Microbial Risk Assessment of Thermophilic Campylobacter SPP. in Raw Vegetables from Farm to Table

The first aim of this study was to determine the prevalence and number of thermophilic Campylobacter spp. (Campylobacter jejuni, Campylobacter coli and Campylobacter fetus) in raw vegetables (ulam) at pre-harvest and retail level, soil and animal manure in an organic and a traditional vegetable farm. The biosafety of Campylobacter jejuni was assessed by phenotypic (antibiotic resistance) and genotypic (presence of virulent and toxin genes) as well as RAPD-PCR characteristics of the strains isolated from vegetables. A kitchen simulation study was conducted to provide decontamination and crosscontamination data and information for estimation of the risk of acquiring campylobacteriosis from consumption of ulam using a step-wise risk assessment. The prevalence of thermophilic Campylobacter spp. in 309 (number of samples) raw vegetables purchased from two supermarkets and a wet market was relatively high, 29% to 68%. Campylobacter jejuni (25.5% to 67.7%) and C. coli (21.6% to 65.7%) were predominant species isolated; while C. fetus was only detected in two samples (1.9%) from one of the supermarkets. Only 18.3% of Campylobacter-MPN-PCR positive samples were recovered by enumeration-plating method indicating that routine enumeration-plating methods has very low recovery rate for Campylobacter spp. from vegetables. The study was extended to investigate the level of contamination with Campylobacter spp. in vegetables farms. A total of 172 samples of animal manure (n=18), soil (n=60), irrigation water (n=45) and vegetables (n=49) samples were collected from both an organic and a conventional vegetable farm. The organic vegetable farm (20.5%) was found to have a higher prevalence of Campylobacter spp. compared to the vegetable farm practicing conventional farming (2%). The low contamination level in the conventional farm was most probably due to the bed-burning practice and the use of composted manure in the farm. Campylobacter coli was not detected in all the samples from both farms. Soil (30.4%) and animal manure (57.1%) sampled from the organic vegetable farm were found to harbor Campylobacter spp. and C. jejuni. However, none of the irrigation water samples examined from both farms were positive for Campylobacter spp. RAPD-PCR fingerprinting and antibiotic resistance profiling indicated that multi-resistant Campylobacter spp. might be wide-spread in the study area. Clustering of C. jejuni isolates based on RAPD-PCR profiles suggested that some isolates from different sources and locations were genotypically closely related. Clusters A2, A3, A5 and A6 comprised C. jejuni strains isolated from raw vegetables in the supermarkets and a wet market. All clusters including B1 and B3, which comprised strains only from supermarkets, were actually consisted of isolates from different sources. The isolates showed multi-resistance to as many as 10 antibiotics tested. All the isolates were detected to carry the virulent genes, cadF, ceuE and flaA. However, toxin genes detection indicated only 16.1% and 10.7% of the isolates carry cdtB and cdtC toxin genes, respectively; while none of the isolates carry cdtA gene. The potential of raw salad vegetables as a vehicle in C. jejuni transmission was demonstrated by a step-wise risk assessment. Based on the assumptions used in the step-wise risk assessment, the annual number of cases of campylobacteriosis acquired from the consumption of ulam is estimated to be 4992/100,000 of Malaysian population, assuming that 10% of Campylobacter spp. infection translates into illness. However, the risk estimate was predicted to reduce to 175/100,000 if an extra blanching step was incorporated into the model. In conclusion, there is an immediate need for further investigation to look into the wide-spread problem of Campylobacter spp. in ready-to-eat foods, such as salad and ulam, in Malaysia.

Preview PDF FSTM_2008_5a.pdf Download (304kB)

Actions (login required)

View Item