Original Studies
Increased Risk for Respiratory Syncytial Virus-associated, Community-acquired Alveolar Pneumonia in Infants Born at 31–36 Weeks of Gestation David Greenberg, MD,*† Ron Dagan, MD,*† Eilon Shany, MD,†‡ Jacob Bar-Ziv, MD,§ and Noga Givon-Lavi, PhD*† Background: We compared hospitalization and pediatric intensive care unit (PICU) admission rates for community-acquired alveolar pneumonia (CAAP) and respiratory syncytial virus (RSV)-associated CAAP (RSV-CAAP) in non-RSV–immunized children 36 weeks GA). Methods: Nasopharyngeal samples for RSV were obtained prospectively (2004–2011) during RSV season, from hospitalized children with radiographic-diagnosed CAAP. Soroka University Medical Center is the only hospital in the region, enabling population-based rate calculation. Relative risks (RR) and 95% confidence intervals (95% CI) were calculated comparing RSV-CAAP incidence in 31–36 weeks GA with >36 weeks GA children. Results: CAAP hospitalization incidences (per 1000 population) were 23.6 and 9.4 (RR: 2.52; 95% CI: 2.13–2.68), respectively; the respective incidences of PICU admission for overall CAAP were 1.8 and 0.2 (RR: 7.88; 95% CI: 4.59–11.83). The RRs (and 95% CI) for RSV-CAAP hospitalizations and PICU admission rates were (after extrapolation) 15.2 and 5.8 (RR: 2.79; 95% CI: 2.31–3.06) and 1.1 and 0.1 (RR: 9.14; 95% CI: 4.93–16.96), respectively. In a multiregression analysis in patients with RSV-CAAP versus CAAP, 31–36 weeks GA was an independent risk factor for hospitalization (RR: 1.485; 95% CI: 1.03–2.14). Conclusions: Children 36 weeks GA. Moreover, in late premature children, RSV represented a 50% increased risk for CAAP compared with infants born at term. Key Words: respiratory syncytial virus, premature infants, communityacquired pneumonia, children. (Pediatr Infect Dis J 2014;33:381–386)
R
espiratory syncytial virus (RSV) is the most common cause of acute lower respiratory infection (ALRI) leading to hospitalization in infancy.1,2 It has been estimated that 33.8 million new episodes of RSV-associated ALRI (RSV-ALRI) occurred worldwide in 2005 with 3.4 million episodes of severe RSV-ALRI necessitating hospital admission and 66,000–199,000 deaths in children 150 million episodes of pneumonia occur yearly in children worldwide, causing 11–20 million hospitalizations accounting for approximately 2 million deaths.7,8 Alveolar pneumonia is often considered a bacterial infection and pneumococcus is the leading pathogen in all age groups.9 Viral pathogens, including RSV, are commonly detected in patients with community-acquired infections.10,11 Immunization with a pneumococcal conjugate vaccine resulted in a substantial reduction of virus-associated pneumonia, including pneumonia associated with RSV,12,13 implicating RSV as a frequent copathogen in childhood pneumonia. Understanding the epidemiology of pneumonia depends on the use of standardized pneumonia case definitions. The limitations of the current etiologic diagnostic approaches favor the use of a case definition that measures reduction in clinical pneumonia rather than bacteriologically or virally proven disease. In response, the World Health Organization (WHO) developed standardized radiological definitions for alveolar and non-alveolar pneumonia for use in epidemiological and vaccine trials.14 This tool allows comparisons of clinical and epidemiological characteristics between populations. Using the WHO radiological alveolar pneumonia definition as an endpoint for pneumococcal conjugated vaccine trials, the efficacy was higher than that in other endpoint definitions, suggesting that the role of Streptococcus pneumoniae in these cases is substantial.15,16 We hypothesized that moderately and late preterm infants are at higher risk of community-acquired alveolar pneumonia (CAAP) and RSV-CAAP. The aim of the present study was to determine the risk of hospitalization and pediatric intensive care unit (PICU) admission due to CAAP and RSV-CAAP in children 36 weeks GA.
