Viral pneumonias

Multifaceted approach to an elusive diagnosis

Jason W. Chien, MD; John L. Johnson, MD

VOL 107 / NO 1 / JANUARY 2000 / POSTGRADUATE MEDICINE

CME learning objectives

• To understand the importance of basing a diagnosis of viral pneumonia on a combination of epidemiologic, clinical, radiographic, and laboratory characteristics
• To review the various laboratory tests available to aid in diagnosis
• To recognize that a virus identified by laboratory testing may not be the pathogen responsible for the pneumonia

Preview: Determining whether a patient has viral pneumonia can be tricky. Even when a virus is isolated, it may not be the cause of the pneumonia. Here, Drs Chien and Johnson discuss the means of making this elusive diagnosis with confidence. Part 2 and Part 3 of this three-part article, which focus on viral pneumonia in the immunocompromised host and viral pneumonia due to epidemic respiratory viruses, will appear in our February and March issues.
Chien JW, Johnson JL. Viral pneumonias: multifaceted approach to an elusive diagnosis. Postgrad Med 2000;107(1):67-72

Viral pneumonias have been reported with increasing frequency during the past decade. This increase is most likely due to a combination of more sensitive diagnostic techniques and the greater risk for viral pneumonias among the growing immunocompromised population.

Several studies have shown that bacterial and viral pathogens are identified in 26% to 77% of cases of community-acquired pneumonia (1-3). Influenza A and B and parainfluenza viruses are the most common causes of respiratory illness among adults (4) (table 1).

Table 1. Causes of viral pneumonia in adults and children

Adults Adenovirus Cytomegalovirus Herpes simplex virus Influenza A and B virus Measles Parainfluenza virus Respiratory syncytial virus Varicella-zoster virus Children Influenza A and B virus Measles Parainfluenza virus Respiratory syncytial virus

Up to 50% of all pneumonias occurring in children under 3 years of age may have a viral origin (1). Among children, the most commonly identified pathogens are respiratory syncytial virus, parainfluenza viruses, and influenza viruses (4) (table 1). Faced with this growing population at risk for serious viral infections of the lower respiratory tract, clinicians need to understand the methods currently available for detecting respiratory viruses and be able to correlate clinical, radiographic, and laboratory findings to determine the causative role of recovered viruses in lower respiratory tract disease.

Clinical diagnosis of viral pneumonia

Before modern laboratory diagnostic tests were developed, viral pneumonia was diagnosed primarily on clinical grounds. This was often challenging because the clinical presentation of viral pneumonia varies and is often nonspecific. Suspicion of a viral cause of pneumonia in a patient should be based on a combination of epidemiologic, clinical, radiographic, and laboratory characteristics.

Several epidemiologic characteristics suggest a viral infection. For instance, various viral pneumonias typically occur during specific times of the year, among closed populations, or in populations with underlying cardiopulmonary or immunocompromising disease (1,5).

Clinical manifestations of viral pneumonia typically include constitutional symptoms such as fever, chills, nonproductive cough, rhinitis, myalgias, headaches, and fatigue. These symptoms, together with physical findings such as wheezing, rales, increased fremitus, and signs of widespread bronchial inflammation, often accompany viral infections of the lower respiratory tract. However, they are also seen in pyogenic pneumonia and are nonspecific. Characteristic rashes of measles or varicella or associated mucocutaneous lesions of herpes simplex infection are important diagnostic clues.

Unfortunately, the radiographic features of viral pneumonia are also nonspecific and difficult to differentiate from those of bacterial pneumonia. Findings are often similar and can range from patchy bronchopneumonia to fleeting infiltrates to more characteristic diffuse interstitial infiltrates (figure 1: not shown). Cavitation and pleural effusions are rare.

Laboratory diagnosis of viral pneumonia

Developments in diagnostic techniques have led to significant improvement in the ability to detect viruses in the respiratory tract. Laboratory methods can be categorized as cytologic evaluation, viral culture, rapid antigen detection, and gene amplification (table 2).

Table 2. Diagnostic techniques for viral infections
Virus	Cytologic evaluation	Viral culture	Antigen detection	Gene amplification
Herpesviruses
Herpes simplex virus	Cowdry type A bodies	CPE, SVA	IFA, ELISA	PCR
Varicella-zoster virus	Cowdry type A bodies	CPE	IFA	MRT-PCR
Cytomegalovirus	"Owl's eye" cells	CPE, SVA	IFA, ELISA	MRT-PCR
Paramyxoviruses
Respiratory syncytial virus	Highly eosinophilic intracytoplasmic inclusions	CPE, SVA	IFA, ELISA	MRT-PCR
Parainfluenza virus	Large cells with single nucleus and multiple small eosinophilic inclusions	HA,SVA	IFA, ELISA	MRT-PCR
Measles		HA	IFA, ELISA
Influenza virus		HA, SVA	IFA, ELISA
Adenovirus	Intranuclear inclusions	CPE, SVA	IFA, ELISA	MRT-PCR
CPE, cytopathogenic effect; ELISA, enzyme-linked immunosorbent assay; HA, hemadsorption; IFA, immunofluorescence assay; MRT-PCR, multiplex reverse transcriptase-polymerase chain reaction; SVA, shell vial assay.

