Respiratory Tract Viruses

Respiratory tract infection in brief:

This is infection of the breathing passages, which extend from the nose to the alveoli (the tiny, balloon like sacs in the lungs). This type of infection is divided into upper and lower respiratory tract infections.

Upper respiratory tract infections affect the nose, the pharynx (throat), the sinuses, and the larynx (voice-box). Examples are the common cold, (see cold, common) and inflammatory conditions such as pharyngitis, tonsillitis, sinusitis, laryngitis, and croup.

Lower respiratory tract infections, which involve inflammation of the trachea (windpipe), bronchi, and lungs, include acute bronchitis, acute bronchiolitis, and pneumonia.

Respiratory tract viruses in detail – technical

Topics covered:

  • Essentials
  • Introduction
  • Transmission
  • Seasonality
  • Laboratory diagnosis
  • Rhinoviruses
  • Enteroviruses
  • Coronaviruses
  • Adenoviruses
  • Respiratory syncytial virus
  • Parainfluenza virus
  • Human metapneumovirus
  • Influenza viruses
  • Bocavirus and polyomavirus KI and WU
  • Nosocomial infection
  • Further reading

Essentials

Viral respiratory infections, including rhinovirus, coronavirus, adenovirus, respiratory syncytial virus, human metapneumovirus, parainfluenza viruses, and influenza viruses, are a substantial cause of morbidity worldwide. Transmission occurs through direct contact, contaminated fomites, and large airborne droplets, with long-range transmission by small particle aerosols reported in at least some instances of influenza and severe acute respiratory syndrome (SARS).

Clinical syndromes affect the upper and/or lower respiratory tract, including coryza, pharyngitis, croup, bronchiolitis, and pneumonia. Each syndrome can potentially be caused by a number of viruses, and each respiratory virus can be associated with different clinical syndromes. Measles is a major cause of lower respiratory tract infections and fatality in tropical countries.

Diagnosis—nasopharyngeal aspirates, washes and swabs are superior to throat and nose swabs for diagnosis, with virus detected by culture or detection of antigen or nucleic acid (e.g. PCR-based methods). New respiratory viruses continue to be discovered, but some acute respiratory infections have no identifiable aetiology, and some patients have multiple respiratory viruses detectable in the respiratory tract in association with their disease—whether these have a synergistic role in pathogenesis remains unclear.

Particular respiratory tract viruses

Influenza—types A and B are clinically important causes of human disease; the viral envelope contains two glycoproteins, haemagglutinin (H) and neuraminidase (N), which are critical in host immunity and used to designate viral subtype, e.g. H1N1. Potential to cause pandemics makes influenza a unique challenge for global public health. Typically causes an illness associated with fever, chills, headache, sore throat, coryza, nonproductive cough, myalgia, and sometimes prostration. Can cause pneumonia directly or by secondary bacterial infections. Oseltamivir and zanamivir result in a reduction of 1 to 2 days in the time to alleviation of symptoms when administered within the first 48 h of illness, but recent emergence of resistance to antivirals is a cause for concern. Can be prevented by influenza vaccine, which contains antigens from the two subtypes of human influenza A (H3N2 and H1N1) and B viruses, but the composition of the vaccine must be updated on an annual basis to keep abreast of change in the surface antigens of the virus, and annual reimmunization is required. Synergic interaction with Streptococcus pneumoniae enhances pathogenesis, and pneumococcal conjugate vaccine reduces hospitalization associated with respiratory viruses.

Respiratory syncytial virus (RSV)—a major cause of bronchiolitis and pneumonia in infants. Infection in adults is often asymptomatic, but during the RSV season (winter months) it is an important cause of lower respiratory tract infection in adults, particularly elderly people. May be lethal (as can other respiratory viruses) in patients immunocompromised following organ or blood and marrow transplants (but is not a significant problem in patients with AIDS).

Severe acute respiratory syndrome (SARS)—this novel coronavirus of animals adapted to efficient human transmission and spread worldwide, causing a global outbreak in 2003 of an illness characterized by lower respiratory tract manifestations, severe respiratory failure, and death in about 10% of cases. Public health interventions interrupted viral transmission and it is no longer transmitting within humans, but the precursor virus remains in the animal reservoir (bats, Rhinolophus spp.) and may readapt to cause human disease in the future.

Introduction

Viral respiratory infections are one of the most common afflictions of humankind. They are the most frequent reasons for medical consultations, are believed to account for 30% of work absences and school absenteeism, and are a major reason for antibiotic prescriptions. Longitudinal family studies suggest that a person has on average 2.4 respiratory viral infections per year, a quarter of them leading to a medical consultation. The synergistic interaction between viruses and bacteria in pathogenesis are being increasingly recognized, for example that between influenza virus and Streptococcus pneumoniae and Staphylococcus aureus. With the exception of influenza in elderly people, these viral infections are not a major cause of mortality in otherwise healthy people in the developed world, but it is estimated that they contribute to over 1 million deaths annually in the developing world.

The term ‘respiratory virus’ is imprecise, but for the purpose of this discussion it will include those that have the respiratory tract as their primary site of clinically relevant pathology. Taxonomically, they belong to six virus families (Table 1 below) and are global in distribution. Other viruses cause systemic disease with respiratory tract involvement as part of an overall disseminated disease process in patients who are immunocompetent (e.g. measles, Hantavirus pulmonary syndrome) or immunocompromised (e.g. cytomegalovirus). These are dealt with elsewhere.

A respiratory virus may cause a range of clinical syndromes. Conversely, a respiratory syndrome may be caused by more than one virus. The major viral respiratory syndromes and their common aetiological agents are shown in Table 2 below. The pattern seen in tropical countries is similar, but a notable difference is the role of measles as a major cause of lower respiratory tract infections and fatality.

The anatomical demarcation between upper (URTI) and lower respiratory tract infections (LRTI) is the larynx. Influenza, respiratory syncytial virus, parainfluenza virus and adenoviruses are well-recognized causes of LRTI in adults as well as in children, although many other respiratory viruses may do so occasionally. Severe acute respiratory syndrome coronavirus (SARS CoV) and avian influenza H5N1 are unusual in that lower respiratory manifestations predominate over the involvement of the upper respiratory tract.

With newer molecular-based approaches to pathogen discovery, new respiratory viruses continue to be recognized. Some recently recognized viruses have been long endemic in humans (e.g. human metapneumovirus, coronavirus NL-63, HKU1, bocavirus) while others are novel pathogens, newly emergent as causes of human infections such as SARS and avian flu H5N1.

