Risk factors for coagulopathy after liver resection.

STUDY OBJECTIVE: To identify risk factors for coagulopathy in patients undergoing liver resection. DESIGN: A retrospective cohort study. SETTING: Patients who underwent liver resection at a university hospital between April 2010 and May 2011 were evaluated within seven days after surgery. PATIENTS: One hundred forty-seven patients were assessed for eligibility. Thirty needed to be excluded because of incomplete data (23) or a preexisting coagulopathy (7). MEASUREMENTS: Coagulopathy was defined as 1 or more of the following events: international normalized ratio ≥1.4, platelet count <80,000/µL, and partial thromboplastin time >38 seconds. Related to the time course and coagulation profile thresholds, 3 different groups could be distinguished: no coagulopathy, temporary coagulopathy, and persistent coagulopathy. MAIN RESULTS: Seventy-seven patients (65.8%) had no coagulopathy, whereas 33 (28.2%) developed temporary coagulopathy and 7 (6%) developed persistent coagulopathy until day 7. Preoperative international normalized ratio (P = .001), postoperative peak lactate levels (P = .012), and resected liver weight (P = .005) were identified as independent predictors. Preoperative liver transaminases and transfusion volumes of red blood cells and fresh frozen plasma were significantly higher in patients with persistent coagulopathy. CONCLUSIONS: Epidural anesthesia is feasible in patients scheduled for liver resection. Caution should be observed for patients with extended resection (≥3 segments) and increased postoperative lactate. In patients with preexisting liver disease, epidural catheters should be avoided.

Heating capabilities of the Hotline and Autoline at low flow rates.

BACKGROUND: At low flow rates, fluid warmers using coaxial warming tubes are superior in preventing heat loss. This laboratory investigation was performed in order to compare the heating capabilities of two coaxial fluid warmers. METHODS: The Hotline and the Autoline were investigated by using normal saline at various flow rates (10-99 ml x h(-1)). Final infusion temperatures were measured six times in a row at the end of the tubing by using a rapid-response thermometer. Final temperatures were compared with those of infusions, which passed through disposable i.v. tubing covered and warmed using an 'off label' convective air warming system (WarmTouch). Measurements were performed at two different room temperatures (20 and 24 degrees C). Each group was analyzed with respect to differences between various flow rates as well as differences between the groups at comparable flow rates by using a three-way anova with multiple comparisons according to Tukey's procedure. Significance was defined at P < 0.05. RESULTS: Both devices heat infusions at low flow rates efficiently above 34 degrees C, with the Hotline being more effective than the Autoline (P < 0.0001). Except for the lowest flow rate (10 ml x h(-1)), the Hotline delivered infusion temperatures between 38 and 39 degrees C, while the Autoline warmed the infusions upto 36 degrees C. While heating capability of the Hotline was improved with elevated room temperatures at low flow rates (10-60 and 80 ml x h(-1)), the Autoline demonstrated lower infusion temperatures throughout elevated room temperature at flow rates between 20 and 90 ml x h(-1). Both devices heated infusions more efficiently compared with 'off label used' convective air warmer (each with P < 0.0001). CONCLUSIONS: Both the Hotline and the Autoline heated infusions sufficiently at low flow rates. However, the heating capability of the Hotline was superior and can further be increased at low flow rates by increasing the room temperature.