When using these highly sensitive and specific methods, one must realize that the detection of certain viral pathogens does not always suggest active disease. Herpesvirus, for example, can establish a lifelong latent infection that may become reactivated without causing significant active disease. Some viruses, such as respiratory syncytial virus and cytomegalovirus, can be detected in the presence of other bacterial pathogens, making it hard to decide which pathogen is the true causative agent. In these situations, histopathologic findings are critical to definitive diagnosis.

Cytologic evaluation
Respiratory secretions, samples obtained from bronchoalveolar lavage, and tissue specimens can be examined by using cytologic and histologic methods (6). While intranuclear inclusions are often seen in cells infected with DNA viruses, cytoplasmic inclusions are usually present in cells infected with RNA viruses (6). For instance, cytomegalovirus infection is characterized by the presence of "owl's eye" cells. These are large cells with basophilic intranuclear inclusions and a surrounding clear zone (figure 2: not shown). Although the presence of viral inclusions is diagnostic of a viral infection, cytologic methods have low sensitivity. Therefore, the absence of inclusions does not reliably exclude infection or active disease.

Viral culture
Viral pneumonia can be definitively diagnosed by isolation and identification of the pathogen in viral culture (6). Most respiratory viruses can be isolated by appropriate tissue culture of specimens from the upper and lower respiratory tract, including sputum and samples obtained by nasopharyngeal washing, bronchoalveolar lavage, and biopsy.

Proper preservation of specimens for viral culture depends on the use of an appropriate transport medium. Transport media consist of enriched broth containing antibiotics and a protein substrate that protects the viruses and prevents bacterial or fungal overgrowth during transport. Once the specimens are obtained, they should be immediately placed in the viral transport medium, kept on ice or refrigerated, and transported quickly to the laboratory, where they are centrifuged and filtered to remove mucus and other debris. The supernatant or filtrate is then used for culture.

Viral cultures are performed on various cell lines, such as monkey kidney cells or diploid fibroblasts. Cell cultures are incubated at 35°C and examined microscopically on alternate days during a predetermined incubation period (usually 14 days). Changes in appearance, called the viral cytopathogenic effect, are evidence of viral growth. The viral cytopathogenic effect is manifested as degenerative changes or the formation of syncytial collections of multinucleated giant cells. These changes are rarely virus-specific.

Viral growth also can be detected by hemadsorption testing. The antigens of certain viruses, such as influenza and parainfluenza, have an affinity for erythrocytes. By adding guinea pig erythrocytes to cultured cell monolayers, the adherence of red blood cells to the cultured cell monolayer of infected tissue may be noted, demonstrating the presence of these viruses.

When the cytopathogenic effect is observed or hemadsorption tests are positive, the responsible virus is identified by further testing. One common technique is immunofluorescence by direct or indirect methods; another is the use of nucleic acid probes. These techniques can be used to identify specific viruses such as influenza A and B, parainfluenza virus, adenoviruses, respiratory syncytial virus, herpes simplex virus, and cytomegalovirus in cell cultures.

Modified cell culture methods, called shell vial culture systems, have been developed to detect viruses, such as cytomegalovirus, that grow very slowly in conventional fibroblast cell culture, often requiring 14 to 18 days to produce the cytopathogenic effect (6). With such systems, prepared clinical specimens are inoculated onto adherent cell monolayers grown on round coverslips in small vials. The vials are centrifuged at low speed for 1 hour, after which fresh culture medium is added. The vials are then incubated and examined serially to detect viral antigen or DNA expression. Shell vial culture systems are now widely used to speed the detection of cytomegalovirus, respiratory syncytial virus, herpes simplex virus, adenovirus, influenza virus, parainfluenza virus, and other pathogens.

Rapid antigen detection
Rapid antigen detection tests are able to provide fast results because they can be applied directly to patient specimens (5,6). Often, these tests can be completed in 1 to 3 hours after sample collection. Immunofluorescence assay and enzyme-linked immunosorbent assay (ELISA) are available for the diagnosis of herpes simplex virus, respiratory syncytial virus, influenza A and B, parainfluenza types 1 through 4, cytomegalovirus, and other respiratory viruses (6). Unlike ELISA, immunofluorescence assay requires that intact infected cells be prepared in the same manner as for cell culture. ELISA can detect viral antigens that are suspended in solution or associated with inflammatory material. The sensitivity of these methods varies depending on the virus being sought and the specific diagnostic assay used.

