PCR testing can be as accurate as culture for diagnosis of Ichthyophonus hoferi in Yukon River Chinook salmon Oncorhynchus tshawytscha .

We evaluated the comparability of culture and PCR tests for detecting Ichthyophonus in Yukon River Chinook salmon Oncorhynchus tshawytscha from field samples collected at 3 locations (Emmonak, Chena, and Salcha, Alaska, USA) in 2004, 2005, and 2006. Assuming diagnosis by culture as the 'true' infection status, we calculated the sensitivity (correctly identifying fish positive for Ichthyophonus), specificity (correctly identifying fish negative for Ichthyophonus), and accuracy (correctly identifying both positive and negative fish) of PCR. Regardless of sampling locations and years, sensitivity, specificity, and accuracy exceeded 90%. Estimates of infection prevalence by PCR were similar to those by culture, except for Salcha 2005, where prevalence by PCR was significantly higher than that by culture (p < 0.0001). These results show that the PCR test is comparable to the culture test for diagnosing Ichthyophonus infection.

Diagnostic methodology is critical for accurately determining the prevalence of Ichthyophonus infections in wild fish populations.

Several different techniques have been employed to detect and identify Ichthyophonus spp. in infected fish hosts; these include macroscopic observation, microscopic examination of tissue squashes, histological evaluation, in vitro culture, and molecular techniques. Examination of the peer-reviewed literature revealed that when more than 1 diagnostic method is used, they often result in significantly different results; for example, when in vitro culture was used to identify infected trout in an experimentally exposed population, 98.7% of infected trout were detected, but when standard histology was used to confirm known infected tissues from wild salmon, it detected ~50% of low-intensity infections and ~85% of high-intensity infections. Other studies on different species reported similar differences. When we examined a possible mechanism to explain the disparity between different diagnostic techniques, we observed non-random distribution of the parasite in 3-dimensionally visualized tissue sections from infected hosts, thus providing a possible explanation for the different sensitivities of commonly used diagnostic techniques. Based on experimental evidence and a review of the peer-reviewed literature, we have concluded that in vitro culture is currently the most accurate diagnostic technique for determining infection prevalence of Ichthyophonus , particularly when the exposure history of the population is not known.

Assessing the accuracy of a polymerase chain reaction test for Ichthyophonus hoferi in Yukon River Chinook salmon Oncorhynchus tshawytscha.

Ichthyophonus hoferi Plehn & Mulsow, 1911, is a cosmopolitan, protistan pathogen of marine fishes. It is prevalent in mature returning Chinook salmon Oncorhynchus tshawytscha in the Yukon River watershed, and may be associated with prespawning mortality. We developed and evaluated a polymerase chain reaction (PCR) test for I. hoferi using primers specific to the parasite's small subunit rDNA. The test has a minimum detection limit of approximately 10(-5) parasite spores per reaction and does not cross-react with the closely related salmon parasites Dermocystidium salmonis or Sphaerothecum destruens. Sensitivity and specificity of the PCR test used on somatic muscle and heart tissue for detecting infected fish were determined using 334 Chinook salmon collected from the Yukon River at 2 locations (Tanana and Emmonak) in 2003 and 2004. The true infection status of the fish was determined by testing somatic muscle, heart and kidney tissue using histological evaluation, culture, and PCR. The severity of infection was grouped into 2 categories, light and heavy infection. The probability of detecting a heavily infected fish (sensitivity of the test) was generally much higher than the probability of detecting light infection, suggesting that more than one tissue and/or method should be used to accurately detect light or early infection by I. hoferi. The probability of correctly identifying a negative fish (specificity of the test) was always greater than 94% regardless of the tissue used, infection severity, sampling site or year of collection.