In the perioperative period colloids are used to obtain circulatory stabilization. Distinction is made between natural (Albumin) and artificial colloids (e.g., Hydroxyethyl Starch, HES). Recently, HES has come under scrutiny after several trials suggested it to be associated with increased risk of mortality and acute renal injury in critically ill patients. While both major trials and large-scale observational data are lacking, the debate continues regarding the safety of perioperative HES use. Using a large national database we aimed to study the use and safety of HES vs Albumin in elective surgery.
Methods: After IRB approval, data on patients undergoing hysterectomies were accessed from the Premier Perspective database (Premier Inc., 2006-2012). Use of HES and Albumin was determined for the day of surgery and the day after surgery creating four groups: HES use only (HES), Albumin use only (ALB), HES/Albumin both used (COMB), no HES or Albumin used (NONE). Primary outcomes of interest were acute renal failure, need for blood transfusion, 30-day mortality, costs of hospitalization (COH), and length of stay (LOS). These were assessed in the four intervention groups, overall, and by patient subgroups based on intensive care unit admission, advanced age (75+ years), and cardiovascular compromise.
Results: Our analysis included 520,476 patients from 515 hospitals. HES was used in 2.9%, Albumin in 0.9% and both were used in 0.2% of the cases, respectively. Mean age for the HES group was 50.9 (SD 13.2) vs 56.4 (SD15.1) for ALB, 59.0 (SD 15.0) for COMB, and 47.2 (SD 11.6) years for the NONE group. A similar pattern was found for the primary outcomes: acute renal failure 1.8% vs 4.9%, 9.7% and 0.3%; blood transfusion 12.0% vs 25.6%, 29.1% and 2.6%; 30-day mortality 0.4% vs 1.0%, 2.5% and 0.05% (all P< 0.001). COH and LOHS were $13,417 vs $26,249, $33,837 and $7,579; 4.1 days versus 8.1 days, 10.8 days and 2.0 days, respectively. Patterns did not change when analyzing the patient subgroups.
Discussion: While there have been safety concerns on HES use in critically ill patients, in this ongoing analysis we were able to show that patients receiving various types of colloids differed significantly in characteristics and more importantly in complication rates. Ongoing regression analysis is targeted to determine the independent impact of HES and Albumin on perioperative outcomes in this elective surgical population.