Table 1 Respiratory tract viruses: summary of classification, incubation period, duration of infectivity, and diagnostic options
Virus Classification (virus family) and composition of virus Subgroups, serotypes, and subtypes Incubation period (days) Duration of virus shedding in immunocompetent patients (days) Options for laboratory diagnosisa
Rhinovirus
  • Picornaviridae
  • Nonenveloped RNA viruses
>102 serotypes phylo-genetically divided into 3 groups A, B and C 1–2 days 5–6 days by culture; 50% remain positive by RT-PCR 2 weeks later RT-PCR or viral culture (less sensitive)
Enterovirus
  • Picornaviridae
  • Nonenveloped RNA viruses
65 serotypes Few days Up to 2 weeks from respiratory tract, much longer in faeces RT-PCR. Viral culture less sensitive and not possible for some types unless animal inoculation is used.
Coronavirus
  • Coronaviridae.
  • Enveloped RNA viruses
5 types (OC43, 229E, NL-63, HKU-1, SARS CoV 4–5 days 5–8 days RT-PCR
Respiratory syncytial virus (RSV)
  • Paramyxoviridae
  • Enveloped RNA virus
Subgroup A and B 5 days 6–7 days
  • Culture
  • Rapid antigen detection,a RT-PCR
  • Serology: useful in adults but less so in infants
Human metapneumovirus
  • Paramyxoviridae
  • Enveloped RNA virus
Serotypes A and B ND ND
  • RT-PCR
  • Viral antigen detection
Parainfluenza
  • Paramyxoviridae
  • Enveloped RNA virus
Type 1, 2, 3, 4a, 4b 3–6 days 7 days
  • Culture
  • Rapid antigen detection,a RT-PCR
  • Serology: useful in adults but less so in infants
Influenza
  • Orthomyxoviridae
  • Enveloped RNA virus
  • Types A, B, C
  • Human influenza A subtypes currently in circulation are H1N1 and H3N2
  • Average 2–3
  • (range 1–7)
  • c.5 days in adults
  • c.7 days in children
  • Culture
  • Rapid antigen detection2, RT-PCR
  • Serology
Adenovirus
  • Adenoviridae
  • Nonenveloped DNA virus
  • Subgroups A–F
  • Types 1–51
  • Average 10
  • (range 2–15)
Days–weeks (from respiratory tract), weeks–months (in faeces)
  • Culture
  • Rapid antigen detection,a RT-PCR, Serology
Bocavirus
  • Parvoviridae
  • Nonenveloped DNA virus
One phylogenetic group ND ND RT-PCR

ND, not defined.

a Best sensitivity from nasopharyngeal aspirates or nasopharyngeal swabs (in that order). Throat swabs give lower sensitivity.

Table 2 Viral aetiology of common respiratory syndromes

 

Virus Coryza Pharyngitis Croup Bronchiolitis Pneumonia
Rhinovirus +++a ++ + + Rare
Coronavirus ++ + + (NL-63)   SARS CoV, HKU-1
Adenoviruses (+) ++ ++ ++ ++ (all ages)
RSV ++ + ++ +++
  • +++ (children);
  • + (elderly)
Human metapneumovirus + + + ++ ++ (children)
Parainfluenza 1 + ++ +++ +  
Parainfluenza 2 + ++ ++ +  
Parainfluenza 3 + ++ ++ ++ ++ (children)
Influenza A/B + ++ ++ + ++ (all ages)

a Frequency of cases caused by the virus: +++ the major cause (>25%); ++ a common cause (5–25%); + an occasional cause; blank, rare cause or not reported.

(Data adapted from Treanor 2009).

Transmission

The routes of respiratory virus transmission are through direct contact, contaminated fomites, and large airborne droplets (mean diameter >5 µm, range of transmission <1 m). There remains controversy over the potential for the spread of viruses such as influenza over longer distances by small particle aerosol (mean diameter <5 µm), but even here, large droplets, direct contact, and fomites are probably more important. Occasionally, SARS CoV appears to have spread by small particle aerosols, although droplets and fomites probably contributed to the major part of the transmission of this disease. Adenoviruses are transmitted by the faeco-oral route as well as by direct contact and large droplets.

Factors increasing transmission of respiratory viruses include the time of exposure, close contact (e.g. spouse, mother), crowding, family size, and lack of pre-existing immunity (including lack of breastfeeding). School-age children often introduce an infection into the family and the beginning of school term may affect transmission patterns in the community. Infected children shed higher titres of viruses than adults. The duration of virus excretion is shown in Table 1. Infectivity usually precedes the onset of clinical symptoms. Immunocompromised patients shed virus for a longer time.

Seasonality

Some respiratory viruses have a predictable seasonality, which varies regionally. For example, influenza A is a typically winter disease in temperate regions, a spring/summer disease in the subtropics (e.g. Hong Kong) and occurs all year round (e.g. Singapore) or predominantly in the rainy season (e.g. Thailand) in the tropics. The basis for such seasonality is unclear, but climatic factors such as high humidity and temperature may help virus survival in small particle aerosols or droplets, and on contaminated surfaces. Factors affecting population congregation such as commencement of school term and seasonal effects on social behaviour may also play a role.

Laboratory diagnosis

A well-collected specimen is the first and often most important determinant in successful laboratory diagnosis. Nasopharyngeal aspirates (secretions aspirated from the back of the nose into a mucus trap), nasopharyngeal washes, and nasopharyngeal swabs are superior to throat and nose swabs for the diagnosis of many respiratory viruses. They offer the advantage that rapid (‘same day’) diagnosis for a number of viruses is possible provided the appropriate methods are available. Swabs for viral culture are placed in viral transport medium immediately upon collection and kept cool (around 4°C) until processed. More invasive specimens such as endotracheal aspirates, bronchoalveolar lavage, or lung biopsy, when available, usually provide better information. However, the likely site of pathology must be kept in mind—the more invasive specimen is not always better.

Laboratory methods used for detecting a virus in clinical specimen/s are viral culture, antigen detection, and, more recently, nucleic acid detection (e.g. polymerase chain reaction (PCR)-based methods). The widespread use of molecular methods for viral detection has led to recognition that some viruses that are difficult to culture (e.g. coronaviruses and some rhinoviruses and enteroviruses) are found more often in patients with acute respiratory disease than previously recognized. Similarly, these methods have allowed the discovery of novel viruses associated with respiratory disease (e.g. coronaviruses NL-63, HKU1, bocavirus). They have also revealed that infection with multiple viruses is relatively common. These findings necessitate a reassessment of the clinical relevance of positive PCR results. Relevant questions include how commonly these viruses are detectable by these methods in age-matched healthy controls and how long viruses remain detectable after infection. It is important to understand the relevance of detection of multiple pathogens in a respiratory specimen. Are these viruses synergistic in pathogenesis or is one more important than another? Many of these questions remain to be resolved.

