Fish disease monitoring along a pollution transect: a case study using dab <i>Limanda limanda</i> in the German Bight

A survey of dab Limanda limanda at 6 stations along a pollution gradient in the German Bight was conducted as part of the Bremerhaven Workshop in order to quantify spatial patterns of 5 selected gross pathologies. The objective was to test the suitability of disease epidemiology and related histological investigations in this species as a technique for monitoring environmental quality in offshore waters. In 2 surveys. 7565 dab (length 2 15 cm) were examined post-mortern and the following externally visible diseases (with overall prevalences) were recorded: lymphocystis (10.5 %), epidermal hyperplasia/papilloma (6.2 "61, acute and healing ulcerations (1 7 %) and X-cell gill disease (0.4 %). In the 1st survey, a total of 686 dab (length 225 cm) were screened Internally for nodular lesions (diameter 22 mm) on the liver surface These liver nodules represented presumptive neoplasia (5.8 %). Histological examination confirmed 30 % of these lesions as neoplastic (hepatocellular adenoma), 27.5 % as foci of cellular alteration (putative pre-neoplastic lesions). 35 % as other lesions, including cysts, parasitic lesions, fibrosis and necrosis, whereas the remaining 7.5 % revealed no significant abnormalities. Histological investigation of 110 dab livers displaying no gross nodules revealed a similar spectrum of lesions, but without any evidence of neoplasia. The spatial patterns for the 2 most common diseases were analysed using a linear logistic model to incorporate main effects and interaction effects of length, age and sex. The prevalence of epidermal hyperplasia/papilloma decreased with distance offshore, in accordance with contaminant concentrations, whereas lymphocystis showed the opposite trend. The spatial patterns of the other disorders were not statistically analysed due to low prevalence values and revealed no clear trend, with the possible exception of skin ulcerations which showed high prevalences at the outermost station Significant positive correlations were found between the prevalence of epidermal hyperplasia/papilloma and contaminants in the llver of dab, but again lymphocystis showed the opposite trend. None of the other disorders, including the gross and histologically identlfled liver lesions, showed any apparent relationship with the chemical data. The findings of the workshop, in conjunction with the available evidence from other studies, indicated that the most promising external disease for biological effects monitoring appeared to be epidermal hyperplasia/papilloma. Putative pre-neoplastic and neoplastic liver lesions, although only found at low prevalences in this study, have been recommended as a suitable internal disease for use in b~ological effects monitoring. However, before these diseases can be used as reliable indicators, further research is required on migration patterns of the dab, the natural background levels of disease, and on disease induction in the laboratory. Future monitoring studies using dab diseases may require larger sample sizes than the minimum recommended by ICES. In studies of liver nodules, the inclusion of histological techniques is strongly recommended.