Serious Epidemic Viral Pneumonias
Daniel H. Libraty
There are a number of established, emerging, and reemerging viruses that can lead to severe respiratory illness in immunocompetent individuals. The etiologic agents of serious viral pneumonias can generally be divided into three groups:
Human-adapted respiratory viruses. The primary site of entry, replication, and disease for these viruses is the human respiratory tract. They are spread efficiently by person-to-person transmission. The most significant members of this group are the human influenza A and B viruses; others are respiratory syncytial virus (RSV) and adenovirus.
Human-adapted viruses—respiratory disease after a viremic phase. Viral entry and person-to-person spread of these viruses is via the respiratory tract. However, these viruses cause respiratory illness after a phase of systemic viral replication and dissemination. Members of this group include varicella zoster virus (chickenpox) and rubeola virus (measles).
Zoonotic viruses. Viruses in this group include the severe acute respiratory syndrome (SARS) coronavirus, New World hantaviruses producing the hantavirus cardiopulmonary syndrome (HCPS), and the H5N1 avian influenza A virus.
Pathogenesis
A virus must first gain access to the lower respiratory tract in order to produce severe pneumonia. The most common mode of entry is via droplet transmission. Airborne virus-containing droplets 5 to 10 μm in diameter are filtered and deposited in the upper respiratory tract. Virus reaches the lower respiratory tract after efficient replication and spread within squamous epithelial cells, often in the setting of impaired mucociliary clearance (due to extremes of age, antecedent or concurrent infections, and drugs). This is the usual mode of entry for many human-adapted respiratory viruses, such as influenza, RSV, adenovirus, and coronavirus. Person-to-person spread via droplets is limited to a distance of approximately 1 m. Other viruses such as varicella and rubeola are transmitted via aerosols (particles 1 to 5 μm in diameter) that can deposit directly in the lower respiratory tract. As such, they are highly infectious and can be transmitted over greater distances and time than agents transmitted by droplets. Although deposited directly in alveoli, viral dissemination in the lung typically occurs hematogenously after a viremic phase [1,2].
Once in the lower respiratory tract, there are a limited number of ways that the lung can respond to a viral infection and produce respiratory illness. Viral invasion and replication can directly produce a necrotizing bronchopneumonia with highly inflammatory, purulent, and exudative reactions. This is not common, but can be seen with influenza and adenovirus infections. Respiratory viral infections can impair host lung defenses in a way that leads to secondary bacterial pneumonias, particularly with Streptococcus pneumoniae or Staphylococcus aureus. The classic examples are postinfluenza or measles pneumonias. Finally, viral infection of the lower respiratory tract may produce severe disease by triggering a common tissue response to acute lung injuries termed diffuse alveolar damage or acute respiratory distress syndrome. The acute lung injury may progress from an early exudative phase, often with profound noncardiogenic pulmonary edema (especially in HCPS), to a proliferative or organizing phase that produces interstitial inflammation, and a late resolving phase [3].
Clinical Manifestations
The limited host response patterns to virus-induced lung injury means that there is significant overlap in the clinical manifestations of viral pneumonias. The clues to a specific viral etiology are often found in assessing host risk factors and epidemiology on presentation. A summary of the common clinical manifestations for specific viral pneumonias is presented in Table 90.1. Many of the viral infections discussed in this chapter are characterized by a “flu-like illness” prodrome. Symptoms begin with the acute onset of headache, chills, and myalgias. Within a few days, a cough and sore throat develop along with upper respiratory tract infection. The presence or absence of upper respiratory symptoms at this stage may provide one clue to the specific viral etiology. The human-adapted respiratory viruses (human influenza, RSV, adenovirus, non-SARS coronavirus) generally all produce upper respiratory symptoms. Measles is characterized by coryza and conjunctivitis in the prodrome. The absence of upper respiratory symptoms has been reported to be characteristic of infections with several of the zoonotic viruses: SARS coronavirus, hantavirus, and the H5N1 avian influenza virus [4,5,6]. The lower respiratory tract signs and symptoms in viral pneumonias are generally nonspecific and progress to dyspnea, tachypnea, and inspiratory crackles. Sputum production is variable. If the clinical course is biphasic (dyspnea and productive cough after improvement of a flu-like illness), then a secondary bacterial pneumonia should be suspected.
