Diagnostic accuracy of gray-scale ultrasonography for the detection of hepatic and splenic lymphoma in dogs.

Gray-scale ultrasonography is often used to screen for involvement of the liver and spleen in canine lymphoma patients but the utility of sonography for staging lymphoma has not been evaluated quantitatively. We performed abdominal sonography in 28 dogs with a confirmed diagnosis of lymphoma. Needle aspirates were obtained for cytology from three separate sites in the liver and three sites in the spleen and the sonographic appearance was noted at each site. Our hypothesis was that in dogs newly diagnosed with lymphoma, abnormal appearance of the liver or spleen on ultrasound examination is an indication that lymphoma is present in that organ. Cytologic evaluation was used as the gold standard. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of hepatic and splenic abnormalities seen on ultrasound for diagnosis of lymphoma were 72.7%, 80.6%, 77.4%, 76.3%, and 76.8% for the liver, respectively, and 100%, 23.3%, 64.6%, 100%, and 68.1% for the spleen, respectively. Based on these results, we recommend that aspirates be performed for detection of lymphoma in the spleen of dogs only when the spleen appears abnormal ultrasonographically and that cytology of the liver be performed, regardless of ultrasonographic appearance, to determine the presence or absence of lymphoma.

Relationship between radiographic hilar lymphadenopathy and serologic titers for Coccidioides sp. in dogs in an endemic region.

Hilar lymphadenopathy is a common radiographic finding in coccidioides infections. Serologic studies are used most often for the diagnosis of coccidioidomycosis in endemic areas, with IgG titers > 1:8 considered positive for infection and lower IgG titers of < 1:8 considered indicative of exposure and not necessarily related to organism presence. The objective of this study was to determine the relationship of hilar lymphadenopathy to coccidioidomycosis titers for dogs in an endemic area. A positive association between these parameters would allow treatment to be initiated before obtaining titer results. Thoracic radiographs of 131 dogs from an endemic area were reviewed for evidence of hilar lymphadenopathy. These results were compared with serology results. There was a significant association between hilar lymphadenopathy and a positive serology result (P < 0.001). With hilar lymphadenopathy as a predictor of a positive titer result, sensitivity was 28.0%, specificity was 91.5%, the positive predictive value was 43.8%; and the negative predictive value was 84.4%. There was no association between the titer result and gender, age, or weight. The radiographic finding of hilar lymphadenopathy appears to be a useful indicator of coccidioidomycosis infection in an endemic population of dogs supporting the treatment of patients for coccidioidomycosis when hilar lymphadenopathy is present and before obtaining serology results.

Binding of ciprofloxacin labelled with technetium Tc 99m versus 99mTc-pertechnetate to a live and killed equine isolate of Escherichia coil.

This paper describes a simple methodology for evaluating the bacterial binding of ciprofloxacin labelled with technetium Tc 99m. Using this methodology, the binding of 99mTc-ciprofloxacin by live Escherichia coli was compared with the binding of 99mTc-ciprofloxacin by killed E. coli and the binding of 99mTc-pertechnetate (99mTcO4-) by live E. coli. The antimicrobial effect of 99mTc-ciprofloxacin on E. coli was evaluated. Four groups were defined: live E. coli with 99mTc-ciprofloxacin, live E. coli with 99mTcO4 , killed E. coli with 99mTc-ciprofloxacin, and killed E. coli with 99mTcO4-. After 0, 2, and 4 h of incubation of 1 x 10(8) colony-forming units of E. coli suspended in 5 mL of sterile distilled water with 1.85 MBq of 99mTc-ciprofloxacin or 99mTcO4, 1 mL from each sample was centrifuged. The radioactivity of the bacterial pellet and that of the supernatant were measured separately, and the percentage of sample radioactivity attributable to bacterial binding was calculated. Of the 99mTc-ciprofloxacin, 3.6% to 5.9% was bound to live or killed E. coli; only 0.1% to 0.2% of the 99mTcO4- was bound to live E. coli (P < 0.0001). No significant difference in 99mTc-ciprofloxacin binding was found between live and killed E. coli (P = 0.887). An antimicrobial effect on E. coli was seen with 99mTc-ciprofloxacin: colony counts were reduced after 4 h. The small amount of 99mTc-ciprofloxacin binding and the lack of difference in binding between live and killed E. coli may limit the utility of this methodology in evaluating the presence of E. coli infection.