Respiratory Syncytial Virus

Viral Structure and Characteristics

RSV is an enveloped, medium-sized (120–300 nm) RNA virus with a nonsegmented, single-stranded, negative-sense genome (Figs. 158.1 and 158.2).^16,17 The viral envelope has transmembrane surface glycoprotein spikes 12 nm in length, and on electron microscopy the virus can be identified as early tubular structures and later more mature spherical structures. (Fig. 158.3).^16,17

The viral RNA consists of approximately 15,222 nucleotides containing 10 genes (see Fig. 158.2).^16,18 Each gene encodes a single protein except for the M2 gene, which possesses two overlapping open reading frames encoding two separate proteins (M2-1 and M2-2, the transcription processivity factor and a transcriptional regulatory protein, respectively). Four proteins, N (nucleoprotein), P (phosphoprotein), L (polymerase), and M2-1, are associated with the RNA-containing nucleocapsid complex. The envelope has three transmembrane surface proteins, F (fusion), G (attachment), and SH (proposed viroporin protein), that are important for viral infectivity.^16,19 In addition, a truncated secreted form of G (Gs) is transcribed from a second start codon. The M (matrix) protein accumulates at the inner surface of the envelope and is important in viral morphogenesis.¹⁷ Two proteins, NS1 and NS2, are nonstructural proteins that inhibit cellular type I and III interferon activity and subsequently affect the adaptive immune response to RSV.^16,20

The two glycosylated surface proteins, F and G, are major immunoprotective antigens and targets for antibody-mediated neutralization.²¹ The G protein is the primary mediator of attachment of the virus by binding to the CX3C receptor on respiratory epithelial cells and immune cells, although the F protein also can facilitate viral attachment by binding to cell surface nucleolin.^22–24 After attachment in the prefusion form, F undergoes structural changes into a postfusion form that initiates viral penetration by fusing viral and host cell membranes.²¹ The prefusion F displays a greater number of neutralizing epitopes than the postfusion F, some of which are highly potent.²¹

Laboratory Properties

RSV poorly withstands slow freezing, thawing, or changes in pH. At 55°C, infectivity rapidly diminishes; at room temperature, only 10% infectivity remains at 48 hours; at 4°C, only 1% remains after 7 days.²⁵ RSV is inactivated quickly at pH <5 or contact with ether, chloroform, and detergents. At room temperature, RSV in secretions of patients may survive on nonporous surfaces such as countertops for 3 to 30 hours and on porous surfaces such as cloth and paper tissue usually for less than 1 hour.²⁶ The infectivity of RSV on the hands is variable but is usually less than 1 hour.

Human heteroploid cell lines (HEp-2, HeLa, and A549) are usually preferred for primary isolation, but RSV may also be recovered in human kidney, amnion, and diploid fibroblastic cells and monkey kidney cells. Characteristic cytopathic effect may be first detected after an average of 3 to 5 days, and typical syncytia develop approximately 10 to 24 hours later. RSV can also infect macrophages, dendritic cells, and T and B cells.

Infection in Animals

Humans and chimpanzees are the only natural hosts for human RSV, although a variety of small animal species may be experimentally infected.²⁷ Although many animal models develop upper respiratory tract infection, their lack of symptomatic lower respiratory tract limits their utility. Rodents are the most commonly used models, particularly cotton rats and mice, but replication of RSV is only semipermissive. The ferret model has been used recently, as it demonstrates several characteristics of human illness and transmission.²⁸

Antigenic Variation

Strain differences among RSV isolates may also affect the intensity, severity, and diversity of RSV outbreaks.^38–47 RSV isolates are divided into two major antigenic groups, A and B, each with multiple genotypes—up to 11 for group A and 23 for group B viruses.⁴¹ The two major groups have 81% nucleotide identity, but some proteins between A and B strains vary appreciably, with the most diversity in the G protein, followed by M2-2 and SH proteins. This is reflected in antigenic relatedness between the two groups of only 1% to 7% for G proteins compared with 50% for F proteins.

Strains of both groups circulate simultaneously, but the proportions of A and B vary, as do genotypes.^39,40,42,43 Analyses of strains collected over decades and from diverse areas suggest that pressure of the population's immunity may result in a selective advantage for dominance of strains most divergent from those that have recently circulated. Recently emerged RSV viruses, containing unique 20–amino-acid and 26–amino-acid duplications in the group B and A G proteins, respectively, have spread throughout the world, suggesting transmission advantage over prior strains.^44,45 The relationship of circulating genotypes to illness severity and manifestations in young children has been inconsistent and thus inconclusive.⁴¹ Several reports indicate that group A ON1 strains, originally isolated in Ontario, Canada, are less severe than strains lacking the 24-amino acid duplication.⁴¹ The relationship between the substantial antigenic diversity of the G protein to reinfection with RSV has not been established.

