Respiratory viral infections in hospitalized children: implications for infection control.

ABSTRACT

Background. Identification of children with respiratory viral infections may augment infection-control practices on inpatient units. There are clinical syndromes leading to morbidity among hospitalized children, however, in which a viral etiology of the illness might not be considered.

Methods. Virus infection rates among 243 children aged <1 to 19 years hospitalized between October 1993 and April 1994 with asthma, pneumonia, bronchiolitis, fever, apnea, croup, or respiratory distress were evaluated as part of a University of Maryland Medical Center infection-control protocol. Anonymous data collected included admission diagnoses, age, and virus-identification result.

Results. Seventy-one children (29%) had a virus identified, including 19 of 123 (15%) with asthma, 4 of 12 (33%) with pneumonia, 27 of 47 (57%) with bronchiolitis, 13 of 41 (32%) with fever, 4 of 9 (44%) with apnea, 2 of 3 (67%) with croup, and 2 of 8 (25%) with unspecified respiratory distress.

Conclusion. This study reinforces the concept that clinicians should consider respiratory viruses for a broad range of diagnoses. This heightened awareness may help reduce the number of nosocomial respiratory viral infections.

* * * * * * * * * *

NOSOCOMIAL respiratory viral infections cause significant morbidity and mortality, particularly among hospitalized children with underlying cardiorespiratory disease or immunodeficiencies. (1,2) Hospital infection-control policies have been traditionally oriented toward clinical syndromes and conditions that have been associated with high rates of respiratory virus shedding (eg, bronchiolitis and croup) . (3) Hospitalized children with asthma, apnea, or fever without an apparent source may not be routinely considered to be a source of nosocomial viral respiratory infection. Although these latter clinical syndromes may be associated with viral pathogens, hospital infection-control protocols may not be uniformly instituted for these children. This issue is important, because infection-control measures, such as cohort nursing, wearing gowns, and using gloves for all encounters with virus-infected children, has been shown to reduce nosocomial spread of disease. (4)

The purpose of this study was to estimate respiratory virus infection rates associated with certain admission diagnoses during the respiratory viral infection season.

METHODS

Study Design

We analyzed a cross-sectional convenience sample of children admitted to a pediatric inpatient service to describe respiratory-virus infection rates associated with clinical syndromes and diagnoses.

Subjects

To more appropriately target infection-control practices, a new protocol to identify respiratory-virus-infected children was initiated by the University of Maryland Department of Pediatrics, in conjunction with the University of Maryland Medical System infection control department, for the respiratory-infection season of October 1993 to April 1994. Children admitted to the pediatric inpatient service during this period with clinical "trigger" diagnoses were to have nasopharyngeal aspirates (NA) obtained for viral antigen detection and culture. Trigger diagnoses included asthma, pneumonia, bronchiolitis (wheezing-associated illness in children <2 years of age with no history of asthma), fever, apnea, croup, and respiratory distress not otherwise classified. All pediatric admission diagnoses for this study period were obtained from a nurse employed by the department of pediatrics, who was responsible for inpatient quality control for hospitalized children. She maintained a computer file of all admissions to the inpatient units, along with admission diagnoses. Two of us (R.L. and J.K.) reviewed medical charts for all admissions.

Nasal Aspirate Samples

Nurses and respiratory therapists from the University of Maryland Medical System pediatric emergency department and inpatient wards were trained by 2 of us (R.L. and J.K.) to obtain NA samples in a uniform manner. Nasopharyngeal aspirate samples were obtained by first instilling 1 to 2 drops of sterile saline without preservatives into each naris. Nasal contents were then gently suctioned through both nares, using an 8-French catheter connected to a specimen trap. Viral transport medium with antibiotics was then aspirated through the same catheter into the specimen trap. The trap was kept refrigerated (but not frozen) until it was transported to the hospital virology laboratory.

