Coagulation profile changes and safety of epidural analgesia after hepatectomy: a retrospective study

We aimed to review post-hepatectomy coagulation profile changes, to assess outcomes of epidural catheter placement in post-hepatectomy patients, and to make justifications regarding use of epidural analgesia in patients undergoing hepatectomy.

We performed a retrospective study of 141 patients undergoing liver resection at Brigham and Women’s Hospital between January 1, 2007 and December 31, 2011. All patients were between 21 and 85 years old, with ASA physical status classification of II or III, and Child-Pugh scores ≤6. Patients undergoing laparoscopy or resection of less than three hepatic segments were excluded. We examined pre-operative hematocrit, platelet count, coagulation studies, and liver function tests, and trended values for 7 post-operative days. We examined frequency of epidural placement, use of peri-operative anticoagulation, and incidence of epidural-related complications.

We demonstrated statistically significant decreases in hematocrit and platelet counts, as well as statistically significant increases in prothrombin time and international normalized ratio (INR) values. Thirty-two percent of patients required vitamin K or fresh frozen plasma to achieve an INR ≤1.3. No patient required platelet transfusion to achieve platelets ≥100,000 prior to catheter removal. Changes in post-operative partial thromboplastin time were not significant. Epidural catheters were placed in 90 % of liver resections performed at our institution. We noted no epidural hematomas, even in the 7 % of patients in whom the epidural catheter was inadvertently removed before coagulation criteria were met. The latter group was monitored with hourly neurologic exams for 24 h.

Epidural analgesia may be safely used in patients undergoing major hepatic resection, providing that they have normal pre-operative coagulation and catheters are removed only when resection-induced perioperative coagulopathy has resolved or has been corrected.