RESUMO
Ascaridoids are one of the main parasitic hazards in commercial fish. Candling is the current industrial screening method whereby visible ascaridoid larvae are detected on a light table and manually removed. The aim of this study was to assess the sensitivity (Se) and negative predictive value (NPV) of this method. To make targeted recommendations to the fish industry, the Se was calculated per fish part, larval genus, and fish species. All fish parts (n = 615) were first candled, and larvae were collected, followed by enzymatic digestion to recover the remaining larvae. A fish part was considered positive if at least one larva was detected using candling and/or enzymatic digestion, with both methods combined as reference standard. The overall Se of candling was 31% (95% CI 23-41%) and NPV was 87% (95% CI 85-90%). The Se increased with higher numbers of larvae/100 g infected muscle. A low NPV was found for the belly flaps, therefore we either advise the removal or proper freezing of this part. Lastly, the Se and larval recovery was the highest for the darker and larger Pseudoterranova spp. larvae. Due to the low overall efficacy of candling, further assessment of its cost-benefit and impact on consumers' health risk should be conducted.
