A New Culture Method for the Detection of Non-Tuberculous Mycobacteria in Water Samples from Heater-Cooler Units and Extracorporeal Membrane Oxygenation Machines

The isolation of non-tuberculous mycobacteria (NTM) from cultures is particularly laborious due to the potential overgrowth of coexisting non-acid fast bacilli. To reduce the overgrowth of these non-mycobacterial organisms, a decontamination step with NaOH or cetylpyridinium chloride is highly recommended before plating the samples on the culture medium. However, due to their toxicity, decontamination solutions tend to decrease NTM recovery from clinical and environmental samples. Here, we tested an alternative method for NTM recovery based on the use of NTM Elite agar, a selective medium that does not require a decontamination step. Using NTM Elite agar, we were able to detect non-tuberculous mycobacteria in 27.7% (30/108) of water samples analyzed. The average time to NTM detection was 18 days, but some strains required longer to grow, perhaps due to the stressful environmental conditions (periodical disinfection of devices). NTM Elite agar's effectiveness in inhibiting background flora was proven by the isolation of NTM from samples with and without background flora, showing no statistically significant differences in detection rates for different total viable counts of background flora ( = 0.4989). In conclusion, our findings indicate that effective NTM recovery from HCU- and ECMO-derived water samples can be achieved via filtration and direct culture of the filters on NTM Elite agar. This simple procedure can speed up laboratory work and provide an improved method, successfully resulting in low contamination and high detection rate, in addition to being less time-consuming. Its sensitivity and lack of a decontamination step make this protocol particularly useful for monitoring the effectiveness of device disinfection in hospital settings, even in the presence of low NTM loads. Reading timeframes should probably be extended to 7 weeks (i.e., well beyond the standard 4 weeks advised by the manufacturer), in order to isolate even the slow-growing mycobacteria. However, an extended incubation period is not necessary for exclusion of contamination of the devices, as isolation times do not generally exceed 3 weeks.