- Tree nuts have been implicated in a number of foodborne outbreaks and recalls in recent years linked to enteric pathogens, particularly Salmonella. Therefore, prior to distribution and marketing, it is necessary to understand the biological risks associated with the consumption of tree nuts and to find effective methods to inactivate foodborne pathogens. Steam treatment processes have been validated for use on California almonds, but little research has been conducted for in-shell hazelnuts. Hazelnuts grown in Oregon were recalled nearly annually from 2009 to 2017 and were implicated in an outbreak of E. coli O157:H7 that sickened 10 people in the Midwest and Canada in 2011 (Miller et al. 2014). In 2017, an outbreak of Salmonella Typhirium sickened 5 people and was traced to an 80-acre Oregon farm and nursery that sold between 32,000 to 48,000 pounds of raw in-shell hazelnuts directly to consumers from a road-side stand (Yada et al. 2019). To help characterize the biological risk of in-shell hazelnuts, we conducted a prevalence and amounts survey of Salmonella on in-shell hazelnuts grown in Oregon over two harvest years (2013-2014). In a separate study, we developed a steam treatment process that inactivates a 5-log reduction of Salmonella spp. on the surface of in-shell hazelnuts with minimal impact on final product quality.
For the Salmonella prevalence study, raw, green-dried in-shell hazelnut samples (n = 472) were collected by six of the largest hazelnut handlers in Oregon’s Willamette Valley and tested for the presence of Salmonella spp. using a modified method from the Food and Drug Administration’s (FDA) Bacteriological Analytical Manual (BAM). In-shell hazelnut samples (375 g) were enriched in 1:10 (w/v) lactose broth followed by selective enrichment in Rappaport-Vassiliadis Broth (RV) and Tetrathionate Broth (TT). Selective enrichments were streaked for isolation onto Hektoen Enteric (HE) and Xylose Lysine Desoxycholate (XLD) Agars. Suspected colonies displaying Salmonella spp. morphology were confirmed on CHROMagar Salmonella Plus. A most-probable-number (MPN) method (3 x 333, 33.3, 3.3 g) using the same cultural steps as the initial testing was used to determine Salmonella population levels on naturally-contaminated in-shell hazelnuts. The prevalence of Salmonella spp. by year was 21.7% (55/254) and 46.8% (102/218) for 2013 and 2014, respectively. Salmonella population levels ranged from 0.092 to 30.7 MPN/100 g, with an average of 2.6 MPN/100 g. These data will help support risk assessment strategies for the Oregon hazelnut industry.
In our second study, we evaluated the efficacy of steam blanching on the reduction of Salmonella spp. on the surface of in-shell hazelnuts as a potential thermal postharvest treatment for hazelnuts. A pilot-scale steam blancher was used to deliver a continuous steam treatment at atmospheric pressure. In-shell hazelnuts were inoculated (~8.5 log CFU/g) with a five-strain Salmonella spp. cocktail and exposed to steam (88°C) for 15 s, 1, 3, 5, and 10 min. Following steam treatment, hazelnut samples were transferred to 0.1% peptone water
(24°C), hand agitated 1 min, serially diluted, plated on Hektoen Enteric agar and incubated at 37°C for 24 h. D-values (0.82 to 1.53 min) were calculated based on plate counts. Salmonella spp. could not be recovered by enrichment after hazelnuts inoculated at 5 log CFU/g were treated with steam at 88°C for 10 min. These data will be useful when developing validated postharvest steam treatments for the hazelnut industry.