INITIAL EGFR DECLINE AND LONG-TERM RENAL FUNCTION DURING BLOOD PRESSURE LOWERING THERAPY: A PATIENT LEVEL META-ANALYSIS OF THE SPRINT AND ACCORD-BP RANDOMIZED CONTROLLED TRIALS

Abstract

Objective: Lowering blood pressure (BP) can lead to an initial decline in renal function. Guidelines recommend treatment adjustment when the estimated glomerular filtration rate (eGFR) decreases by more than 20%. However, there is ongoing debate how much can be accepted. We determined the relation between initial eGFR and BP decrease, and subsequent eGFR during follow-up in participants from ACCORD-BP and SPRINT. Design and method: Diabetic and non-diabetic patients at increased cardiovascular risk were randomized to intensive (<120) or standard (<140 mmHg) systolic BP-targets. We determined the relation between initial decline in mean arterial pressure (MAP) and eGFR. Subsequently, we stratified patients to BP-target and initial eGFR decrease and assessed the relation with annual eGFR-decline after 1 year. Results: In total 13266 patients, aged 66.2 year, 39% female and with mean eGFR 76.4 ml/min/1.73m2, with 41126 eGFR measurements were analyzed. The >20% stratum consisted of 669 (10%) in the intensive and 334 (5%) participants in the standard group. Up to 10 mmHg MAP decrease, there was no significant change in eGFR, with a lower bound of 26% based on 95% of the participants. Hereafter, there was a linear association between MAP and eGFR with an initial eGFR decrease of 3.4% (95%CI 2.9–3.9%) per 10 mmHg MAP decrease, with a lower eGFR bound increasing to 46% after 40 mmHg MAP decrease (figure 1). There was no significant difference in eGFR slope (p = 0.37) after stratification to treatment allocation and initial eGFR decline, with an annual eGFR decrease of 1.24 (95%CI 1.09–1.39), 1.20 (95%CI 0.97–1.43) and 1.14 (95%CI 0.77–1.50) in the 5, 5–20% and >20% stratum during intensive and 0.95 (95%CI 0.81–1.09), 1.23 (95%CI 0.97–1.49) and 1.17 (95%CI 0.65–1.69) ml/min/1.73m2/year during standard treatment. Conclusions: In patients at high cardiovascular risk, we found no association between initial eGFR and annual eGFR decline during BP-lowering treatment. The observed degree of initial eGFR decline was strongly dependent on the achieved BP decrease, suggesting that the achieved BP difference should be taking into account when interpreting the eGFR decline during antihypertensive therapy.