Objective. Preterm infants, especially those with chronic lung disease (CLD), are considered more susceptible to severe respiratory illness from respiratory syncytial virus (RSV) infection than healthy term infants, and are therefore targeted for prophylactic administration of immune globulins. The impact of this practice on the more severe cases of bronchiolitis (ie, pediatric intensive care unit [PICU] admission, mechanical ventilation, mortality) has not been reported to date. The aim of this study was to evaluate PICU admissions, need for mechanical ventilation, and mortality attributable to RSV bronchiolitis in Israel before the introduction of RSV prophylaxis to the country.
Design and Setting. Prospective survey of 11 PICUs in Israel during the RSV season (November 2000-March 2001).
Patients. All patients admitted to the PICU because of bronchiolitis, and the subgroups who needed mechanical ventilation or who died, were analyzed for known risk factors, namely, prematurity, CLD, and chronic oxygen dependence.
Results. A total of 105 patients with RSV bronchiolitis met the inclusion criteria, of whom 33 were mechanically ventilated. Most of the patients (84% total admission, 88% ventilated) were born after 32 weeks’ gestation, and 89% and 88%, respectively, did not have CLD. Only 16% and 9%, respectively, met any of the American Academy of Pediatrics criteria for RSV prophylaxis, such that 84% of the whole sample and 91% of the ventilated patients were not candidates for RSV prophylaxis. Five patients died, 2 of them with cyanotic heart disease.
Conclusions. Most of the infants with severe RSV bronchiolitis were born at term and did not have CLD. The great majority of patients admitted to the PICU for bronchiolitis were not candidates for RSV prophylaxis. Administration of RSV prophylaxis to the predefined high-risk population could be expected to yield no significant change in PICU admissions or number of infants needing mechanical ventilation. New risk-stratified guidelines for RSV prophylaxis are needed.
- respiratory syncytial virus
- pediatric intensive care unit
- chronic lung disease
- mechanical ventilation
Respiratory syncytial virus (RSV) infection is the main reason for hospitalization for respiratory tract illness in young children.1 Infants with primary RSV infection are rarely asymptomatic. Preterm infants, especially those with chronic lung disease (CLD) (previously termed bronchopulmonary dysplasia), are thought to be more susceptible to severe respiratory illness from RSV than healthy term infants,2,3 and therefore are currently targeted for prophylactic administration of immune globulins just before and during the RSV season. The Food and Drug Administration has approved 2 preparations for this purpose: RSV intravenous immune globulin ([IVIG] RespiGam, Massachusetts Public Health Biological Laboratories; and MedImmune, Inc, Gaithersburg, MD),4,5 and a monoclonal antibody preparation against the F glycoprotein of RSV (Palivizumab, Synagis; MedImmune, Inc).6 According to the recommendations of the American Academy of Pediatrics (AAP),7 RSV prophylaxis should be considered for the following patients:
Infants younger than 2 years with CLD who required medical therapy (supplemental oxygen, bronchodilators, diuretics, or corticosteroids) for CLD within 6 months before the anticipated RSV season.
Infants born at 28 weeks of gestation or earlier who do not meet the first criteria, up to 12 months of age.
Infants born at 29 to 32 weeks of gestation who do not meet the first criteria, up to 6 months of age.
Several recent cost-effectiveness studies have questioned the broadness of these guidelines and suggested that RSV prophylaxis should be restricted to preterm infants with CLD who are oxygen-dependent.8 The aim of the present study was to evaluate pediatric intensive care unit (PICU) admissions, need mechanical ventilation, and mortality attributable to RSV bronchiolitis in Israel before the introduction of RSV prophylaxis to the country.
This prospective study was conducted at 11 of the 13 PICUs in Israel during 1 RSV season (November 2000 to March 2001). RSV prophylaxis was not in use in Israel during this period. A detailed questionnaire was completed for all patients admitted because of acute bronchiolitis during or immediately after the RSV season. The data were collected by the authors by personal communication with a senior staff member of every PICU. Weekly discussions were held with a senior staff member of every PICU during the RSV season. All patients admitted because of clinical symptoms compatible with acute bronchiolitis during the winter season were included. The PICU admissions registry was periodically examined to prevent omissions or errors. The following parameters were recorded: age, gender, gestational age, birth weight, perinatal medical history, medical treatment since discharge from the neonatal department, present medical treatment, presence of CLD (defined as oxygen dependence at 36 weeks’ gestational age), oxygen requirement before PICU admission, and indication for PICU admission. All children were monitored to determine the current medical treatment, RSV status, total days of PICU hospitalization, total days of mechanical ventilation, and mortality rate. RSV status was determined by nasopharyngeal aspirate and standard RSV enzyme immunoassay detection, as previously reported.9
Individual hospital policy was followed with regard to specific indications for PICU admission and medical management, which were unrelated to the study. Analyses were performed on the total patients admitted to the PICUs and, separately, on the patients who were mechanically ventilated and the patients who died.