Demonstration of rising antibody titres in paired sera is used to diagnose some respiratory virus diseases, but serology is impracticable for others such as rhinoviruses where the large number of antigenically distinct serotypes have no common immunodominant antigen(s). However, adenoviruses and influenza viruses, though having many antigenic types or variants, have common antigen(s) and a single antigen can detect serological responses to many of them. IgM assays are not routinely available for diagnosis of respiratory viral diseases. Serology is also helpful in assessing the clinical relevance of a virus detected in a respiratory specimen (see above) by helping differentiating recent infection from more remote events.

‘Near patient testing’ is becoming a reality for some viruses (e.g. influenza, RSV) with availability of tests that can be performed in a general practice setting. These become more relevant with the greater availability of antiviral drugs.

Rhinoviruses

Rhinoviruses belong to the Picornavirus family and are adapted to replicate at temperatures of 33–35°C, as found in the external airways. Until recently, 102 serotypes of rhinoviruses were recognized phylogenetically clustered into two groups A and B. Recent studies have revealed at least one additional phylogenetic group (group C) and many more rhinovirus types. But only a few rhinovirus types will circulate in a region at any given time.

Epidemiology

Rhinoviruses remain one of the commonest infections of humans: 0.5 infections per person per year is a conservative estimate. Secondary attack rates in families may be around 50% overall and 70% in those who are antibody negative. They were thought to cause mainly mild community infections, but are being recognized increasingly as the commonest viral agent detected by RT-PCR in children hospitalized with acute respiratory illness. Many of these represent coinfections with other potential respiratory pathogens. As rhinoviruses are often detectable by RT-PCR for weeks after initial infection (50% remain positive at 2 weeks), the aetiological significance of this finding is unresolved and more studies with relevant control populations are needed.

Immunity

In experimental challenges, immunity is serotype specific. Homologous type specific protection lasts for at least 1 year and correlates with serum IgA, IgG, and secretory IgA antibody levels.

Pathogenesis

Viral replication occurs predominantly in the ciliated epithelial cells of the nasopharynx. The structure of the epithelium is preserved. Mucosal secretions associated with coryza appear to be due to the release of inflammatory mediators and neurogenic reflexes.

It was thought that the preference of the virus for a lower temperature for replication restricted it to the upper respiratory tract. However, this is not strictly true. The virus has been isolated from the lower respiratory tract (including bronchial brushings) and viral RNA has been demonstrated by in situ hybridization in bronchial epithelial cells. Rarely, the virus has been isolated post-mortem from lungs of immunocompromised patients.

Clinical manifestations

Rhinorrhoea, nasal obstruction, pharyngitis, and a cough are common features of rhinovirus infections. Fever and systemic symptoms are rare, but more common in the elderly in whom disease can be more severe. Rhinoviruses are a major cause of exacerbations of asthma and chronic obstructive respiratory disease in adults. Lower respiratory tract symptoms are uncommon in healthy young adults, but may occur in children (bronchiolitis), the immunocompromised, and the elderly. Rhinovirus infections associated with wheezing in the first 3 years of life is predictive of asthma in later childhood.

Treatment and prevention

There are no established antiviral drugs for treatment and management is symptomatic. Topical interferon-α prevents symptoms if given before onset of disease, but cannot be used for prophylaxis over prolonged periods because of side effects. Pleconaril is a viral capsid-binding agent that blocks viral attachment and uncoating and has had modest benefit in clinical trials, but concerns over side effects have prevented its licensing. Antibiotics are ineffective in preventing bacterial complications of the common cold. Mucopurulent discharges are part of the natural course of the common cold and are not an indication for antimicrobial treatment, unless it persists (e.g. >10 days). Given the large number of rhinovirus serotypes, vaccination is not an option.

Enteroviruses

Enteroviruses and rhinoviruses (see above) are genera within the family Picornaviridae. Enteroviruses have long been known as causes of central nervous system infections, myocarditis, or exanthema rather than as a respiratory pathogen, the latter role being assigned to rhinoviruses. As many enteroviruses fail to replicate in cell culture, the wider use of molecular diagnosis has revealed an increased role of enteroviruses in acute respiratory infections. Clinically, patients present with rhinitis, cough, fever, sore throat, or otitis media. There remains a need for studies of age-matched controls to better establish the clinical relevance of these molecular tests. In comparative studies done on the duration of shedding of enteroviruses and rhinoviruses, fewer enterovirus infected children continue to shed virus for longer than 2 weeks while 50% of rhinovirus infections do. This suggests that a positive enterovirus RT-PCR result in the respiratory tract is probably more likely to be clinically relevant than one for rhinovirus.

Coronaviruses

Five human coronaviruses are currently known, three of them being new viruses discovered since the SARS outbreak in 2003. Coronaviruses are taxonomically subdivided into three groups and the human coronaviruses 229E and NL-63 belong to group 1 while OC43, HKU1, and SARS CoV belong to group 2. There are no known human group 3 coronaviruses. Human coronaviruses OC43 and 229E have long been recognized as important causes of the common cold but coronaviruses cause a range of respiratory illnesses. SARS CoV is a newly emerged pathogen. Human coronaviruses are difficult to culture from clinical specimens and laboratory diagnosis largely relies on molecular methods.

Epidemiology

Infection with OC43 and 229E occur in early childhood and 85 to 100% of adults have antibody to both virus types. NL-63 has a similar epidemiology but less is presently known of HKU1. SARS CoV emerged from an animal reservoir, adapted to human transmission and caused a global outbreak in 2003 that affected 29 countries across 5 continents. However, determined public health interventions interrupted transmission of this virus and it is no longer transmitting within humans. However, the precursor virus remains in the animal reservoir (bats, Rhinolophus spp.) and these may at some future date, readapt to cause human disease.

Immunity

Volunteer reinfection studies with 229E show that 1 year after initial infection, protection from reinfection and illness following a challenge from the homologous virus is incomplete. Comparable data are not available for the newly recognized NL-63, HKU1, or SARS CoV.

