Use of perioperative epidural analgesia among Medicare patients undergoing hepatic and pancreatic surgery

Background: We sought to characterize epidural analgesia (EA) use among Medicare patients undergoing hepatopancreatic (HP) procedures, identify factors associated with EA use and asses perioperative outcomes. Methods: Patients undergoing HP surgery were identified using the Inpatient Standard Analytic Files. Logistic regression was utilized to identify factors associated with EA receipt, and assess associations of EA with in-hospital outcomes and Medicare expenditures. Results: Among 20,562 patients included in the study, 6.7% (n =1362) had EA. There was no difference in the odds of complications (OR 1.05, 95% CI 0.93–1.19) or blood transfusions (OR 0.90, 95% CI 0.79–1.03) with EA versus conventional analgesia (CA). The odds of prolonged LOS (OR 1.16, 95% CI 1.03–1.30) were higher with EA; the odds of in-hospital mortality were higher with conventional analgesia (OR 1.90, 95% CI 1.28–2.83). Medicare payments for liver surgery were comparable among EA ($19,500) versus conventional analgesia ($19,300, p = 0.85) and slightly higher for EA ($23,600) versus conventional analgesia ($22,000, p < 0.001) for pancreatic procedures. Conclusion: EA utilization among Medicare patients undergoing HP was low. While EA was not associated with morbidity, it resulted in an average additional one day LOS and slightly higher expenditures in pancreatic surgery