C. George Ray

Raised serum levels of tumour necrosis factor in parasitic infections.

In a study of serum levels of endogenous tumour necrosis factor (TNF) in healthy people and patients with neoplastic or infectious disease, only patients with kala-azar (visceral leishmaniasis) and malaria were found to have a strikingly increased frequency of raised TNF levels (66.6% and 70.0%, respectively). 7.9% of samples from both healthy subjects and patients with neoplastic disease contained measurable TNF. The discovery of elevated TNF levels in the sera of patients with parasitic diseases suggests that this cytokine may play a part in host defences against parasitic infections.

Respiratory syncytial virus-associated lower respiratory illnesses : possible influence of other agents

Acute lower respiratory illnesses were prospectively investigated in a cohort of 1246 healthy infants who were enrolled at birth in the Tucson Childrens Respiratory Study and followed through the first 3 years of life. Respiratory syncytial virus (RSV) infection was documented by culture, antigen detection or both in 276 episodes. In 21 (7.6%) of these 276, other viruses were simultaneously detected. Further serologic studies of 50 episodes in which RSV had been found increased the apparent viral codetection rate to 24%. When culture results for Chlamydia trachomatis and Mycoplasma pneumoniae were also considered, the rate of codetection was found to be 10.9% (30 of 276); this increased to 28% for the subgroup of episodes (14 of 50) that was further studied serologically. Illnesses associated with more than one agent were not significantly different from those involving RSV alone, with respect to month of onset, age at illness, illness type or duration of illness. We conclude that when RSV has been detected in previously healthy infants, routine searches for the concomitant presence of other viruses are usually not warranted.

Acute lower respiratory illnesses during the first three years of life : potential roles for various etiologic agents

Lower respiratory tract illnesses (LRIs) occurring during the first 3 years of life among children enrolled in the Tucson Childrens Respiratory Study have been studied for evidence of viral, mycoplasmal and Chlamydia trachomatis infections. This report examines those from whom adequate acute and convalescent sera were available at the time of the LRI. Two groups were compared: those in whom culture and/or antigen detection yielded an etiologic agent (N = 110); and those who did not (culture negative, N = 124). Seroconversions (fold titer rise) to respiratory syncytial virus; influenza virus types A and B; parainfluenza virus types 1, 2 and 3; or adenovirus were found in only 0 to 5% of the culture negative group. No significant differences between groups with regard to frequencies of seroconversion to influenza type C, parainfluenza virus type 4, human coronaviruses 229E and OC43 or cytomegalovirus were detected, which suggests that these agents may not be frequent primary causes of LRIs among otherwise healthy children. Significant differences in seroconversions to Epstein Barr virus were detected, suggesting that Epstein-Barr virus may contribute to LRI morbidity; however, its exact role remains to be defined.

Herpes simplex virus infection in infants: A spectrum of disease

Disseminated herpes simplex virus infection in the infant may result in death or permanent central nervous system damage. However, morbidity and mortality rates are variable, a and complete recovery, as this report illustrates, is apparently possible. Until adequate controlled studies are undertaken, the therapeutic efficacy of 5-iodo-2′-deoxyuridine (IUDR) and other antiviral drugs should be considered in the light of this variation in the severity of disseminated herpes simplex infections.

An eight-year study of the viral agents of acute gastroenteritis in humans: ultrastructural observations and seasonal distribution with a major emphasis on coronavirus-like particles.

Abstract During an 8-yr period, 862 stool specimens from patients with gastroenteritis were examined by electron microscopy after negative staining with 2% phosphotungstic acid (pH 6.5). Forty-one percent of the specimens submitted over an 8-yr period were determined to be positive for virus or viruslike particles belonging to one or more of seven morphologically distinct viral groups. Coronavirus-like particles (CVLPs) were present in 69.8% of the positive stool specimens. Membranous profiles containing “complement-type” holes (10 nm in diameter) were identified in some preparations containing CVLPs. The second most prevalent viral agent found in stool specimens was the rotavirus (17% of all positive stools). The incidence of other viruses identified in the survey were as follows: adenovirus 4.5%, picorna/parvovirus agents 2.9%, Norwalk-like agent 2.9%, astrovirus 1.9%, and calicivirus 0.5%. Unclassified small round viruses (≈25–30 nm in diameter) represented 0.5%. It was also determined that there was a seasonal distribution in excretion of all viruses except for CVLPs. A greater number of viruses were identified in the cooler, drier months of the year.