Pathogenesis

In common with rhinoviruses, coronaviruses 229E induce little or no damage to the respiratory mucosa. The mucosal discharge is caused by the release of mediators from affected host cells. SARS CoV had a predilection to involve alveolar pneumocytes in the lower respiratory tract and consequently caused a severe viral pneumonia. Disease severity of SARS was markedly age related. Children had mild disease whereas those over 50 years had a poor prognosis. The basis for this age-related pathogenesis is unknown. The virus receptor for 229E is CD13, while both SARS CoV and NL-63 utilize the human ACE-2 molecule for virus entry.

Clinical findings

Coronaviruses 229E and OC43 typically cause URTI and the common cold but also cause a range of other respiratory manifestations and are significant pathogens in elderly people. NL-63 and HKU1 cause both upper and lower respiratory disease. NL-63 appears to be an important cause of croup, bronchiolitis, and pneumonia. HKU1 appears to be an important pathogen particularly in those with underlying respiratory complications.

SARS typically presented with lower respiratory tract manifestations and radiological changes with minimum involvement of the upper respiratory tract. Many patients had diarrhoea resulting from viral replication in the gastrointestinal tract. Overall case fatality was 9.6%. Terminal events were severe respiratory failure associated with acute respiratory distress syndrome (ARDS) and multiple organ failure. Autopsies showed diffuse alveolar damage corresponding to the clinical presentation of acute respiratory distress syndrome. Age, comorbidities, and viral load in the nasopharynx and serum during the first 5 days of illness correlated with an adverse prognosis.

Treatment and prevention

There are presently no clinically validated antiviral treatments for human coronaviruses disease, although a number of drugs have been documented to have in vitro activity against SARS CoV. A number of experimental vaccines were developed for SARS, but with its disappearance from the human population, the incentive to take these forward to human clinical trials and licensing has waned.

Adenoviruses

Currently there are 51 adenovirus types classified in six groups (A–F). Adenoviruses in subgroups A to D cause respiratory, ocular, hepatic, genitourinary, or gastrointestinal system disease in immunocompetent or immunocompromised individuals. Only respiratory diseases are considered here.

Productive replication and excretion of infectious virus can occur for a prolonged period (see below). In addition, adenoviruses can establish chronic persistence or ‘latency’, the virological basis and clinical significance of which is poorly understood.

Epidemiology

Adenovirus infections are common during childhood (usually serotypes 1, 2, 5 in early childhood, 3 and 7 during school years or later), but continue to occur throughout life. Reinfection with the same serotype occurs but is usually asymptomatic. Serotypes 1, 2, 5, and 6 are typically endemic, types 4 and 7 more typically associated with outbreaks, and type 3 can occur in either situation. Recently, adenovirus 14 has been spreading in the United States of America.

Clinical features

Adenovirus respiratory illness often leads to URTI with coryza and sore throat. Fever may last up to 2 weeks. The sore throat may be exudative and clinically difficult to differentiate from streptococcal infection. Adenoviral infection may present as pharyngoconjunctival fever. Otitis media is a complication in children. Unlike other respiratory viral infections, adenoviruses may be associated with elevated white blood cell counts (exceeding 15 × 109/litre), C-reactive protein, or ESR and thus more easily confused with bacterial diseases.

Though uncommon, pneumonia may occur sporadically or in epidemics (e.g. caused by serotypes 4 and 7), particularly in closed communities such as the military where stress and physical exertion may predispose to lower respiratory tract involvement. Community outbreaks of adenoviral pneumonia have been reported. Radiological appearance varies from diffuse to patchy interstitial infiltrates and pleural effusion may be present. Adenovirus type 7 pneumonia can lead to permanent lung damage, including bronchiectasis, bronchiolitis obliterans, and unilateral hyperlucent lung syndrome.

Adenoviral infection may disseminate and present as ‘septic shock’ in neonates. Manifestations in immunocompromised patients include hepatitis (especially in liver transplant recipients), colitis,and haemorrhagic cystitis (in renal and bone marrow transplant recipients) in addition to pneumonia. The serotypes associated with disease in these patients may differ from those typically found in the immunocompetent patient, and include the subgroup B2 serotypes 11, 34, and 35. With improving control of other common viral diseases of immunocompromised patients (e.g. cytomegalovirus), the role of adenovirus infections is being increasingly appreciated.

Isolation of an adenovirus from a clinical specimen presents a challenge in interpretation. Adenoviruses are excreted for a prolonged period after initial infection, especially, but not exclusively, from faeces. In children, one-third of patients shed viruses for longer than 1 month and 14% longer than 1 year. The clinical significance of a positive result depends on the specimen, the method, and the serotype. Isolation of viruses from the respiratory tract carries greater significance than that from faeces. Patients who have symptomatic adenoviral diseases have higher viral loads than those with asymptomatic carriage. Thus, a rapidly growing virus, a positive antigen detection test from a respiratory specimen (both reflecting higher virus load), or a detectable serological response all point to greater clinical significance.

Immunocompromised patients may be infected with unusual serotypes. The detection of the virus in the peripheral blood or in multiple body sites suggests greater clinical significance and is an indication that therapeutic intervention needs to be considered.

Treatment and prevention

Most adenoviral infections in immunocompetent patients are self-limited and require no specific therapy; however, some infections, especially but not exclusively in immunocompromised patients, are severe and life threatening. Ribavirin, vidarabine, cidofovir, and ganciclovir are active against adenoviruses in vitro. Although there are anecdotal reports of the therapeutic use of each of these drugs with variable success, on the basis of limited clinical studies cidofovir appears to be the antiviral of choice.

Live attenuated oral vaccines containing serotypes 4 and 7 (associated with outbreaks in military conscripts) are safe and effective, but not licensed for general use.

Respiratory syncitial virus

Respiratory syncytial virus (RSV) infects human and nonhuman primates and was first isolated from a chimpanzee with a ‘cold’. The virus has two surface glycoproteins on its envelope (G and F) and the immune responses to them correlate with protection. Two subgroups (A and B) are recognized on the basis of antigenic differences of the G glycoprotein.

Epidemiology

Over two-thirds of infants acquire RSV infection during the first year of life. Of patients hospitalized with RSV disease, 75% are younger than 5 months. The peak of morbidity occurs around 2 to 4 months of age, a time when passive maternal antibodies protect against most other viral infections. Primary infection does not lead to solid immunity and reinfection is common. The first reinfection can still be associated with lower respiratory tract involvement. Subsequent reinfection occurs throughout life leading to asymptomatic or URTI. However, significant diseases may result in the immunocompromised or elderly.

Immunity

Both antibody and cell mediated immunity are important in protection. Antibody to the G protein prevents attachment of viruses to the cellular receptor, but immunity to the F protein is required to prevent cell to cell spread via fusion of virally infected cells. Cell mediated immunity is important in eliminating established viral infection.

