Updated Guidance for Palivizumab Prophylaxis Among Infants and Young Children at Increased Risk of Hospitalization for Respiratory Syncytial Virus Infection

Burden of RSV Disease

In the United States, RSV remains an important cause of hospitalization in the first months of life. Retrospective analyses using national databases and International Classification of Diseases, Ninth Revision discharge diagnoses have shown considerable variation in estimates of annual hospitalization rates attributable to RSV for infants.^30–33 More recent prospective population-based studies of laboratory confirmed cases demonstrate that RSV hospitalization rates are approximately half the rates reported in retrospective studies, yet rates remain high.^14,34–36

It is estimated that approximately 2.1 million children younger than 5 years require medical care as inpatients or as outpatients for RSV infection annually. Among outpatients, 60% (1.3 million) are 2 through 5 years of age, and the remaining 40% are between birth and 2 years of age.³⁵ Approximately 25% of RSV-infected children younger than 5 years are assessed and treated in an emergency department, and approximately 70% are assessed and treated in a pediatric office. It is estimated that each year, nearly 58 000 children in the first few years of life are hospitalized because of RSV infection.³⁵ Infants in the second month after birth have the highest RSV hospitalization rate, a rate that is almost twice that of the next highest risk group (infants in the first month after birth).³⁴

Preterm Infants Without CLD

In 2012, 3.95 million births were reported in the United States. Preterm infants (singleton and multiple births) born at less than 37 weeks’ gestation represented 11.6% of all births.³⁷ Preterm births of infants at less than 28 weeks’ gestation accounted for 0.7% of the annual birth cohort.³⁷ Infants born from 32 weeks to 35 weeks’ gestation represented approximately 9% of the birth cohort.

Beginning with the first AAP statement in 1998,⁷ the high cost of palivizumab influenced recommendations for use by the COID, leading to attempts to identify risk factors for RSV hospitalization among the large number of moderately preterm infants. The New Vaccine Surveillance Network (NVSN), sponsored by the Centers for Disease Control and Prevention (CDC) was a prospective population-based surveillance program from 3 geographically diverse locations in the United States for young children hospitalized with laboratory-confirmed RSV respiratory illness. Several studies were published summarizing data from the NVSN. One conducted during the RSV seasons from 2000 through 2005 using multiple logistic-regression analyses of data revealed that some of the previously reported potential risk factors, including siblings in the household and child care attendance, were not associated with a significantly increased risk of RSV hospitalization.³⁴ In this study, only young chronologic age was significantly correlated with risk of hospitalization for RSV illness. The association between preterm birth and increased risk of severe illness was not specific for RSV.³⁵

In addition, data from the NVSN study revealed that for all preterm infants (<37 weeks’ gestation), the RSV hospitalization rate was 4.6/1000 children, which was not significantly different from the hospitalization rate for term infants, which was 5.3/1000 children (Table 1).³⁴ Rates were derived from 132 085 children born during the study period, among whom 2149 were hospitalized with acute respiratory illness, and 559 of the hospitalized children had laboratory-confirmed RSV (Table 1). Infants born at <30 weeks’ gestation experienced a higher RSV hospitalization rate (18.7/1000 children) than early preterm infants (30–33 weeks),³⁴ although the small number of infants born before 30 weeks’ gestation limits the generalizability of this data. Late preterm infants were hospitalized significantly less often than term infants for RSV infection.³⁴

An analysis of Tennessee Medicaid data for children younger than 3 years conducted from July 1989 to June 1993 (preimmunoprophylaxis era) included 248 652 child-years of follow-up. The retrospective cohort analysis was conducted to determine RSV hospitalization rates among infants with different degrees of prematurity and other comorbidities.³⁸ Within each age group, preterm infants had similar rates of RSV hospitalization, regardless of the degree of prematurity (Table 2). In this analysis, preterm infants with CLD were categorized separately to reduce confounding.

A historical cohort analysis from Rochester, New York, reported on RSV hospitalization rates among 1029 consecutive preterm infants born before or at 32 weeks’ gestation during a 5-year period.³⁹ The RSV hospitalization rate increased with decreasing gestational age, with a breakpoint at ≤28 weeks’ gestation (Table 3). Among infants born at or before 26 weeks’ gestation, the risk of RSV-associated hospitalization was 13.9% vs 4.4% among children hospitalized at >30 to 32 weeks’ gestation. There was not a statistically significant difference in RSV hospitalization rates between infants with gestational ages of >28 to 30 weeks and infants with gestational ages of >30 to 32 weeks.

A retrospective cohort study of infants enrolled in Medicaid in Texas and Florida between 1999 and 2004 examined RSV hospitalization rates in moderately preterm infants 32 to 34 weeks’ gestation⁴⁰ Less than 20% of each cohort received palivizumab prophylaxis. In Florida, 71 (3.1%) of the moderately preterm infants were hospitalized compared with 1246 (1.5%) of term infants, and in Texas 164 (4.5%) of the preterm infants were hospitalized compared with 3815 (2.5%) of term infants. Palivizumab prophylaxis was associated with decreased hospitalization in moderately preterm infants in Texas but not in Florida. The risk of RSV hospitalization in moderately preterm infants was similar to 1-month-old term infants by 4.2 months in Florida (95% confidence interval [CI], 2.5–5.7) and by 4.5 months in Texas (95% CI, 2.8–6.4).⁴⁰

Choosing an appropriate cutoff for gestational age for which palivizumab prophylaxis may be considered for preterm infants without other indications is challenging. Data consistently demonstrate the greatest increase in risk for hospitalization is in preterm infants born before 29 weeks’ gestation. These infants have hospitalization rates 2 to 4 times higher than later preterm infants (Tables 1, 2, and 3). The consensus of the COID and the Bronchiolitis Guidelines Committee is that palivizumab prophylaxis may be considered for infants, without other indications, whose gestational age is less than 29 weeks (28 weeks, 6 days or fewer). The available data do not support universal recommendations for palivizumab prophylaxis for preterm infants born at or after 29 weeks’ gestation.

