Comments on the renal aspects of the ALLHAT study

A growing interest in the investigation of renal function in big trials dealing with different aspects of cardiovascular disease has developed in recent years. This is justified by different reasons, including the high prevalence of nephrosclerosis as a cause of end-stage renal disease (ESRD) [1] and the capacity of minor alterations of renal function to predict a poor outcome for the patients [2]. The finding of a small increase in serum creatinine values (above 1.3 mg/dl, 115 μmol/l), of a diminished estimated creatinine clearance (< 60 ml/min), of microalbuminuria and/or proteinuria, heralds a significant elevation in cardiovascular events and death as well as in total mortality. According to the data available, one or more of these alterations can be present in up to 40% of high-risk hypertensive patients [2,3]. Interestingly, the prevalence of a diminished renal function in the general population in USA is elevated, with more than one-third of the general population presenting values of estimated creatinine clearance lower than 60 ml/min at ages above 60 years [4]. The initial design of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) study contemplated renal function among the final outcomes of the trial [5]. Two aspects were then considered, the evolution of the slope of the reciprocal of serum creatinine and the development of ESRD (defined as need for dialysis, renal transplant or death). The recent publication of the main results of the study also included the post hoc analysis of the change in estimated glomerular filtration rate [6]. The results show that a total of 448 patients developed ESRD without significant differences between the three arms of the study. The slopes of the reciprocal of serum creatinine over time were virtually identical in the chlortalidone and lisinopril groups, whereas the decline in the amlodipine slope was significantly less than that of the chlortalidone arm. Finally, the estimated creatinine clearance (erroneously referred in the paper as estimated glomerular filtration rate) exhibited a significantly better preservation in the amlodipine arm that had a final mean value of 75.1 ml/min, that compared with 70.0 and 70.7 ml/min in the chlortalidone and lisinopril groups, respectively. These results seem to differ from the available evidence indicating that the administration of an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker (ARB) is required to protect renal function beyond the benefit obtained by blood pressure control [7]. The better evolution observed in both the slope of the reciprocal of serum creatinine and the estimated creatinine clearance with amlodipine also argue against recently published comparative studies in which an ACE inhibitor or an ARB have been shown to be better than a calcium channel blocker (CCB) in primary renal disease, type 2 diabetic nephropathy, and in African American patients with nephrosclerosis [8–10]. How can these differences be explained? Renal protection requires, among other measures, a strict control of blood pressure with a goal below 130/80 mmHg, or even lower if proteinuria above 1 g/day is present, and second to diminish proteinuria, if present, to values as near to zero as possible [7]. Nothing can be said about the second measure because neither proteinuria nor microalbuminuria were determined in the ALLHAT. However, it is of interest to remember that proteinuria can be present in up to 22% of high-risk hypertensive patients, as seen in the VALUE study [11], and that proteinuria, even in small amounts, seems to be the most potent predictor of a rapid fall in renal function [12]. On the other hand, the presence of microalbuminuria has equally been shown to predict a poor outcome of glomerular filtration rate in essential hypertension [13]. The ALLHAT study demonstrated a very good control of blood pressure with final mean values for systolic blood pressure/diastolic blood pressure of 133.9/75.4, 134.7/74.6, 135.9/75.4 mmHg for chlortalidone, amlodipine, and lisinopril, respectively, and with 66% of patients exhibiting blood pressure values below 140/90 mmHg [2]. Small but significant differences in the control of SBP were seen in favour of the chlortalidone arm that exhibited SBP values 1 mmHg lower than the amlodipine arm and 2 mmHg lower than the lisinopril arm [2]. The difference seemed to be particularly marked in African Americans treated with lisinopril who exhibited a difference of 4 mmHg compared to those treated with the diuretic. Such a good control has undoubtedly contributed to the findings related to renal function observed in ALLHAT. The good effects of ACE inhibitors and ARBs on established renal disease have mostly been demonstrated in patients with proteinuria and in the absence of a strict blood pressure control, in particular in diabetic patients [14]. A recent publication indicated that a very strict blood pressure control attained a similar degree of renal protection in type 2 diabetic patients, matching the control attained with enalapril or with the CCB nisoldipine [15]. Furthermore, as proposed some years ago [16], such a good control of systemic blood pressure counteracts the risk of afferent arteriolar vasodilation induced by a CCB by impeding the transmission of a still elevated systemic blood pressure. When a more complete analysis of the evolution of renal function is performed, it will be interesting to see how many cases of crossover to an ACE inhibitor performed in patients receiving the diuretic or the CCB were due to the presence of a diminished renal function, and whether this could have contributed to create some bias in the final results. Recently published data also indicate that, to prove the efficacy of an ACE inhibitor in the presence of nephrosclerosis without proteinuria, follow-up periods longer than 4–5 years may be required [12]. The differences in SBP control observed when lisinopril was administered, in particular to African American patients, could have influenced the final results of renal function. A blood pressure of 4 mmHg, with respect to a mean drop of 10.5 mmHg, represents 39% of the total. However, recent data from the AASKD have shown that, in African American patients, attained mean blood pressure values of 128/78 mmHg did not differ from 141/85 mmHg in affording additional benefit of slowing the progression of hypertensive nephrosclerosis [6]. In summary, the ALLHAT study confirms the adequacy of strict blood pressure control in order to protect renal function in hypertensive patients. Further subanalysis are required to clarify the role of the different monotherapies on renal function, in particular in type 2 diabetics, African Americans and those patients with minor alterations in renal function at baseline.