood-pressure-lowering effect of ARB-intervention. This was further substantiated in a substudy, in which it was documented that there was no difference in 24-h blood pressure in patients treated with ARB or standard antihypertensive therapy [ 4 ]. In patients with type 2 diabetes and overt diabetic nephropathy, the IDNT study with a mean follow-up time of 2.6 years demonstrated a significant 20% reduction in the development of the combined endpoint doubling of S-creatinine, end-stage renal disease (ESRD) or death in the patients treated with irbesartan compared with placebo (standard antihypertensive therapy), and 23% risk reduction compared with the calcium channel blocker amlodipine [ 5 ]. It should be stressed that, particularly in comparison with the amlodipine arm, the systemic blood pressure was completely identical. The RENAAL study with a mean follow-up time of 3.4 years (range, 2.3–4.6) similarly demonstrated that ARB losartan could significantly reduce the development of doubling of creatinine, ESRD and death by 16% compared with standard antihypertensive therapy [ 6 ]. Following these studies, FDA and EMEA approved irbesartan and losartan for renoprotection in diabetic patients (beyond blood pressure). Clinical guidelines recommended the use of inhibition of the RAS for prevention of diabetic nephropathy and ESRD. Despite all this evidence, the renoprotective effects of RAS blockade in diabetic nephropathy have recently been questioned on several occasions. In this article, we critically assess three recent publications that have challenged the specific renoprotective role of ACEI/ARBs in proteinuric diabetic patients suffering from diabetic kidney disease. We will not discuss the role of ACE inhibition or ARB in non-proteinuric kidney diseases in diabetic patients. First, a post hoc analysis from the large randomized controlled hypertension trial the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), with a mean follow-up of 4.9 years, failed to find any difference between the effect of treatment with the ACEI lisinopril or the calcium channel blocker amlodipine compared with chlortalidone (diuretic) on the development of ESRD or a 50% or greater reduction in estimated glomerular filtration rate (GFR) [ 7 ]. ALLHAT was not designed as a renal study, there was no information on proteinuria/albuminuria at baseline or during follow-up, there was no information on the cause of renal failure (acute or chronic renal failure, and if renal failure in diabetic patients were due to diabetic nephropathy, hypertension, other renal diseases or nephrotoxic agents). In a recent meticulous analysis of the ALLHAT study, it was concluded that it cannot be used as a renal study, and due to differences in study populations and antihypertensive regimens it is not in contrast with conclusions from previous chronic kidney disease studies [ 8 ]. The usual renoprotective treatment in diabetic nephropathy includes the combination of an agent blocking RAS and a diuretic [ 1–3 , 5 , 6 , 9 ], but this was not allowed in ALLHAT, as one active arm included a diuretic [ 8 ]. Furthermore, 50% of the patients received no medication or only insufficiently low doses of the ACEI, as they remained on the starting dose of 10 mg per day [ 10 ]. The latter may be important, since IRMA-2 showed that the degree of renal protection is dependent on the dose of the given renoprotective intervention. In addition, we showed the importance of optimal dosing as ultrahigh doses of Irbesartan (900 mg once daily) offers additional renoprotection compared with the usual dose of 300 mg daily, independent of changes in 24 h systemic blood pressures [ 9 ], and similar results were found with 64 mg candesartan daily [ 11 ]. In a meta-analysis, Casas et al . [ 12 ] stated that ‘The benefits of ACE-inhibitors or ARBs on renal outcomes in placebo-controlled trials probably result from a blood-pressure-lowering effect. In patients with diabetes, additional renoprotective action of these substances beyond lowering blood pressure remains unproven’. When RAS blockade was compared with active comparators in the meta-analysis, no difference between treatments was observed, but these analyses were entirely dominated by the aforementioned post hoc analysis of the ALLHAT study [ 12 ]. It was recognized that blocking the RAS was superior to placebo, although most of the ‘placebo’ controlled trials used active treatment consisting of standard conventional antihypertensive therapy. It was suggested in the meta-analysis that the observed effects were due to observed differences in blood pressure. In a previous meta-analysis that did not include the ALLHAT study, the authors concluded that blocking the RAS does offer renoprotection, mediated by factors in addition to decreasing blood pressure [ 13 ]. It is well-known that meta-analyses may reach the wrong conclusion [ 14 ], and obviously they can never be better than the included studies. The idea of using the format of a meta-analysis is to obtain clarity when smaller studies give no clear answer. It should not be used to create confusion by the inclusion of large but less well-suited studies, when well-designed and performed studies give a uniform answer. Most recently, Suissa et al . [ 15 ] presented a population-based study from Canada, suggesting that ACEIs do not appear to decrease the long-term risk of ESRD in diabetes; this study suggested that ACEIs might actually increase the risk for ESRD, contributing to the increasing incidence of ESRD owing to diabetes. This study was based on a registry of medication prescription, including diabetic patients (according to the prescription of insulin or oral hypoglycaemic agents) who were prescribed antihypertensive agents from 1982 to 1986. The 6102 patients were followed to the end of 1997 with respect to development of ESRD, which occurred in 102 patients, of whom 21 had been treated with an ACEI within the initial 3 years of follow-up. The adjusted rate ratio for renal failure was 2.5 in patients initially treated with ACEIs compared with patients treated with diuretics, when compared with the control patients. Although the study is population-based and has a long follow-up regarding development of ESRD, several limitations need to be taken into consideration. First, there is no information on the indication for the medication in the patients, making bias by indication a major problem. Already in 1985, Taguma [ 16 ] described the antiproteinuric effect of ACEIs in diabetic patients in