Data regarding the risk of RSV hospitalization for most preterm infants do not support a benefit from prophylaxis. In recent large cohort studies of moderately preterm infants, the majority of whom did not receive palivizumab, 2.5% to 4.9% required hospitalization for RSV infection during the RSV season indicating that more than 95% did not require hospitalization.⁴⁰ The rate of hospitalization among infants ≥35 weeks’ gestation (5.1/1000) was no different than the rate for term infants (5.3/1000; Table 1). The hospitalization rate of infants ≥30 weeks to 35 weeks’ gestation indicate only a slight increase in risk (less than twofold; Tables 1, 2, and 3). Data concerning host or environmental risk factors for hospitalization in preterm infants without CLD or CHD are inconsistent, with the exception of age younger than 3 months at the start of the RSV season, which has been associated with an increased risk of hospitalization.³⁴

For all infants, particularly those who are preterm, the environment should be optimized to prevent RSV and other viral respiratory infections by offering breast milk feeds, immunizing household contacts with influenza vaccine, practicing hand and cough hygiene, and by avoiding tobacco or other smoke exposure and attendance in large group child care during the first winter season, whenever possible.

Preterm Infants With CLD

Studies have documented that infants and young children with CLD have increased rates of RSV hospitalization.^38,39 Results from the IMpact-RSV trial evaluating all preterm infants with CLD (n = 762 randomized preterm infants) demonstrated that the RSV hospitalization rate among placebo recipients was 12.8% and 7.9% among palivizumab recipients (P = .038).³

Infants With Hemodynamically Significant CHD

The results of an industry-funded multicenter, randomized, double-blind, placebo-controlled trial of palivizumab prophylaxis among 1287 children (639 palivizumab recipients, 648 placebo recipients) younger than 24 months with hemodynamically significant CHD was published in 2003.⁴ Results from this study demonstrated a reduction in RSV hospitalization rates of 4.4% (9.7% among placebo recipients and 5.3% among palivizumab recipients; P = .003).⁴ The reduction in the number of RSV hospitalizations between the 2 study groups was 29 fewer RSV hospitalizations among palivizumab recipients during the 4 years of the study.⁴ Prophylaxis with palivizumab appeared to have less benefit among cyanotic children than among acyanotic children. Among children in the cyanotic group, there were 23 fewer RSV hospitalizations per 1000 palivizumab recipients (7.9% vs 5.6%, P = .285). Among children in the acyanotic group, there were 68 fewer RSV hospitalizations per 1000 prophylaxis recipients (11.8% vs 5.0%; P = .003). Despite enrolling 1287 subjects, the trial did not have sufficient power to detect statistically significant differences among subgroups of children with different cardiac lesions.

A retrospective analysis of the effect of palivizumab prophylaxis on RSV hospitalizations among children with hemodynamically significant CHD was conducted in California. The authors estimated a statewide 19% reduction in RSV hospitalization between 2000 and 2002 (preprophylaxis era) and 2004 and 2006 (prophylaxis era) after the licensure of palivizumab for children with CHD. The authors concluded that in the state of California, 7 fewer RSV hospitalizations per year occurred among children younger than 2 years with hemodynamically significant CHD following recommendations for palivizumab prophylaxis in this group.⁴¹

Other investigators describe rates of RSV hospitalizations among patients with hemodynamically significant CHD (2%–3%) who do not receive prophylaxis as lower than the 9.7% rate reported in the placebo arm of the cardiac study.^42–46 As the rate of RSV hospitalization decreases among children who do not receive prophylaxis, the cost to prevent 1 hospitalization with prophylaxis increases.

A retrospective analysis of children younger than 3 years (248 652 child-years) in the Tennessee Medicaid program revealed that the RSV hospitalization rate for children with CHD in the second year of life (18.2/1000) was less than half the hospitalization rate for low-risk infants in the first 5 months after birth (44.1/1000), a group for whom palivizumab prophylaxis is not recommended³⁸ (Table 2). Thus, prophylaxis is not recommended during the second year of life.

Children With Anatomic Pulmonary Abnormalities or Neuromuscular Disorder

The risk of RSV hospitalization is not well defined in children with neuromuscular disorders that impair the ability to clear secretions from the upper airway because of ineffective cough, recurrent gastroesophageal tract reflux, pulmonary malformations, tracheoesophageal fistula, upper airway conditions, or conditions requiring tracheostomy.

Studies suggest children and infants with neuromuscular disease who are hospitalized with RSV infection tend to be older compared with other groups of patients hospitalized with RSV infection and are more likely to have preexisting immunity to RSV.^47,48 This may reflect the progressive nature of neuromuscular disease, with susceptibility to respiratory disease tract increasing with age.

Immunocompromised Children

Population-based data are not available on the incidence or severity of RSV disease among children who receive solid organ transplants (SOTs) or hematopoietic stem cell transplants (HSCTs), children who receive chemotherapy, or children who are immunocompromised because of other conditions. Progression of RSV infection from upper to lower respiratory tract disease depends on the virulence of the RSV strain, as well as specific abnormalities in the immunocompromised host’s immune response attributable to the underlying disease or to chemotherapy.

