Evaluation and meta-analysis of test accuracy of direct PCR and bioassay methods for detecting Toxoplasma gondii in meat samples

摘要

Toxoplasma gondii infects humans from consumption of raw or undercooked contaminated meat or ingestion of oocysts present in water, soil or on vegetables. Several PCR techniques have been developed to detect T. gondii in animal samples but bioassay method is still considered as the 'gold standard'. The aim of the study was to assess the relationship of detection accuracy of direct PCR in varying amounts of meat samples. Data were collected from peer-reviewed literature. The relationship was fitted in a logistic regression model for different sample sizes with 75% of diagnostic accuracy between predicted and calculated PCR results as compared with bioassay results. We estimated the pooled odds ratio of direct PCR detecting T. gondii over bioassay using a random-effects model which suggested that there is 1.96 times (OR 1.96, 95% CI [1.33-2.86]) higher chance of PCR results being false positives as compared to bioassay method. Heterogeneity was reported significant (Q, p < 0.05 and I-2 > 50%). The results showed that the direct PCR has different accuracies in detecting T. gondii in different sample sizes irrespective of sample source. Despite its limited reliability in detecting the parasite, direct PCR has been used more as compared to the bioassay method.