Unplanned splenectomy during oesophagectomy does not affect survival.

OBJECTIVE: There are limited and conflicting data available concerning the incidence of inadvertent splenectomy and its impact on the outcome in patients who have undergone oesophagectomy. The aim of this study is to identify the factors associated with a likelihood of inadvertent splenectomy and its influence on early and long-term outcome in patients having oesophagectomy for oesophageal carcinoma. METHODS: A consecutive series of 738 oesophagectomies performed between 1991 and 2004 was analysed. In our practice, the spleen was removed only if damaged intraoperatively. Routine chemo- and immunoprophylaxis would subsequently be used. Multivariate analysis with logistic and Cox models determined significant variables. RESULTS: Of the 738 oesophagectomies, 48 (6.5%) had splenectomy. Neoadjuvant chemotherapy was administered to a minority of patients; none subsequently had splenectomy. There were significant differences between types of operation (Ivor-Lewis 18 (9.0%), left thoracolaparotomy 14 (9.9%) and left thoracophrenotomy 15 (3.9%), p=0.01). Splenectomy was more common with advanced N stage disease (OR=0.44 0.20-0.95]; p=0.04). Splenectomy resulted in more blood transfusions (median, 2 units vs 0 units; p=0.03) more anastomotic leaks (7 14.6%] vs 42 6.1%]; p=0.02) but not an increase in pulmonary complications (p=0.64) or in-hospital mortality (1 4.6%] vs 37 5.4%]; p=0.30). Splenectomy did not significantly affect median survival (551 332-770] days vs 627 554-700] days; p=0.63). CONCLUSION: Although inadvertent splenectomy increased the morbidity of oesophagectomy, it did not impair survival. Type of operation and advanced N stage are important risks for splenectomy. Though best avoided, most of the consequences of splenectomy can be managed. An unexpected relationship between splenectomy and anastomotic leaks needs further investigation.