RSV infection in immunocompromised children and adults can progress to respiratory failure and death.^49,50 Lymphopenia has been recognized as a risk factor for disease progression in several studies of immunocompromised patients. One study in adults noted that progression of RSV to lower respiratory tract disease did not occur in patients with a lymphocyte count greater than 1000 cells/mm³ at time of onset of upper respiratory tract infection.⁵¹ An absolute lymphocyte count of 100 cells/mm³ or less at the time of RSV upper tract infection was associated with progression to lower respiratory tract disease. In contrast to lymphopenia, analysis of antibody concentration in these adult HSCT recipients indicated no correlation between preexisting anti-RSV antibody concentration and progression from upper to lower respiratory tract disease.⁵¹

A retrospective report from 1 institution described 58 immunocompromised children with RSV infection between 1997 and 2005.⁵² Sixty-five percent of the RSV-infected children were managed as outpatients. No deaths occurred among 28 children infected with RSV who were receiving chemotherapy for acute lymphoblastic leukemia or among 11 immunosuppressed SOT recipients. Five of 58 patients with lower respiratory tract infection died (8.6%), including 4 children who were allogeneic HSCT recipients and 1 child with severe combined immune deficiency. One of the deaths occurred in a 10-year-old child, and a second death occurred in a child with Aspergillus species coinfection. Profound lymphopenia (<100 cells/mm³) was associated with progression to lower respiratory tract disease.⁵²

Another retrospective report from 1 institution noted 5 deaths among 117 RSV-infected, immunocompromised patients between 2006 and 2011 (2 with severe combined immunodeficiency, 1 with uncharacterized immunodeficiency, 1 with chronic granulomatous disease, and 1 SOT recipient).⁵³ The deaths occurred in children who presented with community-acquired RSV lower respiratory tract infection. No deaths occurred among children who were HSCT recipients or those with leukemia or lymphoma.

A third retrospective review of RSV infections in children with cancer was conducted from 1998 to 2009.⁵⁴ Among 57 patients, 37% experienced progression to lower respiratory tract disease. Three patients died of respiratory failure within 60 days of RSV diagnosis (1 had concomitant bacteremia and fungemia and 1 had concomitant herpes simplex pneumonia).

In a review of 208 viral respiratory infections among 166 patients over a 13-year period who received HSCTs, SOTs, or chemotherapy for malignancy, RSV infection accounted for 43% of the infections.⁵⁵ The mean and median ages of patients at the time of infection were 6.1 years and 4.3 years, with a range of 2 months to 21 years of age. Death occurred in 17 (8%) of patients with viral respiratory infection, including 6 of 88 (7%) RSV-infected children who received allogeneic HSCTs or SOTs. No infection resulted in death among patients who received chemotherapy, despite being severely immunosuppressed. Mortality and morbidity did not have a statistically significant correlation with the degree of immune suppression.

Risk factors for a poor outcome after RSV infection in an immunosuppressed host include age younger than 2 years, presence of lower respiratory tract symptoms at presentation (particularly in the absence of symptoms of upper respiratory infection), corticosteroid therapy, and varying degrees of lymphopenia. Underlying diagnosis, degree of immune suppression, RSV load in bronchoalveolar lavage, or specific humoral immunity to RSV have not been found to correlate with outcome.⁵⁶ This inability to correlate degree of immunosuppression with disease severity indicates an incomplete understanding of the immune response to viral respiratory infections in an immunocompromised host.

Antibody-based treatments, including immune globulin and palivizumab, have not been associated with improved outcome in HSCT recipients.⁵⁷ No data are available to suggest benefit from immunoprophylaxis among immunocompromised patients, and practices vary nationwide.^58,59 Further research is required before definitive recommendations can be made for the use of palivizumab in this heterogeneous group of children.

Children With Down Syndrome

Several factors appear to place children with Down syndrome at increased risk of RSV lower respiratory tract disease than children without Down syndrome.^18,19,60,61 CHD, with or without pulmonary hypertension, occurs in approximately 45% of children with Down syndrome, and lesions include atrioventricular canal, ventricular septal defect, patent ductus arteriosus, and tetralogy of Fallot. Anatomic abnormalities of the upper or lower respiratory tract, muscle dystonia, and intrinsic immune dysfunction may contribute to viral respiratory disease in this population.

One population-based cohort study over an 11-year period in Colorado revealed a statewide total of 85 RSV hospitalizations among 680 children with Down syndrome during their first 2 years of life. Concurrent risk factors were present in 35 of the 85 (41%) hospitalized children, indicating they would have qualified for prophylaxis for other reasons. RSV hospitalization rates were 67/1000 child-years for children with Down syndrome and other risk factors, 42/1000 child-years for children with Down syndrome without cardiopulmonary disease, and 12/1000 child-years for children in a control group.⁶¹ These data suggest an estimated overall 7.7 RSV admissions per year (85 admissions per 11 years) in the state of Colorado for children with Down syndrome. Using these figures, 4.6 RSV hospitalizations per year occur among children with Down syndrome without concurrent factors (50 admissions per 11 years) in Colorado. Assuming a 55% reduction in hospitalization, approximately 2 to 3 hospitalizations per year might have been avoided from prophylaxis administered to 680 children. Although children with Down syndrome were more likely to experience a temperature >38°C, the median duration of stay for children younger than 1 year of age with Down syndrome was 4 days, and for children without Down syndrome the median was 3 days. No deaths were reported in this study.⁶¹ These data suggest children with Down syndrome have a slightly higher hospitalization rate, but the absolute number of RSV hospitalizations is small, and a number of children with Down syndrome are at increased risk because of qualifying heart disease or other factors.