PATIENTS AND METHODS Accepted for publication October 10, 2013. From the *Pediatric Infectious Disease Unit, Soroka University Medical Center; †Ben-Gurion University of the Negev; ‡Neonatal Department, Soroka University Medical Center, Beer-Sheva; and §Department of Radiology, Hadassah University Medical Center, Jerusalem, Israel. This study was supported by a grant from Abbott Ltd. David Greenberg is the recipient of a grant from and has been a speaker for Abbott Ltd. The authors have no other funding or conflicts of interest to disclose. Address for correspondence: David Greenberg, MD, The Pediatric Infectious Disease Unit, Soroka University Medical Center, Beer-Sheva, Israel. E-mail: [email protected]. Copyright © 2013 by Lippincott Williams & Wilkins ISSN: 0891-3668/14/3304-0381 DOI: 10.1097/INF.0000000000000130
Setting The Soroka University Medical Center (SUMC) is the only hospital in the Negev district of Southern Israel providing primary and referral health services to the entire population of the Southern region of Israel (>580,000 inhabitants in 2008).17 Over 95% of the children living in the region are served by the SUMC-enabling incidence figures calculations. Two ethnic populations reside side by side in Southern Israel: a Bedouin Muslim population, in transition from a seminomadic to urban lifestyle, resembling a developing population and a Jewish population whose lifestyle is similar to a developed population.17,18 Annual birth rate from 2004 to 2011 was 6816–7796 for the Bedouin population and 7502–8049 for
The Pediatric Infectious Disease Journal • Volume 33, Number 4, April 2014
www.pidj.com | 381
Greenberg et al
The Pediatric Infectious Disease Journal • Volume 33, Number 4, April 2014
the Jewish population.17 Hospitalization rates for respiratory infectious diseases and especially for community-acquired pneumonia are higher among the Bedouin population.18,19 As medical insurance for children in Israel is universal and is free of charge, there are no financial barriers for healthcare service use in the region. The study was approved by the Institutional Ethics Committees of the SUMC. Monoclonal anti-RSV–specific immunoglobulin (Palivizumab [Synagis®; MedImmune, Gaithersburg, MD])immunization in Israel is reimbursed and widely used for premature babies born at 95% effectiveness.20 The 7-valent pneumococcal conjugated vaccine was introduced to the national immunization program in July 2009 and replaced by the 13-valent pneumococcal conjugated vaccine in December 2010. Therefore, most children in this study had not received any pneumococcal vaccination. In addition, no significant reduction in alveolar pneumonia in children 36 weeks GA were obtained from the Central Bureau of Statistics.17 All children admitted to the SUMC pediatric emergency room, pediatric wards and PICU diagnosed with alveolar pneumonia were included in the study. The study was conducted all year round but because RSV season is only from November to March,11 analysis of cases was performed only during the RSV season. There were no changes with regard to policy, technique of nasopharyngeal sampling or chest X-ray procedure during the study period.
Case Definition A patient was enrolled into the study during the aforementioned period if all of the following criteria were fulfilled and the patient: (1) was 36 weeks GA is equal to the relative birth rate ratio of these 2 groups. Contingency table analyses for comparing rates between unmatched samples were performed using the 2-tailed χ2 tests or Fisher’s exact test as appropriate. Student’s independent samples t-tests were used to compare continuous variables. Annual CAAP incidences were calculated per 1000 children at risk. Relative risks (RR) and their 95% confidence intervals (CIs) were calculated using the Mantel-Haenszel–weighted RRs stratified analysis. Data were analyzed for the entire population group and also separately for each ethnic group (Bedouin and Jewish children) as well as for 3 GA groups: 31–32, 33–34 and 35–36 weeks. Data were analyzed for children enrolled during the RSV season (November to March). Because only 50% of all eligible children provided RSV NPW samples, extrapolation for the missing patients and specimens was needed to calculate incidence figures. We examined 3 options of extrapolation models for each gestational group: (1) all missing cases are considered RSV-positive cases; (2) all missing cases are considered RSV-negative cases and (3) missing cases are assumed to be equally distributed between enrolled children who provided samples and all other eligible children. All 3 analyses demonstrated similar results regarding the differences of risk to be hospitalized due to CAAP and RSV-associated CAAP in cases among different GA groups versus children born at term (>36 weeks GA). Thus, we chose to use the third approach in which the assumption was that the RSV-positive cases among eligible children without a sample were equally distributed compared with children who provided samples. Logistic regression models were used to evaluate potential risk factors, covariate and confounders. Variables implicated in the literature and those that were statistically significant at the level of P < 0.1 in the univariate analyses were included in the multivariate logistic regression models (underlying diseases, age, ethnicity, season at birth and gender). Because the © 2013 Lippincott Williams & Wilkins
The Pediatric Infectious Disease Journal • Volume 33, Number 4, April 2014
2 different ethnic (Jewish and Bedouin children) populations living in the region differ in their higher rates of respiratory infections and other morbidity rates, all RRs were adjusted for ethnicity.24 RRs (approximated from the odds ratio results)25 and 95% CIs were calculated to estimate risks of CAAP cases among different GA groups versus children born at term (>36 weeks GA).
