key: cord-024093-5dplc9xr authors: Sizun, J; Soupre, D; Legrand, MC; Giroux, JD; Rubio, S; Cauvin, JM; Chastel, C; Mix, D; de Parscau, L title: Neonatal nosocomial respiratory infection with coronavirus: a prospective study in a neonatal intensive care unit date: 2008-01-21 journal: Acta Paediatr DOI: 10.1111/j.1651-2227.1995.tb13710.x sha: doc_id: 24093 cord_uid: 5dplc9xr The aim of this prospective study was to evaluate the incidence of viral respiratory infection in hospitalized premature newborn infants and to assess the role of coronaviruses. All hospitalized premature infants with a gestational age less than or equal to 32 weeks were included. Tracheal or nasopharyngal specimens were studied by immunofluorescence for coronaviruses, respiratory syncytial virus, adenoviruses, influenza and parainfluenza viruses. Forty premature infants were included; 13 samples were positive in 10 newborns (coronaviruses n = 10; influenza 1 n= 2; adenovirus n= 1). None was positive at admission. All premature infants infected with coronaviruses had symptoms of bradycardia, apnea, hypoxemia, fever or abdominal distension. Chest X‐ray revealed diffuse infiltrates in two cases. However, no significant difference was observed between infected and non‐infected premature infants for gestational age, birth weight, duration of ventilation, age at discharge, incidence of apnea or bradycardia. Nosocomial respiratory tract infection with coronaviruses appears to be frequent. The clinical consequences should be evaluated in a larger population. Viral pneumonia which develops in neonatal intensive care units is potentially life-threatening. Indeed, young age (less than 1 month) and prematurity are two major risk factors (1). Nosocomial respiratory infections can occur sporadically or as an epidemic. The precise incidence has not been established. The variability or even absence of clinical symptomatology, especially in very immature newborn infants ( 2 ) , makes all retrospective evaluations difficult. Numerous viral agents have been implicated. The most frequently reported viruses are respiratory syncytial virus (RSV) (1, 3) , adenoviruses (4), influenza viruses A and B, and parainfluenza viruses I, I1 and 111 (5). The coronavirus (CV) has also been implicated. We have previously reported four cases of upper respiratory tract infection in premature newborns presenting with attacks of apnea of sudden onset (6). The pathogenic role of CVs during the neonatal period is, however, poorly established. The aim of this prospective study undertaken from November 1, 1991 to March 1, 1993, was to evaluate, in the pediatric intensive care unit, the incidence of upper respiratory tract viral infection in hospitalized premature newborn infants, to assess the role of CV and to analyze the associated symptomatology. The study was approved by the Hospital Center Ethics Committee. All newborn infants less than 3 days of age, with a gestational age less than or equal to 32 weeks and hospitalized in a resuscitation unit throughout the duration of the study, were entered into the trial. Infants who died before day 7 of life were excluded a posteriori. Samples were obtained on admission and then weekly. The methodology was as follows: administration of 0.5 ml of physiological saline via the intranasal route in non-intubated newborns or via the tracheal route in intubated newborns, followed by aspiration using a disposable Vygon-type catheter. Samples were transported at room temperature to the microbiology department and stored at +4"C until analysis. Samples were treated with a mucolytic solution (Mucomyst, Eurobio, France). Cells were then washed twice in phosphate buffered saline (PBS). After the last months = influenza virus. According to the manufacturer, monoclonal antibody against CV is a 110/120 kDa peplomer group specific and recognizes bovine, porcine and human CV strains. Clinical data and data obtained from the infant's family were recorded at the time of entry into the study and weekly by the same observer. Gestational age was established from obstetric data. Apnea was defined as absence of respiratory movements over a 20-s period and bradycardia as heart rate less than 90 beats/min, detected by a Hewlett-Packard cardiorespiratory monitor. Days on which more than six episodes of bradycardia of less than 60 beats/min, or more than eight episodes greater than 60 beats/min, or more than five apneic events occurred, were noted. Forty infants were included in the study. Initial samples were negative in all infants. Thirteen subsequent samples were found to be positive in 10 newborns: 10 for CV, 2 for myxovirus influenzae type 1 and 1 for adenovirus. Two children were CV positive on two successive samples. Another was positive for CV and then positive for myxovirus 4 weeks later. No sample was doubly positive. The seasonal distribution is shown in Fig. 1 . Previous perinatal histories are summarized in Table 1 . The group infected with CV was comparable to the non-infected group in terms of gestational age, birth weight, sex ratio and past history of hyaline membrane disease. All infants infected with CV were symptomatic at the time of infection ( Table 2) . Symptomatology was predominantly bradycardia and apnea. Digestive signs were present in five newborns but virology was not performed on the stools. Mean age at the time of infection was 17 f 7 days. Chest X-ray revealed bilateral interstitial infiltrate in two infants, 3 and 6 days after the date of the first positive CV sample. No statistically significant differences with respect to frequency of bradycardia or apnea were observed between the CV-infected and non-infected infants ( Table 3) . Laboratory signs of an inflammatory syndrome were not observed in any CV-infected newborns (normal C-reactive protein concentration). Blood cultures were negative. The