key: cord-0724168-tqruna9t
authors: KAYE, H. S.; MARSH, H. B.; DOWDLE, W. R.
title: SEROEPIDEMIOLOGIC SURVEY OF CORONAVIRUS (STRAIN OC 43) RELATED INFECTIONS IN A CHILDREN'S POPULATION
date: 1971-07-03
journal: Am J Epidemiol
DOI: 10.1093/oxfordjournals.aje.a121293
sha: b213429f602cee321dec1ddd556264021a3d629a
doc_id: 724168
cord_uid: tqruna9t

Kaye, H. S. (CDC, Atlanta, Ga. 30333), H. B. Marsh and W. R. Dowdle. Sero-epidemiologic survey of coronavirus (strain OC 43) related infections in a children's population. Amer J Epid 94: 43–49, 1971.—Acute and convalescent serum pairs and control sera collected from subjects living in a children's home over a 7-year period (1960–1967) were examined by hemagglutination-inhibition (HI) test with coronavirus strain OC 43. Ninety-three serologic conversions were observed; 44 were associated with reported illnesses and 49 with no reported illnesses. In three distinct outbreaks during the winter and spring quarter of 1960–1961, 1964–1965, and 1966–1967, 67 conversions occurred. Seroconversions to strain OC 43 were associated with as much as 19% of the respiratory diseases in a single season. Over the 7-year period coronavirus strain OC 43 accounted for 3% of the total 1328 respiratory illnesses. Evidence of preexisting antibody was apparent in one-third of the children showing seroconversions. The HI test was more sensitive for serodiagnosis than the complement-fixaticn test. The major presenting complaints of the children with respiratory disease associated with coronavirus strain OC 43 were sore throat, cough and coryza; the predominating symptoms were pharyngitis, coryza, fever and cervical adenitis.

The newly identified coronavirus group, studies have been limited because of the which includes the prototype strains B814 fastidious growth requirement of certain of (1), 229E (2) , and "IBV-like" viruses (3), these viruses (1, 3) . may be emerging as an important cause of T h e adaptation to growth in suckling common cold-like respiratory illness in mouse brain of "IBV-like" strains OC 38 and OC 43, previously grown only in organ Abbreviations: CF, complement-fixation; GMT, cu it U re, resulted in the recognition of comgeometric mean HI antibody titer; HI hemagglu-]ement . fixi (CF) anti (4) and led tination-inhibition; MOT, mouse hepatitis virus. ' • , , \ »Fh>m the Respiratory Virology Unit. Labora-fc o the demonstration of hemagglutmins for tory Division, Center for Disease Control, Health these apparently serologically identical Services and Mental Health Administration, strains (5, 6) . The hemagglutination-inhibi-Public Health Sen-ice, U. S. Department of fcion ( H j) tesfc as we]1 as the C F test are ^t tl » EdU ? tiOn a "f 7f^% Atlanta period. The number of illnesses and infections, seasonal distribution, antibody prevalence, relationship of antibody to protection, and clinical syndrome have been determined, and the results are presented.

A longitudinal survey of respiratory illness was conducted from 1960 to 1968 in a church-sponsored children's home in the Atlanta area. This population was described in detail previously (7) . The study group consisted of healthy Caucasian children ranging in age from five to 19 years who were admitted to the home for socioeconomic reasons. The children lived in cottages which housed from eight to 12 each. Assignment to a cottage was made on the basis of age and sex. Meals were eaten in a common dining room, recreational facilities were shared and the children attended public schools in the community. A study year covered the period between September 1 and August 31. The total number of children, median age, percentage of turnover and the percentage of children from the original study group per year are shown in table 1.

Collection of specimens. Children with symptoms of respiratory illness were sent by house mothers to the home clinic for examination and treatment by the attending physician. Throat swab specimens were collected for bacterial and viral isolation. Techniques for isolation of coronavirus were not available at the time of the study. Acute and convalescent blood specimens were collected two to three weeks apart. Control sera were collected three times per year during the first two years of this study. Less frequent collections were made in succeeding years.

Virus. The coronavirus strain OC 43, which had been isolated in organ culture and subsequently adapted to the suckling mouse brain, was used for antigen production (4) .

Production of antigen. Antigens were prepared from infected and normal suckling mouse brain by making a 10 per cent suspension in phosphate-buffered saline, pH 7.2 (HI), and Veronal-buffered diluent, pH 7.3 (CF). They then were clarified by refrigerated centrifugation at 600 X g for 20 minutes (5) . The antigens were stored at -70 C.

Production of antisera. Immune sera were prepared by intraperitoneal and intracerebral inoculation of adult Swiss white (ICR) mice (5) .

Serologic tests. During the seven-year study all acute, convalescent and control sera were tested by CF for diagnostic rises in antibody titer to influenza A and B, parainfluenza types 1, 2, and 3, adenovirus, mumps vims, respiratory syncytial virus, herpesvirus and Mycoplasma pneumoniae. In this present study sera were tested only for antibody to coronavirus OC 43 antigens. HI tests were performed by the microtiter technique with phosphate-buffered saline diluent and 0.5 per cent adult chicken erythrocytes (8) . The CF test in the present study was also performed by microtiter (9) . All sera were inactivated at 56 C for 30 minutes. A fourfold or greater increase in antibody titer was considered to constitute a seroconversion and to be indicative of infection. § Detected by rises in serum antibody titer between normal bleedings.

"f Three cases of apparent reinfection.

Incidence of respiratory illness and senconversion to coronavirus. Among 1328 respiratory illnesses which occurred over the seven-year period (table 2), 44 seroconversions to coronavirus strain OC 43 were found. An additional 49 seroconversions were detected between normal bleedings, making a total of 93.

