Predictive Indices of Morbidity and Mortality After Liver Resection

To the Editor:

The article by Schroeder et al¹ is a very interesting retrospective study evaluating predictive indices of morbidity and mortality in a large cohort of patients requiring partial hepatectomy.

Among 587 patients undergoing elective liver resections, the authors investigated the possibility of MELD score, Child-Pugh score, ASA classification, Charlson index of comorbidity, and age to predict postoperative morbidity and mortality. From their analysis, they concluded that preoperative MELD score should not be used to predict outcome in patients requiring liver resections.

Postoperative liver failure remains one of the most dreadful complications after partial hepatectomy and the precise evaluation of liver function is one of the key point to select patients undergoing liver surgery in the intent to reduce postoperative morbidity and mortality. In the last 20 years, several approaches have been introduced to assess hepatic function, such as common biochemical tests (total bilirubin, prothrombin time, serum albumin, transaminases), quantitative tests (galactose elimination capacity, indocyanine green clearance test, lidocaine MEGX test),^2,3 and predictive score such as Child-Pugh score.⁴

Actually, in patients with preserved liver function, it is well known that the balance between extension of liver resection and remaining liver volume is the key point to minimize postoperative risks.⁵ The assessment of total liver volume and remaining liver volume through the CT scan,⁶ concept of standard liver volume or GRWR coming from experience with living donor liver transplantation, and introduction of selective portal vein embolization to increase the remaining liver⁷ have dramatically reduced postoperative morbidity and mortality in liver surgery in the last 10 years⁸; in case of normal livers, the limit to the extension of partial hepatectomy is strongly related to the remaining liver functional volume and it cannot be derived by the common biochemical laboratory tests or quantitative tests.⁵

On the other hand, postoperative morbidity and mortality remain significant in cirrhotic patients and, in these cases, a precise evaluation of functional reserve is essential. Child-Pugh classification⁴ remains the most widely accepted system that provides an initial clue to the extent of resection that a cirrhotic patient can tolerate.^8,9 Child-Pugh class C cirrhosis is considered an absolute contraindication for hepatic resection in most of the major hepatobiliary centers, and only minor resection would be considered for Child-Pugh class B cirrhotic patients. For patients with Child-Pugh class A cirrhosis, the decision for hepatectomy often requires additional liver function quantitative tests that provide more refined evaluation of functional reserve.^2,3

Extrahepatic surgery, in particular cardiovascular and abdominal surgery, is associated with an increased incidence of postoperative morbidity and mortality in cirrhotic patients^10,11; large studies investigated the comorbidities that increased significantly postoperative risks after extrahepatic surgery, such as preoperative ascites, INR, creatinemia, ASA status, and total bilirubin.^11,12 In the intent to summarize these factors, MELD score and Child-Pugh classification have been reported to accurately predict postoperative outcome in cirrhotic patents undergoing extrahepatic surgery.^11,13,14

The model for end-stage liver disease (MELD) was developed to predict mortality of cirrhotic patients undergoing transjugular intrahepatic portosystemic shunts¹⁵ and subsequently applied as a disease severity index for priority in waiting list for liver transplantation.¹⁶ Since MELD score is not a reliable index in case of patients with preserved hepatic function, several modifications of MELD score have been introduced in the selection of the best candidates to liver transplantation.¹⁷

We have recently reported that MELD score is a reliable index in the preoperative evaluation of liver function in cirrhotic patients undergoing partial hepatectomy for hepatocellular carcinoma (HCC). In this study, we analyzed the postoperative outcome on 154 resected cirrhotic patients (92% Child A and 8% Child B patients). MELD score was below 9 in 49% of cases, between 9 and 10 in 35%, and 11 or above in 16%. A significant different in postoperative mortality, morbidity, hospital stay, and 1-year survival was reported in these 3 groups.¹⁸ We have further shown that MELD score provides a more accurate partition of Child-Pugh class A patients, and it is able to identify those patients who are at high risk of postoperative liver failure and those who can be safely treated with partial hepatectomy.

A recent similar study was reported by the Mayo Clinic group. Among 82 cirrhotic patients, 45 (54.9%) showed a preoperative MELD score above 9; 43 of 80 Child A patients had a MELD score above 9. As in our study, a significant difference in morbidity, mortality, and 1-year survival depending on MELD score was shown.¹⁹

In the reported study by Schroeder et al, the mean ± SD MELD score was 6.5 ± 4.5; in the session “discussion,” the same authors stated that 91% of patients had minimal or completely no evidence of liver disease. It is evident that in this series the common population was not cirrhotic.

In our series, as well as in the study from the Mayo Clinic, the median MELD score was 9, ie, there was a significant difference between Child A and Child B patients. Furthermore, in our experience of liver resection for HCC in patients with chronic hepatitis, the median MELD score was 8 and all of them below 10 (for this reason, they were excluded from the analysis of our study).

Even if other large controlled studied are needed, from the data available in the literature and our experience, MELD score seems to predict accurately postoperative liver failure and morbidity in patients with HCC on cirrhosis undergoing hepatic resection, and it is recommended in the preoperative assessment of liver function prior to hepatic resection in cirrhotic patients as well as in cirrhotic patients requiring extrahepatic surgery.

It should not be applied to predict outcome in the setting of noncirrhotic patients since it has not developed for these patients.

G. Ercolani