Viral Detection

Aliquots of NA specimen were tested according to the manufacturers' specifications using 2 enzyme immunoassays; one assay was performed for respiratory syncytial virus (RSV) (TestPack RSV, Abbott Laboratories, North Chicago, Ill), and one for influenza A (Directigen-Flu A, Becton Dickinson, Cockeysville, Md). The NA sample was then centrifuged at 3,400 revolutions per minute for 10 minutes at 4[degrees]C. The supernatant was used to inoculate tissue monolayers for viral culture, and the remaining cell button was used for immunofluorescent assay (Respiratory Panel, Viral Screening and Identification Indirect Immunofluorescent Assay, Chemicon International Inc, Temecula, Calif). This fluorescent assay screens for 7 respiratory viruses, including RSV, influenza A and B, parainfluenza types 1, 2, and 3, and adenovirus (common antigen).

Data Collection and Statistics

Demographic information and admitting diagnoses on all inpatient admissions were recorded anonymously, as were the results of respiratory viral assays. The University of Maryland internal review board approved this study, and because subject information was recorded anonymously, signed written informed consent was not obtained. Rates of respiratory virus identification associated with each clinical diagnosis were estimated with 95% confidence intervals (CIs).

RESULTS

A total of 703 children aged <1 to 19 years were admitted to the University of Maryland Medical System during the 7-month study period. On retrospective review, 401 (57%) of these 703 children met the criteria for obtaining NA for virus antigen detection and tissue culture. Of the 401 eligible children, 243 (61%) had NA obtained for virus identification, and 158 (39%) did not. Reasons for incomplete adherence to the protocol were unknown, but the primary reason was thought to be the house staff's unfamiliarity with the new infection-control procedure. The admission diagnoses of the children whose NA were cultured and those who were not included in this study are shown in Table 1; they were not significantly different.

Of the 243 children with NA cultures, 71 (29%) had a respiratory virus detected, which represented approximately 10% of the 703 admissions during this period. Table 2 illustrates the distribution of the 71 viruses detected, which were RSV (n = 52; 73%); influenza A (n = 9; 13%); parainfluenza viruses, types 2 and 3 (n = 4; 6%); adenoviruses, not typed (n = 4; 6%); and enteroviruses, not typed (n = 2; 3%).

Table 2 also shows the rates of respiratory virus identification among hospitalized children with various admission diagnoses. The distribution of virus types within each diagnostic category is also shown in this table. The highest rate of virus identification was observed for croup and bronchiolitis. Although asthma was associated with the lowest rate of virus identification, a respiratory virus was identified in more than 15% of sampled asthmatic children. Of note, for diagnoses such as fever in young infants, and apnea, where a bacterial etiology is often the focus of clinical attention, there was a substantial rate (15% to 44%) of associated respiratory-virus infection.

In children with asthma exacerbations, virus identification rates were inversely proportional to age. When admitted with a diagnosis of asthma, viruses were identified in 10 of 34 children <2 years of age (29%; 95% CI, 14%-44%); 7 of 57 children aged 2 to 5 years (12%; 95% CI, 2%-22%); and 2 of 32 children aged 6 to 19 years (6%; 95% CI, 2%-10%). Of note, both of the children in the oldest group were >13 years of age.

DISCUSSION

Of 703 children admitted during the fall/winter respiratory-infection season, 71 (10%) had a respiratory virus identified from their nasopharynx, and potentially more could have been infected. This observation reinforces the importance of increased vigilance regarding infection-control practices during this high-risk season to prevent nosocomial spread of these pathogens. This study estimated the rates of respiratory-virus infection among hospitalized children with various diagnoses that may be overlooked as viral in etiology. It also reconfirms the importance of RSV as the most significant pathogen associated with pediatric hospitalization, with lesser but noteworthy roles played by influenza, parainfluenza, and adenovirus.

The main limitation of this study was the incomplete sampling of all children with the identified criteria. For example, there may have been a bias to sample children with asthma who had rhinorrhea or other prominent upper airway signs or symptoms. The distribution of diagnoses in Table 1, however, indicates that NA samples were obtained with similar frequencies among the various groups. It is, therefore, likely that the virus isolation rates in the nonsampled children would have been similar to the group whose NA were cultured.

Of note, diagnoses that have not been traditionally identified as warranting infection-control policies, such as asthma, fever without an apparent source, and apnea, (3) were associated with a significant rate of respiratory-virus identification in this study. Even adolescents admitted with asthma had a substantial rate of virus identified from their nasopharynx. This information should be helpful to those responsible for developing infection-control protocols for hospital inpatient units.