Another report described 39 of 395 (9.9%) children with Down syndrome hospitalized because of RSV infection in the first 2 years of life.⁶⁰ Among hospitalized children, 38% had hemodynamically significant heart disease. This study had insufficient power to differentiate among subgroups, meaning the increased RSV hospitalization rate may have been explained by concurrent risk factors and not Down syndrome.⁶⁰ Another study of 41 children with Down syndrome hospitalized with RSV infection noted that 51% had underlying CHD.¹⁸ An additional report of 222 children with Down syndrome hospitalized with RSV infection noted the mean age of hospitalized children (1.3 years; range, 0–6.1 years) was significantly older than the age of children hospitalized with RSV who did not have Down syndrome. A similar finding was noted in the Colorado report, with a mean age of 9.6 months at admission for RSV infected patients with Down syndrome and no other risk factors. In this study, the age range for hospitalization extended through 17 years.⁶¹ RSV prophylaxis for the first year of life would have limited effect on RSV hospitalization for children with Down syndrome without other risk factors for RSV.^19,61

Children With Cystic Fibrosis

Available studies indicate the incidence of RSV hospitalization in children with cystic fibrosis is uncommon and unlikely to be different from children without cystic fibrosis. Evidence to support a benefit from palivizumab prophylaxis in patients with cystic fibrosis is not available.^20,62,63

A randomized clinical trial with palivizumab prophylaxis included 186 children with cystic fibrosis from 40 centers. One subject in the untreated group and 1 subject in the palivizumab group were hospitalized for RSV infection.⁶⁴ Although this study was not powered for efficacy, no clinically meaningful differences in outcome between the 2 groups were reported. At the 12-month follow-up, there was no significant difference between the treated and untreated groups in number of Pseudomonas colonizations or change in weight-to height ratio. A case-control study of palivizumab in 75 children with cystic fibrosis noted a possible trend toward a potential clinical benefit of palivizumab prophylaxis, but the difference was not statistically significant.⁶²

A large study of RSV hospitalizations occurring between 1997 and 2003 in Danish children with chronic medical conditions identified 72 children with cystic fibrosis.⁶⁵ There were 13 RSV-related hospitalizations, which resulted in an adjusted incidence rate ratio for risk of RSV hospitalization of 4.32 (95% CI, 2.42–7.71). The geometric mean ratio for duration of RSV hospitalization in these children with cystic fibrosis was 1.3 days (95% CI, 0.81–2.11 days).

Two recent reviews^20,66 of RSV infection in infants with cystic fibrosis acknowledged that infants with cystic fibrosis may have a slightly increased risk for hospitalization with RSV. However, they both stated that there is insufficient evidence related to safety and efficacy in infants with cystic fibrosis to support a recommendation of palivizumab prophylaxis.^20,66 A survey of cystic fibrosis center directors published in 2008 noted that palivizumab prophylaxis is not the standard of care for patients with cystic fibrosis.⁶⁷

Discontinuation of Palivizumab Prophylaxis Among Children Who Experience Breakthrough RSV Hospitalization

RSV is classified into subgroups A and B, based on antigenic differences in the surface G glycoprotein. Subgroups are classified further into genotypes based on genetic analysis. The ability of RSV to cause reinfections throughout life likely is attributable both to strain variability and to an immune response that does not fully protect against subsequent infection. Reinfections with both heterologous and homologous strains occur. More than 1 RSV strain may circulate concurrently in a community. However, repeat RSV hospitalizations during 1 season are rare.

One study identified 726 RSV lower respiratory tract infections among 1560 children younger than 5 years over 8 successive RSV seasons in an outpatient setting.⁶⁸ Only 1 instance of repeat RSV infection occurred during the same season. Furthermore, it is well established that repeat RSV infections are associated with less severe clinical illness than the initial RSV infection.^69,70

In the blinded and randomized cardiac trial that involved 1287 children younger than 24 months with hemodynamically significant CHD, a total of 5 readmissions for a second RSV hospitalization occurred (a rate of less than 0.5%). Three of 648 children in the placebo group and 2 of 639 children who received palivizumab had more than 1 RSV hospitalization over 4 years.⁴

In the Dutch trial of 429 preterm infants randomly assigned to receive either palivizumab prophylaxis or placebo between April 2008 and December 2010, infants were followed for recurrent wheezing. No RSV reinfections were detected in either group, again indicating that repeat RSV infections in the same year seldom occur.¹⁶

Use of Palivizumab in the Second Year of Life

A prospective population-based surveillance study of 5067 children younger than 5 years evaluated 564 children hospitalized with laboratory-confirmed RSV infection. Among the children hospitalized with RSV infection, 75% were younger than 12 months. Less than 20% of all pediatric RSV hospitalizations occurred during the second year of life.³⁵

Limited safety data and no efficacy data are available regarding palivizumab prophylaxis in the second year of life.^71,72 Regardless of the presence or absence of comorbidities, RSV hospitalization rates decline during the second RSV season for all children.^34,38 In a retrospective cohort study conducted over 4 years and involving 248 652 child-years, RSV hospitalization rates in the second year of life for children with comorbidities were lower than the rate for healthy term infants in the first 12 months of life, a group for whom prophylaxis is not recommended³⁸ (Table 2).

Lack of Therapeutic Efficacy of Palivizumab

Controlled studies have demonstrated that monoclonal antibodies have no therapeutic benefit in the treatment of RSV infected children. One randomized study determined that intravenous palivizumab administered to RSV-infected, intubated infants reduced the RSV viral load in the lower respiratory tract but had no effect on RSV concentration in the upper respiratory tract.⁷³ Despite a reduction in viral load in the lower respiratory tract, no difference in disease severity was found between palivizumab recipients and placebo recipients. A phase 2 therapeutic trial involving 118 hospitalized, RSV infected infants evaluated the outcome among recipients of intravenous motavizumab at 30 mg/kg or at 100 mg/kg compared with placebo.⁷⁴ Motavizumab is an investigational monoclonal antibody with enhanced potency relative to palivizumab. No significant effect on RSV viral load in the upper respiratory tract was detected among motavizumab recipients. No difference in duration of hospitalization, requirement for supplemental oxygen, ICU admission, or need for mechanical ventilation between groups was detected.

