Are angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs) equally effective in diabetic nephropathy?
This question came up during an evening symposium I attended at the American Society of Nephrology Renal Week last November in Philadelphia. It resulted in some heated discussion. It is necessary to review the differences between these agents and the rationale for their use in diabetes.
An enormous body of work in experimental and human diabetes has established that blockade of the renin-angiotensin system is effective in reducing proteinuria and preserving renal function. Although there is some debate, it is likely that the primary effect is reduction of intraglomerular capillary pressure by reversing angiotensin II (AII)-mediated constriction of the efferent glomerular arteriole.
The first drug in this class was saralasin, a partial AT1 blocker/agonist that was, unfortunately, too toxic for human use.
ACE inhibitors prevent conversion of AI to AII. Different drugs in this class have different organ penetration, and it is now realized that the ability to block tissue ACE (particularly in the kidney) is more important than the ability to block circulating ACE. In addition, ACE inhibitors also block kininases and thus increase bradykinin levels. This action probably mediates the side effect of cough, but is also thought by some to be contributory to their effectiveness in reducing albuminuria. Theoretically, blocking AII production would tend to upregulate AII receptors. Because AII can be produced from non-ACE-mediated pathways (eg, enzymes such as chymases), there is a potential to enhance AII action, particularly if doses are missed.
The vasoconstricting actions of AII are mediated via the type 1 receptor, and a relatively new class of drugs (the sartans) specifically blocks this. These drugs have no effect on kininase and thus should cause no cough. They will, of course, not have any beneficial effect that might result from increased bradykinin. Theoretically they might also tend to raise AII levels as a result of a classic negative feedback.
Several head-to-head studies have been published, but these have usually been of relatively short duration (up to a year or so), had relatively small numbers of patients, and used the intermediate end point of reduction in albuminuria. These studies have shown no significant difference between the 2 classes of agent, although for the most part the magnitude of the reduction in albumin excretion rate has tended to be less with the ARB.
Last year 3 studies were published simultaneously on the effect of ARBs in type 2 diabetes patients with microalbuminuria (Irbesartan Microalbuminuria Type 2 Diabetes Mellitus in Hypertensive Patients [IRMA 2]) or clinical nephropathy (Irbesartan Diabetic Nephropathy Trial [IDNT] and Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan [RENAAL]).
IRMA 2 compared 2 doses of the ARB irbesartan with a control of any antihypertensive agent except ACE inhibitors, ARB, or dihydropyridine calcium channel blockers (CCBs) in 590 microalbuminuric, hypertensive type 2 diabetes patients. The adjusted hazard ratio for the development of nephropathy was 0.32 for the group on 300 mg/day compared with controls (5% progressing compared with 15%). Of interest, this was almost identical to the effectiveness of ACE inhibitors in normotensive type 1 diabetic patients described in a meta-analysis of over 600 patients.
RENAAL and IDNT studied a combined cohort of over 3200 nephropathic type 2 patients with a reduced creatinine clearance. The studies differed in one important respect: IDNT had 3 arms with a comparison between irbesartan, amlodipine, and a control group in whom any other class of drug could be used except ACE inhibitors; whereas RENAAL compared 2 groups -- losartan vs any other agent except ACE inhibitors/ARBs. The effectiveness of both ARBs was almost identical, with an adjusted relative risk for the primary combined outcome (doubling of baseline serum creatinine/end-stage renal disease/dialysis or transplantation/all-cause mortality) of 0.81 for IDNT and 0.84 for RENAAL.
However, the reduction in doubling of serum creatinine was approximately one half that seen in the Collaborative Study Group Trial of captopril in type 1 diabetic nephropathy. In IDNT, the amlodipine and control groups had similar rates of primary outcome. Of note, the magnitude of reduction of proteinuria for each 10-mmHg lowering of blood pressure in the irbesartan-treated patients in IDNT was about one half that reported in type 1 patients, which is consistent with the short-term studies of ACE inhibitors vs ARB.
Why was the question of equality of effectiveness asked in the first place? ACE inhibitors are now coming off product license and are, therefore, cheaper, and in our own cash-strapped health service this is an important consideration. Moreover, many clinicians are familiar with them and feel comfortable in their use.
Strictly speaking, the hard evidence, therefore, is with ACE inhibitors in type 1 and ARBs in type 2 nephropathy; the quantity of head-to-head comparative information is inadequate to draw a firm conclusion. However, irbesartan in IRMA 2 had an almost identical magnitude of effect to that seen with an ACE inhibitors in type 1, which suggests that at least at this early stage of nephropathy, the condition responds to renin-angiotensin blockade of any kind.
IDNT and RENAAL patients had much more advanced disease and greater cardiovascular comorbidity than the type 1 patients in the Collaborative Study Group Trial, thus making direct comparisons difficult. Of note, the magnitude of reduction of proteinuria for each 10-mmHg lowering of blood pressure in IDNT was about one half that reported in type 1 patients, which is consistent with the short-term studies. The lack of positive effect of amlodipine, despite blood pressure reduction virtually identical to irbesartan, confirmed the specific benefit of renin-angiotensin system blockade and it is likely -- though unproven -- that this benefit would be seen with ACE inhibitors. A comparative study is under way but will not report for some time. ARBs are very well tolerated, and the side-effect rate was not greater than controls in IDNT and RENAAL. At the moment, the evidence base rests with them in type 2 diabetic nephropathy, but the implications suggest that renin-angiotensin blockade by any means is likely to be of equal benefit.