Epidemiology Manifestations

RSV outbreaks may vary year to year in size and intensity.^1–3,48,49 Severe lower respiratory tract illness from RSV in previously healthy children occurs most frequently in the first year of life and is almost always associated with primary infection.^50,51 Essentially all children experience RSV infection within the first several years of life.^52–54

Prevalence and Incidence

RSV is the most frequent cause of bronchiolitis and is estimated to cause 40% to 90% of bronchiolitis hospitalizations and up to 50% of pneumonia admissions among infants.^5,49–51,55 The remaining cases are caused by rhinovirus, parainfluenza viruses, human metapneumovirus, and influenza virus.⁵⁶ Of tracheobronchitis cases, 10% to 30% have been associated with RSV infection, but only 2% to 10% of croup cases have been associated with RSV infection. Bronchiolitis is the leading cause of all hospitalizations among infants in the United States. Overall, 172,000 children younger than 5 years of age are reported to be hospitalized annually with RSV infection in the United States.^4,6,7 The yearly rates of RSV hospitalizations estimated from national databases have been variable, ranging from 2 to 44 per 1000 children within the first 2 to 5 years of life.^{48,50,51,57,58} Children within the first year of life consistently had the highest rates of hospitalization for bronchiolitis and other RSV-associated illnesses, and the preponderance of admissions were in infants younger than 6 months.^7,51 Population-based studies in the United States indicated current hospitalization rates of 17 per 1000 children younger than 6 months and 3 per 1000 children younger than 5 years.^3,49,51 The highest hospitalization rates (25.9/1000) were in infants 1 month of age, slowly decreasing thereafter.⁵¹ These rates are similar to rates derived from RSV-coded hospitalization data and are more than three times the rates from parainfluenza or influenza viral infections over 4 years of surveillance among the same population.^59,60 RSV hospitalization rates from European countries have been generally similar, ranging from 2.5 to 11 per 1000 children within the first 4 years of life and highest among infants younger than 12 months of age at 19 to 22 per 1000 children.^8,50,61–63

Lower respiratory tract infections from RSV in developing countries are estimated to be twice as prevalent as in developed countries.⁸ Worldwide, RSV is estimated to cause 33.1 million RSV acute lower respiratory infections annually and 3.2 million hospitalizations in children younger than 5 years, most of which are in developing and low-income countries.⁸ Despite its frequency, mortality from RSV in developed countries is conspicuously less common.⁶⁴ US data indicate approximately 42 annual deaths in children with RSV, whereas estimates for developing countries suggest approximately 59,000 infant deaths annually are due to RSV.^8,64

Many host, socioeconomic, and environmental factors have been associated with a greater likelihood of young children developing more severe RSV infection and requiring hospitalization (see “Patients at High Risk for Severe Infection”).^2,6 Population-based surveillance of RSV hospitalizations and emergency department and outpatient visits in the counties surrounding Rochester, New York; Nashville, Tennessee; and Cincinnati, Ohio, during 2000 to 2004 indicated that outpatient visits constituted a major proportion of the health care burden attributable to RSV.³ Of all visits for acute respiratory illnesses (ARI) among children within the first 5 years of life, RSV infection was documented among 20% of ARI hospitalizations, 18% of emergency department ARI visits, and 15% of office ARI visits. Estimated rates of ARI visits from RSV among children younger than 5 years in pediatric practices (80 per 1000 children) were approximately 3 times that observed among emergency department patients and 26 times hospitalization rates.

Fewer data exist on the health care burden imposed by RSV infection among older children and adults.^58,65 In the Houston family study of children followed from birth, the infection rate was 68.8 per 100 children in the first year of life, and at least half were reinfected during the second year of life.⁵² In urban Rochester, 44% of families with young children became infected with RSV during winter months when RSV was prevalent.⁵⁴ Of exposed family members, 46% were infected with RSV. Although the attack rate was highest among infants, 38% to 47% of older children and adults developed RSV infection. Similar findings were noted in family studies in Finland and rural Kenya.^66,67

The incidence of RSV infections among older adults and the resulting appreciable clinical and economic impact are increasingly recognized.^9,12,68–72 In one study spanning 4 years, rates of RSV infection ranged from 2% to 10% per year in elderly adults 65 years of age or older and in high-risk individuals with underlying cardiopulmonary disease.⁹ RSV infection among elderly adults is remarkably similar to influenza infection with respect to clinical manifestations and as a cause of hospitalization.^68,73 The rate of hospitalization attributable to RSV among elderly adults ranges from 1.5 to 23 per 10,000 in various studies.

Immunity and Pathogenesis of Disease

Much of the knowledge regarding immune responses to RSV is from in vitro studies and in animal models. In humans, the immune response is confounded by the influence of genetics, presence of maternal antibody, environmental exposures including concurrent respiratory tract microbiota, age, and the virus.^92–95 More severe disease in the youngest infants is thought to be related to decreased levels of maternally derived RSV-specific antibody as well as physical, immune, and viral factors. The severity of RSV infection in a young infant with augmented disease induced by the inactivated RSV vaccine developed in the 1960s first suggested a role of the immune response in pathogenesis of RSV in infants.^95–98

The potential importance of the host's immune response to disease has been supported by the observation that RSV is not generally invasive or cytopathic.⁹⁵ However, one report of fatal RSV noted that RSV antigen was extensively present in pulmonary tissue, indicating abundant viral replication.⁸⁹ Cytokine production was nearly absent, and the expression of apoptosis was increased, with the conclusion that the patient had an inadequate immune response and unchecked viral replication.

