Incidence and predictors of portal and splenic vein thrombosis after pure laparoscopic splenectomy

Background. Optimal modalities for diagnosis, treatment, and surveillance of portal or splenic vein thrombosis have not yet been defined. The present retrospective study aimed to investigate the role of computed tomography performed systematically before and after laparoscopic splenectomy to assess the incidence of portal or splenic vein thrombosis, predictors, and outcomes. Methods. Computed tomography scans were obtained from 170 patients undergoing elective laparoscopic splenectomy between 2005 and 2015. Pre- and postoperative splenic vein diameter was measured at the splenoportal junction and at a distance of 2, 4, 6 cm from it. Univariate and multivariate analyses were used to identify portal or splenic vein thrombosis risk factors and predictors of treatment outcome. Results. Overall, 68.2% of patients had benign hematologic diseases; 64.1% showed splenomegaly. Portal or splenic vein thrombosis occurred in 53.5% of patients (91/170), of whom 49.5% were asymptomatic. Preoperative splenic vein diameter measurements at 2, 4, and 6 cm from the splenoportal junction were significantly greater in portal or splenic vein thrombosis patients than in no-portal or splenic vein thrombosis patients. Patients with splenic vein diameter $8 mm at all measured sites had a greater risk of developing portal or splenic vein thrombosis (P = .009; odds ratio, 2.57; 95% confidence interval, 1.26–5.23). The majority of thromboses involved the distal splenic vein (45.1%, 41/91), and 41.7% of patients had thromboses located in multiple sites. Fully 71.4% showed complete resolution of portal or splenic vein thrombosis. Thrombus location at a single site predicted a favorable treatment outcome (P < .0001). Conclusion. Portal or splenic vein thrombosis is a frequent complication of splenectomy that occurs asymptomatically in half of cases. Computed tomography could have an important role in identifying patients at risk of developing portal or splenic vein thrombosis as well as in predicting portal or splenic vein thrombosis resolution. (Surgery 2017;162:1219-30.)