RESULTS During the study period, 119,871 children were born in Southern Israel: 715 at 36 weeks GA. A total of 2240 children 36 weeks GA. Nasopharyngeal washes for RSV were obtained from 1058 of 2240 (47.2%) patients: 40 of 86 (46.5%) from children born at 36 weeks GA. The mean age (± standard deviation) at hospitalization was 7.2 ± 5.8 months for children from whom a nasopharyngeal wash was obtained versus 10.0 ± 6.7 months for children from whom a nasopharyngeal wash was not obtained (P < 0.001). Differences in GA, gender, ethnicity or mortality rates were not significantly different between these 2 groups (Table 1). Demographic data such as age, ethnic origin and gender were similar in patients born at 31–36 weeks GA versus patients born at > 36 weeks GA with CAAP and with RSV-CAAP from whom a nasopharyngeal wash was obtained (Table 2). Underlying diseases and PICU admissions were significantly more common in the 31–36 weeks GA group with CAAP and with RSV-CAAP than in the >36 weeks GA group (Table 2). Incidences of hospitalization for CAAP (per 1000 population 36 weeks GA were 23.6 and 9.4, respectively (RR: 2.52; 95% CI: 2.13–2.68; Fig. 1, Table 3). The respective incidences of PICU admission for overall CAAP were 1.8 and 0.2 (RR: 7.88; 95% CI: 4.59–11.83). The incidence of RSV-CAAP (after extrapolation) was 15.2 and 5.8 (RR: 2.79; 95% CI: 2.31–3.06) for children born at 31– 36 weeks GA and those born at >36 weeks GA, respectively. The respective figures for PICU admission for RSV-CAAP were 1.1 and 0.1 (RR: 9.14; 95% CI: 4.93–16.96; Fig. 1, Table 3). As stated in the methods, all RRs were controlled for ethnicity.
RSV-associated Pneumonia
To examine whether younger GA was a risk factor within the group of children born at 31–36 weeks GA, we calculated RRs for hospitalization and for PICU admissions of GA groups 31–32, 33–34 and 35–36 weeks, in relation to those born at >36 weeks GA (Fig. 1, Table 4). All were significantly higher compared with infants born at >36 weeks GA. Independent risk factors for hospitalization due to RSV-CAAP in children
Increased Risk for Respiratory Syncytial Virus-associated, Community-acquired Alveolar Pneumonia in Infants Born at 31–36 Weeks of Gestation David Greenberg, MD,*† Ron Dagan, MD,*† Eilon Shany, MD,†‡ Jacob Bar-Ziv, MD,§ and Noga Givon-Lavi, PhD*† Background: We compared hospitalization and pediatric intensive care unit (PICU) admission rates for community-acquired alveolar pneumonia (CAAP) and respiratory syncytial virus (RSV)-associated CAAP (RSV-CAAP) in non-RSV–immunized children 36 weeks GA). Methods: Nasopharyngeal samples for RSV were obtained prospectively (2004–2011) during RSV season, from hospitalized children with radiographic-diagnosed CAAP. Soroka University Medical Center is the only hospital in the region, enabling population-based rate calculation. Relative risks (RR) and 95% confidence intervals (95% CI) were calculated comparing RSV-CAAP incidence in 31–36 weeks GA with >36 weeks GA children. Results: CAAP hospitalization incidences (per 1000 population) were 23.6 and 9.4 (RR: 2.52; 95% CI: 2.13–2.68), respectively; the respective incidences of PICU admission for overall CAAP were 1.8 and 0.2 (RR: 7.88; 95% CI: 4.59–11.83). The RRs (and 95% CI) for RSV-CAAP hospitalizations and PICU admission rates were (after extrapolation) 15.2 and 5.8 (RR: 2.79; 95% CI: 2.31–3.06) and 1.1 and 0.1 (RR: 9.14; 95% CI: 4.93–16.96), respectively. In a multiregression analysis in patients with RSV-CAAP versus CAAP, 31–36 weeks GA was an independent risk factor for hospitalization (RR: 1.485; 95% CI: 1.03–2.14). Conclusions: Children 36 weeks GA. Moreover, in late premature children, RSV represented a 50% increased risk for CAAP compared with infants born at term. Key Words: respiratory syncytial virus, premature infants, communityacquired pneumonia, children. (Pediatr Infect Dis J 2014;33:381–386)
R
espiratory syncytial virus (RSV) is the most common cause of acute lower respiratory infection (ALRI) leading to hospitalization in infancy.1,2 It has been estimated that 33.8 million new episodes of RSV-associated ALRI (RSV-ALRI) occurred worldwide in 2005 with 3.4 million episodes of severe RSV-ALRI necessitating hospital admission and 66,000–199,000 deaths in children 150 million episodes of pneumonia occur yearly in children worldwide, causing 11–20 million hospitalizations accounting for approximately 2 million deaths.7,8 Alveolar pneumonia is often considered a bacterial infection and pneumococcus is the leading pathogen in all age groups.9 Viral pathogens, including RSV, are commonly detected in patients