The Effect of a Lower Target Blood Pressure on the Progression of Kidney Disease: Long-Term Follow-up of the Modification of Diet in Renal Disease Study

Context We do not know the optimal blood pressure needed to slow progression of chronic kidney disease. Contribution In this multicenter trial, 840 adults with mostly nondiabetic kidney disease and moderately to severely decreased glomerular filtration rate were randomly assigned to usual blood pressure control (target: mean arterial pressure < 107 mm Hg) or a low blood pressure goal (target: mean arterial pressure < 92 mm Hg). Approximately 10 years later, the hazard ratio for kidney failure in the low compared with the usual blood pressure group was 0.68 (95% CI, 0.57 to 0.82). Implications A low blood pressure target slows progression of nondiabetic kidney disease. The Editors The Seventh Report of the Joint National Commission on Prevention, Detection, Assessment and Treatment of Hypertension and the National Kidney Foundation Kidney Disease Outcome Quality Initiative Clinical Practice Guidelines on Hypertension and Antihypertensive Agents in Chronic Kidney Disease recommend a target blood pressure of less than 130/80 mm Hg in patients with chronic kidney disease (1, 2). These guidelines are based on data from observational studies demonstrating that persons with chronic kidney disease are at high risk for cardiovascular disease and from subgroup analyses of the Modification of Diet in Renal Disease (MDRD) Study suggesting that a low target blood pressure may slow the decline in kidney function in persons with moderate to high levels of proteinuria (3). However, no published trials in diabetic or nondiabetic kidney disease have demonstrated that a lower target blood pressure reduces the incidence of kidney failure. Many types of chronic kidney disease progress slowly, requiring years for the onset of kidney failure. In clinical trials, progression of kidney disease is usually assessed by change in kidney function, which is measured as the glomerular filtration rate (GFR) or serum creatinine concentration. Two randomized, controlled trials in nondiabetic kidney disease have compared the effects of a low target blood pressure with those of a usual target blood pressure on the decline in GFR (3, 4). The MDRD Study enrolled participants with predominantly nondiabetic kidney diseases, a baseline GFR of 13 to 55 mL/min per 1.73 m2, and median proteinuria of 0.35 g/d (3). In that study, a low target blood pressure did not produce a slower projected mean decline in GFR at 3 years compared with the usual target blood pressure. However, participants with greater proteinuria significantly benefited from the low target blood pressure (3). On the basis of these findings, it was recommended that participants with proteinuria greater than 1 g/d have a lower target blood pressure. The African American Study of Kidney Disease and Hypertension (AASK) enrolled participants with hypertensive nephrosclerosis, a baseline GFR of 20 to 70 mL/min per 1.73 m2, and median baseline proteinuria of 0.081 g/d (4). In that study, a lower target blood pressure did not reduce the mean decline in GFR or the risk for the composite outcome of reduction in GFR of at least 50%, kidney failure, or all-cause mortality during a median follow-up of 4 years. The average rates of decline in GFR in the MDRD Study and AASK were 4 mL/min per 1.73 m2 per year and 2 mL/min per 1.73 m2 per year, respectively, and relatively few participants developed kidney failure during the observation period. The long-term effect of a low target blood pressure on the onset of kidney failure may differ from the effect on decline in GFR observed during the trial. Therefore, we evaluated the effects of a low target blood pressure on the onset of kidney failure and all-cause mortality through extended follow-up after completion of the MDRD Study. Methods The MDRD Randomized, Controlled Trial The MDRD Study was a randomized, controlled trial of the effect of dietary protein restriction and strict control of blood pressure on the progression of kidney disease. Details of the design and randomization criteria are provided elsewhere (5, 6). The trial was conducted from 1989 to 1993. Eligibility criteria for enrollment were age 18 to 70 years and presence of chronic kidney disease, with a serum creatinine concentration of 123.8 mol/L (1.4 mg/dL) to 618.8 mol/L (7.0 mg/dL) in men and 106.1 mol/L (1.2 mg/dL) to 618.8 mol/L (7.0 mg/dL) in women. Persons with diabetes requiring therapy with insulin, class III or IV congestive heart failure, renal artery stenosis, history of kidney transplantation, or frequent hospitalizations were excluded. Eight hundred forty participants were enrolled: Study A included 585 participants with a baseline GFR of 25 to 55 mL/min per 1.73 m2, and study B included 255 participants with a baseline GFR of 13 to 24 mL/min per 1.73 m2. Participants in both studies were randomly assigned to a low or a usual target blood pressure. The low target blood pressure was a mean arterial