Venous thromboembolism (VTE) is a significant preventable cause of postoperative morbidity and mortality after major abdominopelvic surgery that calls for extended VTE prophylaxis (eVTEp). Literature suggests that significant racial disparities may exist in post-operative care. The study sought to examine if racial disparities exist in the administration of eVTEp after hysterectomy in a statewide collaborative. We conducted a retrospective cohort study of post-hysterectomy patients across 69 hospitals in the Michigan Surgical Quality Collaborative from January 2016 to February 2020. The variable of interest was race (Black/African or White American). The primary outcome was administration or absence of eVTEp. Descriptive statistics and mixed effects logistic regression were performed for risk adjustment with covariates such as age, cancer occurrence, inflammatory bowel disease, American Society of Anesthesiologists physical status classification, perioperative VTE prophylaxis, postoperative VTE prophylaxis, surgical approach, and surgical duration, among other variables. In total, 24,513 patients underwent hysterectomy. Of these patients, 1,107 (4.45%) received eVTEp, 153 (13.24%) of which were Black and 954 (82.53%) of which were White. Mixed effects logistic regression analysis suggested that Black patients were significantly less likely to receive eVTEp than White patients (odds ratio = 0.776; 95% CI: 0.615-0.979; P = 0.039). Additionally, tobacco use, coronary artery disease, bleeding disorder, cancer occurrence, functional status, perioperative VTE prophylaxis, surgical duration, length of stay, and surgical approach were associated with a higher likelihood of receiving eVTEp. eVTEp is recommended for the prevention of post-discharge VTE in select patients after hysterectomy. Regression analysis showed that, compared to their White counterparts, Black females were significantly less likely to receive eVTEp. The underlying reasons for this disparity require further investigation into possible socioeconomic influences and inherent biases.
Read full abstract