Prevention of Health Care-Associated RSV Disease

Strict infection-control practices, including restriction of visitors to the neonatal ICU during respiratory virus season, will decrease health care-associated RSV disease. Evidence does not support the use of palivizumab among hospitalized preterm infants to prevent health care-associated spread of RSV.^75,76 If an RSV outbreak occurs in a high-risk unit (eg, pediatric or neonatal ICU or HSCT unit), primary emphasis should be placed on proper infection-control practices, especially hand hygiene. No rigorous data exist to support palivizumab use in controlling outbreaks of health care-associated disease. Further, hospitalization rates for RSV infection do not differ among infants who receive inpatient palivizumab prophylaxis while in the neonatal ICU compared with those who initiate prophylaxis at hospital discharge.⁷⁷

Risk Factors for RSV Hospitalization

Overall, approximately 2% to 3% of infants in the first 12 months of life are hospitalized with RSV infection each year in the United States. Children with certain comorbidities are at increased risk of severe RSV disease relative to children without these comorbidities.^35,38 Chronologic age is the single most important risk factor for RSV hospitalization on the basis of the observation that more than 58% to 64% of pediatric RSV hospitalizations occur in the first 5 months after birth.^34,35,38 Most of these hospitalizations occur in the first 90 days after birth.^22,34,40 Certain subgroups of infants with comorbidities such as prematurity, CLD, or hemodynamically significant CHD have increased risks for RSV hospitalization, although the degree of risk varies among studies.^30,78 The risk of RSV hospitalization associated with most other risk factors has been difficult to determine because of low rates of occurrence. Most host and environmental factors increase the risk for RSV hospitalization by only a small magnitude, so their contribution to overall disease burden is limited.⁷⁹ In addition, these factors are not identified consistently from study to study. These inconsistencies likely reflect variation in practice patterns in different countries, variation in living conditions and climate, variation in health care coverage, yearly variation in disease severity, and incompletely understood genetic factors. Differences in study design may contribute to these differences.

Reported host risk factors of limited (inconsistent) impact include the following: congenital malformations, congenital airway anomaly, neuromuscular impairment, birth weight, gender, lack of breastfeeding, duration of breastfeeding, cord serum anti-RSV antibody concentration, small for gestational age, Down syndrome, epilepsy, cord blood vitamin D concentration, family history of atopy, viral load, malnutrition, multiple births, and singletons versus multiple birth subjects. Environmental risk factors of limited (variable) impact include the following: environmental pollution, crowded living conditions, living at increased altitude, meteorological conditions, low parental education, low socioeconomic status, child care attendance, size of child care facility, month of birth, smoke exposure, maternal smoking during pregnancy, and proximity to hospital care.^{30,35,38,47,48,60,65,78,80–94} One publication suggested that malformations of the urinary tract increase the risk of RSV hospitalization.⁶⁵

Multiple logistic-regression analyses of data from the 4-year population-based prospective study revealed that of the evaluated risk factors (male gender, child care attendance, smoke exposure, lack of breastfeeding, and other children in the house), only preterm birth and young chronologic age independently correlated with more severe RSV disease after adjusting for other covariates.³⁵

RSV Seasonality

During the 6 RSV seasons from July 2007 to January 2013, the median duration of the RSV season ranged from 13 to 23 weeks, with median peak activity from mid-December to early February, with the exception of Florida and Alaska (see later discussions for each).^12,13 Within the 10 Health and Human Services Regions, in the few regions when the RSV season began in October, the season ended in March or early April. In regions where the RSV season began in November or December, the season ended by April or early May. Because 5 monthly doses of palivizumab at 15 mg/kg per dose will provide more than 6 months of serum palivizumab concentrations above the desired serum concentration for most infants, administration of more than 5 monthly doses is not recommended within the continental United States.¹¹ Children who qualify for 5 monthly doses of palivizumab prophylaxis should receive the first dose at the time of onset of the RSV season. For qualifying infants born during the RSV season, fewer than 5 doses will be needed to provide protection until the RSV season ends in their region (maximum of 5 doses).

A small number of sporadic RSV hospitalizations occur before or after the main season in many areas of the United States,^79,95 but maximum benefit from prophylaxis is derived during the peak of the season and not when the incidence of RSV hospitalization is low.

Prophylaxis for Alaska Native/American Indian Children and Timing of Palivizumab Initiation

Hospitalization rates for all causes of bronchiolitis as high as 484 to 590/1000 infants have been described in isolated Inuit populations.^96–99 Alaska Native infants in southwestern Alaska experience higher RSV hospitalization rates and a longer RSV season. On the basis of the epidemiology of RSV in Alaska, particularly in remote regions, the selection of infants eligible for prophylaxis may differ from the remainder of the United States. Clinicians may wish to use RSV surveillance data generated by the state of Alaska to assist in determining onset and end of the RSV season for appropriate timing of palivizumab administration.¹⁰⁰

Two published studies have documented a bronchiolitis hospitalization rate in Navajo populations that was 91.3 to 96.3/1000 infants younger than 1 year.^101,102 This rate was similar to those seen with high-risk groups, such as infants born preterm and those with CLD. There are no data on efficacy of palivizumab in this population. However, if local data support a high burden of RSV disease in select American Indian populations, selection of infants eligible for prophylaxis may differ from the remainder of the United States for infants in the first year of life.