a small study, and it is possible that the ACEIs were given to patients at particularly high risk for development of ESRD. No information regarding well-known risk factors for development of ESRD due to diabetes were available, such as proteinuria/albuminuria, presence of retinopathy, lipids, smoking, blood pressure or glycaemic control. In addition, development of renal failure requiring dialysis for at least 6 weeks was the endpoint, without any information regarding the cause for ESRD. In other words, numerous renal and non-renal conditions including nephrotoxic drugs may have resulted in ESRD, in contrast to the RENAAL [ 6 ] and IDNT [ 5 ]. The finding of a particular high risk of ESRD for patients treated with ACEIs after more than 3 years of follow-up (when more data regarding a specific renoprotective effect became available) suggests that bias by indication is of concern. This has also been a problem in previous studies suggesting that diuretics [ 17 ] or ACEIs had a deleterious effect in patients with diabetic nephropathy [ 18 ]. Randomized controlled trials have furthermore demonstrated the beneficial effect of RAS blockade in early diabetic nephropathy [ 3 , 19 ] as well as in later stages [ 1 , 2 , 5 , 6 ]. In addition, long-term follow-up studies of type 1 diabetic patients with diabetic nephropathy have demonstrated improved median survival from 5 years before antihypertensive treatment up to 21 years from onset of nephropathy, as arterial blood pressure is being treated more aggressively and to a larger extent (>85%) with RAS blocking agents [ 20 ]. Finally, there is no information on the doses of ACEIs used. Initially, potentially nephrotoxic doses were used irrespective of renal function (maximum dose of captopril 450 mg per day) [ 21 ]. Subsequently doses were markedly reduced, but the effect is dose-dependent, thus it is important that renoprotective doses be applied comparably to the doses in the trials demonstrating a beneficial effect. Unfortunately, the study gives no information regarding this, or on the use of the combination of diuretics with ACEIs. The authors conclude, they are in accordance with the literature when stating that ACEIs are not associated with a decrease of the risk of renal failure compared with other antihypertensive agents. They refer to studies that were not powered to evaluate an effect on ESRD: the UKPDS study including patients with newly diagnosed diabetes where only eight patients developed ESRD [ 22 ], and the MICRO–HOPE study [ 23 ], with 18 events of ESRD; however, the MICRO–HOPE study did observe a reduction in the development of overt diabetic nephropathy. In the aforementioned studies designed to evaluate renoprotection, ACEIs were renoprotective in type 1 diabetic patients and a significant reduction in the development of diabetic nephropathy and ESRD was observed in type 2 diabetic patients, when the RAS was inhibited with ARBs independent of the beneficial effect on arterial blood pressure. The majority of the patients in the study by Suissa et al . [ 15 ] were type 2 diabetic patients (treated with oral hypoglycaemic agents alone or in combination with insulin) and the renoprotective effect of blocking the RAS in type 2 diabetic patients has been documented for ARBs in randomized controlled studies. Due to lack of studies, this is not the case for ACEIs (only demonstrated to be renoprotective in type 1 diabetic patients). The DETAIL study suggested that there was no difference in the effect on renal function of ARBs and ACEIs in type 2 diabetic patients with early nephropathy [ 24 ]. However, due to an endpoint which could only be evaluated in 54% of the patients and lack of power, this study could not settle this issue [ 25 ]. In summary, the study by Suissa et al . [ 15 ] should be interpreted with caution due to confounding by indication, lack of information regarding risk factors for development of diabetic nephropathy, and absence of information regarding the cause of renal failure. Finally there was no information regarding the used doses of ACE inhibition. Thus, despite the fact that we have specifically designed randomized double masked appropriately-powered, well-conducted clinical trials on the renoprotective effect of both ACEIs and ARBs in diabetic patients, we still have colleagues that challenge these results and they hamper the already difficult implementation in daily practice. Of course, we must always be open-minded, the drugs (and or results) of today may not be those of tomorrow. However, one may not draw any therapeutic guideline conclusions based on post hoc analyses of trials, neither on meta-analysis nor on cohort studies, in case appropriate RCT are available. We wish to stress that the only data against using RAS blockade in diabetic patients should come from well-designed clinical trials, and we know of none that has not shown renoprotection beyond blood pressure lowering. There is a need for further improvement in the therapy of patients with overt diabetic nephropathy, despite the significant improvements obtained by blocking the RAS, as a substantial number of patients in the trials, and in the clinic, continue to develop ESRD [ 26 ]. It is however promising that, at least at specialized clinics, there is a decline in the development of diabetic nephropathy [ 27 , 28 ], and improved survival due to aggressive antihypertensive therapy from a median survival after onset of diabetic nephropathy of 5 to 21 years [ 20 ]. This dramatic increase in the incidence of diabetic ESRD seems to level off, although at a higher level than 20 years ago [ 26 ]. In agreement with the large randomized controlled clinical trials, we therefore believe that inhibition of the RAS by the use of ACEIs or angiotensin II receptor blockers in the optimal doses, usually combined with diuretics, should at present be considered as first-line treatment modality for delaying renal insufficiency in diabetic patients with elevated urinary albumin excretion. Conflict of interest statement . H.-H.P. has equity in Novo Nordisk and Merck, and has received consulting and lecture fees from Merck, Bristol-Myers Squibb, Pfizer and Sanofi, and grants from Merck and Bristol-Myers Squibb. D.d.Z. has been adviser to and received grants from Bristol-Myers Squibb and Merck Sharp and Dohme. 1Steno Diabetes Center, Gentofte and 2Faculty of Health Sciences, University of Aarhus, Aarhus, Denmark and 3Department of Clinical Pharmacology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands