Abstract:
A case control study was performed in the conservancy area of the Kruger National Park (KNP), South Africa to find out whether the faeces of impala (Aepyceros melampus) were more likely to contain tetracycline resistant Escherichia coli (TREC) when they drank from rivers that contained these bacteria, compared to rivers that were uncontaminated with TREC. Five perennial rivers (Crocodile, Letaba, Olifants, Sabie and Sand) were selected. A total of 11 points in these rivers were sampled on three separate occasions and cultured for E. coli. Impala herds within 5 kilometres of each water collection site were identified and between 5 and 10 fresh faeces were collected for each collection period (n=209 faecal specimens). Selective culturing of E. coli was done and the resulting colonies were screened for tetracycline resistance by using the Lederberg Replica Plating (LRP) method. Resistant colonies were those that grew in the presence of 4 mg/L of tetracycline. Both a resistant and/or a susceptible isolate were then selected from each specimen, and subjected to the minimum inhibitory concentration (MIC) micro-broth dilution test for tetracyclines. The breakpoint for the MIC method was considered ≥ 8 mg/L (which is higher than for the LRP method). Twenty one of the 33 water specimens examined were found to be contaminated by E. coli. Among them (n=21), 76.19% (n=16) were resistant to tetracycline using the LRP method, although using the MIC method only 19.05% (n=4) were resistant. All of the resistant strains originated from the Letaba, Olifants and Crocodile rivers (TRECpos rivers). Among the 209 impala faeces sampled, 191 were positive for the presence of E. coli (91.38%). Within these (n=191), 36.64% (n=70) showed TREC using the RPL method, while using the MIC, 9.95% (n=19) were found to be TREC. The RPL and MIC methods showed a concordance of only 48%. Resistant isolates screened by PCR for tet(A) and tet(B) genes were found to be negative and it was concluded that other resistant genes were responsible. The odds ratios (OR) showed that impala drinking from TRECpos rivers were 19.3 (2. 63-141.68) times more likely to be infected with TREC than unexposed impala
