Effects of an Angiotensin-Converting Enzyme Inhibitor on Residual Renal Function in Patients Receiving Peritoneal Dialysis

Context Few studies assess preservation of residual renal function after initiation of dialysis. Contribution This open-label randomized trial in patients receiving peritoneal dialysis showed that ramipril reduced declines in glomerular filtration rate and decreased the hazard rate of anuria at 1 year. Five of 30 patients stopped taking ramipril because of dizziness or cough; none withdrew as a result of hyperkalemia. Cautions The trial did not use a placebo comparison group, involved patients from a single university teaching hospital, and was not powered to detect differences in health care utilization or morbidity or mortality. The Editors Residual renal function is an important determinant of mortality and morbidity in patients receiving peritoneal dialysis (1, 2). It contributes to measures of dialysis adequacy, including Kt/V (dialyzer clearance multiplied by time over volume) and creatinine clearance (3, 4), and accounts for most of the variability in dialysis requirement (5). Previous studies of patients receiving peritoneal dialysis showed that nutritional indices gradually deteriorate when residual renal function declines (2, 6). More important, the ADEMEX (Adequacy of Peritoneal Dialysis in Mexico) study, a recently published randomized, controlled trial, found that increases in doses of fluid for peritoneal dialysis had no effect on patient survival (7). Available data suggest that renal and peritoneal clearances are not equivalent (2, 8, 9) and that an increase in the exchange volume or frequency of peritoneal dialysis cannot completely compensate for loss of residual renal function. As a result, measures to preserve residual renal function are an important target in the treatment of patients receiving dialysis. Residual renal function is better preserved with peritoneal dialysis than with hemodialysis (10, 11), but few studies have evaluated therapeutic approaches for preserving residual renal function after the initiation of dialysis. In one study, furosemide maintained the urine output of patients receiving peritoneal dialysis, but the rate of residual renal function decline was not altered (12). Recently, a retrospective study found that female sex, nonwhite race, history of diabetes, and history of congestive heart failure were predictors of loss of residual renal function (13). Patients treated with peritoneal dialysis had a 65% lower risk for losing residual renal function than did patients receiving hemodialysis. In addition, higher serum calcium levels, use of a calcium-channel blocker, and use of an angiotensin-converting enzyme (ACE) inhibitor were independently associated with decreased risk for loss of residual renal function (13). Several trials have shown that ACE inhibitors reduce the rates of renal function deterioration in patients with diabetic nephropathy (14, 15) and chronic proteinuric nephropathy (16, 17). We present results of a randomized, open-label study that examined the efficacy of ramipril, an ACE inhibitor, in preserving the residual renal function of patients receiving peritoneal dialysis. Methods Patients The Clinical Research Ethical Committee of the Chinese University of Hong Kong approved the study. We enrolled 60 stable patients from our hospital who were receiving peritoneal dialysis. All patients received traditional continuous ambulatory peritoneal dialysis, and all had standard peritoneal equilibration tests to determine peritoneal transport characteristics 1 month after dialysis was started. Within 3 months before randomization, blood pressure was measured, and serum and 24-hour urine samples were collected for the measurement of proteinuria and residual renal function to determine eligibility for the trial. We screened 217 patients in our dialysis unit; 72 met the enrollment criteria. On the basis of the sample size estimates, we invited 62 patients to participate in the study. Two declined for personal reasons. Residual glomerular filtration rate (GFR) was defined as the average of 24-hour urinary urea and creatinine clearances [18]. Enrollment criteria were as follows: 1) residual GFR of 2 mL/min per 1.73 m2 or more, 2) blood pressure of at least 120/70 mm Hg, and 3) no history of taking an ACE inhibitor or angiotensin-receptor blockers for at least 6 months. Since all patients had been followed in our unit for at least 6 months before randomization, we had exact knowledge about the use of an ACE inhibitor or angiotensin-receptor blocker before the study. Exclusion criteria were as follows: 1) underlying medical conditions, such as congestive heart failure, that mandate therapy with an ACE inhibitor or angiotensin IIreceptor antagonist; 2) myocardial infarction within the preceding 6 months; 3) clinically significant valvular disease; 4) malignant hypertension or KeithWagener grade III or IV hypertensive retinopathy; 5) history of hypertensive encephalopathy or cerebrovascular accident