Modeling Transfer of Vibrio Parahaemolyticus During Peeling of Raw Shrimp

Abstract

This study aimed to qualify the transfer of Vibrio parahaemolyticus during the shrimp peeling process via gloves under 3 different scenarios. The 1st 2 scenarios provided quantitative information for the probability distribution of bacterial transfer rates from (i) contaminated shrimp (6 log CFU/g) to non-contaminated gloves (Scenario 1) and (ii) contaminated gloves (6 log CFU/per pair) to non-contaminated shrimp (Scenario 2). In Scenario 3, bacterial transfer from contaminated shrimp to non-contaminated shrimp in the shrimp peeling process via gloves was investigated to develop a predictive model for describing the successive bacterial transfer. The range of bacterial transfer rate (%) in Scenarios 1 and 2 was 7% to 91.95% and 0.04% to 12.87%, respectively, indicating that the bacteria can be transferred from shrimp to gloves much easier than that from gloves to shrimp. A Logistic (1.59, 0.14) and Triangle distribution (-1.61, 0.12, 1.32) could be used to describe the bacterial transfer rate in Scenarios 1 and 2, respectively. In Scenario 3, a continuously decay patterning with fluctuations as the peeling progressed has been observed at all inoculation levels of the 1st shrimp (5, 6, and 7 log CFU/g). The bacteria could be transferred easier at 1st few peels, and the decreasing bacterial transfer was found in later phase. Two models (exponential and Weibull) could describe the successive bacterial transfer satisfactorily (pseudo-R² > 0.84, RMSE < 1.23, SEP < 10.37). The result of this study can provide information regarding cross-contamination events in the seafood factory.

Practical application: This study presented that Vibrio parahaemolyticus cross-contamination could be caused by gloves during the shrimp peeling process. The bacterial transfer rate distribution and predictive model derived from this work could be used in risk assessment of V. parahaemolyticus to ensure peeled shrimp safety.

Keywords: cross-contamination; distribution; glove; predictive model; shrimp peeling.