Seroconversions to strain OC 43 were observed each year of the study, but the largest number totalling 75 was recorded for the years 1960-1961, 1964-1965, and 1966-1967 . About one-half of the 75 were associated with reported illnesses. Eighteen seroconversions were scattered throughout the remaining four years of the study, although only seven of these occurred among children reporting respiratory disease.

Sixteen other subjects with antibody titer of < 1:10 in two consecutive serum specimens later acquired antibody titers of 1:10 in at least two consecutive serum specimens. Because these subjects had less than fourfold rises in antibody titers, they were not included in the total seroconversions. Seasonal distribution. In three distinct outbreaks which occurred during the winter and spring quarters of 1960-1961, 1964-1965, and 1966-1967 Evidence of -preexisting antibody. Of the total number of the children with seroconversions (93), 33% had preexisting antibody titers of 1:10 or greater in at least two sequential serum specimens collected prior to antibody rise ( In our present study HI antibody seroconversions to coronavirus strain OC 43 accounted for 44 (3 per cent) of the 1328 respiratory illnesses reported. Thirty-seven of these seroconversions were found in three distinct outbreaks which occurred in the winter and spring quarters of the years 1960-1961, 1964-1965, and 1966-1967 . The highest incidence of seroconversions without reported illness also occurred during these same periods. Although the average incidence of illness associated with coronavirus strain OC 43 was 5 per cent for all winter seasons, 19 per cent of respiratory illnesses occurring in the winter of 1960-1961 showed diagnostic coronavirus antibody rises. These findings suggest that a significant percentage of respiratory illness may be due to infection with this virus. Like Mclntosh and associates, we also found that the highest number of seroconversions to OC 43 occurred at a time (1960) (1961) when respiratory diseases associated with other etiologic agents were absent.

HI antibody prevalence in our children's population rose almost continually from 1960-1961 to the termination of the study in December 1967. The level of antibody in the home was slightly higher than has been previously reported for other populations. Mclntosh and coworkers (10) found that 47 per cent of the sera collected in 1965-1967 from children between the ages of five and seven years had CF antibody titers of 1:4 or greater against OC 38 and OC 43. The measurable CF antibody in adults during 1962-1964 and 1965-1967 was 72 and 67 per cent, respectively. Unreported results from HI tests performed in our laboratory on sera collected from the general population in 1963-1965 also confirm these general findings. The high level of antibody prevailing in the children's home population at the termination of the survey may be attributed to the increased median age of the population from age nine in 1960 to age 11 in 1967, and the outbreaks occurring in the latter part of the study. The increased prevalence of antibody among the 34 children remaining in the study over the entire seven-year period reflects both the long-term persistence of HI antibody and the number of seroconversions over the same period of time.

The significance of preexisting antibody to OC 43 is not clear. Approximately onehalf of the older age groups demonstrating seroconversions were found to have preexisting antibody, and diagnostic antibody rises were recorded on two separate occasions for three subjects. These findings suggest that either protection against reinfection is short-lived or that such antibody reflects previous infection with a closely related strain. Other studies have also noted the heterogeneity of antibody responses with known or presumed coronavirus infection. Mclntosh (12) reports that fourfold or greater HI antibody rises to OC 43 were found among 14 per cent of 70 paired sera obtained from volunteers before and after infection with coronaviruses other than OC 43. Extension of these observations of heterotypic rises to yet unknown strains cannot be excluded. The absence of data on isolation of coronaviruses during our study places severe restrictions on assigning OC 43 as the sole etiologic agent. Regardless of the origin of preexisting coronavirus antibody, our finding that only seven of the 31 cases with preexisting antibody had titers of 1:20 or above suggests that reinfection is limited largely to those subjects with low levels of antibody.

Because of the limited quantity of sera, CF as well as HI tests were not performed on all specimens. However, the small number of CF tests performed on sera showing HI seroconversion confirm a previous report by Kaye and Dowdle (5) that CF is less sensitive for diagnosis. The possibility that improved antigens could increase the sensitivity of the CF test should not be discounted.

The sparse amount of information available on the clinical aspects of infection with OC 38 and OC 43 does not permit a comparison of our data with those of others. Our findings, based only on patients with seroconversion to coronavirus strain OC 43, indicate a close parallel with B814 and 229E (1, 13, 14) , particularly in regard to nasal involvement. However, our attending physicians saw considerably more fever than has been noted for other members of the human coronavirus group. This discrepancy may be explained by the age differences in the population under study or by the conditions of infection. Because of the absence of virus isolation, further studies are necessary to confirm the signs and symptoms of this disease under natural conditions.

Cultivation of a novel type of common-cold virus in organ cultures

A new virus isolated from the human respiratory tract

Recovery in tracheal organ cultures of novel viruses from patients with respiratory disease

CORONAVIEUS RELATED INFECTIONS IN CHILDREN Growth in suckling-mouse brain of "IBV-like" viruses from patients with upper respiratory tract disease

Some characteristics of hemagglutination of certain strains of "IBV-like" virus

Antigenic relationships among the coronaviruses of man and between human and animal coronaviruses

Robinson RQ: Mycoplasma pneumoniae infections in a children's population: a five-year study

Application of a microtechnique to viral serological investigations

US Public Health Service: Standardized diagnostic complement-fixation method and adaptation to microtechnique

Seroepidemiologic studies of coronavirus infection inadults and children

Antibodies to mouse hepatitis virus in human sera

Antigenic relationship amongst coronaviruses

Isolation from man of "avian infectious bronchitis virus-like" viruses (coronavirus) similar to 229E virus, with some epidemiological observations

Effects of a "new" human respiratory virus in volunteers