Choosing Wisely: Decreasing the incidence of perioperative blood transfusions in gynecologic oncology.

Abstract

OBJECTIVE\nTo evaluate the efficacy and economic impact of a transfusion reduction initiative for patients undergoing gynecologic surgery.\n\n\nMETHODS\nWe conducted a prospective healthcare improvement study to align transfusion practices with the American Society of Hematology s Choosing Wisely® campaign. Baseline transfusion rates were determined retrospectively for all major gynecologic surgical cases from 3/1/14 to 6/30/14. Data for the post-intervention period from 5/15/15 to 5/16/16 were captured prospectively. The primary outcome was transfusion within 72\u202fh of surgery. Secondary outcomes included perioperative morbidity, mortality, number of units ordered per transfusion episode and cost.\n\n\nRESULTS\nWe identified 1281 surgical cases, 334 in the baseline and 947 in the post-implementation cohort. The baseline cohort was noted to have a higher median estimated blood loss (100 v. 75\u202fmL, P\u202f<\u202f0.01). Otherwise, there were no differences in clinical or perioperative characteristics between the two cohorts. The perioperative transfusion rate decreased from 24% to 11% (adjusted OR 0.27, 95% CI 0.16 to 0.45; P\u202f<\u202f0.001). The perioperative laparotomy transfusion rate decreased from 48% to 23% (adjusted OR 0.21, 95% CI 0.12, 0.37; P\u202f<\u202f0.001). The number of occurrences in which more than one unit of blood was ordered at a time decreased from 65% to 23%, P\u202f<\u202f0.001. The incidence of surgical site infections declined in the post-intervention group, otherwise there were no differences in 30-day mortality, cardiac, venous thromboembolism or readmission rates between the groups. The projected cost savings was $161,112 over the 12-month intervention period.\n\n\nCONCLUSIONS\nImplementation of an educational based transfusion reduction program was associated with substantial reductions in perioperative transfusions and cost without significant changes in morbidity or mortality.