Pathogenesis

The virus leads to a ballooning degeneration of the ciliated epithelial cells, lymphocytic infiltration, and necrosis of the epithelium. There is oedema and increased secretion from the mucous cells and the formation of plugs of mucous and cellular debris in the bronchioles. This results in obstruction and air trapping leading to collapse or over-distension of the distal alveoli. Cells throughout the respiratory tract are affected but the alveoli are spared unless there is RSV pneumonia. The pathogenesis of RSV bronchiolitis still remains controversial.

Severe RSV bronchiolitis is strongly associated with subsequent childhood asthma. RSV appears to promote type 1 hypersensitivity responses following subsequent exposure to unrelated antigens.

Clinical features

RSV infections of infants may lead to bronchiolitis and pneumonia. Bronchiolitis in infants is associated with expiratory wheeze, subcostal recession, hyperinflation of the chest, nasal flaring, and hypoxia with or without cyanosis. Fever is not prominent in one-half of the patients. Complete obstruction of a small airway leads to subsegmental atelectasis. Apnoea may occur (particularly in premature infants or in those <3 months of age) and may precede the development of bronchiolitis. Interstitial pneumonitis is uncommon but carries a bad prognosis. Otitis media is a common complication of RSV infection in children. Infants at highest risk from severe RSV disease are those <6 months, those with pre-existing congenital heart disease, chronic lung diseases (e.g. bronchopulmonary dysplasia), and those born premature.

Infection in adults is often asymptomatic or leads to URTI. However, during the RSV season, it is an important cause of LRTI in adults and elderly people and it is estimated to cause 2 to 9% of the hospitalizations and deaths associated with pneumonia in elderly individuals. Much of this morbidity is clinically indistinguishable from influenza.

RSV (as well as parainfluenza and influenza) infections in the immunocompromised patient can be life threatening. They usually occur during community outbreaks, but a significant proportion are nosocomially acquired. The disease typically commences as an URTI but may progress to involve the lower respiratory tract with more serious consequences. Factors that increase risk of disease progression appear to include bone marrow transplant recipients who acquire the infection in the period prior to engraftment and oncology patients with neutrophil counts less than 0.5 × 109/litre. Those immunocompromised by HIV appear to tolerate community acquired respiratory viruses better than oncology patients and transplant recipients.

Treatment and prevention

Ribavirin has activity against RSV in vitro. Administration of small particle aerosols via a mist tent, mask, oxygen hood, or ventilator has been recommended because it results in much higher concentrations in the respiratory tract than can be achieved by intravenous administration. There seems little therapeutic benefit of ribavirin therapy in RSV disease in immunocompetent children or adults. However, in patients at high risk for severe RSV disease such as adult bone marrow transplant recipients, an uncontrolled study of ribavirin together with intravenous immune globulin (selected batches with high neutralizing antibody titre) appeared to be beneficial when compared to historical controls. More information is required for deciding the best management strategy.

Monthly intravenous administration of a polyclonal immune globulin enriched in neutralizing antibodies to RSV (RespiGam) or a humanized monoclonal antibody to RSV (palivizumab) during the RSV season protects against disease of the lower respiratory tract and otitis media in children with pre-existing risk factors. Palivizumab appears to be more effective than RespiGam and there is less of a problem with fluid overload in children with chronic heart disease. High-titre RSV intravenous immunoglobulin by itself is ineffective in treatment of established RSV disease.

Candidate vaccines for RSV are undergoing clinical trials at present but none is yet available for routine use. Experience of early trials with inactivated RSV vaccines that led to enhanced RSV disease, rather than protection continues to haunt the field.

Parainfluenza virus

Parainfluenza viruses, despite their name, are not related to influenza viruses, and are more akin to respiratory syncytial virus with which they are classified (Table 1). They carry two envelope glycoproteins: HN containing both haemagglutinin and neuraminidase activity and F carrying fusion activity.

Epidemiology

The total impact on hospitalization of children by all four types of parainfluenza viruses taken together is similar to that of RSV but, in contrast to RSV, their impact is in later infancy and childhood. In temperate countries, parainfluenza virus type 3 occurs annually and infects two-thirds of all infants in their first year of life. Parainfluenza types 1 and 2 tend to occur in alternate years and infection is acquired more slowly over childhood. Reinfection with parainfluenza viruses occurs, but rarely leads to LRTI.

Pathogenesis

The virus is confined to the respiratory epithelial cells, macrophages, and dendritic cells within the respiratory tract. Dissemination is rarely documented even in immunocompromised patients.

Immunity

Reinfection with parainfluenza viruses continues throughout life. Presence of virus-specific IgE in nasopharyngeal secretions has been implicated in the development of parainfluenza croup or bronchiolitis.

Clinical features

Parainfluenza type 1 predominantly causes croup, while types 2 and 3 also cause bronchiolitis and pneumonia. Croup (or laryngotracheobronchitis) in children is associated with fever, hoarseness, and a barking cough and may progress to inspiratory stridor due to narrowing of the subglottic area of the trachea. The differential diagnosis is epiglottitis due to Haemophilus influenzae type b. Parainfluenza type 4 infection is less common, but causes bronchiolitis and pneumonia in children, often in those with underlying disease.

Reinfection in adults, when symptomatic, is a coryzal illness with hoarseness being prominent. Parainfluenza viruses (type 3 in particular) are significant causes of LRTI in adults when the virus is active in the community.

As with RSV, parainfluenza viruses cause problems in immunocompromised patients. Lower respiratory tract involvement is associated with wheezing, rales, dyspnoea, and diffuse interstitial infiltrates, and a fatal outcome in one-third of patients with allogenic bone marrow transplants. When pneumonia occurs, the histological appearance of the lung is that of a giant cell or an interstitial pneumonia.

Treatment and prevention

The need for specific antiviral therapy arises, particularly in the immunocompromised. Ribavirin is effective in vitro and was associated with a reduction of viral replication in vivo in anecdotal cases but there are no controlled trials documenting its clinical efficacy.

There are no options for prevention at present, either using vaccines or passive immunization. A live attenuated bovine-derived vaccine strain is currently undergoing clinical trials.

Human metapneumovirus

Human metapneumovirus (HMPV) belongs to the genus Metapneumovirus within the virus family Paramyxoviridae, subfamily Pneumovirinae. It closest known relative is the avian pneumovirus, an upper respiratory tract disease of turkeys and among human viruses is RSV which also belongs to the subfamily Pneumovirinae. It was first recognized in 2001 but is a virus that has circulated unrecognized in humans for many decades. There are at least two serotypes A and B which are antigenically distinct and appear to provide partial cross-protection.