Although viral load seems to correlate with disease severity in young children and adults, and disease symptoms decrease as viral load decreases, reducing viral replication by administration of neutralizing antibody to F protein has not ameliorated clinical disease in infants.^100,101 Data in murine models of RSV indicate that nonneutralizing antibodies to the centrally conserved CX3C chemokine motif of G protein can reduce inflammatory responses even after infection is established and in the absence of reduced viral load.^102–104 Whether innate or adaptive immune responses or both are enhanced or suppressed during more severe disease remains controversial.

The influence of the nasal microbiota and their distinct metabolic pathways on clinical manifestations of RSV infection has been appreciated.^105–107 It has also been recognized that RSV infection can influence the pathogenicity of Streptococcus pneumoniae, and RSV epidemics are associated with an increase in pneumococcal pneumonia in young children.¹⁰⁸ The presence of S. pneumoniae or Haemophilus influenzae increases the susceptibility and inflammatory response of airway epithelial cells in vitro.^109,110 Even in the absence of overt bacterial infection, the nasal microbiome may affect infant peripheral blood transcriptomic response to infection and is associated with illness severity.¹⁰⁶

Maternally Derived Immunity

The first barrier of defense against RSV infection in infants is maternally derived RSV-specific serum antibody. Despite a relatively short half-life of 28 to 40 days, maternally derived antibodies have been correlated with protection in some, but not all, studies.^111–116 An early study noted that higher levels of cord blood neutralizing antibody were associated with reduced risk of hospitalization and were directly correlated with older age at hospitalization.¹¹¹ A similar analysis from Denmark calculated a 26% reduction in hospitalization during the first 6 months of life for every twofold increase in cord blood neutralizing antibody.¹¹⁵ In addition, a study found that higher levels of antibody to prefusion-F and G proteins of RSV and neutralizing antibody were associated with less severe disease in the first few months of life.¹¹³ Supplementing these observational studies, and perhaps the most compelling case for the beneficial effects of antibody, results from administration of RSV polyclonal or monoclonal antibody to high-risk infants have demonstrated protection against severe RSV disease.^99,117,118

Innate Immunity

A rapid and vigorous innate immune response is initiated when infection of the infant's respiratory epithelium occurs.^{95,98,119,120} Early events include RSV interaction with Toll-like receptors 2, 3, 4, and 7, which triggers secretion of inflammatory cytokines and chemokines such as interleukin (IL)-8, monocyte chemoattractant protein 1, macrophage inflammatory proteins MIP-1α and MIP-1β, RANTES (regulated on activation, normal T-cell expressed and secreted), and eotaxin with recruitment of macrophages, mononuclear cells, natural killer cells, and eosinophils.^98,121–123 RSV infection induces cellular production of innate interferons, which are counterbalanced by viral synthesis of the NS1 and NS2 proteins that are potent inhibitors of antiviral type I and III interferons (α, β, and λ).²⁰ An early and robust neutrophilic infiltration into the airway is correlated with a decline in viral load before the development of T-cell responses.^124–127 Dendritic cells also infiltrate the nasal mucosa early in infection and can be noted in lower airway secretions associated with a number of proinflammatory cytokines (e.g., IL-6, tumor necrosis factor-α, IL-8). Concentrations of many of these cell types as well as levels of cytokines and chemokines have been linked to clinical disease phenotypes and severity.¹²⁰ The variability in the endowed innate defense and susceptibility of the host are being increasingly correlated with polymorphisms in genes that are integral to various components of innate immunity.⁹⁴

Adaptive Immunity

An effective immune response to RSV infection requires a fine balance of multiple components of immunity, a balance likely determined by both host and viral factors.^92,95,120 During primary infection, serum immunoglobulin M (IgM) antibody appears within several days but is transient and detectable usually only for a few weeks.⁹⁵ IgG antibody appears during the second week, peaks in the fourth week, and begins to decline after 1 to 2 months. An anamnestic response involving all three immunoglobulin classes occurs after reinfection, and after about three infections, titers reach levels similar to those in adults.

Primary and subsequent infections result in antibody production to many of the RSV proteins including the major immunoprotective surface glycoproteins F and G.¹⁶ Both contain neutralizing epitopes, but those on the F protein are conserved between the two viral groups. The response to the variable G protein is group and genotype specific.¹²⁸ Most adults with RSV infection develop IgG responses to the centrally conserved chemokine motif of G, although their role in recovery or protection from infection or illness is not clear.^116,129

Although young children are able to produce neutralizing antibodies directed against both the F and G proteins, neutralizing antibody responses are blunted in infants younger than 6 months of age owing to a dampening effect of maternal antibody.^130,131 In infants, antibody responses to F and G proteins mainly involve the subclasses IgG1 and IgG3. Adults respond to the G protein with both IgG1 and IgG2 subclass antibodies, and the adult response to the F protein is predominantly IgG1. After primary and recurrent infections, antibodies usually decline substantially within months. Following natural infection, 75% of adults demonstrate a fourfold or greater decrease in titer, returning to preinfection titers within 2 years in most cases.¹³² Higher titers of antibody generally correlate with better resistance to infection, but no defined level of neutralizing antibody is predictive of the risk for infection, severity of illness, or recovery in children or adults.^{10,111,133–135}