Respiratory virus infection has long been thought to trigger asthma exacerbations. Johnston et al (5) detected respiratory viruses in the nasopharynx of 80% of patients with asthma exacerbations in prospectively followed children with known asthma. Other studies have shown lower but substantial rates (19% to 42%) of respiratory viruses isolated from children hospitalized with asthma exacerbations. (6,7) The lower rate of virus identification from asthmatics encountered in the present study may partially reflect that rhinoviruses were not routinely identified in our hospital laboratory. Rhinoviruses have often been identified from individuals with asthma exacerbations. (5-7)

The present study showed decreased virus-shedding rates with increased age in children with asthma. Lower rates of respiratory virus infection have previously been described in adults than in children with asthma exacerbations. (8) Still, several adolescent asthmatics in the present study had a virus identified from their nasal secretions.

Traditionally, the clinical and therapeutic focus on an infant admitted with fever or apnea has been on possible bacterial etiologies of the illness. This study confirms previous observations of the high rate of viral infection in these infants. (9-11) This study also confirms a high rate of virus identification in patients with diagnoses ofbronchiolitis, croup, and pneumonia, which traditionally have been considered in most respiratory infection-control protocols.

Although the published hospital isolation recommendations do not specifically include children with asthma exacerbations, apnea, or fever without a source, (3) these diagnoses have been included in a recent National Association of Children's Hospitals and Related Institutions Infection Control Cohort Guidelines. (12) These guidelines specifically refer to "virus-induced asthma," which we believe may be difficult to distinguish from other types of asthma in hospitalized children.

Obtaining samples for virus identification on all children with possible respiratory viral infections is costly and may be impractical. In general, length of hospital stays is decreasing; therefore, patients may be discharged before the virus identification is complete. Consequently, the practice of grouping together children with similar diagnoses, such as asthma, and extending respiratory-virus isolation procedures during the fall/winter season to patients with the broad range of diagnoses included in this study would seem an optimal management strategy.

At the time this study was done (1993 to 1994), our policy was to culture the NA of all children with evidence of possible respiratory viral infection. Based on the results of this study and the considerable added expense of viral culture and antigen testing, the current infection-control policy at our institution was changed after 1994. Patients with similar diagnoses share a common room without viral testing. In addition, the trend toward briefer hospital admissions makes this cohort practice a more efficient, less expensive option than viral testing. As of 2001, viral cultures and antigen testing are more specifically used for patients with prolonged hospital stays or for therapeutic considerations, such as the cessation of antibiotics.

TABLE 1.

Age and Admission Diagnoses of 401 Children Admitted With Possible
Respiratory Viral Illness

                                     Children With  Children Without
                                       NA Sample       NA Sample
          Characteristic               (n = 243)       (n = 158)

Asthma                                 123 (51%)        85 (54%)
Pneumonia                               12 (5%)          8 (5%)
Bronchiolitis                           47 (19%)        17 (11%)
Fever (>3 months of age)                20 (8%)         28 (18%)
Fever (<3 months of age)                21 (9%)          8 (5%)
Apnea (<1 year of age)                   9 (4%)          5 (3%)
Group                                    3 (1%)          5 (3%)
Respiratory distress not classified      8 (3%)          2 (1%)

NA = Nasal aspirate for virus detection.
TABLE 2.

Virus Identification Rates Among Various Admission Diagnostic Categories

Admission                             Any
Diagnosis                        No.  Virus *              RSV

Asthma                           123  19 (15%)           18 (15%)
                                      (95% CI, 9%-21%)
Pneumonia *                       12   4 (33%)            2 (17%)
Bronchiolitis                     47  27 (57%)           23 (49%)
                                      (95% CI, 19%-47%)
Fever (>3 months                  20   3 (15%)            0
  of age)                             (95% CI, 0%-31%)
Fever ([less than or equal to]3   21  10 (48%)            5 (24%)
  months of age)                      (95% CI, 27%-62%)
Apnea ([less than or equal to]1    9   4 (44%)            3 (33%)
  months of age)
Croup **                           3   2 (67%)            1 (33%)
Respiratory distresss **           8   2 (25%)            0

  Total                          243  71 (29%)           52 (21%)
                                      (95% CI, 23%-35%)

Admission
Diagnosis                        Influenza A  Parainfluenza  Adenovirus

