The influence of chemotherapy-associated sinusoidal dilatation on short-term outcome after partial hepatectomy for colorectal liver metastases: A systematic review with meta-analysis

A Summary background data: Hepatic sinusoidal dilatation (SD) is a histopathological entity that occurs in up to 75% of patients undergoing oxaliplatin-based chemotherapy for colorectal liver metastases (CRLM). Objective: To study the influence of SD on outcome after partial hepatectomy in patients with CRLM. Methods: Medline, Embase, CENTRAL, LILACS and CINAHL were searched for studies published between 01.01.2004 and 09.06.2015 with keywords: "sinusoidal obstruction syndrome", "hepatic veno-occlusive disease", and "Stuart-Bras syndrome". Studies comprising adults who underwent partial hepatectomy for CRLM with grading of SD and registration of postoperative morbidity and/or mortality were included. Risk of bias and quality of studies were evaluated with the Quality In Prognosis Studies Instrument (QUIPS) and modified GRADE framework. Results: Search strategies produced 2007 hits from which 23 and 13 articles were extracted for qualitative and quantitative analyses, respectively. Meta-analysis on the influence of SD grade 2-3 vs. SD grade 0-1 on postoperative overall morbidity showed an odds ratio (OR) of 1.26 [95% CI 0.74, 2.15](p = 0.40), an OR of 1.03 [0.15, 6.89]( p = 0.98) for liver failure, an OR of 1.21 [0.23, 6.35](p = 0.82) for overall mortality, and an OR of 3.52 [0.31, 39.91](p = 0.31) for liver-related morbidity. QUIPS showed a low to high risk of bias for studies, and GRADE showed very low quality of evidence per outcome. Conclusions: No significant effect of SD grade 2-3 on short-term outcome after partial hepatectomy was found. However, the data on which this conclusionwas based were not very robust and therefore no solid conclusions could be drawn.