Data on the prevalence and populations of pathogens in individual foods are critical to the development of product-specific quantitative microbial risk assessments. An outbreak of salmonellosis associated with the consumption of raw almonds in 2000 to 2001 provided an opportunity to evaluate the levels of Salmonella in the recalled product. Duplicate 100-g samples from each of fifty 22.7-kg boxes of recalled almonds were enriched by one of two methods. Salmonella was isolated by at least one method from 42 boxes (84% positive). The levels of Salmonella determined by a three-tube most-probable-number (MPN) method were 8.5+/-1.3 MPN/100 g. In a subsequent study, raw almonds that arrived at almond processors were sampled from 2001 through 2005 to determine the overall prevalence and levels of Salmonella and to characterize the Salmonella isolates obtained. Aerobic plate counts, coliform counts, and MPN levels of Escherichia coli were also determined on positive samples. An isolation frequency for Salmonella of 81 (0.87%+/-0.2%) of 9,274 samples tested (100 g) was determined for raw almonds sampled from throughout California over the 5-year period. Salmonella was not isolated upon retesting in 59 of 65 positive samples. When detected, levels were 1.2 to 2.9 MPN/100 g. Of the 81 total isolates, 35 different serotypes of Salmonella were represented. Aerobic plate counts, coliform counts, and E. coli levels did not correlate with the presence of Salmonella.

The heat resistance of Salmonella inoculated onto almonds was determined after immersion in hot oil. Whole almonds were inoculated with Salmonella Enteritidis PT 30 or Salmonella Senftenberg 775W and heated in oil. After heating, almonds were drained, transferred to cold tryptic soy broth, and mixed with a stomacher, and samples were plated onto tryptic soy and bismuth sulfite agars. Salmonella survivor inactivation curves were upwardly concave. Rapid reductions of 2.9, 3.0, or 3.6 log CFU/g for Salmonella Enteritidis were observed after 30 s of exposure to oil at 116, 121, or 127 degrees C, respectively. Thereafter, reduction occurred at a much slower rate. Similar reductions were observed at 127 degrees C for Salmonella Senftenberg. The Weibull model was used to predict 4- and 5-log reductions of Salmonella Enteritidis after 0.74 and 1.3 min at 127 degrees C, respectively. Neither Salmonella serovar could be recovered by enrichment of 1-g samples after almonds inoculated at 5 log CFU/g were exposed to oil at 127 degrees C for 1.5 min. Standard oil roasting times and temperatures that achieve acceptable kernel color and texture should result in much greater than 5-log reductions of Salmonella in almonds.

Pathogens occurring in particulate foods may be unevenly distributed, which may impact interpretation of most-probable-number (MPN) values. The MPN analysis of Salmonella in naturally contaminated raw almonds was conducted using two sample preparation methods. Raw almond kernels (3,698 samples) and inshell almonds (455 samples) were collected from almond processors throughout California during the 2006 and 2007 harvests, and 100-g samples were enriched for Salmonella. The prevalence of Salmonella on kernels and inshell almonds was 1.6 and 0.9%, respectively, in 2006, and 0.83 and 2.2%, respectively, in 2007. Almond kernel samples from 2006 were further enriched for Salmonella, and levels of the organism were determined for positive samples by three-tube MPN analysis (25 g, 2.5 g, 0.25 g). Almonds were either divided into subsamples prior to blending and enrichment (method A), or samples were blended in enrichment broth prior to preparation of subsamples (method B). Salmonella was not isolated (<1.2 MPN/100 g) upon retesting of 19 of 31 (method A) or 23 of 29 (method B) positive samples. When detected, levels were 1.4 to 15.5 MPN/100 g (average 2.3 MPN/100 g) or 1.4 to 18.3 MPN/100 g (average 2.1 MPN/100 g) using methods A or B, respectively. A total of 23 different Salmonella serovars were identified from the original almond samples. Salmonella Muenchen was the most frequently isolated serovar (15%) from the 53 Salmonella-positive samples, followed by Newport (12%), Enteritidis (10%), and Typhimurium (8%). No correlation was found between presence of Salmonella and E. coli levels, aerobic plate counts, or counts of yeasts or molds.

No abstract

This research provides information of value in performing or evaluating validation studies for thermally processed almonds. The sensitivity of Salmonella to oil roasting is demonstrated during typical commercial almond storage times and temperatures.