Localized hepatic ischemia after liver resection: a prospective evaluation.

Objective: To prospectively assess the frequency, severity, and extension of localized ischemia in the remaining liver parenchyma after hepatectomy.

Background: Major blood loss and postoperative ischemia of the remnant liver are known factors contributing to morbidity after liver surgery. The segmental anatomy of the liver and the techniques of selective hilar or suprahilar clamping of the Glissonian sheaths permit identification of ischemia on the surface of the corresponding segments for precise section of the parenchyma. Incomplete resection of a segment, or compromised blood supply to the remaining liver, may result in ischemia of various extension and severity.

Methods: Patients undergoing hepatectomy received enhanced computerized tomodensitometry with study of the arterial and venous phases within 48 hours after resection. We defined hepatic ischemia as reduced or absent contrast enhancement during the venous phase. We classified the severity of ischemia as hypoperfusion, nonperfusion, or necrosis. The extension of ischemia was identified as marginal, partial, or segmental. Factors that may influence postoperative ischemia were analyzed by univariate and multivariate analyses.

Results: One hundred fifty consecutive patients (70 F, 80 M, mean age 62 +/- 12 years) underwent 64 major and 81 minor hepatectomies and 5 wedge resections. We observed radiologic signs of ischemia in 38 patients (25.3%): 33 hypoperfusions (17 marginal, 12 partial, and 4 segmental), 3 nonperfusions (1 marginal, 1 partial, and 1 segmental), and 2 necroses (1 partial, 1 segmental). One patient with a segmental necrosis underwent an early reoperation. In all other cases, the evolution was spontaneously favorable. Postoperative peak levels of serum aspartate aminotransferase and alanine aminotransferase were significantly higher in patients with ischemia. Patients with ischemia had a significantly higher risk of developing a biliary leak (18.4% vs. 2.6%, P < 0.001). There was no correlation between liver ischemia and mortality (2%). None of the following factors were associated with ischemia after univariate and multivariate analysis: age, preoperative bilirubin level, liver fibrosis, malignant tumor, type of hepatectomy, surface of transection, weight of resected liver, Pringle maneuver, blood loss, and number of transfusions.

Conclusions: Some form of localized ischemia after hepatectomy was detected in 1 of 4 of our patients. Its clinical expression was discreet in the large majority of cases, even if it might have been one of the underlying causes of postoperative biliary fistulas. Clinical observation is sufficient to detect the rare patient with suspected postoperative liver ischemia that will require active treatment.