Comparison of direct electron microscopy, immune electron microscopy, and rotavirus enzyme-linked immunosorbent assay for detection of gastroenteritis viruses in children.

An approximate 10% suspension in water of the first available stool sample from 411 infants and young children with acute gastroenteritis was examined by electron microscopy (EM) after 2 min of negative staining. This procedure enabled the detection of 88% of the 199 rotavirus infections, all of the 22 adenovirus infections, and 47% of the 15 approximately 27-nm virus infections ultimately detected by a combination of techniques, including immune electron microscopy (IEM) and rotavirus enzyme-linked immunosorbent assay (ELISA). Of the 204 infections detected by direct EM of stools, 76% were detected within 2 min of viewing, and 94% were detected within 6 min of viewing. Type 1 and type 2 rotavirus particles were visualized with approximately equal efficiency, although type 2 rotavirus infections were more common. Rectal swab preparations were clearly inferior to stool preparations for the detection of virus infection by direct EM. IEM examination was required for efficient visualization of viruses in rectal swab specimens. ELISA was the most sensitive method for the detection of rotaviruses; with this method, all infections in which rotavirus particles were visualized by EM or IEM were detected. However, 73% of the 1,834 specimens which were presumptively positive for rotavirus by conventional indirect ELISA proved to be falsely positive on the basis of EM, IEM, blocking ELISA, confirmatory ELISA, or a combination of these methods. False-positive rotavirus ELISA reactions apparently were eliminated when fecal specimens were tested in a modified confirmatory ELISA with a lower dilution of rotavirus-negative (pre-immunization) than rotavirus-positive (post-immunization) capture antibody from the same animal.