Hepatic resection for hepatocellular carcinoma (HCC) in patients with cirrhosis is generally recommended for patients with Child-Turcotte-Pugh (CTP) Class A liver disease and early tumor stage. The Model for End-Stage Liver Disease (MELD) has been shown to accurately predict survival in patients with cirrhosis, but whether MELD is useful for selection of patients with cirrhosis for hepatic resection is unknown. We examined whether MELD was predictive of perioperative mortality and correlated MELD with other potential clinicopathologic factors to overall survival in patients with cirrhosis undergoing hepatic resection for HCC. A retrospective chart review was undertaken of patients with HCC and cirrhosis undergoing hepatic resection between 1993 and 2003. Eighty-two patients (62 men, 20 women; mean age, 62 years) were identified. Forty-five patients had MELD score > or =9 (range, 9-15) and CTP score ranged from 5 to 9 points. Fifty-nine patients underwent minor (<3 segments) hepatic resections (MELD < or =8, n = 29; MELD > or =9, n = 30) and 23 underwent major (> or =3 segments) hepatic resections (MELD < or =8, n = 8; MELD > or =9, n = 15). Perioperative mortality rate was 16%. MELD score < or =8 was associated with no perioperative mortality versus 29% for patients with an MELD score > or =9 (P < 0.01). Multivariate analysis demonstrated that MELD score > or =9 (P < 0.01), clinical tumor symptoms (P < 0.01), and ASA score (P = 0.046) are independent predictors of perioperative mortality. Multivariate analysis showed MELD > or =9 (P < 0.01), tumor size >5 cm (P < 0.01), high tumor grade (P = 0.03), and absence of tumor capsule (P < 0.01) as independent predictors of decreased long-term survival. MELD score was a strong predictor of both perioperative mortality and long-term survival in patients with cirrhosis undergoing hepatic resection for HCC. In patients with cirrhosis, hepatic resection (minor or major) for HCC is recommended if the MELD score is < or =8. In patients with MELD score > or =9, other treatment modalities should be considered.
View details for DOI 10.1016/j.gassur.2005.09.008
View details for Web of Science ID 000234336100003
View details for PubMedID 16332475