No abstract

SUMMARY In a prospective one year study, comprising children with acute gastroenteritis admitted to hospital or treated as outpatients, the clinical and laboratory features of rotavirus diarrhoea (168 cases) were compared with those of enteric adenovirus (32 cases), bacterial (42), mixed (16), and non-specific (135) infections. The rotavirus disease was remarkably consistent, with a sudden onset of vomiting, a high frequency of fever and dehydration, and a mean duration of diarrhoea of 5-9 days. Outpatients excreting rotavirus had a similar but milder illness, mainly on account of less pronounced vomiting. The predominant symptom of enteric adenoviruses was long lasting diarrhoea (mean 10-8 days). Abdominal pain, bloody stools, prolonged diarrhoea (mean 14*1 days), leucocytosis, and a raised erythrocyte sedimentation rate strongly suggested a bacterial aetiology. Mixed infections caused longer lasting diarrhoea (mean 8-0 days) than rotavirus alone, but the severity of the illness was not increased. The clinical features of infection with unidentified pathogens most resembled those of bacterial infections. Respiratory symptoms were not significantly associated with any particular pathogen. Hypernatraemia and complications were uncommon.This study showed that the clinical features of gastroenteritis with rotavirus, enteric adenoviruses, and bacteria each exhibited patterns that could guide the experienced clinician to a presumptive diagnosis.

Fiber-optic bronchoscopy (FOB) and bronchoalveolar lavage (BAL) were performed on 67 occasions in 57 immunocompromised patients with symptoms consistent with pulmonary infection. Diagnosis was achieved more often in renal transplant patients than in patients with hematological malignancies (85% versus 28%). Culture (bacteria, virus, fungi), staining and microscopy (bacteria, fungi, Pneumocystis carinii (PC)) and antigen detection by indirect immunofluorescence (cytomegalovirus (CMV), respiratory viruses, PC, Legionella) were used for diagnosis. On 20 occasions transbronchial biopsies with histopathologic examination were performed. In addition, serology comprising the herpes group (HHV-6) and respiratory viruses was done. A microbial diagnosis was obtained on 45% of occasions. The most common pathogens found were CMV (31%) and PC (25%). On 22 (33%) occasions a rapid diagnosis of 1 or more microbial agents was obtained within 24 h by conventional staining or indirect immunofluorescence. The clinical relevance of findings of CMV, HHV-6, and Epstein-Barr virus in BAL by polymerase chain detection on 18, 6 and 3 occasions is discussed. On 4 occasions pathogenic bacteria were found. It was not possible to relate findings of coagulase-negative staphylococci, alpha-streptococci and Candida albicans to the pulmonary infection.

Bronchoalveolar lavage (BAL) products from 52 immunocompromised patients with symptoms of pulmonary infection was examined for cytomegalovirus (CMV) by virus isolation, polymerase chain reaction (PCR) and detection of CMV antigen by immunofluorescence or immunoperoxidase staining after short-term incubation in tissue culture and directly in BAL cells. We found that PCR detected all cases positive by virus isolation (15/52 samples) and the result was obtained within 5 h. PCR detected more cases of CMV than did virus isolation (22/52 samples). Positive PCR and negative virus isolation were consistent with probable CMV infection in 3/7 patients when other clinical and laboratory parameters of CMV infection were considered. The negative predictive value of PCR was high; none of 30 patients negative by PCR developed CMV pneumonia within the subsequent 2 months. Detection of CMV antigen after short-term incubation was rapid enough to be used in clinical practice, specific (100%) and with a sensitivity of 60%. Demonstration of CMV antigen in alveolar cells was highly specific (100%) but had too low a sensitivity (26.7%) to be used as the only rapid method. Our conclusion is that a combination of PCR and detection of CMV antigen after short-term incubation and directly in alveolar cells is optimal for rapid identification of CMV.

Specific immunofluorescence was used to detect measles virus antigen in skin rashes. Cryostat sections of punch-biopsy specimens of the skin were stained with use of hyperimmune rabbit and horse antisera to measles virus, also conjugated with fluorescein isothiocyanate, served as controls. Measles virus was specifically demonstrated in 20 of 21 biopsy specimens taken within four days after the onset of exanthema. Measles virus antigen was also found in three of five biopsy specimens from nonexanthematous skin during the first four days of the exanthematous phase. The viral antigen was found in single cells or in clusters of cells in the surface epithelium, skin appendages, and corium. No viral antigen was detected in biopsy specimens taken five to six days after the onset of rash.