Purpose: Investigate the spread of Salmonella during postharvest washing of contaminated lettuce and determine factors that affect the efficacy of chlorine in preventing cross-contamination.
Methods: Eight g of cut romaine lettuce inoculated with 7 log CFU/g of Salmonella Typhimurium expressing green fluorescent protein were added to 40 l of sterile tap water together with 800 g of uninoculated cut lettuce and washed for 2 minutes. Washing trials were performed at 3ºC with different levels of sodium hypochlorite (0, 5, 10, and 20 ppm). Small-scale (100 ml) experiments were performed to determine the effects of organic load (lettuce juice) and solid (sandy soil) on the efficacy of chlorine in wash water disinfection.
Results: Without chlorine treatment, Salmonella transfer occurred and resulted in the contamination of uninoculated lettuce and wash water at levels of 3.0 ± 0.1 log CFU/g and 3.2 ± 0.1 log CFU/mL, respectively. With 5 ppm chlorine, although no Salmonella was detected in the wash water, the pathogen was found in the uninoculated lettuce after culture enrichments. At either 10 ppm or 20 ppm of chlorine, Salmonella was not detected in the wash water or the uninoculated lettuce samples. Increases in the total organic carbon (TOC) in wash water resulted in a lower level of free chlorine and greater survival of Salmonella. At 5 ppm of chlorine, the level of Salmonella varied from not detectable (< 0 log CFU/ml) to completely unaffected (3.3 ± 0.4 log CFU/ml) in the absence of added lettuce juice (TOC = 14 ± 9 mg/l) vs. in the presence of 20 % (v/v) of lettuce juice (TOC = 159 ± 17 mg/l).
Significance: The use of sufficient chlorine could prevent Salmonella cross-contamination during postharvest washing but the effective chlorine level is influenced by the organic load in the wash water.