RSV-specific IgA antibody, produced in nasal secretions during primary and subsequent infection, is associated with protecting the upper respiratory tract from either natural infection or experimental challenge in adults and has been correlated with viral clearance in infants.^135–137 In the adult RSV challenge model, impaired induction of IgA-specific memory B cells was noted, which may account for the ease of reinfection.¹³⁸ Children with RSV infection may also produce transient specific IgE antibody responses in the respiratory tract. Higher levels of nasal RSV-specific IgE antibody and cysteinyl leukotrienes have been correlated with increased risk for more severe illness and wheezing and with later episodes of recurrent wheezing.^120,139,140

Cell-mediated immunity is considered pivotal for clearance of virus and clinical recovery. Adults and children with deficiencies of cellular immunity have more severe disease and prolonged virus shedding.^141,142 Most data delineating the specific components of the cellular immune response induced by RSV are derived from rodent models and, to a lesser extent, from humans.^92,95,120 RSV infection has multiple inhibitory effects on the cellular immune response. Diminished in vitro lymphoproliferative responses during initial and repeated infections suggest impaired RSV-specific helper T-cell responses. Furthermore, RSV-infected dendritic cells have diminished ability to activate CD4⁺ T cells, and enhanced apoptosis of CD4⁺ and CD8⁺ lymphocytes is observed among infants with bronchiolitis.^120,143,144

The relationship between disease manifestations and the balance between Th1 and Th2 T-cell responses during infection has been an area of great interest. RSV infection in animals and humans engenders both Th1 and Th2 responses. Th1-dominant responses, characterized by the production of CD8⁺ cytotoxic lymphocyte and Th1 CD4⁺ cells secreting IL-2, IFN-γ, and tumor necrosis factor-α, are associated with viral clearance and minimal pulmonary cytopathology in animal models. In contrast, Th2 CD4⁺ cells, with associated IL-4, IL-5, IL-10, and IL-13 secretion, impairs CD8⁺ T-cell function and viral clearance.¹⁴⁵ An overexuberant memory CD8⁺ T-cell response can mediate severe immunopathology in murine models of RSV infection.¹⁴⁶ IL-4 and IL-13 also augment isotype switching to IgE, and a Th2-biased response has been correlated with wheezing, more severe disease, and greater cellular inflammation and eosinophilia in the lung.¹³⁹

Infants infected in the first several months of life have higher levels of Th2-type cytokines in nasal secretions than older infants.^147,148 Transcriptomic analysis of isolated circulating CD4 T cells during primary RSV infection in young infants suggested a pattern of activation consistent with a Th2 environment, including evidence of Th9 activation, that was associated with more severe disease.¹⁴⁹ Inactivated virus, as used in the initial formalin-inactivated vaccine and even in subunit vaccines, is more likely than live virus to induce a Th2-like response in experiments with unprimed animals.¹⁴⁵

Otitis Media

Acute otitis media is a common complication of RSV infection among young children.^154–156 Among children within the first 3 years of life who developed acute otitis media, 74% had RSV detected in middle ear fluid. Acute otitis media usually develops about 5 days after the onset of the respiratory illness and is more common among children older than 1 year. Otitis and earache are common complications even among older children and previously healthy adults with RSV infection; these complications also were noted in approximately half of hospitalized infants in one study.^54,65,156 RSV has been recovered from the middle ear fluid as the sole pathogen, with bacteria, or with another virus in 2% to 22% of cases.^154,156 The most frequent concurrent bacterial pathogens are S. pneumoniae, H. influenzae, and Moraxella catarrhalis.¹⁵⁶ Clinical and experimental evidence suggests that coinfection of RSV with a bacterial pathogen may prolong the duration and worsen the outcome of otitis media, resulting in a greater chance of treatment failure with antibiotics and persistent effusion.

Infections Among Older Children

The frequency of RSV infections at all ages is well illustrated among families with young children and among persons in contact with young children, as in schools and child care facilities.^52–54 Among children attending child care who had primary RSV infection in their first winter, 75% and 65% develop infection during their second and third years, respectively, using serologic and viral detection methods.⁵³ Recurrent infections commonly are upper respiratory tract illnesses, but 20% to 50% of recurrent infections among preschool-aged children involve the lower respiratory tract, including wheezing, although generally less severe than initial infection.⁵² The overall burden of repeated infection in children younger than 5 years of age is substantial.³

Infections Among Adults

RSV infection in adults typically manifests with upper respiratory symptoms, nasal congestion, and scratchy throat preceding lower respiratory symptoms by several days. Wheezing is more common with RSV than with other respiratory viruses. Changes on the chest radiograph are seen in about half of hospitalized patients, with minimal lower lobe infiltrates or ground-glass appearance most common.¹⁵⁷ On computed tomography scan, ground-glass changes and bronchiole wall thickening are noted in 67%.¹⁵⁸ Adults also may have repetitive RSV infection occurring in sequential years, especially individuals living or working with children.^54,65,74,133 Despite the lack of durable immunity evoked by natural infection, both infection risk and severity of infection in adults are reduced by higher titers of serum and mucosal antibody in observational studies.^10,135

