Factors associated with morbitity in liver hydatid surgery

Abstract

The purpose of the present paper was to determine the association between clinical evolutionary and laboratory variables with postoperative morbidity in patients surgically treated for liver hydatidosis (LH). A case-control nested cohort study was undertaken. Patients were studied and surgically treated in Hospital Regional of Temuco between 1994 and 2001. The morbidity variable registered with a minimum 12-month follow-up period was considered in a dichotomous way. The association with clinical, laboratory, evolutionary, and surgical variables was studied. Descriptive statistics, bivariate analysis using chi(2) and Fisher's exact test for categorical variables; t-test, anova, and Kruskal-Wallis for continuous variables; odds ratio calculations, and ordinal and multivariate logistic regression models were applied. The cohort was composed of 202 patients surgically treated for LH, 112 women (55.5%) and 90 men (44.5%), with an average age of 45.2 years. Average morbidity of the series, with a median follow-up period of 53 months, was 16.4%. Six variables were significantly associated in the bivariate analysis (age, haematocrit, total leucocyte count, alkaline phosphatase, history of previous surgery for LH and existence of biliary communications), but only two achieved statistical significance in the multivariate model (age and history of previous surgery for LH; P = 0.017 and 0.018; and odds ratio of 1.0 and 4.1, respectively). In surgical decision making for patients with LH, the history of previous surgery for LH must be considered as a risk factor for postoperative morbidity.