Epidemiology

The virus is ubiquitous and most children have been infected with one or both serotypes by the age of 5 years. Symptomatic reinfection is common through life. Infection is commonest in the winter months in temperate regions and in late spring or summer in subtropical areas.

Clinical manifestations

HMPV is one of the common causes of hospitalization of children under 5 years of age and accounted for 12% of all LRTI hospitalization in one long-term study. However, the incidence in any given year may vary widely. The peak age for HMPV morbidity is between 6 and 12 months, which is later than that for RSV (2–4 months). Clinical features of HMPV are similar to that of RSV and range from URTI to bronchiolitis and pneumonia. In common with rhinovirus and RSV, HMPV appears to trigger exacerbations of asthma. Diarrhoea, vomiting, rash, febrile seizures, conjunctivitis, and otitis media have been reported. HMPV has on one occasion been isolated as the sole pathogen from the brain in a patient with encephalitis.

HMPV can cause respiratory disease in elderly or immunocompromised individuals, and those with underlying conditions at any age.

Since HMPV is difficult to grow in vitro, laboratory diagnosis is reliant on the detection of viral RNA in clinical specimens by molecular methods.

Treatment and prevention

There are currently no available vaccines. As with RSV, the F and G proteins are the main targets of the neutralizing antibody response and while the former is antigenically conserved, the latter is more variable. Thus the F protein has been the focus of vaccine development. Ribavirin has comparable in vitro activity against HMPV as against RSV but there is no clinical trial data that demonstrates therapeutic efficacy.

Influenza viruses

Influenza viruses contain a segmented RNA genome. Types A, B, and C are antigenically distinct; of these, types A and B are clinically important causes of human disease. The viral envelope contains two glycoproteins, the haemagglutinin (H) and neuraminidase (N) which are critical in host immunity. The M2 transmembrane protein is also found on the virion surface but does not appear to elicit a significantly protective host response following natural infection. Human influenza viruses are designated by the virus type, place of isolation, strain designation, year of isolation, and the H and N antigen subtype, e.g. A/Sydney/5/95 (H3N2).

Epidemiology

The H and N genes of influenza types A and B undergo mutational change resulting in the emergence of antigenic variants (‘antigenic drift’). Every few years, a variant successful in evading the prior immunity of the human population emerges, to cause a global epidemic. Influenza viruses have a marked winter seasonality in temperate regions, making the disease burden of the virus more obvious. The more diffuse seasonality in tropical and subtropical regions leads to an obscuring of the clinical impact of the virus, leading to the illusion in some quarters that influenza is less significant in warmer climates. However, careful epidemiological studies demonstrate that the burden of mortality and morbidity in temperate and tropical regions are very similar. In those 65 years or older, influenza is associated with approximately 1 excess death per 1000 population annually in both the temperate and tropical regions.

In aquatic birds, the natural reservoir of the virus, 16 H and 9 N subtypes of influenza A are found. From 1918 till 1957, human influenza A viruses carried H1N1 surface antigens. In 1957, this virus acquired the novel H, N, and additional polymerase gene (PB1) from an avian influenza virus through genetic reassortment of its segmented genome giving rise to the H2N2 subtype virus (‘antigenic shift’). As the human population lacked immunity to these novel viral antigens, this led to the ‘Asian flu’ pandemic. A similar reassortment event gave rise to the H3N2 virus and the ‘Hong Kong influenza’ pandemic of 1968. In contrast, the pandemic of 1918 is believed to have arisen by the direct adaptation of an avian influenza virus without reassortment with the pre-existing human influenza virus. Although all three influenza pandemics of the 20th century resulted in significant morbidity and mortality, the toll exacted by the ‘Spanish flu’ of 1918 was particularly horrendous—over 40 million deaths, greater than that of both World Wars combined. Since influenza B (and C) have no significant zoonotic reservoirs, antigenic shift and pandemics do not occur.

In early 2009, a novel H1N1 virus of swine-origin gave rise to the first pandemic of the 21st century. The pandemic arose in Mexico and rapidly spread worldwide along routes of air-travel. Unlike the two previous pandemics (1957, 1968) that arose through genetic reassortment of an avian virus with the prevailing human seasonal influenza virus, the pandemic virus of 2009 arose through reassortment between swine viruses previously documented in North America (so called ‘triple reassortant’ swine viruses that contained virus gene segments of swine, avian and human origin) and ‘Eurasian-swine’ viruses. Although the H1 haemagglutinin of both human and swine influenza viruses was originally derived from the 1918 ‘Spanish flu’ H1N1 virus, they had antigenically diverged during their subsequent evolution in these two hosts so that the contemporary seasonal human H1N1 virus offered little cross-protection against the pandemic H1N1 virus of swine-origin. However, people born prior to the 1950s had substantial cross-protection against the novel pandemic virus, presumably derived by infection with H1N1 viruses circulating in the first half of the 20th century. Thus the pandemic was associated with explosive outbreaks in children and young adults while there was less infection in older adults. The disease was largely a mild-influenza-like illness comparable with seasonal influenza, sometimes associated with gastrointestinal symptoms of diarrhoea and vomiting. However, complications, severe illness, and fatalities did occur, especially in those who were pregnant or with underlying comorbidities including asthma and other lung disease, cardiovascular diseases, diabetes, neurological disorders, autoimmune disorders, and morbid obesity. While some of those with severe disease had secondary bacterial infections, others developed a primary viral pneumonia leading to acute respiratory distress syndrome.

Avian viruses (e.g. subtype H5N1, H9N2, H7N7) can zoonotically infect humans occasionally without undergoing prior reassortment with existing human strains. Currently, an H5N1 virus that is highly pathogenic for chickens has become entrenched in poultry flocks in a number of Asian and African countries and continues to zoonotically transmit to humans, often causing severe disease and pandemic concern. However, such transmission has so far not led to sustained human-to-human transmission that is the prerequisite for the generation of a new pandemic.

Pathogenesis

Viral replication occurs in the columnar epithelial cells leading to its desquamation down to the basal cell layer. The pathology typically involves the upper respiratory tract and the tracheobronchial tree. Infection results in decreased ciliary clearance, impaired phagocyte function, and increased adherence of bacteria to viral infected cells, all of which promote the occurrence of secondary bacterial infection.