Timing of Prophylaxis for the State of Florida

Variation in the onset and offset of the RSV season in different regions of Florida may affect the timing of palivizumab administration. Florida Department of Health data may be used to determine the appropriate timing for administration of the first dose of palivizumab for qualifying infants. Despite varying onset and offset dates of the RSV season in different regions of Florida, a maximum of 5 monthly doses of palivizumab will be adequate for qualifying infants for most RSV seasons in Florida. Even if the first of 5 monthly doses is administered in July, protective serum concentrations of palivizumab will be present for most infants and young children for at least 6 months and likely into February. More than 5 monthly doses are not recommended, despite the detection of a small number of cases of RSV infection outside this time window. A small number of sporadic RSV hospitalizations occur before or after the main season in many areas of the United States, but maximum benefit from prophylaxis is derived during the peak of the season and not when the incidence of RSV hospitalization is low.

Pharmacokinetics of Palivizumab

A threshold protective serum palivizumab concentration in humans has not been established. On the basis of studies of palivizumab prophylaxis with the cotton rat, serum concentrations of 25 to 30 mcg/mL produced a mean reduction in pulmonary RSV concentrations of 99% (2 log₁₀).^103,104 Because of the reliability of the cotton rat model to predict results in humans, this serum concentration became the target trough concentration in the randomized clinical trials.^3,4 The package label states “palivizumab serum concentrations of greater than or equal to 40 mcg/mL have been shown to reduce pulmonary RSV replication in the cotton rat model of RSV infection by 100 fold.”¹

Palivizumab pharmacokinetic data published by the manufacturer in 2012 demonstrate that after 5 monthly doses, serum concentrations of palivizumab remain at or above protective levels for most children for at least 6 months (>24 weeks).¹¹ There is seldom justification to administer more than 5 doses within the continental United States. These data were derived from a computer model based on 22 clinical trials.

Weight dosing at 15 mg/kg resulted in similar palivizumab concentrations in healthy term infants as well as preterm infants.¹¹

Race and Sex

Data from the NVSN demonstrate that the overall rate of RSV hospitalization does not differ between African American and white children younger than 24 months of age (5.4/1000 for African American children and 4.7/1000 for white children).³⁴ Among infants younger than 6 months, RSV hospitalization rates for African American and white children were not significantly different.³⁴ Another report from NVSN evaluated 564 children hospitalized with RSV infection. Neither race nor ethnic group was found to be an independent risk factor for hospitalization.³⁵ A third CDC report revealed rates of RSV coded illness in African American and white children to be similar at 5.3/1000 (95% CI, 3.7–6.9/1000) and 5.3/1000 (95% CI, 4.3–6.3/1000), respectively.³³ Hospitalization rates did not differ during the years of this study (1997–1999 and 2004–2006) by African American or white race. In this study, hospitalization rates for RSV-coded illness were 7.5/1000 boys (95% CI, 5.9–9.1) and 5.9/1000 girls (95% CI, 4.5–7.3).

Another population-based study evaluated racial disparities between African American and white children who were hospitalized because of RSV infection in 3 large United States counties.¹⁰⁵ No disparity was found in any county for any of the 7 years studied among children in the first 12 months of life or between infants 0 to 2 months or infants 3 to 5 months of age. Because of small numbers, reliable estimates for other race groups are not available.³³ RSV hospitalization rates among boys exceed that of girls in some, but not all studies.^34,35

Mortality Rates Among Hospitalized Children With RSV Infection

Using 2 national databases (the Pediatric Health Information System [PHIS] data for 2004–2011 and the Health Cost and Utilization Project Kids’ Inpatient Database [KID] for 2009), mortality rates associated with hospitalized infants with RSV infection were lower than previously estimated. The PHIS data set from 44 children’s hospitals identified 33 deaths per year (13.7 deaths/10 000 RSV admissions during the RSV season), but only 8.5 deaths/10 000 admissions were coded with RSV as the primary diagnosis, suggesting that other comorbidities were involved. In the KID data set from more than 4000 hospitals in 44 states, RSV was estimated to account for 121 deaths annually in the United States (9/10 000 RSV admissions) with 84 deaths per year occurring during the typical RSV season. In addition, as suggested by the PHIS database, nearly 80% of deaths occurred in children with complex chronic medical conditions. The mean age at time of death was 7.5 months for infants in the PHIS data set and 6.2 months in the KID data set.¹⁵

An industry-sponsored meta-analysis of published reports of fatality rates attributable to RSV infection in children suggested higher rates of death based on estimates from earlier years, likely because supportive care received in intensive care units was less effective in earlier years.¹⁰⁶

A statistically significant reduction in RSV mortality has not been demonstrated in any randomized clinical trial with palivizumab or motavizumab. Thus, inclusion of mortality reduction or life years saved is difficult to justify in a cost analysis of palivizumab prophylaxis.

Motavizumab

Motavizumab is a second-generation monoclonal antibody that differs from palivizumab by 13 amino acids. Motavizumab was developed by affinity maturation of the complementarity determining regions of palivizumab to improve binding affinity to F protein. In cell culture, motavizumab has approximately 10-fold greater neutralizing activity than palivizumab against RSV clinical isolates of both A and B subtypes.¹⁰⁷ Unlike palivizumab, motavizumab may reduce RSV vial load in the upper respiratory tract, as well as the lower respiratory tract.¹⁰⁸

Data from several clinical trials, including a noninferiority trial between palivizumab and motavizumab, were submitted to the FDA as part of the licensing application for motavizumab.^109–111 In August 2010, the FDA rejected the license application for motavizumab. Among a number of concerns noted by the FDA was lack of greater clinical efficacy from motavizumab and a threefold increase in hypersensitivity reactions among motavizumab recipients relative to palivizumab recipients.^107,109,112 Methodologic concerns were raised by the FDA in regard to laboratory testing procedures and geographic variation in results from the noninferiority trial. Specifically, the generalizability of the main study finding was uncertain, because the results of the primary end point differed when stratified by geographic location. The results did not reach the noninferiority threshold for the northern hemisphere, where approximately 90% of subjects were enrolled.¹⁰⁷