National statistics on number of annual live births, rate and incidence of preterm deliveries, and neonatal intensive care unit (NICU) admissions and discharges were obtained from the Israel Neonatal Network (Israel Center for Disease Control).10
The study was approved by the institutional review board. Statistical analysis was performed with Pearson χ2 and Fisher exact tests.
Full data were obtained from 11 of the 13 PICUs in Israel. In the other 2 units, only the total number of admissions for bronchiolitis was known (8 cases); there were no RSV-related deaths in either one. Because these data were incomplete, these units were not included in our analysis.
Questionnaires were completed for a total of 129 patients admitted to the PICU because of acute bronchiolitis during the RSV season of winter 2000–2001; about half (51.2%) were males. Mean age at admission was 4.7 months and median age, 2.3 months. Seventy-eight percent were Jewish and 28% were Arabic. The most common indication for PICU admission was severe respiratory distress (82%), which included severe tachypnea, poor air entry, increased respiratory effort, hypoxemia despite oxygen supplementation, or hypercarbia. Apnea was present in 32% of the admitted patients. The average duration of PICU stay was 7.7 days (range: 2–40). Thirty-seven patients (28.7%) received mechanical ventilation; 35% of them were born prematurely. The average duration of mechanical ventilation was 8.9 days (range: 1–27).
Twenty-four of the 129 admissions to PICU (19%) were classified as non-RSV. In 20, the RSV test was negative, and in the other 4, RSV status was unknown (test not performed or inconclusive). Because our study focuses on candidates for monoclonal RSV prophylaxis (palivizumab), the non-RSV patients were excluded from the analysis.
In 105 of the 129 patients (81%), RSV was identified in nasopharyngeal secretions. The data on PICU admissions, mechanical ventilation, and mortality in the children with RSV bronchiolitis according to gestational age and presence of CLD are presented in Tables 1 and 2, respectively. Eighty-four percent of the patients admitted to the PICU and 88% of the patients who were mechanically ventilated were born after 32 weeks’ gestation. In addition, 89% of the patients admitted to the PICU and 88% of the patients who were mechanically ventilated did not have CLD.
The classification of the infants according to the AAP guidelines for the use of RSV prophylaxis7 is presented in Table 3. Only 16% of all admitted patients and 9% of the mechanically ventilated patients met any of the AAP criteria; that is, 84% of all patients and 91% of the ventilated patients were not candidates for RSV prophylaxis.
The infants predefined as high risk had a lower rate of mechanical ventilation (3/17, 17%) than the low-risk group (30/88, 34%), but the difference did not reach statistical significance (P = .25; Table 3).
Congenital Heart Disease and RSV Bronchiolitis
Five patients with RSV bronchiolitis and cyanotic congenital heart disease were admitted to the PICU (Table 4); 3 of them were mechanically ventilated and 2 died. There were 4 additional patients with acyanotic congenital heart disease, 1 of whom was mechanically ventilated.
Overall, 5 children with acute bronchiolitis died during the RSV season (Table 5). All were younger than 4 months at admission, and in all of them, nasopharyngeal secretions were positive for RSV. Two had cyanotic congenital heart disease (hypoplastic left heart, tetralogy of Fallot). Two were born at 27 weeks’ gestation: 1 was receiving oxygen therapy at home, and 1 had an intestinal obstruction which was the direct cause of death. The fifth patient was born at 31 weeks and died of Candida sepsis and polymicrobial infection following prolonged mechanical ventilation.
According to the Israel Neonatal Network10 (Israel Center for Disease Control), an annual average of 442 infants born before 28 weeks’ gestation and 703 infants born between 29 and 32 weeks’ gestation were discharged from the NICUs in Israel. Overall, 7.8% of all preterm infants were diagnosed with CLD, but only 2.2% were discharged from the hospital with oxygen therapy. Table 6 shows the national annual delivery data and the expected prevention of PICU admission were RSV prophylaxis to be administered to the high-risk population.
During the study period, 404 patients were hospitalized because of bronchiolitis in 1 of the participating centers (Schneider Children’s Medical Center of Israel). Twenty of the hospitalized children were admitted to the intensive care unit ([ICU] 20/404, 5%), of whom 6 needed mechanical ventilation.