Routine laboratory tests are generally of little help in distinguishing among the viruses that can produce severe respiratory illness. Total leukocyte counts are typically within the normal range or slightly elevated. One exception is measles virus infection, which can produce a marked leukopenia [7]. The most common hematologic finding in the viral pneumonias is a relative lymphopenia. The complete blood count may be useful for diagnosing HCPS. In HCPS caused by Sin Nombre virus (a New World hantavirus), the triad of thrombocytopenia (platelet count less than 150 K per mm3), absolute neutrophilia, and the presence of immunoblasts was a sensitive and specific predictor of HCPS in one study [5]. Electrolyte abnormalities and hepatic transaminase elevation can occur among any of the severe viral pneumonias.
Table 90.1 Presentation and Manifestations of Specific Viral Pneumonias | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
The radiographic findings in viral pneumonias are also broad and nonspecific. Radiographic infiltrates can have interstitial, alveolar, or combined patterns. The presence of only a diffuse alveolar pattern might suggest a primary influenza pneumonia with hemorrhagic alveolitis [8] or the capillary leak syndrome of acute respiratory distress syndrome, especially due to HCPS. Peribronchial nodular infiltrates is a pattern often reported with varicella pneumonia [9]. Computed tomography (CT) scans are better at detecting the presence, extent, and complications of respiratory infections than chest radiographs. However, they are no better at defining particular radiographic patterns of specific viral or bacterial causes [9].
Diagnosis
The diagnostic modalities available for viral pneumonias rely on detection of a viral component (nucleic acid or protein), growth of the virus in vitro, or development of a virus-specific antibody response. Definitive serologic evidence of a viral infection requires a rise in virus-specific antibody titers between paired acute illness and convalescent sera. With a few exceptions, serologic assays are therefore not generally helpful for the clinician in the acute setting of a viral pneumonia. This section will focus on diagnostic tests that may assist the clinician faced with a critically ill patient and suspected viral pneumonia.
Human Influenza A and B
Rapid, direct, antigen-detection assays are commercially available for diagnosing human influenza A and B virus infections. These assays rely on detection of the influenza virus nucleoprotein in respiratory secretions, and results can be obtained within 1 hour. Because they are based on the viral nucleoprotein, none of the rapid antigen tests provide information about influenza A hemagglutinin subtypes (e.g., H1, H3). Details regarding the available rapid antigen tests for influenza are provided by the Centers for Disease Control (CDC) (http://www.cdc.gov/flu/professionals/diagnosis/rapidclin.htm). The test specificities for diagnosing an influenza virus infection are generally high (more than 90%), but reported sensitivities are lower (33% to 80%) and may vary for different human influenza A virus subtypes [10,11,12]. In clinical practice, the timing and method of sample collection can greatly affect test sensitivity. Influenza A virus shedding from the upper respiratory tract typically peaks 2 to 3 days after symptom onset [13,14]. The window available to reliably detect viral antigen from upper respiratory tract secretions may extend only 5 to 6 days after symptom onset. Influenza virus nucleoprotein is most abundant in the columnar respiratory epithelium. Posterior nasopharyngeal swabs or aspirates that collect columnar epithelial cells are usually the preferred samples for rapid antigen detection assays [10,15,16], even for mechanically ventilated patients in the intensive care unit (ICU).
Reverse transcriptase polymerase chain reaction (RT-PCR) assays and viral culture are the next most commonly used diagnostic tests for human influenza virus infections. Posterior nasopharyngeal swabs or washes, and samples of lower respiratory tract secretions such as endotracheal aspirates or bronchoalveolar lavages, are acceptable samples. Virus typing and influenza A subtyping can be accomplished with either method. Due to its high sensitivity, specificity, and throughput, RT-PCR assays have generally supplanted virus culture in many clinical microbiology laboratories. Unlike viral culture, the detection of influenza viral RNA by RT-PCR cannot assess the presence of live virus in respiratory secretions.
Respiratory Syncytial Virus
Rapid antigen-detection assays and direct immunofluorescent staining for RSV from respiratory secretions have been the primary diagnostic tests used in children. These tests have > 80% sensitivity and > 90% specificity [17]. RT-PCR assays and respiratory viral culture are the other common diagnostic approaches in pediatric populations. In adults, the RSV rapid antigen assays and viral culture are generally insensitive due to low virus shedding and preexisting anti-RSV antibody in respiratory secretions [18,19]. Direct fluorescent antibody staining in nasopharyngeal specimens was reported to be the only rapid assay at least equivalent to viral culture in adults [17]. A RT-PCR assay on respiratory secretions is the preferred acute illness diagnostic method for RSV infection in adults [20,21].