with community-acquired infections.10,11 Immunization with a pneumococcal conjugate vaccine resulted in a substantial reduction of virus-associated pneumonia, including pneumonia associated with RSV,12,13 implicating RSV as a frequent copathogen in childhood pneumonia. Understanding the epidemiology of pneumonia depends on the use of standardized pneumonia case definitions. The limitations of the current etiologic diagnostic approaches favor the use of a case definition that measures reduction in clinical pneumonia rather than bacteriologically or virally proven disease. In response, the World Health Organization (WHO) developed standardized radiological definitions for alveolar and non-alveolar pneumonia for use in epidemiological and vaccine trials.14 This tool allows comparisons of clinical and epidemiological characteristics between populations. Using the WHO radiological alveolar pneumonia definition as an endpoint for pneumococcal conjugated vaccine trials, the efficacy was higher than that in other endpoint definitions, suggesting that the role of Streptococcus pneumoniae in these cases is substantial.15,16 We hypothesized that moderately and late preterm infants are at higher risk of community-acquired alveolar pneumonia (CAAP) and RSV-CAAP. The aim of the present study was to determine the risk of hospitalization and pediatric intensive care unit (PICU) admission due to CAAP and RSV-CAAP in children 36 weeks GA.
PATIENTS AND METHODS Accepted for publication October 10, 2013. From the *Pediatric Infectious Disease Unit, Soroka University Medical Center; †Ben-Gurion University of the Negev; ‡Neonatal Department, Soroka University Medical Center, Beer-Sheva; and §Department of Radiology, Hadassah University Medical Center, Jerusalem, Israel. This study was supported by a grant from Abbott Ltd. David Greenberg is the recipient of a grant from and has been a speaker for Abbott Ltd. The authors have no other funding or conflicts of interest to disclose. Address for correspondence: David Greenberg, MD, The Pediatric Infectious Disease Unit, Soroka University Medical Center, Beer-Sheva, Israel. E-mail: [email protected]. Copyright © 2013 by Lippincott Williams & Wilkins ISSN: 0891-3668/14/3304-0381 DOI: 10.1097/INF.0000000000000130
Setting The Soroka University Medical Center (SUMC) is the only hospital in the Negev district of Southern Israel providing primary and referral health services to the entire population of the Southern region of Israel (>580,000 inhabitants in 2008).17 Over 95% of the children living in the region are served by the SUMC-enabling incidence figures calculations. Two ethnic populations reside side by side in Southern Israel: a Bedouin Muslim population, in transition from a seminomadic to urban lifestyle, resembling a developing population and a Jewish population whose lifestyle is similar to a developed population.17,18 Annual birth rate from 2004 to 2011 was 6816–7796 for the Bedouin population and 7502–8049 for
The Pediatric Infectious Disease Journal • Volume 33, Number 4, April 2014
www.pidj.com | 381
Greenberg et al
The Pediatric Infectious Disease Journal • Volume 33, Number 4, April 2014
the Jewish population.17 Hospitalization rates for respiratory infectious diseases and especially for community-acquired pneumonia are higher among the Bedouin population.18,19 As medical insurance for children in Israel is universal and is free of charge, there are no financial barriers for healthcare service use in the region. The study was approved by the Institutional Ethics Committees of the SUMC. Monoclonal anti-RSV–specific immunoglobulin (Palivizumab [Synagis®; MedImmune, Gaithersburg, MD])immunization in Israel is reimbursed and widely used for premature babies born at 95% effectiveness.20 The 7-valent pneumococcal conjugated vaccine was introduced to the national immunization program in July 2009 and replaced by the 13-valent pneumococcal conjugated vaccine in December 2010. Therefore, most children in this study had not received any pneumococcal vaccination. In addition, no significant reduction in alveolar pneumonia in children 36 weeks GA were obtained from the Central Bureau of Statistics.17 All children admitted to the SUMC pediatric emergency room, pediatric wards and PICU diagnosed with alveolar pneumonia were included in the study. The study was conducted all year round but because RSV season is only from November to March,11 analysis of cases was performed only during the RSV season. There were no changes with regard to policy, technique of nasopharyngeal sampling or chest X-ray procedure during the study period.
