Utilization of Enterococcus faecium as a Salmonella spp. Surrogate for Thermal Treatment in Selected Low-moisture Food Products

Introduction: Composition and structure of low moisture foods may affect Salmonella thermal resistance. Therefore, it is important to qualify the appropriateness of surrogates prior to use in validating preventive controls.

Purpose: This study aimed to compare thermal resistance of Enterococcus faecium and Salmonella in peanut butter, almond meal, wheat flour, nonfat dried milk powder, date paste, and ground black pepper; and to determine reproducibility of results across five laboratories.

Methods: Each of six low-moisture foods were assigned to two laboratories, with the lead laboratory inoculating samples, testing homogeneity, and equilibrating to the desired water activity (a_w) prior to shipping. The products were inoculated with E. faecium NRRL B-2354 or a five-strain Salmonella cocktail (Agona, Typhimurium, Montevideo, Mbandaka, Enteritidis - all from low-moisture products). Both laboratories subjected samples to three isothermal heat treatments and survivors were enumerated on differential media.

Results: Overall, Salmonella showed lower (P<0.05) thermal resistance than E. faecium in almond meal, peanut butter, and nonfat dried milk powder. Although almond meal and peanut butter have a similar fat content, E. faecium and Salmonella showed less thermal resistance in almond meal (D_{90°C Ef}: 8.55±0.13 min; D_{90°C Sal} : 7.24±0.08 min at 0.45 a_w) than in peanut butter (D_{90°C Ef}: 21.17±0.21; D_{90°C Sal}: 11.45 ± 0.25 min at 0.25 a_w). Milk powder, at 0.25 a_w , showed greater thermal resistance of E. faecium (D_90°C: 25.20 ± 0.94 min) and less thermal resistance for Salmonella (D_90°C: 6.69±0.13 min at a_w0.25) than in almond meal and peanut butter.

Significance: Overall, E. faecium appears to be a robust surrogate for Salmonella spp. across multiple low moisture products (with multiple cross-laboratory validations); however, product characteristics significantly impact that relationship. It is, therefore, necessary to consider product composition when validating thermal processes for low a_w foods.