pressure less than 92 mm Hg (equivalent to a blood pressure less than 125/75 mm Hg) for participants 60 years of age or younger and less than 98 mm Hg for participants 61 years of age or older. The usual target blood pressure was a mean arterial pressure less than 107 mm Hg (equivalent to a blood pressure of 140/90 mm Hg) for participants 60 years of age or younger and less than 113 mm Hg for participants 61 years of age or older. Participants in study A were randomly assigned to receive a usual-protein diet (1.0 g/kg of body weight daily) or a low-protein diet (0.6 g/kg daily), whereas participants in study B were randomly assigned to receive a low-protein diet (0.6 g/kg daily) or a very-low-protein diet (0.28 g/kg daily, with keto acid and amino acid supplementation). No interactions were noted between the dietary intervention and the blood pressure intervention (3). Participants were followed up to 4 years. The rate of decline in GFR, measured as kidney clearance of 125I-iothalamate, was the primary outcome. Long-Term Follow-up (19932000) of the MDRD Study The investigational review boards of the Cleveland Clinic (Data Coordinating Center for the MDRD Study) and Tufts-New England Medical Center approved the long-term follow-up study. Kidney failure, defined by the requirement for dialysis or kidney transplantation, and a composite of kidney failure or all-cause mortality were the outcomes of interest for our analysis. All-cause mortality before kidney failure was included in the composite outcome because it is a competing risk factor for kidney failure. Participants were censored on 31 December 2000 if neither of these outcomes had occurred. Blood Pressure Measurement During the Trial (19891993) Trained personnel measured blood pressure monthly by using a random-zero mercury sphygmomanometer. Therapeutic regimens were modified monthly or more often as needed to achieve target blood pressure. For our analysis, mean follow-up blood pressure during the study was defined as the average of all blood pressure measurements obtained after the 4th month of follow-up. This definition differs from that of previous studies, in which the average of all blood pressure measurements during follow-up was calculated (3, 7). The change in definition was introduced to allow comparison with the AASK Study. During Long-Term Follow-up (19932000) No specific target blood pressure was recommended after completion of the trial. Blood pressure was measured by using the same methods in a large subgroup of participants 9 months after the end of the study (phase V). Thereafter, blood pressure measurements are not available. Antihypertensive Regimens During the Trial (19891993) After random allocation, therapeutic regimens were modified to achieve the target blood pressure. Nonpharmacologic therapy consisted of recommendations for exercise and weight loss and reductions in intake of dietary sodium and alcohol. For pharmacologic therapy, use of all agents was allowed, in the interest of achieving blood pressure goals. However, angiotensin-converting enzyme (ACE) inhibitors, with or without a diuretic, were encouraged as agents of first choice. Calcium-channel blockers, with or without a diuretic, were encouraged as agents of second choice. During Long-Term Follow-up (19932000) Because the MDRD Study did not evaluate the effects of individual antihypertensive agents on decline in kidney function, no specific pharmacologic therapy was recommended after completion of the trial. Outcomes The onset of kidney failure was ascertained from the U.S. Renal Data System by using participants' Social Security number, name, sex, and date of birth. Data on all-cause mortality were obtained from the National Death Index by using the same identifying information. One hundred eighty-five participants were recorded as starting dialysis or receiving a kidney transplant during or shortly after the randomized trial in both the MDRD Study database and the U.S. Renal Data System database. The recorded date of kidney failure agreed between the 2 databases to within 1 day for 50% of these participants, to within 30 days for 90%, and to within 120 days for 97%. Nine participants recorded as reaching kidney failure in the MDRD Study database were not found in the U.S. Renal Data System database. Except in the case of these 9 participants (4 of whom were in the usual target blood pressure group and 5 of whom were in the low target blood pressure group), data obtained from U.S. Renal Data System database were used in all analyses to define the date of kidney failure. All deaths that were coded in the U.S. Renal Data System database were confirmed in the National Death Index. For all analyses, the date in the National Death Index was used as the date of death so as to maintain consistency with participants who died but had not reached kidney failure and therefore were not included in the U.S. Renal Data System database. Statistical Analyses We combined participants from studies A and B to improve statistical power. This approach was justified