Infectious outbreaks associated with the use of gastrointestinal endoscopes have increased in line with the spread of highly resistant bacteria. The aim of this study was to determine the measures required to improve microbial quality surveillance of gastrointestinal endoscopes. We reviewed the results of all microbiological surveillance testing of gastrointestinal endoscopes and automatic endoscope reprocessors (AERs) performed at Brest Teaching Hospital from 1 January 2008 to 1 June 2015. We analyzed the influence of the time of incubation on the rate of positive results using the Kaplan - Meier method. We also studied risk factors for gastrointestinal endoscope contamination using a multivariable logistic regression model. Over the study period, 1100 microbiological tests of gastrointestinal endoscopes (n = 762) and AERs (n = 338) were performed. A total of 264 endoscope tests (34.6 %) showed a level of contamination higher than the target. After 2 days of incubation, contamination was apparent in only 55.5 % of the endoscopes that were later shown to be contaminated (95 % confidence interval [CI] 49.2 - 61.8). Multivariable analysis showed that the use of storage cabinets for heat-sensitive endoscopes significantly reduced the risk of endoscope contamination (odds ratio [OR] 0.23, 95 %CI 0.09 - 0.54; P < 0.001) and that the use of endoscopes older than 4 years significantly increased this risk (OR ≥ 6 vs. < 2 years 2.92, 95 %CI 1.63 - 5.24; P < 0.001). Microbiological culture technique, mainly incubation duration, strongly influenced the results of endoscope sampling. Samples should be cultured for more than 2 days to improve the detection of contaminated endoscopes. Particular attention should be paid to endoscopes older than 2 years and to those that are not stored in storage cabinets for heat-sensitive endoscopes.