SEARCH

SEARCH BY CITATION

The prevalence and cost of chronic kidney disease (CKD) among adults in the United States has continued to increase, and black persons with hypertension have been disproportionately affected compared with whites. Because the relationship between blood pressure (BP) and the progression of CKD is linear, the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) and other guideline committees recommended a more aggressive target, <130/80 mm Hg, for BP control in patients with CKD. This recommendation, however, is based mostly on observational data, with little clinical trial evidence from prospective randomized, controlled studies to support it.

The African American Study of Kidney Disease and Hypertension (AASK) evaluated the effects of intensive vs traditional targets for BP on CKD progression among African Americans with nondiabetic CKD and hypertension. At baseline, all patients in AASK were self-described African Americans aged between 18 and 70 years with a diastolic BP >95 mm Hg and a glomerular filtration rate (GFR) of 20 to 65 mL/min. Major exclusion criteria included diabetes, a urinary protein to creatinine ratio (UP/Cr) >2.5, malignant hypertension during the previous 6 months, secondary hypertension, serious systemic disease, heart failure, or a specific indication for, or contraindication to, a study medication. The study was conducted in two phases. The initial trial phase, previously published (JAMA. 2002;288:2421–2431), consisted of a prospective, randomized clinical trial with a 3×2 factorial design, which was followed by a nonrandomized observational cohort phase (the subject of the present paper). In the initial trial phase, 1094 patients between February 1995 and September 1998 were randomized to receive either intensive BP control (mean arterial pressure [MAP] <92 mm Hg, roughly corresponding to BP <130/80 mm Hg) or standard BP control (MAP, 102–107 mm Hg, roughly corresponding to BP 140/90 mm Hg). Within both BP target groups, patients were randomly assigned to 1 of 3 initial drug therapies: ramipril, an angiotensin-converting enzyme (ACE) inhibitor; metoprolol succinate, a sustained-release β-blocker; or amlodipine, a dihydropyridine calcium channel blocker (CCB). If the BP target was not achieved with the initial medication, other antihypertensive medications, including furosemide, doxazosin, clonidine, hydralazine, and minoxidil, were sequentially added. At the end of the initial randomized trial phase, all patients who had not experienced the primary CKD end point were invited to enroll in an observational cohort phase. Initiated in April 2002, all patients in the cohort phase were switched from their original blinded randomized therapy to open-label ramipril. If ramipril was not tolerated, the patient was switched to an angiotensin receptor blocker (ARB) selected by the clinical site investigator. Once again, as in the trial phase, if the BP target was not achieved with the highest-tolerated dose of ramipril, other medications, including furosemide, β-blockers, CCBs, centrally acting α-adrenergic blockers, and direct-acting vasodilators were added as needed to the regimen. At the start of the cohort phase, the BP target for all patients was <140/90 mm Hg; however, the target was subsequently lowered to <130/80 mm Hg in 2004 after the publication of JNC 7.

Both the initial trial and follow-up cohort phase had the same primary end point, a composite of death, a doubling of the serum creatinine level, or a diagnosis of end-stage renal disease (ESRD), defined as initiation of dialysis or kidney transplant. Based on evidence from other studies showing a beneficial effect of ACE inhibitors on the level of protein excretion and renal disease progression and a deleterious effect of dihydropyridine CCBs on the same two aspects of renal disease, the steering committee (blinded to the AASK data) requested that the Data Safety Monitoring Board be provided with information on the levels of proteinuria in the ramipril and amlodipine-treated patients during the initial trial phase, as an extension of the original primary comparison. They found a significant relationship between baseline proteinuria and the short-term and total mean GFR slopes. Therefore, a subgroup analysis was performed in trial-phase patients with baseline UP/Cr >0.22 (one third of participants, accounting for two thirds of all events) and <0.22 (two thirds of participants, accounting for one third of the events). The UP/Cr cut point of 0.22, corresponding to 300 mg/d of urinary protein, was therefore selected post hoc, independent of the original AASK protocol. Accordingly, as in the trial phase, a post hoc prespecified stratified analysis was performed during the cohort phase of the trial based on the presence or absence of clinical proteinuria (above or below baseline protein to creatinine ratio >0.22).

At baseline, the mean BP in the 1094 trial-phase patients was 150/95 mm Hg. Throughout the 4.1 mean years of follow-up in the initial trial phase, the mean BP was significantly lower in the intensive control group (130/78 mm Hg) as compared with the standard control group (142/86 mm Hg). During the 8 mean years of follow-up in the cohort phase, since all patients had a common target BP of <130/80 mm Hg, the difference was smaller (131/78 mm Hg in the intensively treated group vs 134/78 mm Hg in the standard group). Throughout the entire study, ACE inhibitor and ARB use was similar in both BP target groups.

