Evaluating the association between mean arterial blood pressure through the first 24 hours after cardiac surgery and postoperative acute kidney injury

点击翻译按钮获取中文摘要

Approximately half a million patients undergo cardiac surgery in the United States each year. Acute kidney injury (AKI) complicates 25% of cases, and is associated with increased hospital length of stay, cost, progression to chronic kidney disease and death. Inadequate blood pressure during the early postoperative period is a likely contributor to AKI after cardiac surgery that may be effectively addressed with a higher target mean arterial pressure (MAP) during this period. Current practice typically targets a MAP ≥65 mmHg after cardiac surgery but whether this is the optimal target pressure for minimizing kidney injury is unknown. The long-term objective of our research is to minimize the incidence of AKI after adult cardiac surgery through identification of an optimal postoperative target MAP in this setting. The objective of the proposed cohort study is to harness existing minute-by-minute postoperative MAP data from patients undergoing cardiac surgery to identify a postoperative MAP threshold beyond which the risk of AKI increases and, separately, explore for evidence of association between early postoperative MAP and other non-renal harm that may also need consideration in selecting the most appropriate target-MAP for testing in a clinical trial. Our Specific Aims are to: 1a) Evaluate the association between time spent in each of multiple absolute MAP range-bands (e.g., 65-69, 70-74 mmHg) through the first 24 hours(h) after cardiac surgery and AKI; 1b) Evaluate the association between time spent in each of multiple discrete MAP range- bands defined by the percent change from preoperative baseline MAP (e.g., -6 to -10%, -11 to -15%) through the first 24h after cardiac surgery and AKI; 2a) Evaluate the association between time spent in each of multiple absolute MAP range-bands through the first 24h after cardiac surgery and non-renal adverse outcomes including each of delirium, stroke, blood transfusion and unplanned surgical re- exploration; 2b) Evaluate the association between time spent in each of multiple discrete MAP range- bands defined by percent change from preoperative baseline MAP through the first 24h after cardiac surgery and the specified non-renal adverse outcomes. These aims will be achieved by regression modelling of longitudinal MAP profiles created from minute-by-minute MAP data through 24h postoperatively in patients undergoing cardiac surgery. Combined with validated outcome data and adjusted for key potentially confounding covariates appropriate to each outcome, we will test the hypothesis that: Increased time spent with MAP in each of the multiple discrete range-bands through 24h after surgery (relative to a reference MAP range-band) is associated with an increased odds for AKI and, separately, other non-renal outcomes. Our results will provide critical evidence to guide the data-driven selection of the most appropriate target MAP for a future trial testing the effectiveness of this alternate target MAP to reduce AKI and improve patient outcome after cardiac surgery.