Step-up mini-invasive surgery for infected pancreatic necrosis: Results from prospective cohort study

Abstract

ObjectivesTo investigate the clinical efficacy and success predictors of mini-invasive techniques in the treatment of infected pancreatic necrosis (IPN). MethodsIPN patients admitted to our clinic for treatment by mini-invasive techniques were included in this study prospectively. Treatment was divided into four sequential phases: percutaneous catheter drainage (PCD), mini-incision drainage (MID), video assisted debridement (VAD) and open surgery. Patients progressed to next phase if the infection cannot be controlled. The frequency of surgery, treatment duration, cure rate, incidence of complication and overall mortality were recorded. Risk factors for failure of PCD and MID procedures were detected by logistic regression including demographics, disease severity and morphologic characteristics. ResultsFrom January 2012 to March 2015, a total of 54 consecutive IPN patients were treated, with an average age of 51.2 ± 3.1 years. Of the 54 cases, 18 (33.3%) were cured after PCD; 13 (24.1%) with uncontrolled infection were cured after MID; and the remaining 19 cases (35.2%) were cured after VAD. No open surgery was performed. Overall mortality was 7.4% (4/54), and the incidence of complications was 12.9% (7/54). In multivariable regression, the following factors were associated with high failure rate for both PCD and MID: heterogeneous fluid collection (odds ratio (OR) = 3.14; 95% confidence interval (CI): 1.32 ～ 4.25, P = 0.001 for PCD; OR = 2.99; 95% CI: 1.52 ～ 5.10, P = 0.006 for MID), multiple infected collections (OR = 4.51; 95% CI: 2.94 ～ 8.63; P = 0.000 for PCD; OR = 4.17; 95% CI: 2.77 ～ 8.12, P = 0.000 for MID), CT severity index (0 ～ 3/4 ～ 6/7 ～ 10: OR = 2.16; 95% CI: 1.83 ～ 3.62, P = 0.031 for PCD; OR = 2.72; 95% CI: 1.78 ～ 4.10, P = 0.005 for MID). ConclusionsStep-up mini-invasive techniques can be considered a first choice in the treatment of IPN. CT is effective to predict success of PCD and MID.