Antigen detection tests have several advantages. Unlike viral culture, they are specific for each virus. In addition, many laboratories use panels of antibodies to common respiratory viruses, thus permitting the screening of clinical specimens for many viruses at once. For instance, the respiratory antigen panel routinely includes influenza and parainfluenza viruses, adenovirus, and respiratory syncytial virus. In addition, antigen detection assays remain positive for several days to weeks after viable virus can no longer be detected in culture. This feature is useful in the evaluation of patients who present late after the onset of symptoms or patients who have received prior treatment with antiviral agents.

The disadvantage of viral antigen detection methods is that, overall, their sensitivity is lower than that of viral cultures. Sensitivity also varies depending on the virus sought. For instance, the sensitivity of immunofluorescence assays is high for respiratory syncytial virus and influenza B virus (87% to 93%) but low for adenovirus and parainfluenza virus (51% to 67%) (6). It is always recommended, therefore, that antigen detection methods be used in conjunction with cell culture for optimal diagnosis of viral infections.

Gene amplification
Polymerase chain reaction (PCR) is a highly sensitive and specific technique for amplifying specific genes to detect the presence of a virus. PCR has become especially useful for the detection of cytomegalovirus in various body fluids, such as blood, urine, and respiratory secretions. Unfortunately, owing to the latent nature of cytomegalovirus, its detection does not necessarily prove it to be the pathogen responsible for the associated disease. PCR detection should be used in combination with viral culture and immunocytologic and histochemical staining.

A recently developed molecular diagnostic technique, multiplex reverse transcriptase-PCR (MRT-PCR), overcomes the low sensitivity of antigen detection methods, the delay of viral cultures, and the limitation of assaying for only one virus at a time by traditional PCR (7). This technique permits rapid detection of respiratory syncytial virus, adenovirus, and parainfluenza virus types 1, 2, and 3 in appropriate respiratory tract secretions by using a single-step MRT-PCR with high sensitivity and specificity.

Summary

Despite enhanced laboratory techniques such as viral culture, rapid antigen detection, and gene amplification, a confident diagnosis of viral pneumonia continues to be a challenge. The nonspecific nature of clinical characteristics and the extreme sensitivity of laboratory techniques make the diagnosis difficult, even when a viral agent is detected. Understanding the limitations of these technological advances and the use of histopathologic techniques can greatly enhance a skilled clinician's ability to make an accurate diagnosis.

Part 2 and Part 3 of this article, which focus on viral pneumonia in the immunocompromised host and viral pneumonia due to epidemic respiratory viruses, respectively, will appear in our February and March issues.

References

1. Greenberg SB. Viral pneumonia. Infect Dis Clin North Am 1991;5(3):603-21
2. Marrie TJ, Durant H, Yates L. Community-acquired pneumonia requiring hospitalization: 5-year prospective study. Rev Infect Dis 1989;11(4):586-99
3. Murphy TF, Henderson FW, Clyde WA Jr, et al. Pneumonia: an eleven-year study in a pediatric practice. Am J Epidemiol 1981;113(1):12-21
4. Bartlett JG, Mundy LM. Community-acquired pneumonia. N Engl J Med 1995;333(24):1618-24
5. Sullivan CJ, Jordan MC. Diagnosis of viral pneumonia. Semin Respir Infect 1988;3(2):148-61
6. Leland DS, Emanuel D. Laboratory diagnosis of viral infections of the lung. Semin Respir Infect 1995;10(4):189-98
7. Osiowy C. Direct detection of respiratory syncytial virus, parainfluenza virus, and adenovirus in clinical respiratory specimens by a multiplex reverse transcription-PCR assay. J Clin Microbiol 1998;36(11):3149-54

Dr Chien has completed a fellowship in infectious diseases at Case Western Reserve University School of Medicine, Cleveland, and is now a fellow in the division of pulmonary and critical care medicine, University of Washington School of Medicine, Seattle. Dr Johnson is associate professor of medicine, division of infectious diseases, Case Western Reserve University School of Medicine. Correspondence: Jason W. Chien, MD, University of Washington School of Medicine, Division of Pulmonary and Critical Care Medicine, BB-1253 Health Sciences Center, UW Box 356522, Seattle, WA 98195-6522. E-mail: jasonc@u.washington.edu.

Please send technical questions related to the Web site to Ann Harste