We studied all admissions for bronchiolitis in 11 of 13 PICUs in Israel during 1 RSV season (November 2000 to March 2001) to identify the most critical patients for prophylactic therapy. Our findings showed that the great majority of infants were born at term and did not have any known risk factor. Only 16% met any of the AAP criteria for RSV prophylaxis, including only 3 of the 33 patients who received mechanical ventilation. We identified 1 important criterion for PICU admission in this population: presence of CLD requiring oxygen (2/35, 5.7%) or medical therapy (4/54, 7.4%) within 6 months of the start of the RSV season (Table 6). These infants are probably an appropriate target group for RSV prophylaxis. The rate of ICU admissions in the preterm group was lower; 1% for infants born at 29 to 32 weeks and 2.2% for infants born before 29 weeks, but the small number of patients in each group limits the statistical analysis. Nevertheless, our data agree with previous studies indicating that CLD is an higher risk factor for severe bronchiolitis than prematurity per se.6,11,12
The rate of PICU admissions in one of the participating centers was 5% (20/404). Although the national rate is unavailable, this figure probably reflects the general status. Previous studies have reported a higher rates of PICU admission, from 9.4% to 50%,6,11–14 but most of them were performed in an a priori high-risk population (premature, bronchopulmonary dysplasia, underlying lung diseases, etc). In 1 study of children younger than 6 months, born at 36 gestational weeks or earlier, 6.3% of the hospitalized children were admitted to the PICU.12 Because this population was slightly higher risk than ours, the ICU admission rates are comparable.
The mortality rate of RSV bronchiolitis remains unclear because of the low absolute number of patients and the disparity of medical characteristics among the studies, including ours. Notably, although the 3 children in our sample without congenital heart disease meet the AAP criteria, at least in 2 of them, death was probably not directly caused by RSV infection.
In 1997, the PREVENT study5 reported that the use of intravenous RSV immune globulin (Respigam) in preterm infants with or without bronchopulmonary dysplasia leads to a 41% reduction in RSV-related hospitalizations and a 53% reduction in the total number of hospitalization days, in addition to a reduction in oxygen requirement and in severity of the lower respiratory tract illness.4,5 However, cyanotic children who received RSV-IVIG and underwent cardiac surgery appeared to have an increased surgical mortality rate.15 The Impact study6 tested the use of a humanized monoclonal antibody preparation against the F glycoprotein RSV (palivizumab) and noted similar results—a 55% reduction in RSV-related hospitalizations (10.6% vs 4.8%), in addition to a reduction in number of hospitalization days, oxygen requirement, and severity of disease. Children with congenital heart disease were excluded from the study. Overall, 3% of the placebo-treated children were admitted to the ICU in contrast to 1.3% of the palivizumab group (P = .026). These results seem to contradict our data, but in fact they also demonstrate the low frequency of PICU admissions attributable to RSV bronchiolitis, even in a cohort of premature children. As the impact of this group on the general pediatric population is small because of their reduced absolute numbers, their impact on general morbidity will also be small.
Would the implementation of the AAP guidelines in our population reduce the rate of PICU admissions, the need for mechanical ventilation, and mortality? Our results indicate that the introduction of a prophylaxis program before the RSV season would have only a marginal impact on PICU admissions and need for mechanical ventilation. Following its use in the high-risk population, as proposed by the AAP, we could expect no significant changes in total number or rate of PICU admissions. Accordingly, neither the PREVENT5 nor the Impact study6 showed a reduction either in days spent in the PICU or in the need for mechanical ventilation in the infants receiving prophylaxis, and neither study showed a reduction in mortality from RSV in these infants. Therefore, RSV prophylaxis would apparently not significantly reduce the rate and number of the most severely affected bronchiolitis patients. The high cost of the RSV prophylaxis program together with its relatively low impact on the severe morbidity of the whole population should be carefully considered. Some of the more recent studies8–11,12,16 have concluded that RSV prophylaxis in the available forms would increase the net cost of care if the population were vaccinated according to the AAP recommendations. However, the RSV hospitalization rate and the cost-effectiveness of prophylaxis vary markedly by subgroup, with the infants who are very premature, who require prolonged respiratory support, and who require oxygen supplementation at home constituting the best candidates for RSV prophylaxis.
Children with congenital heart disease are a target group warranting further study with regard to efficacy and safety of this preventive program, and recently, one large unpublished control trial has yielded positive results (H. Cody Meissner, 2000, personal communication). Other populations should probably be included in the prophylaxis program, such as infants with congenital heart disease, CLD unrelated to prematurity, pulmonary hypertension, interstitial lung disease, neuromuscular diseases, and immunocompromised infants. New risk-stratified guidelines for RSV prophylaxis are needed, particularly because of the restricted available resources.
