Diagnostic markers of postoperative morbidity after laparoscopic Roux-en-Y gastric bypass for obesity.

Fiche publication


Date publication

avril 2014

Auteurs

Membres identifiés du Cancéropôle Est :
Pr MEYER Nicolas, Pr ROHR Serge, Dr ROMAIN Benoit, Pr BRIGAND Cécile


Tous les auteurs :
Romain B, Chemaly R, Meyer N, Chilintseva N, Triki E, Brigand C, Rohr S

Résumé

PURPOSE: The main objective of this study was to detect subacute complications that can arise from laparoscopic Roux-en-Y gastric bypass and take a rational approach to manage these complications. METHODS: A prospective observational study was performed from November 2010 to December 2012. All patients undergoing gastric bypass surgery for morbid obesity were included in this study. Patients with complications before day 5 were excluded from the study. Clinical and laboratory data (C-reactive protein, leukocyte count) at postoperative day 5, 30-day morbidity, were recorded. The diagnostic value of C-reactive protein (CRP) and leukocytes were determined by the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. RESULTS: One hundred and twenty-six patients were included. The overall incidence of 30-day morbidity was 8.7 %, and anastomotic leakage rate was 3.2 %. C-reactive protein at postoperative day 5 was a good predictor of complications (AUC was 0.862 (95 % CI [0.76; 0.96]; p < 0.001) and anastomotic leakage (AUC was 0.863 (95 % CI [0.66; 1]; p = 0.014). A CRP cutoff level of 136 mg/l at postoperative day 5 yielded a specificity of 95.5 % and a sensitivity of 57.1 % for the detection of postoperative complications. The negative predictive value was 94.6 %. A CRP level of 136 mg/l at day 5 was significantly associated with postoperative morbidity. CONCLUSIONS: C-reactive protein dosage at postoperative day 5 is a relevant predictor of postoperative complications permitting to select patients at risk. Radiological examination and close monitoring could be restricted to patients with CRP level exceeding 136 mg/l.

Référence

Langenbecks Arch Surg. 2014 Apr;399(4):503-8