Bacteriological Examination of Ice‐cream in The Netherlands: Comparative Studies on Methods

Abstract

Samples (351) of ice‐cream were examined for ‘total aerobic’ colony count, Enterobacteriaceae, coliforms and Escherichia coli. Different methods for the indicator groups were compared. A number of the samples were also examined for Staphylococcus aureus, Bacillus cereus and Salmonella spp. In addition samples were tested microbiologically to detect fraudulently added inhibitory substances. A percentage (31.1) of the samples showed a total count ≤ 103/ml; 11.0% contained > 105/ml. which is the limit laid down in the Dutch Food Law. The MPN of coliforms (without pre‐enrichment) was ≤1/ml in 46.7% of the samples; 7.4% showed MPNs > 103 ml. Corresponding figures for Enterobacteriaceae (with pre‐enrichment) were 25.4% and 16.8% respectively. A percentage (33.0) did not meet the present Dutch standard for coliforms. The figures for samples that did not meet the standards of the Food Law are somewhat higher than those found by the Food Inspection Departments, probably because the latter generally investigate more samples from large factories. Staph. aureus and B. cereus were found only sporadically. None of 36 samples, selected because they contained appreciable numbers of indicator organisms, was found to contain salmonellas. Of 100 samples 86% showed inhibitory properties to one or more test micro‐organisms. There was no correlation between a positive test and microbial quality. Sometimes, however, the flavouring agent (lemon and chocolate) seemed to exert an inhibitory activity. The hygienic quality of ice‐cream prepared in large factories was better than that of the other samples. The poor quality sometimes reported for ‘soft’ ice‐cream was not confirmed in our investigations. A typically favourable influence of flavour on bacteriological quality could be demonstrated only for lemon ice‐cream. The same values for MPNs of Enterobacteriaceae were found when overnight pre‐enrichment in buffered peptone water was replaced by 2 h resuscitation in tryptone soya broth. Resuscitation for only 45 min in peptone saline yielded lower results. When followed by an Enterobacteriaceae colony count, overnight pre‐enrichment, 2 h resuscitation in tryptone soya broth or 1 h resuscitation on tryptone soya agar in Petri dishes gave almost the same results. No differences were found in favour of MPNs when compared with colony counts of Enterobacteriaceae. The method found most efficient for the detection of Esch. coli in ice‐cream relies on resuscitation followed by enrichment in brilliant green bile lactose broth at 44°C.