Mycoplasma and ureaplasma in bronchoalveolar lavage fluids from immunocompromised hosts

The significance of Mycoplasma spp. and Ureaplasma urealyticum infection in immunocompromised patients has not been clearly established. We identified mycoplasma or ureaplasma in bronchoalveolar lavage fluid from 12 of 61 (20%) immunocompromised patients with pulmonary infiltrates. A complete microbiological investigation was made on the bronchoalveolar lavage fluids, and Mycoplasma pneumoniae was the sole agent detected in three instances, suggesting that it may have been the cause of the infiltrates in these immunocompromised patients. Other Mycoplasma spp. and ureaplasma were detected in nine patients, but in eight of these patients other pulmonary pathogens were also recovered.

Atypical neonatal respiratory syncytical virus infection

may include fibrinous pleuritis or pleural effusions, but empyema has not been reported. In several reports, the organisms have been identified in hilar lymph nodes, and in the sinusoids of the spleen and liver1~ no cellular reaction or inflammatory response to the organisms was documented. In our patient, the lesions in the liver and brain resembled those found in the lungs, indicating that under certain circumstances the bacilli may become disseminated and cause multisystem disease. The precipitating factors in our patient likely include the profound immunodeficiency associated with SCID, and potentially the co-presence of parainfluenza virus and Pneumocytis carinii. The present case illustrates severat important points. It documents Legionella pneumophila infection in an infant. The features not reported thus far include disseminated infection with marked tissue reaction in several organs, including liver and brain. Immunocompromise d patients appear to be at the h ighest risk. Since many of these patients develop mixed infections, Legionella pneumophila should be included in the differential diagnosis, with appropriate studies being carried out when confronted with progressive respiratory failure of unknown etiology.

1161 CHLAMYDIA-ASSOCIATED LOWER RESPIRATORY TRACT INFECTIONS (LRI) IN CHILDREN OVER 1 YEAR OF AGE

Chlamydia trachomatis (Ct) is believed to be a cause of LRIs mainly in young infants. Close surveillance for LRIs was maintained since birth on a group of 1182 children. Over a period of 17 months, we obtained nasopharyngeal and throat swabs for acute LRI from 340 children. All specimens were cultured for viruses, Ct and Mycoplasma pneumoniae. One or more of these agents were isolated from 67% of LRIs. Ct was isolated from 3/52 (5.8%) of infants ≤6 months of age, 0/61 (0%) infants 7-12 months and 9/227 (4.0%) children 1-3 years with LRI. Infection with Ct in these 9 older children (ages 17-34 mos, med=21 mos) occured from December to April and was associated with a variety of syndromes and with concurrent isolation of other pathogens.Symptoms and signs included cough (9), rhinitis (8), fever (3), abnormal bronchial breath sounds (3), hoarseness (2), shortness of breath (2), stridor (2) and wheezing (1).These results suggest that Ct is associated with a diversity of LRIs in children >1 year of age, and may often be present with other pathogens.(Supported by NHLBI-SCOR Grant HL-14136)

Application of pooled monoclonal antibodies for 1-hr detection of respiratory syncytial virus antigen in clinical specimens

A fluorescein isothiocyanate-conjugated pool of monoclonal antibodies (MoAb) to respiratory syncytial virus (RSV) was prospectively evaluated for its utility as a direct, 1-hr test for the diagnosis of RSV infection. Direct nasopharyngeal swab smears collected from 109 infants and children with acute respiratory illnesses were studied and compared with results obtained by indirect immunofluorescence using bovine polyclonal anti-RSV antibody on eluted cells derived from pooled nasopharyngeal and throat swab specimens (a 2.5-3 hr procedure), and culture. The MoAb-direct smear method was at least 86%-89% sensitive and 95%-100% specific compared with either of the other procedures. Additional prospective evaluations, as well as retrospective studies on a selected bank of slides stored from the preceding year, established that this MoAb could also be used with confidence in testing where direct smears are not employed.

IMMUNOLOGIC STUDIES OF ENTERIC CORONAVIRUS-LIKE PARTICLES|[lpar]|CVLP|[rpar]|

CVLP have been implicated in gastrointestinal illness sporadically among children and in an intensive care nursery outbreak at our hospital. Fecal examination by electron microscopy(EM) is the only means of diagnosis. We developed an enzyme-linked immunosorbent assay(ELISA) using ammonium sulfate precipitated CVLP from feces. The purified CVLP was immunogenic in rabbits. Sera containing antibodies against bovine enteric, canine, OC43, and 229E coronaviruses(CV) did not block binding of rabbit anti-CVLP to microtiter wells coated with purified CVLP. Immune EM using these antisera and nonaggregated CVLP also suggested that CVLP are antigenically distinct from other CV. False positives occurred when stools negative for CVLP by EM were tested. This nonspecific reaction was not blocked by sonicates of E.coli, Strep. fecalis, B.fragilis, or C.difficile, or by C.difficile antitoxin. Absorption of rabbit anti-CVLP serum with suckling mouse intestine did not improve specificity. More sophisticated purification of fecal CVLP is needed to improve ELISA sensitivity and specificity, and to characterize the antigen.