within the preceding 6 months; 6) any condition that may have precluded a patient from remaining in the study, such as alcohol or drug abuse, chronic liver disease, malignant disease, or psychiatric disorder; 7) known history of bilateral renal artery stenosis; and 8) history of allergy or intolerance to an ACE inhibitor. Patients with poor short-term likelihoods of survival, planned elective living related kidney transplantation, or planned transfer to another renal center within 6 months were also excluded. Design After obtaining informed consent and performing initial evaluations, we randomly assigned the 60 patients to either ramipril or no treatment. A computer-generated list that was maintained by a third party not involved in the conduct of the study was used for randomization. Investigators were unaware of the randomization schedule when recruiting patients, and both investigators and patients were not blinded during the follow-up period. Thirty patients received 5 mg of ramipril daily, the dosage commonly used for treating proteinuric nephropathy (16, 17). Antihypertensive agents other than ACE inhibitors were allowed. Doses were adjusted appropriately to achieve and maintain the target blood pressure of 135/85 mm Hg or to avoid symptomatic hypotension. Thirty patients in the control group received identical clinical management, except that ramipril was not prescribed. After randomization, patients were followed at 0, 3, 6, 9, and 12 months and at any time in between according to clinical need. At each clinic visit, serum creatinine and electrolyte concentrations, complete blood counts, and other serum biochemical values (uric acid, glucose, and liver enzymes) were measured. Residual GFR was assessed at 0, 3, 6, 9, and 12 months by 24-hour urinary collection (18). Indices of the adequacy of dialysis, including Kt/V and weekly creatinine clearance, were assessed at 0, 6, and 12 months by 24-hour dialysate and urinary collection (19). We recommended that all patients limit their sodium intake and that they eat 1.0 to 1.2 g of protein/kg of body weight daily. Persons who performed the 24-hour urinary assessments were unaware of the patients' assignment status. To further avoid bias in outcome assessment, the time at which anuria began was checked by both the investigators and clinic nurses; the nurses were unaware of the treatment assignment of patients. Outcome Measures The primary outcome measures were the longitudinal change in residual GFR and the time to anuria. Anuria was defined as total absence of urine output. Secondary outcome measures included urinary protein excretion, death from any cause, duration of hospitalization for any cause, and cardiovascular events. Cardiovascular events included death from cardiovascular causes, nonfatal myocardial infarction, cerebrovascular events with permanent neurologic deficit, and peripheral vascular disease requiring lower-limb amputation above the ankle. Data for secondary outcomes were assessed by using the computerized Clinical Management System of the Hong Kong Hospital Authority and the Renal Registry Database, developed and maintained by the Central Renal Committee, Hong Kong. Patients assigned to the ramipril group were asked open-ended questions about adverse events at each clinic visit. Patients assigned to no treatment were not routinely asked open-ended questions about adverse events. Statistical Analysis The sample size was estimated before the study with Power Analysis and Sample Size software (PASS 2000, NCSS, Kaysville, Utah). Our previous study on the adequacy of peritoneal dialysis showed that the mean rate of residual GFR decline (SD) in patients not taking an ACE inhibitor was approximately 0.3 0.2 mL/min per 1.73 m2 per month (2). Group sample sizes of 30 each would achieve an 83% power to detect a predefined meaningful difference of 0.15 mL/min per 1.73 m2 per month between the null hypothesis that both group means are 0.3 mL/min per 1.73 m2 per month (that is, a 50% reduction in the rate of decline) and the alternative hypothesis that the mean of the ramipril group is 0.15 mL/min per 1.73 m2 per month, with a known group standard deviation of 0.2 and an level of 0.05 using a two-sided, two-sample t-test. Statistical analyses were performed by using SPSS statistical software, version 9.0 (SPSS, Inc., Chicago, Illinois). Results were expressed as the mean (SD) unless otherwise stated. P values less than 0.05 were considered significant. All P values were two tailed. Analyses were done on an intention-to-treat basis, irrespective of adherence to treatment regimen. After 12 months, 26 patients in the ramipril group and 27 in the control group were available for analysis of longitudinal change in residual GFR. The analysis of the effect of ramipril on longitudinal changes in residual GFR used repeated-measures analysis of covariance. Residual GFR was the repeated measure; treatment group was the between-group factor; a product term for treatment group by time interaction and diabetic status, body weig