Case Definition A patient was enrolled into the study during the aforementioned period if all of the following criteria were fulfilled and the patient: (1) was 36 weeks GA is equal to the relative birth rate ratio of these 2 groups. Contingency table analyses for comparing rates between unmatched samples were performed using the 2-tailed χ2 tests or Fisher’s exact test as appropriate. Student’s independent samples t-tests were used to compare continuous variables. Annual CAAP incidences were calculated per 1000 children at risk. Relative risks (RR) and their 95% confidence intervals (CIs) were calculated using the Mantel-Haenszel–weighted RRs stratified analysis. Data were analyzed for the entire population group and also separately for each ethnic group (Bedouin and Jewish children) as well as for 3 GA groups: 31–32, 33–34 and 35–36 weeks. Data were analyzed for children enrolled during the RSV season (November to March). Because only 50% of all eligible children provided RSV NPW samples, extrapolation for the missing patients and specimens was needed to calculate incidence figures. We examined 3 options of extrapolation models for each gestational group: (1) all missing cases are considered RSV-positive cases; (2) all missing cases are considered RSV-negative cases and (3) missing cases are assumed to be equally distributed between enrolled children who provided samples and all other eligible children. All 3 analyses demonstrated similar results regarding the differences of risk to be hospitalized due to CAAP and RSV-associated CAAP in cases among different GA groups versus children born at term (>36 weeks GA). Thus, we chose to use the third approach in which the assumption was that the RSV-positive cases among eligible children without a sample were equally distributed compared with children who provided samples. Logistic regression models were used to evaluate potential risk factors, covariate and confounders. Variables implicated in the literature and those that were statistically significant at the level of P < 0.1 in the univariate analyses were included in the multivariate logistic regression models (underlying diseases, age, ethnicity, season at birth and gender). Because the © 2013 Lippincott Williams & Wilkins
The Pediatric Infectious Disease Journal • Volume 33, Number 4, April 2014
2 different ethnic (Jewish and Bedouin children) populations living in the region differ in their higher rates of respiratory infections and other morbidity rates, all RRs were adjusted for ethnicity.24 RRs (approximated from the odds ratio results)25 and 95% CIs were calculated to estimate risks of CAAP cases among different GA groups versus children born at term (>36 weeks GA).
RESULTS During the study period, 119,871 children were born in Southern Israel: 715 at 36 weeks GA. A total of 2240 children 36 weeks GA. Nasopharyngeal washes for RSV were obtained from 1058 of 2240 (47.2%) patients: 40 of 86 (46.5%) from children born at 36 weeks GA. The mean age (± standard deviation) at hospitalization was 7.2 ± 5.8 months for children from whom a nasopharyngeal wash was obtained versus 10.0 ± 6.7 months for children from whom a nasopharyngeal wash was not obtained (P < 0.001). Differences in GA, gender, ethnicity or mortality rates were not significantly different between these 2 groups (Table 1). Demographic data such as age, ethnic origin and gender were similar in patients born at 31–36 weeks GA versus patients born at > 36 weeks GA with CAAP and with RSV-CAAP from whom a nasopharyngeal wash was obtained (Table 2). Underlying diseases and PICU admissions were significantly more common in the 31–36 weeks GA group with CAAP and with RSV-CAAP than in the >36 weeks GA group (Table 2). Incidences of hospitalization for CAAP (per 1000 population 36 weeks GA were 23.6 and 9.4, respectively (RR: 2.52; 95% CI: 2.13–2.68; Fig. 1, Table 3). The respective incidences of PICU admission for overall CAAP were 1.8 and 0.2 (RR: 7.88; 95% CI: 4.59–11.83). The incidence of RSV-CAAP (after extrapolation) was 15.2 and 5.8 (RR: 2.79; 95% CI: 2.31–3.06) for children born at 31– 36 weeks GA and those born at >36 weeks GA, respectively. The respective figures for PICU admission for RSV-CAAP were 1.1 and 0.1 (RR: 9.14; 95% CI: 4.93–16.96; Fig. 1, Table 3). As stated in the methods, all RRs were controlled for ethnicity.
RSV-associated Pneumonia
To examine whether younger GA was a risk factor within the group of children born at 31–36 weeks GA, we calculated RRs for hospitalization and for PICU admissions of GA groups 31–32, 33–34 and 35–36 weeks, in relation to those born at >36 weeks GA (Fig. 1, Table 4). All were significantly higher compared with infants born at >36 weeks GA. Independent risk factors for hospitalization due to RSV-CAAP in children
