The liver resection database at Memorial Sloan Kettering (MSK) was queried to identify patients undergoing hepatectomy for malignancy from 1993 to 2012. The MSK Institutional Review Board provided a waiver from IRB review and HIPAA Authorization. Patients were divided into 3 groups of similar size according to time period [early (1993-1999), middle (2000-2006), and recent (2007-2012)].
The general approach to patients under consideration for liver resection has been previously documented.(10 (link)) CT, MR, or PET imaging were used for preoperative radiologic evaluations, and intraoperative ultrasound was used in all cases. Portal inflow occlusion (Pringle maneuver) was used frequently, in 5- to 15-minute intervals. Control of portal, arterial, and biliary in-flow pedicles was performed extrahepatically or intrahepatically, depending on the resection planned, disease location, and surgeon preference. Low central venous pressure (<5 mmHg) was used in all cases when feasible. Parenchymal division was performed via a clamp-crushing technique with sutures or clips to control intrahepatic biliary and vascular structures; more recently, thermal bipolar devices were added. Postoperatively, patients were managed for 12-24 hours in the recovery room and then transferred to the surgical ward, unless clinical factors dictated a monitored setting. Red blood cell transfusions were guided by patient hemodynamic status in combination with Hgb level (<8 mg/dL).
Liver resections were quantified by number of segments resected using the Brisbane 2000 terminology of hepatic anatomy and resection:(11 (link)) enucleations (0), wedge resection and formal segmentectomy (1 ), sectionectomy (left lateral, right anterior or posterior, 2), left hepatectomy (3 (link)), right hepatectomy (4 (link)), and extended hepatectomy (5 (link)). Major hepatectomy was defined as resection of ≥three segments. For our data, “perioperative” referred to the time period that included both operation and recovery room. Complications from 1993 to 2002 were identified retrospectively. From 2002 to 2012 complications were entered prospectively. All complications were graded using a score of 1 to 5.(12 (link))
Complications were analyzed per patient and per resection. Major complications were defined as ≥grade 3. Liver dysfunction was defined by presence of at least one from the following: postoperative prolonged hyperbilirubinemia without obstruction or leak, prolonged coagulopathy, ascites (drainage >500 mL/day), or encephalopathy with hyperbilirubinemia.(5 (link)) Steatosis was defined as percentage of hepatocytes containing lipid vesicles divided by total number of hepatocytes. It was graded using the Kleiner-Brunt histologic scoring system, (13 (link)) where moderate steatosis was defined as 5 - 33%. Mortality was determined at 90-days post procedure.
Summary statistics are reported as median and interquartile range (IQR) for continuous variables, unless otherwise specified. Categorical variables are reported as percentages. Comparisons are made with t-tests or Mann-Whitney tests depending on type of distribution for continuous variables. Categorical variables are compared with Chi-squared or Fisher's exact test depending on number of observations. Multivariate analyses used logistic regression. All reported p values were two-tailed and those ≤0.05 were considered significant. All analyses were conducted using Stata/IC 12.1 (StataCorp LP, College Station, TX).
The general approach to patients under consideration for liver resection has been previously documented.(10 (link)) CT, MR, or PET imaging were used for preoperative radiologic evaluations, and intraoperative ultrasound was used in all cases. Portal inflow occlusion (Pringle maneuver) was used frequently, in 5- to 15-minute intervals. Control of portal, arterial, and biliary in-flow pedicles was performed extrahepatically or intrahepatically, depending on the resection planned, disease location, and surgeon preference. Low central venous pressure (<5 mmHg) was used in all cases when feasible. Parenchymal division was performed via a clamp-crushing technique with sutures or clips to control intrahepatic biliary and vascular structures; more recently, thermal bipolar devices were added. Postoperatively, patients were managed for 12-24 hours in the recovery room and then transferred to the surgical ward, unless clinical factors dictated a monitored setting. Red blood cell transfusions were guided by patient hemodynamic status in combination with Hgb level (<8 mg/dL).
Liver resections were quantified by number of segments resected using the Brisbane 2000 terminology of hepatic anatomy and resection:(11 (link)) enucleations (0), wedge resection and formal segmentectomy (1 ), sectionectomy (left lateral, right anterior or posterior, 2), left hepatectomy (3 (link)), right hepatectomy (4 (link)), and extended hepatectomy (5 (link)). Major hepatectomy was defined as resection of ≥three segments. For our data, “perioperative” referred to the time period that included both operation and recovery room. Complications from 1993 to 2002 were identified retrospectively. From 2002 to 2012 complications were entered prospectively. All complications were graded using a score of 1 to 5.(12 (link))
Complications were analyzed per patient and per resection. Major complications were defined as ≥grade 3. Liver dysfunction was defined by presence of at least one from the following: postoperative prolonged hyperbilirubinemia without obstruction or leak, prolonged coagulopathy, ascites (drainage >500 mL/day), or encephalopathy with hyperbilirubinemia.(5 (link)) Steatosis was defined as percentage of hepatocytes containing lipid vesicles divided by total number of hepatocytes. It was graded using the Kleiner-Brunt histologic scoring system, (13 (link)) where moderate steatosis was defined as 5 - 33%. Mortality was determined at 90-days post procedure.
Summary statistics are reported as median and interquartile range (IQR) for continuous variables, unless otherwise specified. Categorical variables are reported as percentages. Comparisons are made with t-tests or Mann-Whitney tests depending on type of distribution for continuous variables. Categorical variables are compared with Chi-squared or Fisher's exact test depending on number of observations. Multivariate analyses used logistic regression. All reported p values were two-tailed and those ≤0.05 were considered significant. All analyses were conducted using Stata/IC 12.1 (StataCorp LP, College Station, TX).
Free full text: Click here
