OBJECTIVE: To evaluate the frequency of human polyomavirus reactivation in urine specimens from HIV-positive patients; compare the sensitivity of cytology, immunohistochemistry and molecular biology; differentiate viral genotypes; and correlate the results with urinary cytologic abnormalities. STUDY DESIGN: Urine specimens from 78 unselected HIV-positive patients were evaluated by means of cytology, immunohistochemistry and nested polymerase chain reaction (n-PCR) to evaluate the presence of polyomaviruses. Restriction fragment length polymorphism (RFLP) was carried out in positive cases in order to differentiate BK virus (BKV) from JC virus (JCV). CD4 cells and serum creatinine levels were evaluated as indices of immune status and renal function, respectively, whereas the presence of red blood cells was used as an index of urogenital damage. RESULTS: Cytologic evidence of polyomavirus infection was found in 17 samples and immunohistochemically confirmed in 9; another 6 cytologically negative cases were detected by means of immunohistochemistry. In all cases, only one or two cells showed typical viral inclusions or positive staining. n-PCR identified 44 positive samples, thus confirming all of the cytologically and immunohistochemically positive cases and detecting polyomavirus genome in a further 21. RFLP detected 39 JCV, 1 BKV and 4 JCV-BKV infections. No correlation was found between the presence or type of polyomavirus and immune status, but red blood cells were found more frequently in the positive than in the negative samples. Serum creatinine levels fell within the normal range in all cases. CONCLUSION: Molecular biology is the most sensitive tool for detecting polyomavirus urinary infection in HIV-positive patients and the only reliable method of differentiating JCV and BKV viral genotypes.
OBJECTIVE: To evaluate the frequency of humanpolyomavirus reactivation in urine specimens from HIV-positivepatients; compare the sensitivity of cytology, immunohistochemistry and molecular biology; differentiate viral genotypes; and correlate the results with urinary cytologic abnormalities. STUDY DESIGN: Urine specimens from 78 unselected HIV-positivepatients were evaluated by means of cytology, immunohistochemistry and nested polymerase chain reaction (n-PCR) to evaluate the presence of polyomaviruses. Restriction fragment length polymorphism (RFLP) was carried out in positive cases in order to differentiate BK virus (BKV) from JC virus (JCV). CD4 cells and serum creatinine levels were evaluated as indices of immune status and renal function, respectively, whereas the presence of red blood cells was used as an index of urogenital damage. RESULTS: Cytologic evidence of polyomavirus infection was found in 17 samples and immunohistochemically confirmed in 9; another 6 cytologically negative cases were detected by means of immunohistochemistry. In all cases, only one or two cells showed typical viral inclusions or positive staining. n-PCR identified 44 positive samples, thus confirming all of the cytologically and immunohistochemically positive cases and detecting polyomavirus genome in a further 21. RFLP detected 39 JCV, 1 BKV and 4 JCV-BKV infections. No correlation was found between the presence or type of polyomavirus and immune status, but red blood cells were found more frequently in the positive than in the negative samples. Serum creatinine levels fell within the normal range in all cases. CONCLUSION: Molecular biology is the most sensitive tool for detecting polyomavirus urinary infection in HIV-positivepatients and the only reliable method of differentiating JCV and BKV viral genotypes.