Our study had several limitations: 1) despite the efforts, it is possible that some infants with bronchiolitis were missed. Nevertheless, this number is probably low and would not influence the results; 2) as pointed out, there were no national criteria for PICU admissions, and no uniformity. The indications for PICU admissions are mainly clinical and may have varied among the different centers and populations; 3) RSV strains in our population were not tested; and 4) finally, our study was conducted during only one RSV season because it was the last year before the introduction of RSV prophylaxis in Israel, but we know that data from previous seasons were similar to ours (Tommy Schonfeld, 2000, unpublished data).
Most of the infants with severe RSV bronchiolitis were born at term and did not have CLD. The great majority of infants admitted to the PICU for bronchiolitis were not candidates for RSV prophylaxis. The administration of RSV prophylaxis according to the current recommendations could be expected to yield no significant change in PICU admissions or number of infants needing mechanical ventilation. New risk-stratified guidelines for RSV prophylaxis are needed.
Members of the Israeli RSV Monitoring Group are: Jacob Amir, Schneider Children’s Medical Center of Israel, Petah Tiqva; Zohar Barzilay, The Chaim Sheba Medical Center, Tel Hashomer; Tamar Bernstein, Soroka Medical Center, Beer Sheva; Haim Bibi, Barzilai Medical Center, Ashkelon; Gideon Eshel, Assaf Harofeh Medical Center, Zerifin; Avihu Gazit, Western Galilee Hospital, Nahariya; Imhad Kasis, Rambam Medical Center, Haifa; David Kleid, Shaare Zedek Medical Center, Jerusalem; Alik Korneci, Dana Children’s Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv; Yosef Marzel, HaEmek Medical Center, Afula; Gideon Paret, Chaim Sheba Medical Center, Tel Hashomer; Dario Prais, Schneider Children’s Medical Center of Israel, Petah Tiqva; Tommy Schonfeld, Schneider Children’s Medical Center of Israel, Petah Tiqva; Yakov Sivan, Dana Children’s Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv; Shaul Sofer, Soroka Medical Center, Beer Sheva; Eli Somekh, Wolfson Hospital, Holon; Elvan Tabachnik, Kaplan Medical Center, Rehovot; Dana Wolf, Hadassah Medical Center, Jerusalem; Ido Yatziv, Hadassah Medical Center, Jerusalem; and Zeev Zonis, Western Galilee Hospital, Nahariya.
- Received November 25, 2002.
- Accepted March 10, 2003.
- Reprint requests to (J.A.) Department of Pediatrics C, Schneider Children’s Medical Center of Israel Petah Tiqva 49202, Israel. E-mail:
↵‖ The members of the Israeli RSV Monitoring Group are listed in the Appendix.
This work was presented in part at the Third World Congress of Pediatric Infectious Diseases, World Society of Pediatric Infectious Diseases; November 19–23, 2002; Santiago, Chile.
- ↵Groothuis JR, Gutierrez KM, Lauer BA. Respiratory syncytial virus infection in children with bronchopulmonary dysplasia. Pediatrics.1988;82 :199– 203
- ↵Cunningham CK, McMillan JA, Gross SJ. Rehospitalization for respiratory illness in infants less than 32 weeks’ gestation. Pediatrics.1991;88 :527– 532
- ↵The PREVENT Study Group. Reduction of respiratory syncytial virus hospitalization among premature infants and infants with bronchopulmonary dysplasia using respiratory syncytial virus immune globulin prophylaxis. Pediatrics.1997;99 :93– 99
- ↵The Impact-RSV Study Group. Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants. Pediatrics.1998;102 :531– 537
- ↵American Academy of Pediatrics. Prevention of respiratory syncytial virus infections: indications for the use of Palivizumab and update on the use of RSV-IGIV. Pediatrics.1998;102 :1211– 1216
- ↵Thomas M, Bedford-Russell A, Sharland M. Hospitalization for RSV infection in ex-preterm infants—implications for the use of RSV immune globulin. Arch Dis Child.2000;83 :122– 127
- ↵Swenson PD, Kaplan MH. Rapid detection of respiratory syncytial virus in nasopharyngeal aspirates by a commercial enzyme immunoassay. J Clin Microbiol.1986;23 :485– 488
- ↵Very Low Birth Weight Children in Israel. Annual National Report. Tel Hashomer, Israel: Israel Disease Control Center, Israel Neonatal Network; 2002
- ↵Clark SJ, Beresford MW, Subjedar NV, Shaw NJ. Respiratory syncytial virus infection in high-risk infants and the potential of prophylaxis in a United Kingdom cohort. Arch Dis Child.2000;83 :313– 316
- Wang EEL, Law BJ, Stephans D, and other members of PICNIC. Pediatric investigators collaborative network on infections in Canada (PICNIC) prospective study of risk factors and outcomes in patients hospitalized with respiratory syncytial viral lower respiratory tract infection. J Pediatr.1995;126 :212– 219
- Copyright © 2003 by the American Academy of Pediatrics