While there may be differences in viral virulence, pre-existing cross-reactive immunity is a major determinant in reducing disease severity. Virus dissemination outside the respiratory tract is uncommon with human influenza viruses. However, zoonotic infections with the avian H5N1 virus may disseminate, and virus has been often detected in the gastrointestinal tract and occasionally in the central nervous system.

Immunity

Infection by an influenza virus results in long-lived immunity to homologous reinfection. However, the continued antigenic change in the virus allows it to keep ahead of the host immune response. Cross-immunity to ‘drifted’ strains within the same H or N subtype may provide partial protection, but there is believed to be little cross protection between different subtypes. Local and systemic antibody responses and cytotoxic T cells contribute to host protection.

Clinical features

The severity of influenzal disease ranges from asymptomatic infection, through the typical influenza syndrome, to the complications of influenza. Although it cannot always be distinguished from other viral infections on clinical grounds, the typical influenza syndrome is relatively characteristic in the adult. It is associated with fever, chills, headache, sore throat, coryza, nonproductive cough, myalgia, and sometimes prostration. The onset of illness is abrupt and the fever lasts 1 to 5 days. The pharynx is hyperaemic but has no exudate. Cervical lymphadenopathy is often present and crackles or wheezing are heard in around 10% of patients. While the acute illness usually resolves in 4 to 5 days, cough and fatigue may persist for weeks thereafter.

Common (>10% of symptomatic patients) complications of influenza include otitis media (in children) and exacerbation of asthma, chronic airways obstruction, and cystic fibrosis. Less common complications are acute bronchitis, primary (viral) and secondary (bacterial) pneumonia, myocarditis, febrile convulsions, encephalopathy, encephalitis, and myositis (especially in patients with influenza B infection). Age, prior immunity, virus strain, the presence of underlying diseases, pregnancy, and smoking all influence morbidity and severity.

Treatment and prevention

Antiviral therapy

Antiviral drugs with proven clinical efficacy for treatment of influenza A are the ion channel (M2) blockers that interfere with viral uncoating (amantadine, rimantadine) and the neuraminidase inhibitors (e.g. zanamivir, oseltamivir) which block virus release from infected cells. The neuraminidase inhibitors are also active against influenza B, while amantadine and rimantadine are only active against influenza A.

Since 2003, seasonal H3N2 and H1N1 viruses increasingly acquired resistance to amantadine and rimantadine and the 2009 pandemic H1N1 virus is also resistant to these drugs. Thus they are no longer drugs of choice in the treatment or prophylaxis of human influenza. Oseltamivir resistance to seasonal H1N1 viruses emerged in early 2008 and spread worldwide. The pandemic H1N1 virus (as well as seasonal H3N2 viruses) remains sensitive to oseltamivir although resistant pandemic H1N1 viruses have been occasionally reported. Zanamivir remains uniformly effective against seasonal and pandemic influenza viruses.

Zanamivir is administered by inhalation and oseltamivir orally. In patients infected with viruses sensitive to these drugs, zanamivir or oseltamivir treatment commenced within the first 48 h of disease onset leads to a 1 to 2 days reduction in the time to alleviation of clinical symptoms and also reduces incidence of influenza associated complications. Some studies have indicated benefit in reducing the complications of influenza even for patients in whom treatment commenced after the second day of clinical illness.

However, the sooner the drugs are used, the better the chance of clinical benefit. With a virus such as the highly pathogenic H5N1 virus which can disseminate beyond the respiratory tract, a systemically administered drug (oseltamivir) is likely to be superior to one administered by inhalation (zanamivir). However, oseltamivir has had variable success in the treatment of H5N1 influenza and although therapeutic failure may be partly due to late commencement of therapy, poor drug bioavailability in a severely ill patient and emergence of resistance may also contribute. Parenteral therapy is ideal for such patients and such options (e.g. peramivir) are currently undergoing clinical trials.

Aspirin should be avoided in children with influenza because of the increased risk of Reye’s syndrome.

Vaccines

Influenza vaccine is a trivalent vaccine containing antigens from the two subtypes of human influenza A (H3N2 and N1N1) and B viruses. To keep abreast of change in the surface antigens of the virus, its composition must be modified on an annual basis and annual reimmunization is required. This updating of the vaccine is achieved through a collaborative effort of the global influenza virus surveillance network coordinated effort by the World Health Organization (WHO). As a result of this surveillance, the WHO makes recommendations of candidate vaccine viruses twice annually for vaccine production for the northern and southern hemispheres.

Vaccines currently in use are based on antigen derived from viruses grown in embryonated eggs or cell cultures and contain detergent-treated virus (split virus vaccines) or purified surface antigens (subunit of surface antigen vaccines). These vaccines have less side effects than killed vaccines containing the whole virus which were used in the past and are licensed for use in anyone 6 months of age or older. Previously unvaccinated children require two doses at least 1 month apart, whereas a single dose appears adequate for adults. These vaccines are generally safe, the most common side effect being soreness at the injection site lasting a few days. Vaccine efficacy is best when there is a good antigenic match between the vaccine and outbreak virus.

An intranasally administered, cold-adapted, live attenuated vaccine is now also licensed for use in those aged 5 to 49 years and offers the advantages of broader cross-protection across antigenic drifted viruses as well as easier administration and greater patient acceptability.

Inactivated and cold-adapted live attenuated monovalent vaccines containing the pandemic H1N1 were rapidly developed and used in 2009 in response to the pandemic. Some of these inactivated vaccines had adjuvents (MF59; AS03) added to enhance immune response. In 2010 and beyond, it is likely that the pandemic H1N1 virus will be included as one component of the seasonal influenza vaccine.

Immunogenicity and clinical protection are better in healthy young adults compared to patients with chronic renal failure and immunocompromised or elderly patients (all groups most at need of the vaccine). However, the vaccine is still effective in reducing influenza and pneumonia-related hospitalization and mortality in elderly people and is cost-saving. An additional option for protecting such high-risk individuals is the immunization of children and caregivers in contact with these individuals. In young adults, vaccination is associated with decreased absenteeism from work. The duration of protection is limited and therefore vaccine administration should be timed to precede the expected peak of influenza activity.

Influenza vaccine recommendations vary from country to country. In general, vaccine is recommended to those groups at highest risk of influenza related complications including (1) those aged 6 months to 5 years of age; (2) those aged 65 years or older (in the United States of America all those over 50 are recommended for vaccination); (3) pregnant women who will be in the second or third trimester during the influenza season; and (4) those with chronic medial conditions including persons with chronic disorders of pulmonary or cardiovascular systems (except hypertension), those with renal dysfunction, haemoglobinopathies, metabolic disorders, or immunodeficiency and those aged 6 months to 18 years who are on long-term aspirin therapy. Furthermore, vaccine is also recommended for health care workers and for persons living or caring for those at high risk, who may transmit influenza to such high-risk individuals.

