Helicobacter pylori is an important global human pathogen and there is growing evidence from PCR assays that contaminated drinking water might be a possible source of infection in some circumstances. There are no validated protocols for direct isolation but various culture media have been developed for possible environmental sampling. Our aim here was to investigate how inter-strain variation might affect the interpretation of results with such media. Two laboratory adapted reference strains and four recent clinical isolates were tested on four solid media and in ten liquid media. Considerable variation was found between strains in their ability to recover on the different media after stress exposure (suspension in sterile tap water). Generally, clinical isolates were less robust than the laboratory-adapted strains and, overall, the former required longer recovery times. Our findings highlighted the importance of using a range of isolates for evaluations, as examination of laboratory-adapted strains alone did not provide an accurate representation of the utility of media that may be used to recover H. pylori from water.