The burden RSV places on the health care of this growing population has only recently been appreciated fully.^{4,9,12,68,159–162} At greatest risk of severe illness are frail elderly patients; patients with underlying cardiopulmonary disease, especially chronic obstructive pulmonary disease; and severely immunocompromised patients.¹⁰ Adults shed RSV at titers considerably lower than infants,¹⁶³ and thus estimates of RSV burden in this population have been hampered until more recently by lack of sensitive diagnostics. The incidence of hospitalization, found by using regression models using viral culture data from infants and adult hospital coding data, has been estimated at 86 per 100,000, or 28% of the incidence of influenza hospitalizations.⁴ Using similar methods, RSV mortality in The Netherlands was 64% of that attributable to influenza.¹² Similar rates of RSV hospitalization and mortality have been reported by other authors.^{120,161,164–167}

In long-term care facilities, 5% to 27% of respiratory infections have been reported to be caused by RSV, with attack rates estimated to be 1% to 15%, pneumonia rates of 10% to 20%, and mortality of 2% to 5% of infected individuals.¹⁶² In a retrospective cohort analysis of residents of Tennessee nursing homes, RSV was associated with 15 hospitalizations, 17 deaths, and 76 antibiotic courses per 1000 people and 7% of hospitalizations and 9% of deaths from cardiopulmonary disease.¹⁶⁷ Among adults in daycare facilities, 10% of acute respiratory infections were due to RSV, a rate similar to influenza and coronaviruses.¹⁵⁹

The highest risk for severe RSV infection is in patients with underlying cardiopulmonary disease, especially chronic obstructive pulmonary disease, and congestive heart failure.^{9,68,164,168–170} The clinical illness caused by RSV in elderly adults is nonspecific and indistinguishable from other respiratory viruses, such that studies with active prospective diagnosis of RSV infection are required.^73,160 In a 4-year prospective outpatient study of 608 healthy individuals older than 65 years of age, 504 high-risk adults with cardiopulmonary conditions, and 1388 adults hospitalized with acute cardiopulmonary conditions, RSV infection was identified in 3% to 7% of the elderly cohort, 4% to 10% of the high-risk cohort, and 11% of the hospitalized cohort (Table 158.1).⁹ These rates were similar to rates for influenza in the same groups. Among the hospitalized patients with RSV, 15% required intensive care and 8% died compared with 12% and 7%, respectively, of patients with influenza. In a large outpatient study of adults ≥50 years old with medically attended respiratory illness, RSV was identified in 8% compared with influenza identified in 17%.¹⁶⁰ The same investigators estimated the incidence of RSV medically attended respiratory illness to be 154 per 10,000, increasing with each decade of life.¹¹

Peak nasal RSV titers in adults are substantially lower than in infants, ranging from 1 to 10⁵ pfu/mL (mean, 10^2.3 pfu/mL).¹⁶³ When simultaneously measured, virus titers in sputum are generally higher than in nasal secretions.¹⁷¹ The duration of virus shedding in adults averages 10 days but can be 20 to 30 days. Higher viral load has been correlated with more severe disease.^172,173 Bacterial coinfection during RSV infection occurred in 31% of hospitalized patients, similar to the rate with influenza.¹⁷⁴ There are no controlled data on antiviral treatment for RSV in adults, and thus treatment is rarely considered except in immunocompromised patients (see “Therapy”).

RSV severity among older adults is only partly explained by comorbidities and age-associated decline in pulmonary function, and it has been suggested that age-associated decline in immune function may be relevant. However, humoral immunity to RSV among elderly adults is equal to or greater than that among young adults.¹³⁴ In stable uninfected older adults, baseline RSV-specific CD8⁺ T-cell function is diminished.^175–177 However, CD4 and CD8 T-cell responses during RSV infection were not age dependent, and more vigorous responses were noted in the most ill patients regardless of age.¹⁷⁸

The burden of RSV among young healthy adults may be considerable and comparable to the burden influenza.⁶⁵ Among working healthy adults with RSV infection, 22% had lower respiratory tract manifestations. Compared with influenza, fever was less frequent, but earache, sinus pain, persistent productive cough, and wheezing were significantly more common with RSV (Table 158.2). In contrast to influenza, there are few descriptions of RSV infection during pregnancy or adverse impacts of maternal RSV on the fetus.^179–182 Reported rates of RSV infection in pregnant women are low (0.3–3.9/1000 person-years), but results are limited by lack of appropriate prospective viral surveillance. Severe disease with RSV during pregnancy has been reported, with one pregnant woman requiring mechanical ventilation.¹⁸² No fetal adverse outcomes have been noted.¹⁷⁹

