Postoperative renal replacement therapy after hydroxyethyl starch infusion: a meta-analysis of randomised trials

Background: Hydroxyethyl starch (HES) solutions are used for perioperative fluid management. However, robust renal safety data in surgical patients are lacking. Methods: A meta-analysis was performed of randomised clinical trials comparing HES with non-HES control fluid in adult surgical patients. The endpoint was recourse to renal replacement therapy (RRT). Eligible trials were identified by multiple methods including computer database searches. Results: Fifteen randomised trials with a total of 4409 surgical patients were included. The Crystalloid versus Hydroxyethyl Starch Trial (CHEST) was the source for 65.1% of the included surgical patients. Eighty-three of the 2157 patients assigned to HES (3.8%) underwent RRT versus 56 of 2252 control patients (2.5%). HES significantly increased recourse to RRT, with a pooled relative risk (RR) of 1.44 and 95% confidence interval (95% CI) of 1.04-2.01. The absolute risk increase in recourse to RRT attributable to HES was 1.2% (95% CI: 0.1-2.2%), indicating a number needed to treat with HES of 85 to prompt RRT in one additional patient. In a subset of trials comparing HES 130/0.4 with crystalloid, the pooled RR for recourse to RRT (1.47; 95% CI: 1.02-2.12) coincided closely with the overall pooled RR of 1.44. In a subgroup analysis of data from CHEST, the RR for RRT was higher in surgical than non-surgical patients (ratio of RR: 1.19; 95% CI: 0.77-1.83); however, the difference was not significant (p = 0.43). Conclusions: HES increased recourse to RRT among surgical patients. There was no evidence that surgical patients are at lower risk for HES-induced acute kidney injury than non-surgical patients. Introduction Colloids have long been a common choice for perioperative volume expansion. In a 2010 survey of 391 intensive care units (ICUs) in 25 countries, colloids were administered to 67.3% of patients admitted from the operating room after elective Postoperative renal replacement therapy after hydroxyethyl starch infusion: a meta-analysis of randomised trials surgery and 67.8% of those after emergency surgery.1 In the prospective observational Sepsis Occurrence in Acutely ill Patients (SOAP) study at 198 European ICUs, 33.6% of elective surgery patients received the artificial colloid hydroxyethyl starch (HES) in the ICU; in patients undergoing emergency surgery this proportion was 20.7%.2 Exposure to HES often commences in the operating room. In a prospective sequential analysis of perioperative fluid therapy in cardiopulmonary bypass surgery at one centre, all 6478 patients received HES intraoperatively to prime the extracorporeal circuit.3 Perioperative fluid management practices may now be in flux, however, because of safety concerns about HES solutions, most notably their potential to cause postoperative acute kidney injury (AKI).3-6 In a survey of 80 German ICUs treating patients after cardiothoracic surgery, the proportion of respondents favouring HES as the first choice for volume therapy declined from 65.3 to 38.7% in the wake of emerging evidence about HES-related AKI.7 Data implicating exposure to HES solutions as a cause of AKI have accumulated from rigorous large-scale randomised trials8-10 and from meta-analyses of randomised trials.11,12 While it had been speculated that the renal safety profiles of various HES solutions might differ, such differences could not be confirmed in meta-analyses,11,12 and an expert panel convened by the US Food and Drug Administration (FDA) concluded that AKI is a class effect of HES solutions.13 The European Society of Intensive Care has recommended against the routine use of HES solutions in patients with severe sepsis and other ICU patients at increased risk for AKI.14 Evidence of AKI and excess mortality attributable to HES has also prompted regulatory actions. Both the FDA15 and the European Medicines Agency16 (EMA) have determined that HES should no longer be used in critically ill patients, including those with sepsis. With the implementation of new risk minimisation procedures, including monitoring of renal function for 90 days after HES exposure, the perioperative administration of HES will still be permitted, although such continued HES use is controversial.17,18 M.M. Wilkes, R.J. Navickis Hygeia Associates, Grass Valley, California, USA