The overall RSV hospitalization rate was 4.8% among palivizumab recipients in the IMpact-RSV trial compared with a 1.9% RSV hospitalization rate among palivizumab recipients in the palivizumab-motavizumab noninferiority trial. It was concluded that the noninferiority trial may have been conducted in a population with less comorbidity than the IMpact-RSV study. The FDA requested an additional clinical trial to support a satisfactory risk–benefit profile in populations for which prophylaxis is being considered.¹⁰⁷

RSV-specific outpatient medically attended lower respiratory tract infections were reported to be significantly lower among motavizumab recipients relative to palivizumab recipients in the noninferiority trial. However, this outcome was determined in a subset of patients from selected study sites, making the risk of bias high, as noted in the FDA report.¹⁰⁷

Outpatient visits among RSV-infected children exceed the number of outpatient visits attributable to influenza infection by more than twofold.¹¹³ One study of acute respiratory infection in children younger than 8 years estimated children from birth through 23 months of age experienced overall emergency department visits attributable to RSV infection at a rate of 64.4/1000 (95% CI, 45.4–91.3) compared with a rate of 15.0/1000 (95% CI, 4.4–50.6) among influenza-infected children (27% had received influenza vaccine) during 2 respiratory virus seasons between 2003 and 2005.¹¹³ The impact of immunoprophylaxis on outpatient medically attended events attributable to RSV infection is an important consideration but remains unknown because of lack of evaluation in a rigorous, controlled fashion.

Palivizumab-Resistant Isolates

Palivizumab and motavizumab bind to a highly conserved epitope (antigenic site A) on the extracellular domain of the mature F protein that encompasses amino acids 262 to 275. After antibody binding, viral entry into the respiratory epithelial cell is blocked, as is cell-to-cell fusion of infected cells.² RSV escape mutants resistant to palivizumab have been isolated from approximately 5% of children hospitalized with breakthrough RSV infection while receiving monthly palivizumab prophylaxis.^{1, 21,22,23,114} RSV isolates resistant to palivizumab contain mutations in codons encoding the amino acids between 262 and 275 of the F protein. Amino acid sequence variations outside antigenic site A do not appear to confer palivizumab resistance.

Effect of Palivizumab Prophylaxis on Subsequent Wheezing

Numerous studies have documented that infants hospitalized with viral lower airway disease are more likely to experience recurrent wheezing compared with infants who do not experience severe bronchiolitis.^68,115–119 The possible effect of avoidance of RSV lower respiratory tract infection early in life with palivizumab prophylaxis on recurrent wheezing was addressed in 3 industry-sponsored studies.^16,17,120

One trial among preterm infants without CLD describes physician-diagnosed recurrent wheezing in 8% of prophylaxis recipients and in 16% of the control group. This trial was conducted at 27 sites, and the patients were followed for up to 24 months. Participants were not prospectively randomized, the groups were not balanced for birth weight, degree of prematurity or known RSV risk factors, and families were not blinded to study medication, making the results of uncertain significance.¹²⁰

A nonrandomized observational report from Japan compared the incidence of wheezing between 349 preterm infants (33–35 weeks’ gestation) who received palivizumab prophylaxis in the first 6 months of life with 95 infants who did not receive prophylaxis. The primary end point of the study was physician-diagnosed wheezing during a 3-year follow-up. Twenty-two of 345 infants (6.4%) who received immunoprophylaxis experienced wheezing, whereas 18 of 95 infants (18.9%) who did not receive prophylaxis developed recurrent wheezing (P < .001) during the 3-year follow-up period.¹⁷ No difference in hospitalization rates attributable to respiratory disease between the 2 groups was found. Concerns with this study design include the lack of randomization and blinding and the absence of standardized enrollment criteria at the 52 participating sites, each of which may have led to enrollment bias, and no information was provided on the severity of the wheezing episodes. In addition, a history of smokers in the household and family history of allergy was significantly more common in the untreated group.

A double-blind placebo-controlled trial conducted in the Netherlands addressed palivizumab prophylaxis and recurrent wheezing using a different study design.¹⁶ Parent-reported wheezing in 429 otherwise healthy late preterm infants during the first year of life was evaluated in 214 infants who received monthly prophylaxis with palivizumab compared with 215 infants who received placebo. During the first year of life, infants and young children in the placebo group experienced 2309 days with wheezing from a total 51 726 patient days (4.5%), whereas those in the palivizumab group had 930 days of wheezing of 53 075 patient days (1.8%; P = .01). This represents an absolute 2.7 fewer days of wheezing per 100 patient days (17.5 wheezing days/1000 days vs 44.6 wheezing days/1000 days or 27.1 fewer wheezing days/1000 days) among infants who received monthly palivizumab in contrast to those who did not receive prophylaxis.¹⁶ Reliance on parent reporting of wheezing was identified as a potential limitation. The wheezing episodes were not medically attended events but parent-reported wheezing of unknown severity.¹⁶ The proportion of infants using bronchodilators in the placebo group was 23% and 13% among palivizumab recipients.

Cost Analyses

Financial stewardship is a concept that acknowledges a physician’s responsibility to advocate “for a just and cost-effective distribution of finite resources.”^121,122 Use of noncost-effective interventions contribute to maladies within the health care system, including huge deficits, lower quality of care, and inequitable access to health care.¹²³ Financial stewardship has become an essential component of successful reform of the existing health care system.

