Background: Gastric leak is the most common and dreaded post-operative infectious complication (PIC) after laparoscopic sleeve gastrectomy (LSG). Accurate identification of patients at risk postoperatively is of cardinal importance.
Objective: The aim of this study is to assess the diagnostic performance of C-reactive protein (CRP) in predicting PICs and the most optimal time to measure it.
Methods: CRP results were collected in patients undergoing LSG between 2011 and 2015. CRP was systematically measured on post-operative days (POD) 1, 3, and 5.
Results: Of 1326 patients, 42 (3.2%) developed a PIC at a median of 5 days after surgery. The incidence of leakage was 1.9%. The best area under the curve was observed on POD5 (0.87; 95% CI 0.77-0.96). At this time point, a cut-off of 115 mg/L yielded a sensitivity of 66.7% (95% CI 46.5-86.8%), a specificity of 95.1% (95% CI 93.9-96.3%), a positive and negative predictive values of 19.4% (95% CI 10.3-28.6%) and 99.4% (95% CI 99.0-100%), respectively, and a positive and negative likelihood ratios (LRs) of 13.62 and 0.35, respectively. The combination of sequential assessments of CRP on POD3 and 5 provided a sensitivity of 84.4% (95% CI 71.8-97.0%), a specificity of 91.1% (95% CI 89.5-92.8%), a positive and negative predictive values of 20.9% (95% CI 14.0-27.9%) and 99.5% (95% CI 99.1-99.9%), respectively, and a positive and a negative LRs of 9.58 and 0.17, respectively.
Conclusions: CRP may be useful to identify patients at risk of PICs after LSG and, therefore, to prompt early investigation. However, CRP does not help rule out PICs.
Keywords: C-reactive protein; Diagnostic accuracy; Infectious complication; Laparoscopic sleeve gastrectomy; Post-operative.