Organ culture at necropsy is the standard method used for isolating most microbial agents that affect finfishes (Bullock, 1978; McDaniel, 1975). Although this is acceptable in most situations, the 5dCrifice of especially valuable animals such as broodstock is obviously disadvantageous. Consequently, alternate detection methods have been devised which avoid the sacrifice of valuable individuals, These methods, generally used to diagnose chronic viral infections, involve the sampling of various body fluids such as excretory products (Wolf et al., 1968) or blood (Yu et al., 1982). Such fluids, however, are not always as sensitive as internal organs for the detection of pathogens. This is especially true for bacterial fish pathogens, where the kidney is usually the organ of choice for isolation of the agent (McDaniel, 1975).

The kidney of teleost fishes is a long ribbon-shaped organ which lies retroperitoneally along the ventral surface of the vertebral column. At its anterior limit, it curves ventrally and comes to lie just beneath the medial surface of the branchial chamber. The proximity of the kidney at this point to the external body surface facilitates its sampling by needle aspiration. Thus, we compared the diagnostic usefulness of this potentially nonlethal method, sampling of kidney via needle biopsy, with the standard accepted method of sampling at necropsy.

Renal aspirates were collected from fish by lifting the operculum and directing a needle with syringe first dorsally and then dorsocaudally into the kidney, just caudal to the last branchial arch. The syringe was aspirated until 0.10 ml of kidney tissue was obtained. The sample was placed on a sterile bacteriological loop and streaked onto a trypticase soy agar (TSA) slant. Immediately after this kidney biopsy procedure, trout were surface-disinfected and cut transversely at the caudal limit of the operculum. Kidney tissue was again sampled with a bacteriological loop and the inoculum streaked onto a TSA slant.

Kidney biopsy was both sensitive and specific in detecting Yersina ruckeri infection (sensitivity = 0.93; specificity = 0.88). Analysis with d L Test-used for comparing sensitivity or specificity (Galen and Gambino, 1975) revealed no significant difference between kidney biopsy versus kidney necropsy.