Immunocompromised Patients

RSV is a well-recognized cause of morbidity and mortality among immunocompromised patients. Greater awareness of RSV disease in this population is a result of increasing numbers of patients undergoing solid-organ transplant (SOT) and hematopoietic cell transplantation (HCT), the use of more prolonged and intensive chemotherapy regimens, and the wider availability of sensitive diagnostic techniques for respiratory viruses.^193,194 The reported frequency of RSV infection among immunocompromised patients varies widely from about 2% to 50%, depending on the type of surveillance, the population, and the degree of immunosuppression.^{142,193–200} Although rhinovirus, not RSV, is the most common respiratory virus reported following HCT or SOT, RSV is one of the most serious viral causes of morbidity and mortality in pediatric and adult SOT and HCT recipients.^201–203 Reported rates of RSV-associated complications vary widely with reports of RSV progression from upper to lower respiratory tract disease up to 55% and mortality ranging from 7% to 33% in HCT recipients, with lower rates in SOT recipients.²⁰⁴

RSV may be introduced by medical staff or individuals in the community into both inpatient and outpatient settings where immunocompromised patients receive care. Viral spread may be rapid and difficult to control and accompanied by appreciable morbidity and mortality.^{142,195,197–199,205–207} The severity of RSV infection among these patients is related to the duration and degree of immunosuppression, type of transplant or immunosuppressive therapy, and other risk factors including treatment for graft-versus-host disease or organ rejection.^{142,198–200,208–211} Patients with severe combined immunodeficiency states, patients with infection early after HCT, and lung transplant recipients are particularly at risk for a poor outcome. Among transplant recipients, factors associated with poor outcomes include use of bone marrow cells (in contrast to peripheral stem cells), occurrence of RSV infection before engraftment, the presence of acute or chronic graft-versus-host disease, and lymphopenia.²⁰⁹ Among children, young age also is associated with a poorer prognosis.

RSV infection in these patients may clinically mimic other opportunistic agents, and the correct etiology may not be suspected. Sinusitis, otitis media, and wheezing may be more indicative of RSV infection than other respiratory pathogens; fever is not universally present. Concurrent infections by other infectious agents including community-acquired respiratory viruses may further confound or delay the diagnosis of RSV infection. In adult HCT recipients with RSV, upper respiratory tract signs precede pneumonia in 80% to 90% of patients, with progression to pneumonia in 30% to 40% after a median of 7 days.^193,194 The need for oxygen supplementation at the time of presentation is associated with significantly higher rates of respiratory failure and mortality.^211,212 With lower respiratory tract involvement, radiographic findings range from focal interstitial infiltrates, sometimes with hyperinflation or with lobar consolidation, to generalized alveolar and interstitial infiltrates, or even to a picture of acute respiratory distress syndrome.¹⁹⁷ With high-resolution computed tomography, the characteristic findings are airspace consolidation; small centrilobular nodules; ground-glass opacities; and thickening of the bronchial walls, which may be asymmetrical.²¹³

Long-term complications in the immunocompromised host include bronchiolitis obliterans syndrome in nearly 10%, although rates of 50% can be seen in lung transplant recipients with RSV lower respiratory infection.²¹⁴ In the past, overall mortality with RSV pneumonia was reported to be up to 45%, but current mortality rates appear to be substantially less than that (15% for HCT patients with upper respiratory tract presentation; 33% for patients with lower respiratory tract disease).²¹² This has been attributed to better diagnostics, prospective screening before HCT to delay transplant in the face of active RSV infection, advances in supportive care, and potentially treatment with ribavirin with or without immunoglobulin (intravenous immune globulin [IVIG] or monoclonal antibody) (see “Therapy”).

Among children with human immunodeficiency virus (HIV) infection, RSV has been the most frequently identified cause of viral respiratory disease.¹⁹⁵ The manifestations of RSV infection vary according to the stage and severity of the HIV infection, and although most patients develop lower respiratory tract involvement, disease is generally is not as severe as disease among highly immunosuppressed transplant recipients.²¹⁵ Patients with HIV infection may shed RSV for prolonged periods. Confounding this, however, is the observation that children with HIV infection and concurrent viral respiratory infections also have a higher rate of bacterial coinfections.²¹⁵ The clinical outcome of respiratory viral infections has not been consistently different between children with and without HIV infection.

Although treatment of severely immunocompromised patients with aerosolized or even oral ribavirin is recommended by some experts, definitive guidelines are not available because of the lack of highly effective antivirals and prospective controlled trials (see “Therapy”).^216,217 The most important aspect of management of immunocompromised patients is prevention of RSV nosocomial infection by strict adherence to infection control policies; guidelines for preventing opportunistic infections in hematopoietic stem cell transplant recipients are available.^193,218,219 These guidelines emphasize preventing the introduction of community respiratory viruses, including RSV, onto units with immunocompromised patients and stress the importance of early diagnosis.

Acute Complications in Infants

Apnea is one of the most striking and serious acute complications in young infants with RSV. Up to 20% of infants hospitalized with RSV infection have been admitted with apnea.^220,221 Among infants evaluated in the emergency department with bronchiolitis, 3% had a diagnosis of apnea. Infants at highest risk for apnea are preterm infants with a gestational age ≤32 weeks, infants with a history of apnea of prematurity, and infants of postnatal age of less than 44 weeks after conception. Characteristically, apnea occurs at the onset of RSV infection and may precede respiratory symptoms. The pathophysiology of apnea is unclear, although it is nonobstructive. Prognosis is generally good after acute RSV infection with no subsequent episodes, even with respiratory infections.

