Prospective Validation of the “Fifty-Fifty” Criteria as an Early and Accurate Predictor of Death After Liver Resection in Intensive Care Unit Patients

Abstract

Postoperative liver failure after hepatectomy has been identified by the association of prothrombin time 50 micromol/L (the "50-50" criteria). Whether these criteria are of prognostic value in a prospective study remains unknown. To determine prospectively the prognostic value of the 50-50 criteria on day 3 and day 5 in intensive care unit (ICU) patients after hepatectomy. From January 2005 to February 2007, among 436 elective liver resections, 99 (23%) consecutive patients aged 58 +/- 17 years were admitted postoperatively in ICU with a mean SAPSII 25 +/- 10. Malignant disease was present in 87 and major resections (< or =3 segments) were performed in 79 (80%) cases. The underlying liver parenchyma was abnormal in 59 (59%) cases including cirrhosis, fibrosis, or steatosis >30% in 19, 23, and 17 patients, respectively. The 50-50 criteria were present on day 3 in 10 patients and on day 5 in 13. Ten patients (10, 6%) died in ICU. Survivors with these criteria were characterized by early aggressive support including reoperation and/or liver assist system. Nonsurvivors were more often cirrhotic, had significantly higher SAPS II and more frequently postoperative prolonged mechanical ventilation. The 50-50 criteria on days 3 and 5 were predictors of death on multivariate analysis [OR (95% CI): 12.7 (2.3-71.4), OR (95% CI): 29.4 (4.9-167), respectively]. After hepatic resection, results of this prospective study validate the 50-50 criteria as a predictive factor of mortality in ICU on both days 3 and 5. These criteria allow an early diagnosis of postoperative liver failure, which may contribute to reduce mortality in ICU patients after hepatectomy.