The use of mesenchymal stem cells in animal models for gastrointestinal anastomotic leak: A systematic review

Aim Anastomotic leak is the most feared complication of gastrointestinal surgery. Mesenchymal stem cell technology is used clinically to promote wound healing; however, the safety and efficacy of this technology on anastomotic healing has yet to be defined. The aim of this study was to investigate whether mesenchymal stem cells confer any benefit when applied to animal models for gastrointestinal anastomotic leak, identify the methodology and how efficacy is assessed. Methods The MEDLINE, EMBASE, WebofScience and Cochrane Library databases were interrogated between 1 January1947 to 1 May 2020. All studies where mesenchymal stem cells were applied to laboratory animal leak models to demonstrate a healing effect were considered. All experimental and histological outcomes were examined. Compliance to ARRIVE and current International Consensus was assessed. Results A total of 1205 studies were screened. Twelve studies reported on 438 gastrointestinal anastomoses in four species using 11 models; seven in the colon. No studies utilised a model with a known leak rate. Significant variance was observed in histological outcomes with efficacy demonstrated in five out of 12 studies. One study demonstrated a benefit in leak rate. Colorectal studies had a greater median ARRIVE compliance, 60.8% (IQR 63.2-64.5) compared to noncolorectal 45.4% (IQR 43.8-49.0). Conclusions Mesenchymal stem cell delivery to an animal anastomosis is safe and feasible. Use may confer benefit but findings are currently limited to surrogate histological outcomes. There is consistency in outcome measures reported but variance in how this is assessed. Poor compliance to ARRIVE but good compliance to current international consensus in leak models of the colon was observed.

Automated summary

Mesenchymal stem cell delivery to an animal anastomosis is safe and feasible, but current findings are limited to surrogate histological outcomes. There is consistency in reported outcome measures, but variance in assessment methods. Compliance with ARRIVE guidelines is poor, but compliance with current international consensus in colon leak models is good.