RESUMO
Ten laboratories participated in a collaborative study to compare a pectin-based violet red bile (VRB) method with the VRB agar-based standard method for estimation of coliform bacterial counts in 7 different dairy food groups: cream, cheddar cheese, cottage cheese, homogenized milk, raw milk, sour cream, and yogurt. Each laboratory analyzed 8 samples of each food group as sample pairs prepared at high, medium, and low inoculum levels, and an uninoculated control pair. Overall mean log counts were higher for the pectin gel method in 18 of 21 cases (7 samples at 3 inoculum levels); 12 results were significantly higher (P less than 0.01) than those for the VRB agar method. Of the 3 higher VRB agar method means, 2 were not significant at P less than 0.10, and one was of borderline significance (0.05 less than P less than 0.06). Repeatability variation (sr) favored the pectin gel method in 14 of 21 cases; 7 were highly significant differences (P less than 0.01). None of the 7 results favoring the VRB agar method was statistically significant. Reproducibility variation (sR) favored the pectin gel method in 14 of 21 results. These data indicate that the pectin gel method gives higher recovery of coliforms with better precision than does the VRB agar method, and strongly support the suitability of the pectin gel method as an alternative to the agar-based VRB method for coliform counts in dairy products. The pectin gel method has been adopted official first action.