Infants admitted with RSV lower respiratory tract disease may be at increased risk for aspiration, which can appear clinically similar to bronchiolitis with airway hyperreactivity.²²² In a 12-month follow-up of infants hospitalized with RSV bronchiolitis, 83% developed reactive airway disease if they received neither ribavirin nor therapy for aspiration. The decrease in reactive airway episodes was greater in infants receiving ribavirin and thickened feedings than infants who received either therapy alone.

Coexistent bacterial infection is a frequent concern in infants hospitalized with RSV, and many likely receive unnecessary antibiotics. This is due in part to young age, the presence of fever, and the relatively frequent appearance on the chest radiograph of opacities from viral infiltrates and atelectasis commonly mistaken for bacterial pneumonia. Multiple studies have shown that secondary bacterial infection is an unusual complication of RSV infection.^223–225 A 9-year prospective study identified secondary bacterial pneumonia in less than 1% of subjects, and a multicenter Spanish study identified bacteremia in only the most seriously ill children.^223,226 Furthermore, antibiotic therapy has not been shown to improve recovery from RSV lower respiratory tract disease.²²⁷ Urinary tract infections are the most frequently identified concurrent bacterial infections.²²⁸ In developing countries, however, respiratory bacterial coinfections are more common and may contribute appreciably to the high mortality rate from RSV. The most common coinfection in infants is infection with another virus, most commonly rhinoviruses, adenoviruses, coronaviruses, parainfluenza viruses, human metapneumovirus, and bocavirus.²²⁹ There is no definitive evidence that viral coinfection carries a worse outcome; studies indicate that RSV coinfections may be less severe than RSV infections alone.^230,231

Long-Term Complications

Recurrent wheezing after RSV bronchiolitis in infancy has long been recognized as a frequent sequela, but a causal link between the two remains unclear.^231a Approximately 30% to 50% of children hospitalized with RSV infection later develop repeated occurrences of wheezing.^232–234 For many children, the severity of the recurrent wheezing episodes decreases with age, although pulmonary function abnormalities may persist in some without clinical manifestations.^235,236 Others may have persistent wheezing into adolescence or have wheezing cease during childhood and recur in adulthood. The frequency of long-term sequela in the general population is confounded by most studies having focused on children with more severe illness.

Epidemiologic evidence indicates that atopy in the child or family is not a major cause of this link. However, in a murine model of allergen-induced airway inflammation and remodeling, previous RSV infection could induce airway abnormalities in mice exposed to allergen through the airway, even though these mice had not been previously sensitized to the allergens.²³⁷ In addition, RSV infection has been shown both in vitro and in children infected with RSV to produce an immunologic response similar to that observed with allergic sensitization, one with a predominantly Th2 T-cell profile and release of proinflammatory mediators, IgE, and neuropeptides (see “Immunity and Pathogenesis of Disease”).^92,238 However, a similar response may be produced by viruses other than RSV.^239,240 Relevant to the relationship between severe RSV infection and asthma are two studies suggesting that palivizumab prophylaxis of premature infants was associated with a decrease in incidence of recurrent wheeze in the first few years of life,^241,242 although this decrease did not persist at age 6 years.^242a

Prophylaxis

Prophylaxis using the passive administration of RSV-specific antibody currently is available primarily for groups most at risk for developing severe or complicated RSV disease. In high-risk children administered monthly doses of IVIG containing high levels of RSV neutralizing antibody (RSV-IVIG) or intramuscular monoclonal antibody, reduced rates of hospitalization due to RSV infection have been shown. RSV-IVIG has been replaced by palivizumab, a humanized monoclonal antibody developed from a mouse monoclonal antibody directed against a protective epitope of the RSV F protein. Immunoprophylaxis with five monthly intramuscular doses of 15 mg/kg initiated 1 month before the RSV season demonstrated a 4.4% to 5.8% absolute reduction in hospitalization rates among selected high-risk infants.¹¹⁷ It has consistently been demonstrated that preterm infants born before 29 weeks’ gestation are at greatest risk for hospitalization as a result of RSV infection.¹¹⁷ For preterm infants born after 29 weeks’ gestation who are otherwise healthy, there is no distinct cutoff at which the benefits of prophylaxis are clear. Palivizumab administration does not prevent infection with RSV but is associated with decreased clinical severity, risk for developing lower respiratory tract disease, and need for hospitalization.

At the present time, palivizumab is recommended by the American Academy of Pediatrics for a small group of infants in the first year of life who are considered to have a high risk for developing severe RSV disease.²⁷⁸ Included are infants with prematurity (<29 weeks’ gestation), chronic lung disease, functionally important cardiac disease that has not been surgically corrected, and chronic conditions that interfere with the handling of respiratory secretions.¹¹⁷ Palivizumab is not recommended in the second year of life except for preterm children who required at least 28 days of supplemental oxygen at birth and continue to require supplemental oxygen, long-term systemic steroid therapy, or diuretic therapy within 6 months of the start of the second RSV season. Palivizumab is administered at a dose of 15 mg/kg for a maximum of 5 monthly doses during the RSV season. Prophylaxis should be discontinued in any child who experiences a breakthrough RSV hospitalization.¹¹⁷

