Detection of cytomegalovirus infections in specimens other than urine by the shell vial assay and conventional tube cell cultures.

Blood, bronchoscopy-lavage, biopsy (lung, liver, kidney), sputum, and other (cecum, bone) specimens were inoculated into shell vials and conventional cell tube cultures seeded with MRC-5 cells over a 23-month period. Of 1,472 specimens, 182 (12.4%) yielded cytomegalovirus (CMV)-positive results from 81 patients. Significantly more CMV-positive specimens were detected in shell vials (n = 154; 84.6%) than in conventional tube cell cultures (n = 126; 69.2%) (P less than 0.01). We found that 98 (53.8%) of the total 182 and 41 (42.7%) of the 96 blood specimens positive for CMV were detected by both the shell vial assay and conventional tube cell cultures. However, 56 (30.7%) of the total 182 and 31 (32.3%) of the 96 blood specimens positive for CMV were obtained exclusively in shell vials after detection with monoclonal antibody. Alternatively, 28 (15.4%) of the total 182 and 24 (25%) of the 96 blood specimens positive for the virus were isolated only in conventional tube cell cultures. Thus, although the shell vial assay was more sensitive and rapid than the conventional tube cell culture method, both systems must be used, especially for blood specimens, for the laboratory diagnosis of CMV infections.