development of a neonatal viral infection did not significantly alter the therapeutic management of these infants, since the duration of intubation, administration of atropine, need to interrupt feeding and conceptional mean age at the time of discharge were comparable in both groups. Two newborn infants in the non-infected group died , the first on day 13 with multiple organ failure and the second on day 30 due to cardiorespiratory arrest during surgical closure of a persistent ductus arteriosus; no infant in the infected group died. The incidence of viral infection in this study was high, affecting 25% of the newborn infants under investigation. No comparable evaluation exists in the literature. The negative initial sampling performed before the third day of life tends to support the diagnosis of nosocomial rather than materno-fetal contamination. RSV, the primary cause of bronchiolitis in infancy, has also been implicated as the cause of epidemics of chest infections in neonatal intensive care units (2, 3 ) and can worsen pre-existing respiratory disease. In this study, no sample was positive for RSV. On the other hand, we have seen a high rate of occurrence of infection with CV. CVs are RNA viruses with a spherical shape covered with projections similar to a "crown". Two main strains of human CV have been identified: 229 E and OC (7) . These viruses are commonly implicated in benign upper respiratory tract infections in adults (7), reproduced experimentally in healthy volunteers (8) . They are also implicated in acute decompensation in chronic respiratory insufficiency (9) . Epidemiological data in young infants are rare and vary depending on the study. In infants, CVs can be the cause of respiratory infections such as pneumonia or bronchiolitis, without clinical or radiological specificity compared with other respiratory viruses. In a serological study carried out in infants aged less than 18 months and hospitalized for respiratory tract infection, this cause was identified in 8.2% of cases (10). On the Similarly, in a prospective study by Ray et al., based on serological analysis and cell culture from acute lower respiratory tract infections in children less than 3 years of age, human CV viruses, as initial infecting agents, were not found to be responsible (12) . The difficulty in culturing CVs and the absence of serological specificity probably explain these findings (1 3) . The pathogenic role of CVs in the newborn is poorly understood. An association with enterocolitis in fullterm newborn infants has been reported (14) . No data are available in the literature regarding the possibility of nosocomial respiratory tract infections in the newborn. We reported four apneic episodes associated with bradycardia of sudden onset in the absence of an explicable etiology and with positive nasotracheal samples for CV on I F (6). CV has also been isolated using I F in infants with chronic respiratory insufficiency requiring prolonged mechanical ventilation (1 5). All the infected neonates in our prospective study had symptoms at the time of infection, which supports the recent onset of an acute process. However, we were unable to prove any significant difference between infected or non-infected infants with respect to symptomatology and therapeutic measures necessary. Several hypotheses can be proposed: these symptoms (apnea, bradycardia and digestive disorders) are not very specific in the premature infant and can be found in many pathological conditions ; moreover, duration of infection is short compared with duration of hospitalization. The diagnosis of a CV infection is based mainly on I F (1 5). The special culture conditions required make this procedure expensive and difficut to achieve. Serological methods are not very reliable (14) . The polymerase chain reaction has been used in the diagnosis of murine CV by Homberger et al. (16) . This technique appears to be faster and as sensitive as the usual techniques of inoculation of suckling mice or seroconversion after mouse inoculation, but it is not yet applicable to human CV, even though the genetic M sequence of OC43 is the counterpart of that of the murin virus. This study suggests a significant occurrence of upper respiratory tract viral infection in hospitalized premature newborns, mainly with CVs. The pathogenic role of CVs, as suggested by simultaneous viral infection and development of symptoms, requires a larger series of patients to be confirmed. Viral pneumonia in the first month of life Infections a virus respiratoire syncitial du nouveau-nt Neonatal respiratory syncitial virus infection Epidemiology of neonatal enterovirus infection Nasal colonization with coronavirus and apnea of the premature newborn Coronaviruses of man Effects of a "new" human respiratory virus in volunteers Coronavirus infections of man associated with diseases other than the common cold Coronavirus infection in acute lower respiratory tract disease of infants Diagnosis of human coronavirus infection by immunofluorescence : method and application to respiratory disease in hospitalized children Acute lower respiratory illnesses during the first three years of life: potential roles for various etiologic agents Viral infections of humans-epidemiology and control Association of coronavirus infection with neonatal necrotizing enterocolitis RBle pathogene des coronavirus en rtanimation ptdiatrique: analyse rttrospective de 19 prtlkvements positifs en immunofluorescence indirecte Detection of rodent coronaviruses in tissues and cell cultures by using polymerase chain reaction Bone mineralization after treatment of growth hormone deficiency in survivors of childhood malignancy 10) in "The discovery of phenylketonuria" by I. Farlling is given incorrectly in the above supplement A recessive syndrome in man We apologize for this inconvenience