Controversy exists concerning the extent of the use of palivizumab prophylaxis based primarily on concern regarding its considerable cost relative to its benefit. Economic analyses in general have not shown an overall savings in health care costs for prophylaxis of all infants less than 32 weeks’ gestation and with underlying high-risk conditions, but the benefit relative to the cost among those at highest risk increases.^279,280 Mathematical modeling has suggested that four monthly doses of palivizumab initiated later in the season could provide comparable protection to five doses and improve cost-effectiveness.²⁸¹ Although the mortality from RSV infection even among high-risk infants in the United States is very low, studies evaluating the impact of decreased palivizumab administration in infants >29 weeks’ gestation have demonstrated increased rates of hospital admission in infants 29 to 34 weeks’ gestation and demonstrated increased morbidity as well as mortality.^282,283 The effect of palivizumab on the long-term sequelae of RSV infection has not been adequately evaluated.^241,242,284

Children with severe immunodeficiencies may benefit from RSV prophylaxis, particularly when used early in life to prevent pretransplant infections. The use of palivizumab during or after transplant remains controversial. Using a decision analysis model to evaluate palivizumab prophylaxis to prevent RSV mortality after HCT in children, the survival rate was estimated to increase by 10%, and 12 children would need to be treated to prevent one fatal RSV infection.²⁶⁴

Additional products have been and are being evaluated for the prevention of RSV disease in high-risk individuals. The enhanced potency humanized monoclonal antibody against the F protein, motavizumab, had a 20-fold higher rate in RSV neutralization compared with palivizumab and prevented RSV disease in preterm infants, but this agent was not licensed due to safety issues (primarily rash reactions).^285,286 A trial of motavizumab in healthy Native American infants demonstrated an 87% relative reduction in hospitalizations for RSV,²⁸⁷marking the first time an anti-RSV antibody prevented serious RSV disease in healthy term infants. A new potent monoclonal antibody, MEDI18897, also known as nirsevimab, has an active site directed against the prefusion form of the RSV F protein and has approximately 100-fold potency in vitro compared with palivizumab.^288,289 This antibody is a totally human recombinant IgG1 κ monoclonal antibody with an extended half-life due to a 3–amino acid mutation in the Fc region enhancing binding within the lysosome and preventing antibody degradation and increasing recirculation to the cell surface. Safety and tolerability in adults has been established, and this compound is in clinical trials in infants.^290,291 The increased half-life may potentially provide passive protection for infants throughout an entire RSV season following a single intramuscular injection. This compound is intended for use in healthy term and at-risk preterm infants entering their first RSV season in both developed and developing countries.

Immunization

No effective vaccine for prevention of RSV in any population is available. Challenges in vaccine development include concern that an unpredictable abnormal immune-mediated response to subsequent infection could occur, such as was observed during the trials with the initial formalin-inactivated RSV vaccine.⁹⁶ Advances in molecular technology have resulted in new RSV vaccine candidates including subunit, particle, vector-based, and live-attenuated/chimeric vaccines. Recent crystallographic characterization of the RSV F-protein has led to design of a structure-based vaccine consisting of a stable prefusion form of the F-protein. This immunogen induces high levels of neutralizing antibody in mice, hamsters, and macaques and is a promising vaccine candidate.²⁸⁹

Multiple approaches are being pursued for vaccines to protect two major groups at highest risk for severe RSV disease: very young infants and frail elderly adults.^292,293 Immunization for infants would optimally be initiated within the first weeks of life because most hospitalization for RSV occurs in the first several months of life. Another approach for protection of infants is maternal immunization with a subunit vaccine, similar to the approach taken with influenza virus and tetanus.^294,295 Boosting of maternal neutralizing antibodies using the RSV F protein could potentially reduce the severity of infection in the first few months of life or delay infection until the infant is older when disease is less severe.^296,297 A large international clinical trial with a nanoparticle F protein vaccine candidate given to pregnant women at 32 to 36 weeks of gestation is underway (Novavax, Gaithersburg, MD). Infants with decreased transplacental antibody transfer such as premature infants or infants born to mothers with decreased maternal antibody transfer due to HIV infection or maternal hypergammaglobulinemia would benefit less from this strategy.²⁹⁸

An approach for older infants or children is intranasal immunization with live-attenuated vaccine strains.^299,300 The use of reverse genetics has resulted in the generation of candidate strains that contain mutations specifically chosen for their attenuating, immunogenic, and other advantageous characteristics. These candidate designer gene vaccines are potentially safer and have increased breadth of antigenic expression.³⁰⁰ An advantage of this approach is that immunization should induce nasal IgA and cellular responses as well as serum antibody. An alternative approach is the use of live vectored or chimeric vaccines to deliver RSV antigens.

Subunit F or G vaccines are being considered for older adults. Several subunit F protein vaccines have been shown to be safe and immunogenic in individuals seropositive from previous natural RSV infection and should theoretically boost neutralizing serum antibodies, especially in adults, although results from several trials in elderly adults have been disappointing. Such a vaccine may also offer the possibility of boosting immunity among older children and adults who have an increased risk for developing more severe RSV infection.