|
|
Back to 2016 Annual Meeting
A Clinical Score Predicting the Occurence of Liver-related Complications Following Hepatectomy
Andrea Ruzzenente, Fabio Bagante*, Adriano E. Dorna, Simone Conci, Tommaso Campagnaro, Corrado Pedrazzani, Calogero Iacono, Alfredo Guglielmi
G. B. Rossi Hospital, University of Verona, Verona, Italy
Background:
Hepatic resection is the treatment of choice for patients with symptomatic benign and malignant liver tumors. Despite a low rate of mortality, liver surgery is still associated with high rate of post-operative morbidity. Particularly, liver-related complications (LRC) including post-hepatectomy liver failure (PHLF) and ascites are potentially life-threatening complication after hepatectomy. We sought to develop a new prognostic model able to predict the risk of LRC in patients undergoing liver surgery.
Methods:
A prospectively collected institution database was review to identify 138 patients who underwent minor hepatectomy on injured liver (cirrhosis, cholestasis, nonalcholic and post-chemotherapy steatohepatitis) or major hepatectomy for both benign and malignant liver diseases between January 2010 and December 2014. Clinical and pathological information available for these patients included preoperative estimation (CT-volumetry) of the future remnant liver volume (FRLV) and the indocyanine-green retention rate at 15 minutes (ICG). Data on the short-term outcomes comprised rate of overall complications, LRC, and 30-day mortality.
Results: The median age of the study population was 66.2 years (IQR:58.6-72.1) and the majority of patients (n=91, 65.9%) was male. Major (≥3 liver segments) and minor (<3 liver segments) hepatectomy were performed on 73 (52.9%) and 65 (47.1%) patients, respectively. The median ICG was 5.2% (IQR, 3.3-10.6) while the median FRLV resulted 68% (IQR, 45-85). The 30-day mortality rate was 3.6% (n=5), according to Clavien-Dindo classification the grade of complication was ≥3 in 24 (24%) patients. LRC occurred in 29 (21.0%) patients, the most frequent LRC was ascites (n=21, 15.2%). In the univariate analysis, several variables including type of hepatectomy, ICG, FRLV, preoperative level of hemoglobin, GGT, ALT, Bilirubin, Albumin, and Sodium were associated with LRC (all p<0.05). Conversely, in the multivariable analysis, ICG (OR=1.06, 95%CI, 1.01-1.11, p=0.02), and FRLV (OR=0.98, 95%CI, 0.96-0.99, p=0.03) resulted the only independent predictors of LRC. Moreover, the median ICG and FRLV were 4.2% and 74.0% for patients without LRC compared to 7.2% and 62.0% for patients with LRCs (p=0.01 and p=0.05). These variables were used to develop a clinical score to predict LRC. A FRLV of 100% and a ICGR15 of 0% had a baseline score of 0 points. The clinical score increased of 10 points for additional 5% in ICG and of 4 points for a decrease of 5% in FRLV. Among the 60 patients with a score ≥46 points 20 (33.3%) had a LRC compared to 9 (11.5%) among the 78 patients with <46 points (p<0.001).
Conclusions:
Preoperative FRLV and ICG demonstrated the ability to predict the occurrence of LRC. Using these two variables we developed a clinical score able to identify those patients with an increased risk to develop postoperative LRC.
Back to 2016 Annual Meeting
|