Adenovirus
PCR of adenovirus DNA or respiratory viral culture from a nasopharyngeal swab or aspirate, sputum, or lower respiratory tract secretions is the diagnostic test of choice for adenoviral pneumonia. Direct adenovirus antigen assays that cover most serotypes, such as immunofluorescent antibody staining, are not as sensitive as PCR assays or viral culture.
Varicella
Varicella pneumonia typically develops within 1 to 6 days after the characteristic rash of chickenpox has appeared [22]. If desired, a specific microbiological diagnosis can be obtained by PCR assay or viral culture from a swab or scraping at the base of an unroofed vesicle. Viral detection in respiratory secretions is generally not required.
Rubeola (Measles)
Pulmonary involvement with measles is generally diagnosed on the basis of history and physical findings. In outbreak settings, pneumonia should be suspected in patients who develop respiratory distress and persistent or recurrent fevers during the course of typical measles. Measles is characterized by malaise and fever, followed rapidly by coryza, conjunctivitis, and cough [23]. Early in illness, the presence of Koplik spots on the buccal mucosa is pathognomonic of measles. The classic morbilliform rash begins 3 to 4 days after onset of illness and starts to fade after another 3 days. Worsening respiratory symptoms as the rash is fading is suspicious for rubeola pneumonia. Laboratory confirmation may be useful, particularly in suspected sporadic cases within a highly immunized population. Viral isolation or rapid detection of measles antigen in nasopharyngeal secretions is difficult and not readily available. A presumptive serologic diagnosis can be made by detection of serum antimeasles virus immunoglobulin M (IgM) or immunoglobulin G (IgG) in unimmunized individuals. Serum antibodies appear 1 to 3 days after onset of the rash [24]. Definitive serologic diagnosis requires paired acute and convalescent sera. In immunocompromised patients with overwhelming pneumonia, the antibody response may be minimal. Viral antigen staining of cells or RT-PCR assays on nasal exudates or urinary sediment may be useful in this setting [23].
Severe Acute Respiratory Syndrome Coronavirus
The most practical diagnostic approach for SARS coronavirus is a RT-PCR assay on nasopharyngeal specimens within
2 weeks after symptom onset [25]. The other primary site where SARS coronavirus RNA can be detected is stool (week 2 onward). Lower respiratory tract secretions harbor a greater viral load than upper respiratory tract secretions early in illness. However, lower respiratory tract aspiration, lavage, or intubation pose serious nosocomial transmission risks and should not be pursued solely for diagnostic purposes. IgM seroconversion does not occur until after the first week of illness and therefore is also of limited diagnostic utility [4]. With resolution of SARS viral transmission in 2003, and the apparent subsequent mutation of the virus [26], any initial positive test for SARS coronavirus must be viewed as a potential false-positive finding.
2 weeks after symptom onset [25]. The other primary site where SARS coronavirus RNA can be detected is stool (week 2 onward). Lower respiratory tract secretions harbor a greater viral load than upper respiratory tract secretions early in illness. However, lower respiratory tract aspiration, lavage, or intubation pose serious nosocomial transmission risks and should not be pursued solely for diagnostic purposes. IgM seroconversion does not occur until after the first week of illness and therefore is also of limited diagnostic utility [4]. With resolution of SARS viral transmission in 2003, and the apparent subsequent mutation of the virus [26], any initial positive test for SARS coronavirus must be viewed as a potential false-positive finding.
Hantavirus
There are nearly a dozen New World hantaviruses that have been associated with HCPS. Sin Nombre virus (in the southwestern United States) and Andes virus (in South America) are the two best known HCPS-associated hantaviruses. The diagnosis of HCPS can be made by detection of antihantavirus IgM antibodies in acute illness serum. Nearly all patients with HCPS have detectable IgM in their sera at the onset of pulmonary edema. The currently available IgM capture enzyme-linked immunosorbent assay using a recombinant Sin Nombre virus antigen can be used to diagnose all New World hantavirus infections [27]. RT-PCR assay on blood or lung tissue is a research assay of limited utility and not widely available. Because of low yield and biosafety issues, attempted culture of hantaviruses in clinical microbiology laboratories is not recommended.