Among all patients across both phases of the study, there was no significant difference between the intensive and standard BP control groups in the incidence of the primary end point (hazard ratio [HR], 0.91; 95% confidence interval [CI], 0.77–1.08; P=.27) or any secondary end point. However, the effects of the randomized BP target significantly differed according to the baseline protein to creatinine ratio (P=.02 for interaction). Among patients with a baseline protein to creatinine ratio <0.22, there was no difference in the primary end point across the entire study (HR, 1.18; 95% CI, 0.93–1.50; P=.16) and an inconsistent pattern for the outcome of a doubling of the serum creatinine or ESRD (HR, 1.39; 95% CI, 1.04–1.87; P=.03) and the outcome of ESRD or death (HR, 1.12; 95% CI, 0.87–1.45; P=.39). However, among patients with a baseline protein to creatinine ratio >0.22, those in the intensive-control group had a significant reduction in the risk of the primary end point (HR, 0.73; 95% CI, 0.58–0.93; P=.001) as well as the two secondary outcomes: a doubling of the serum creatinine or ESRD (HR, 0.76; 95% CI, 0.58–0.99; P=.04) and ESRD or death (HR, 0.67; 95% CI, 0.52–0.87; P=.002). Furthermore, the beneficial effect in the clinical proteinuria subgroup only reached clinical significance during the cohort follow-up phase of the study. Finally, during the entire course of the trial, while the rate of the composite end point remained similar in the two BP groups (about 7 events per 100 person-years), the overall incidence of the primary end point was nearly 3 times higher in those with, as compared with those without, proteinuria.

Overall, the AASK data do not support the guideline recommendation that a target BP of <130/80 mm Hg prevents progression of nondiabetic hypertensive renal disease in all patients. However, there may be differential effects based on baseline level of proteinuria, whereby more intensive BP reduction is of greater benefit in patients with significant proteinuria. Since this finding was largely based on observational subgroup analysis, further research is needed to clarify this issue.—Appel LJ, Wright JT, Green T, et al; for the AASK Collaborative Research Group. Intensive blood-pressure control in hypertensive chronic kidney disease. N Engl J Med. 2010;363:918–929.

Comment

  1. Top of page
  2. Comment

Guidelines, and the majority of clinicians who follow them, have long been of the opinion that when it comes to BP control in patients with CKD the lower the achieved BP the better. However, this conclusion has been almost exclusively based on observational data rather than prospective, randomized clinical trial evidence. As the authors of this report highlight, this question is very difficult to test in prospective trials because even high-risk patients with relatively advanced CKD generally have a relatively slow progression to ESRD, often measured in decades rather than the 12.2 years of total follow-up in the complete AASK trial.

Prior to these AASK results, the only other major, well-designed, large, prospective, long-term, randomized clinical trial to examine truly intensive BP reduction on the progression of CKD in nondiabetic adults was the Modification of Diet in Renal Disease (MDRD) trial, which compared the effects of an intensive BP target (MAP <92 mm Hg) with a standard BP target (MAP <107 mm Hg). During the initial phase of the MDRD trial, which lasted about 3 years, intensive BP control did not demonstrate a significant effect on GFR in the entire cohort; however, in subgroup analysis it did appear that the group with >1 g/d of protein excretion at baseline had less progression to ESRD with more intensive BP treatment. Additionally, in the long-term observational follow-up phase, patients originally assigned to the intensive BP control group had a lower incidence of ESRD, regardless of whether they had significant proteinuria at baseline. Unfortunately, since ACE inhibitor use was much higher in the intensively treated group, it was impossible to determine whether this beneficial effect in the long-term observational phase of MDRD was due to intensive BP control or the benefits of renin-angiotensin system (RAS) blockade. In AASK, on the other hand, all patients, regardless of original BP target assignment, received equivalent ACE inhibitor (or ARB if ACE inhibitors were not tolerated) treatment in the follow-up cohort phase. Therefore, the investigators were able to distinguish the benefit of BP control from the differential importance of RAS blockade. As seen in the MDRD trial, patients with clinical proteinuria at baseline had a greater risk of clinical progression of CKD and appeared to be the only patients who benefited from more intensive BP control.

The results reported here are subject to a number of potential shortcomings. First, the cohort phase was not randomized, but purely observational. Second, the conclusion reported here is based on subgroup analysis, which optimally should only be considered hypothesis-generating. Third, BP control was based only on office readings, which may not reflect cardiovascular and renal risk as well as out-of-office measurements (home or 24-hour ambulatory BP measurement) in this patient population. Finally, this study included only self-reported African American hypertensive patients without diabetes. Whether results would be similar in other racial and ethnic groups or with other causes of CKD is presently unknown.

So, what do these results mean for treatment of patients with CKD? Just as we have seen for BP goals in the diabetic patient with the publication of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) BP data, these data from AASK should cause us to re-evaluate BP goals in patients with CKD. We would suggest the following recommendations for patients with hypertension and CKD. First, CKD should continue to be a compelling indication for use of an ACE inhibitor or ARB either alone or as part of combination antihypertensive therapy for all patients with hypertension. Second, since this study and others suggest that patients with significant proteinuria have a substantially worse cardiovascular and renal prognosis, our CKD staging system should be revised to rely not solely on GFR but also on the presence or absence of proteinuria (albuminuria). Third, for patients with CKD and no evidence of significant proteinuria, a BP target of <140/90 mm Hg appears sufficient. Fourth, for patients with CKD and clinical proteinuria, the BP target should be <130/80 mm Hg, but we anxiously await future trials that should better elucidate the evidence for this statement. Given the significantly higher rate of CKD progression among patients with proteinuria, future long-term studies should specifically focus on this patient population and examine whether even more intensive BP control, use of dual RAS blockade, or other interventions can help slow the progression to ESRD. While these recommendations may be more complicated than our current guidelines for staging and treating patients with CKD, they more accurately reflect our current state of knowledge about this important patient population whose numbers are steadily increasing.