Heat-selected thermophilic burden in imported skim milk powder: A cluster-aware industrial quality assessment in Libya

Thermophilic bacilli and their spores are important contaminants in dairy-powder manufacture, but their interpretation requires separation of product-quality signals from food-safety hazards. This industrial cross-sectional study examined 380 imported skim milk powder samples from two international suppliers, with 19 production batches and 10 primary samples per batch for each supplier. Aerobic colony counts averaged 5.42 ± 0.44 and 4.59 ± 0.40 log10 colony-forming units per gram for Suppliers A and B, respectively, whereas heat-selected thermophilic counts averaged 3.85 ± 0.33 and 2.82 ± 0.31 log10 colony-forming units per gram. After accounting for within-batch clustering, Supplier B had an aerobic count lower by 0.83 log10 colony-forming units per gram (95% confidence interval, 0.72 to 0.94 lower; p < 0.001) and a heat-selected thermophilic count lower by 1.03 log10 colony-forming units per gram (95% confidence interval, 0.93 to 1.14 lower; p < 0.001). Pooled physicochemical associations attenuated after accounting for batch effects, while within-batch differences in pH, titratable acidity, powder moisture and water activity remained associated with the heat-selected thermophilic count in multivariable analysis (all p < 0.001). The heat-selected thermophilic count provided useful information that complemented general aerobic counts for incoming quality assessment and supplier or batch trending. Because the 80°C/12-min heat selection followed by incubation at 55°C can recover a method-dependent mixture of thermophilic spore-formers, including facultatively thermophilic Bacillus spp., the endpoint should not be interpreted as a universal estimate of strictly thermophilic or highly heat-resistant spores. Species attribution and causal upstream-process interpretation require molecular characterization and direct process data.