Asthma                             1 (1%)        0            0

Pneumonia *                        0             2 (17%)      0
Bronchiolitis                      1 (2%)        1 (2%)       1 (2%)

Fever (>3 months                   1 (5%)        0            1 (5%)
  of age)
Fever ([less than or equal to]3    5 (24%)       0            0
  months of age)
Apnea ([less than or equal to]1    0             0            1 (11%)
  months of age)
Croup **                           0             1 (33%)      0
Respiratory distresss **           1 (13%)       0            1 (13%)

  Total                            9 (4%)        4 (2%)       4 (2%)


Admission
Diagnosis                        Enterovirus

Asthma                             0

Pneumonia *                        0
Bronchiolitis                      1 (2%)

Fever (>3 months                   1 (5%)
  of age)
Fever ([less than or equal to]3    0
  months of age)
Apnea ([less than or equal to]1    0
  months of age)
Croup **                           0
Respiratory distresss **           0

  Total                            2 (1%)


* 95% confidence intervals (CIs) applied when sample size appropriate.

+ Six were <24 months of age, and 3 of these had a respiratory virus
identified.

** All <24 months of age.

RSV = Respiratory syncytial virus.

Acknowledgements. We thank Alice Steffen, RN, Frances Green, MD, and Jennifer Wehberg, MD, for chart review and data collection; and Eric Talley, BA, for virology Laboratory analysis.

References

(1.) Moler FW, Khan AS, Meliones JW, et al: Respiratory syncytial virus morbidity and mortality estimates in congenital heart disease patients: a recent experience. Crit Care Med 1992; 20:1406-1413

(2.) Hall GB, Powell KR, MacDonald NE, et al: Respiratory syncytial viral infection with compromised immune function. N Engl J Med 1986; 315:77-81

(3.) Garner JS: Guideline for isolation precautions in hospitals. The Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1996; 17:53-80

(4.) Madge P, Paton JY, McColl JH, et al: Prospective controlled study of four infection control procedures to prevent nosocomial infection with respiratory syncytial virus. Lancet 1992; 340:1079-1082

(5.) Johnston SL, Pattemore PK, Sanderson G, et al: Community study of role of viral infections in exacerbations of asthma in 9-11 year old children. BMJ 1995; 310:1225-1229

(6.) Jennings LC, Barns G, Dawson KP: The association of viruses with acute asthma. NZ Med J 1987; 100:488-490

(7.) Beasley R, Coleman ED, Hermon Y, et al: Viral respiratory tract infections and exacerbations of asthma in adult patients. Thorax 1988; 43:679-683

(8.) Pattemore PK, Johnston SL, Bardin PG: Viruses as precipitants of asthma. I. Epidemiology. Clin Exp Allergy 1992; 22:325-336

(9.) Dagan R, Hall CB, Powell KR, et al: Epidemiology and laboratory diagnosis of infection with viral and bacterial pathogens in infants hospitalized for suspected sepsis. J Pediatr 1989; 115:351-356

(10.) Bruhn FW, Mokrohisky ST, McIntosh K: Apnea associated with respiratory syncytial virus infection in young infants. J Pediatr 1997; 90:382-386

(11.) Church NR, Anas NG, Hall GB, et al: Respiratory syncytial virus related apnea in infants: demographics and outcome. Am J Dis Child 1984; 138:247-250

(12.) National Association of Children's Hospitals and Related Institutions Infection Control Focus Group: Bed Assignment/Cohort Guidelines. Alexandria, Va, National Association of Children's Hospitals and Related Institutions, 1999

RELATED ARTICLE: KEY POINTS

* Respiratory viral infection is prevalent between October and April among hospitalized children.

* Some diagnoses, such as asthma and fever in young infants, must be considered to be associated with respiratory viral infection, along with diagnoses more commonly associated with these viruses, such as bronchiolitis and croup.

* Infection-control practices for hospitalized children need to incorporate this type of information to reduce the risk of nosocomial infection.

From the Department of Pediatrics and the Virology Laboratory, University of Maryland Medical Systems, Baltimore.

Reprint requests to Richard Lichenstein, MD, University of Maryland Medical Systems, Pediatric Emergency Department (N1W46), 22 S Greene St, Baltimore, MD 21201.