Peritoneal microdialysis as a tool for detecting anastomotic leakage in patients after left-side colon and rectal resection. A systematic review

Abstract

Objective: The objective was to evaluate peritoneal microdialysis in the detection of clinical anastomotic leakage after left-sided colon and rectal resection through a systematic review.

Methods: A systematic review (PRISMA guidelines) based on a systematic search through PubMed, Cochrane Library, and EMBASE (1 February 2017) was performed. Methodological index of non-randomised studies score was selected to assess the methodological quality. Patient demographics and raw data for intraperitoneal microdialysis concentrations of glucose, lactate, glycerol and pyruvate for 5 d postoperative were obtained from the respective study groups.

Results: Ten studies with a total of 128 patients were included. Thirty (23%) patients developed clinical anastomotic leakage. The area under the curve for intraperitoneal lactate concentration was significant higher in patients with anastomotic leakage (58.2; 95% CI 39.2, 77.2) compared with the no leakage group (41.0; 95% CI 35.2, 46.1; p = .007). Receiver operating characteristic curve analysis of the maximum measured lactate concentration demonstrated 25% sensitivity, 88% specificity and 74% accuracy for AL at a cut-off value of 9.8 mmol/L. The odds ratio for a 5 mmol/L increase in lactate in relation to the risk of AL was 2.9 (CI 1.1, 8.0).

Conclusions: Increased intraperitoneal lactate concentration within the first 5 d postoperative was significantly associated with clinical anastomotic leakage, but with low predictive values. The microdialysis method is not yet ready for clinical implication before large prospective studies have defined cut off values for a biologic marker in the setting of a clear definitions of leakage.

Keywords:

• Peritoneal microdialysis
• anastomotic leakage
• colorectal resection
• lactate

Acknowledgements

The authors thank Herdis Foverskov, Librarian, The Medical Research library, The University Library of Southern Denmark, Odense University Hospital and Svendborg Hospital. Data were presented at ESCP 2017 (Poster).

Disclosure statement

All authors declare that they have no conflict of interest in this study. The study was performed without external funding. Research involving human participants and/or animals: No human or animals are directly involved in this work, as it is a systematic review.