Abstract
The authors of the review ‘Combination therapy in chronic kidney disease’ [Juncos and Caputo, 2015] published in the August issue of Therapeutic Advances in Cardiovascular Disease describe with elegance the probable Paleolithic origin of the renin–angiotensin–aldosterone system (RAAS). Due to dietary and behavioral changes over the centuries, the system could have become partially maladaptive, accelerating the course of a number of pathological conditions such as chronic kidney disease (CKD).
This assertion is proved by the well-documented protective effect of single RAAS blockade on cardiovascular diseases of the contemporary human, such as hypertension, coronary heart disease, heart failure, diabetes, and CKD with proteinuria [Lewis et al. 1993]. However, the system blockade obtained by currently available drugs is incomplete. Based on these facts, we might assume that the more intensive RAAS blockade should offer additional benefits.
The hypothesis has excellent rational basis. However, history has shown that rationally perfect assumptions not always resist the test of empirical validation. Such that the Platonic idea of seeking the truth solely by reason has already been questioned by Aristotle in the time of the classical Greco-Roman culture, and continues to be questioned by thinkers over time, as John Locke and David Hume, and even Immanuel Kant.
Through the submission to empirical validation, the hypothesis of additional benefit of intensive RAAS blockade on renal outcomes has resulted in findings less encouraging than expected. At first, there were great expectations due to the detection of an additional effect of dual blockade in reducing proteinuria, a surrogate renal endpoint [Catapano et al. 2008]. Such enthusiasm was exacerbated by the publication of the COOPERATE trial in 2003 [Nakao et al. 2003], the first randomized controlled trial (RCT) to show the benefits of dual blockade on hard renal endpoints.
However, doubts emerged with the publication of the results of the Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial (ONTARGET) [ONTARGET Investigators et al. 2008], an RCT that evaluated more than 25,000 elderly patients at high cardiovascular risk and found an increase of side effects in the combined use of telmisartan and ramipril arm, including hypotensive episodes and need for dialysis, without cardiovascular or renal benefits. Discouragement culminated with the withdrawal of COOPERATE from the Lancet in 2009 [Editors of The Lancet, 2009], after an institutional investigation detected a number of inconsistencies in the study database.
As the authors of the review just published describe, beyond the ONTARGET, two other large RCTs failed to show benefits of the dual when compared with single blockade of RAAS on hard renal endpoints. The Aliskiren Trial in Type 2 Diabetes Using Cardiorenal Endpoints (ALTITUDE) trial [Parving et al. 2012] compared treatment with aliskiren or placebo as an adjunt to an angiotensin-converting-enzyme inhibitor (ACEi) or an angiotensin-receptor blocker (ARB) in more than 8000 high-risk diabetic patients. This study was stopped prematurely after an interim analysis showed an increased occurrence of side effects (hyperkalemia, hypotension and renal impairment) and no cardiovascular or renal benefits from the addition of aliskiren to standard therapy with ACEi or ARB. The Veterans Affairs Nephropathy in Diabetes (VA NEPHRON D) trial [Fried et al. 2013] was designed to evaluate potential additional benefits of dual blockade with ARB and ACEi on cardiovascular and renal outcomes in proteinuric diabetic patients. This trial was also stopped early due to increased rates of hyperkalemia and renal dysfunction in combination therapy group. At the time of interruption, there had been a follow up of 2.2 years, which was shorter than initially planned. At this point, there was no effect on mortality or cardiovascular benefits, but a trend toward renal benefits with dual blockade. However, this trend has decreased over time, meaning that longer follow up would probably not significantly change the results. The very high significance of this trial stands on the specific population studied: diabetics with proteinuria. Taking into account the well-documented additional effect of dual RAAS blockade on proteinuria reduction [Catapano et al. 2008], and considering the positive association between proteinuria degree and steeper loss of glomerular filtration, these patients would have a higher chance of obtain benefits of dual blockade. The criticism of ONTARGET and ALTITUDE almost always includes the small number of proteinuric patients included in these trials. The publication of VA NEPHRON D comes exactly to help clarify this issue, showing that even in proteinuric diabetic patients, the risks of dual RAAS blockade outweigh benefits.
The best currently available evidence comes from RCTs evaluating hard endpoints, such as mortality, cardiovascular events, or need for renal replacement therapy. From the available literature, three RCTs including renal outcomes found negative results with dual RAAS blockage [ONTARGET Investigators, 2008; Parving et al. 2102; Fried et al. 2013]. One found benefits [Nakao et al. 2003], but was retracted by the publishing journal because of serious inconsistencies. However, it is important that healthcare decisions are not based solely on some studies without account being taken of the whole of the research information available. Thus, systematic reviews with meta-analysis of RCTs provide us with this overview. The latest on the issue was published in 2013 [Susantitaphong et al. 2013]. The authors analyzed 59 RCTs (4975 patients) comparing the efficacy and safety of combined versus single RAAS blockade in CKD patients. They found that combined therapy was associated with a decrease in glomerular filtration (−1.8 ml/min/1.73 m2, p = 0.005), a 3.4% higher risk of hyperkalemia and a 4.6% higher rate of hypotension, despite better control of proteinuria (−363 mg/day, p < 0.001) and a 5% higher rate of achieving the blood pressure goal. No effect on hard endpoints such as doubling of serum creatinine [odds ratio (OR) 0.80, 95% confidence interval (CI) 0.52–1.21, p = 0.29] or mortality (OR 0.38 (95% CI 0.06–2.40, p = 0.30) was found.
This amount of evidence, added to the fact that there is also no documented benefit on cardiovascular endpoints, led the scientific committee of a major guideline for the management of high blood pressure, the 8th Joint National Committee [James et al. 2014], to announce categorically that, based on current scientific evidence, dual RAAS blockade should be avoided. This statement makes no exception for CKD patients.
Therefore, the empirical evidence indicates against the rational hypothesis and it seems that more is not always better, at least in terms of RAAS blockade. Due to the dynamic nature of science, it is not unlikely that in future times we will identify subgroups, such as younger and non-diabetic patients, who attain renal benefits from the more aggressive blockade of the RAAS. But before indicating this treatment, even to these subgroups, we need to confirm through RCTs any long-term protective effect that outweighs the risks in such a well-defined population.