Bocavirus and polyomavirus K1 WU

Human bocavirus is a member within a newly discovered genus Bocavirus within the family Parvoviridae. As with other parvoviruses, they are relatively resistant to inactivation by acid or alkaline pH or moderate heat (e.g. 56°C). Molecular detection by PCR in respiratory clinical specimens is the main option for diagnosis. The virus can also be sometimes detected in serum. Using these methods, it is one of the five most commonly detected viral agents in respiratory specimens from children with acute respiratory disease. However, relatively few studies have included age-matched healthy controls to assess the clinical relevance of the detection of these agents and the available data is at present contradictory. The peak age of detection is in children aged 6 months to 2 years and occasionally in adults. These patients presented with rhinitis, a cough that is often paroxysmal or ‘pertussis-like’, and wheezing and were categorized as bronchiolitis, pneumonia, or asthma. Some patients also had diarrhoea, vomiting, and a skin rash. The virus has been detected worldwide and is likely to have been long endemic in humans. Reliable tests to study the seroepidemiology of this infection are still awaited.

KI and WU are two novel polyomaviruses recently discovered in the respiratory tract of patients with acute respiratory infections. There are found in a proportion of children and adults with acute respiratory infection but often found as coinfections with other known respiratory pathogens. Their contribution to disease causation is still unclear.

Nosocomial

Respiratory viruses are efficient nosocomial pathogens. Though paediatric units face the brunt of the problem, adult wards are not exempt. Transmission may occur from patient to patient, patient to staff, and staff to patient, with visitors making their own contribution. Although influenza and RSV are the most notorious among the endemic respiratory viruses, even rhinoviruses cause problems when transmitted to immunocompromised patients. Once infected, immunocompromised patients have a prolonged period of viral shedding and pose a significant risk of transmission to other high-risk patients.

Transmission of many respiratory virus infections occurs by large respiratory droplets gaining access to the mucosa of a susceptible individual. Large respiratory droplets have a relatively short dispersal range (<1 m). On the other hand, direct hand contact is an important means of transmission within health care settings and adherence to strict hand-washing is the most critical preventive measure. Gloves are useful in reinforcing the ‘hand-washing message’, but will only be effective if they are changed between patients. Cohorting infected patients, either by symptoms (during the outbreak season) or by rapid viral diagnostic results, is useful. Influenza A vaccination of health care workers, especially those caring for high-risk children, is to be recommended. Staff education is vital, including awareness of the fact that some of these viruses manifest themselves as a mild ‘cold’ in adults, and that infected staff members can transmit to patients under their care.

The most dramatic example of the impact of nosocomial transmission with a respiratory virus occurred with SARS where health care facilities served as a major hub of virus transmission and health care workers accounted for one-fifth of all documented cases. Much of this transmission was preventable by basic (large) droplet and contact precautions, although protection from small particle aerosols was important when carrying out aerosol-generating procedures such as intubation.

Further reading

Top

Abed Y, Boivin G (2006). Treatment of respiratory virus infections. Antiviral Res, 70, 1–16. [Reviews role of antiviral therapy for respiratory virus infections.]

Centers for Disease Control and Prevention (2007). Prevention and control of influenza: recommendations of the Advisory Committee on Immunisation Practices (ACIP). MMWR, 56, 1–54. Atlanta, GA. [Reviews the disease burden of influenza and the use of vaccines and antiviral therapy.]

Dolin R, Wright PF (eds) (1999). Viral infections of the respiratory tract. Marcel Dekker, Basel, pp. 1–432. [Comprehensive monograph with chapters on each of the respiratory viruses, antiviral therapy, and on infections in immunocompromised patients.]

Dowell SF (ed.) (1998). Principles of judicious use of antimicrobial agents for pediatric upper respiratory tract infections. Pediatrics, 101 Suppl, 163–84. [Journal supplement reviewing the use and abuse of antibiotics in upper respiratory tract infections.]

Falsey AR, Walsh EE (2006). Viral pneumonia in older adults. Clin Infect Dis, 42, 518–24. [Reviews role of virus in lower respiratory tract disease of adults.]

Gern JE, Busse WW (1999). Association of rhinovirus infections with asthma. Clin Microbiol Rev, 12, 9–18.

Kim YJ, Boeckh M, Englund JA (2007). Community respiratory virus infections in immunocompromised patients: hematopoietic stem cell and solid organ transplant recipients, and individuals with human immunodeficiency virus infection. Semin Respir Crit Care Med, 28, 222–42. [Reviews the management of respiratory viral infections in the immunocompromised patient.]

Madeley CR, Peiris JSM, McQuillin J (1996). Adenoviruses. In: Myint S, Taylor-Robinson D (eds) Viral and other infections of the human respiratory tract, pp. 169–90. Chapman & Hall, London. [Reviews the adenoviral respiratory disease and laboratory diagnosis.]

Mallia P, Johnston SL (2006). How viral infections cause exacerbation of airway diseases. Chest, 130, 1203–10.

Nicholson KG, Webster RG, Hay AJ (eds) (1998). Textbook of influenza. Blackwell Scientific, Oxford. [Comprehensive review of the ecology, clinical features, and control of influenza.]

Peiris JSM, De Jong MD, Guan Y (2007). Avian influenza virus (H5N1): a threat to human health. Clin Microbiol Rev, 20, 243–67. [Reviews the threat from emerging zoonotic and potentially pandemic influenza viruses.]

Peiris JSM, et al. (2006). Severe acute respiratory syndrome (SARS). In Scheld WM, Hooper DC, Hughes JM (eds) Emerging infections 7, pp. 23–50. ASM Press, Washington DC. [Reviews the epidemiology, clinical features, pathogenesis and management of SARS.]

Siddell S, Myint S (1996). Coronaviruses. In: Myint S, Taylor-Robinson D (eds) Viral and other infections of the human respiratory tract, pp. 141–67. Chapman & Hall, London.

Treanor J (2009). Respiratory infections. In: Richmond DD, Whitley RJ, Hayden FG (eds) Clinical Virology, 3rd edition, pp. 7–27. ASM Press, Washington, DC. [Reviews viral respiratory infections.]

van den Hoogen BG, Osterhaus ADME, Fouchier RAM. (2006). Human metapneumovirus. In: Scheld WM, Hooper DC, Hughes JM (eds) Emerging Infections 7. pp. 51–68, ASM Press, Washington, DC.