A number of economic analyses from different countries have evaluated immunoprophylaxis use in different age groups, among children with varying comorbidities and from both the payer and the societal perspective.²⁴ Economic evaluations sponsored by the manufacturer of palivizumab suggest cost neutrality or even cost savings.^124–134 In contrast, analyses conducted by independent investigators consistently demonstrate the cost of palivizumab prophylaxis far exceeds the economic benefit of hospital avoidance, even among infants at highest risk.^{24,25,39–42,46,90,135–143} Variation in results are explained by differences in study methodology and different base case assumptions used in the model, such as incidence of RSV hospitalization for different risk groups, effectiveness of prophylaxis in reducing hospitalization rate by risk group, estimates of cost of prophylaxis and RSV hospitalizations avoided, number of doses administered, estimated age and weight of infants, and inclusion of a theoretical benefit on mortality reduction.

The CDC has recommended quality-adjusted life years (QALYs) saved as an optimal approach to evaluate the benefit of an intervention. One analysis using this methodology for hypothetical cohorts of infants without CLD born at 26 to 32 weeks’ gestation revealed the incremental cost-effectiveness ratio to be greater than $200 000 per QALY saved for all gestational ages, a figure not considered to be cost-effective.¹³⁵

An economic analysis funded by the manufacturer of palivizumab and published in the Journal of Medical Economics also examined incremental cost-effectiveness ratios per QALY gained in 4 groups of preterm infants without CLD and revealed the cost of treatment with palivizumab for infants 32 to 35 weeks’ gestation with ≤1 risk factor was $464 476, using an average cost for palivizumab for private and public payers. Medicaid discounts were reported to have resulted in an “approximate 40% reduction in palivizumab cost for approximately 60% of palivizumab recipients.”¹²⁹ Nonetheless, in most state Medicaid programs, palivizumab is one of the most costly medication expenditures.

Cost was considered during deliberations by the COID and the Bronchiolitis Guidelines Committee, but the final guidance as presented in the accompanying Policy Statement is driven by the limited clinical benefit derived from palivizumab prophylaxis.

The American College of Physicians Clinical Guidelines Committee outlines principles to help clinicians define high-value health care by considering key concepts: benefits, harms and cost of the intervention, downstream costs that occur as a result of the intervention, and finally, the incremental cost-effectiveness ratio.¹⁴⁴ On the basis of these principles, the minimal clinical reduction in RSV hospitalizations and reduction in wheezing episodes associated with palivizumab prophylaxis are not of sufficient clinical and societal importance to justify the cost. Although some RSV hospitalizations may be severe and prolonged, the majority of hospitalizations generally last 2 to 3 days. The high cost of palivizumab prophylaxis becomes a cost-inefficient way to prevent a few short hospitalization stays and a small number of longer hospital stays, especially in the absence of evidence of significant long-term benefit and no measurable effect on mortality.¹⁴⁴

Health expenditures should not be based only on cost and benefit but rather on the assessment of the benefit of the intervention relative to the expenditure. High-cost interventions may be appropriate if highly beneficial.¹⁴⁴ Because the high cost of palivizumab prophylaxis is associated with minimal health benefit, this intervention cannot be considered as high-value health care for any group of infants.

Control Measures

A critical aspect of RSV prevention among all infants is education of parents and other caregivers about the importance of decreasing exposure to and transmission of RSV. Preventive measures include limiting, where feasible, exposure to contagious settings (eg, child care centers) and emphasis on hand hygiene in all settings, including the home and the neonatal ICU, especially during periods when contacts of children at high risk have respiratory tract infections. For all children, the importance of breastfeeding, avoidance of crowds, and absence of exposure to tobacco smoke, including second-hand and third-hand exposure, should be emphasized.

Future Possibilities

Continued evaluation of the impact of palivizumab prophylaxis should include other groups considered to be at increased risk of disease to evaluate reductions in RSV hospitalizations, as well as the effect of prophylaxis on medically attended outpatient visits.

Investigation of other types of passive immunoprophylaxis including anti-G monoclonal antibodies, should continue not only in prophylaxis but in treatment studies to modulate RSV disease severity.¹⁴⁵ Modification of the Fc fragment of the motavizumab molecule appears to prolong the half-life and theoretically may enable administration of fewer doses of motavizumab per season, although an increased risk of adverse effects remains a concern.^107,146 Nanobodies are antibody-derived proteins consisting of functional heavy-chain antibodies that lack light chains and were initially found in camels and llamas. Nanobodies directed against the RSV fusion protein have demonstrated RSV-neutralizing activity in experimental models.¹⁴⁷

Ribavirin was licensed in 1986 and remains the only FDA-licensed antiviral agent for therapy but seldom is used because of limited efficacy, cumbersome delivery (aerosol), and high cost.⁵ Interest continues in developing more broadly effective antiviral agents including fusion inhibitors and small interfering RNA.^148,149

Development of a safe and effective RSV vaccine remains a high priority.^150–153 Progress has been achieved with live-attenuated intranasal vaccines,¹⁵⁴ but ensuring adequate attenuation while maintaining immunogenicity remains a challenge. Early experience with a formalin-inactivated whole virus vaccine resulted in enhanced RSV disease in young children in the 1960s, which has complicated development of inactivated RSV vaccines.^150,155 The demonstration of the effectiveness of passive prophylaxis with an anti-F monoclonal provides reassurance that antibody to the F protein is protective. New subunit vaccines and particularly vaccines directed against the F protein administered intramuscularly or intranasally continue to be explored.^150,155 A vaccine against a “prefusion” configuration of the F protein may offer increased efficacy with less risk of disease enhancement.¹⁵⁶ F protein viral-like particle vaccines administered during the latter half of pregnancy might offer passive protection for young children through the first months of life.^157,158