© 2007 American Thoracic Society doi: 10.1164/rccm.200701-133ED
One Swallow Does Not Make a SummerCan Herpes Simplex Virus-1 Cause Pneumonia and Acute Lung Injury?Vrije Universiteit Medical Center, Amsterdam, The Netherlands In this issue of the AJRCCM (pp. 935–942), Luyt and colleagues (1) examine the role in immunocompetent patients of herpes simplex virus type 1 (HSV-1) isolated from the lower respiratory tract. These patients all had new pulmonary infiltrates and clinical deterioration during prolonged mechanical ventilation in the intensive care unit (ICU). The authors tried to link HSV-1 isolation, nuclear inclusions, and viral load (by quantitative polymerase chain reaction [PCR]) to manifestations of pulmonary dysfunction and thus to delineate the potential contribution of HSV-1 infection to ventilator-associated pneumonia (VAP). In examining such associations, the authors conclude that the virus may contribute to clinical deterioration in about 20% of all patients studied and in 40% of those who were HSV-1 positive, even though histologic abnormalities were observed in only 12 of 42 patients with viral pneumonia, as defined by the authors. HSV-1 was detectable in the lower airway of 56 other patients with clinical deterioration during mechanical ventilation, but without clinical criteria for viral pneumonia. The criteria for clinical deterioration were relatively loose, and we are not fully informed on the course of HSV-1–positive patients who did not meet the authors' criteria for viral pneumonia and how the clinical course of such patients compared with that found in HSV-1–positive patients with pneumonia. Information is also lacking about the identity of the cellular populations in which nuclear inclusions suggestive for active HSV-1 were seen. To put the present findings into perspective, the natural course of HSV-1 infection has to be considered (2). After primary infection in the upper respiratory tract in childhood, a period of viral latency is induced in many otherwise healthy individuals by incorporation of the viral genome into trigeminal nerve sensory ganglia. Upon reactivation during stressful conditions, including pregnancy and pneumonia, the neurotropic virus may travel along axons to provoke epithelial lesions, such as herpes labialis and gingivostomatitis. However, oropharyngeal reactivation often results in symptomless viral shedding, detectable by PCR (3). Most healthy individuals develop antibodies against the virus after their initial infection, and primary childhood infections may be more severe than recurrences in later life. Viral recurrences can be contagious: HSV-1 in ICU patients can be transmitted horizontally to health care workers lacking antibodies, resulting in herpetic whitlow (4). The pathogenesis of HSV-1 reactivation and shedding in about 15 to 30% of nonimmunocompromised patients on prolonged mechanical ventilation is incompletely understood (5–7). Moreover, it is largely unknown how HSV-1 reaches the lower respiratory tract and if it replicates there, even though shedding is associated with more severe disease, prolonged length of stay in the ICU, increased duration of mechanical ventilation, and greater mortality in most (5, 6), but not all studies (1). Bruynseels and colleagues (5) previously demonstrated that the presence of HSV-1 in the upper respiratory tract was a major risk factor for recovery of the virus from the lower respiratory tract, as was found in the present study by Luyt and coworkers (1). This suggests that, even in intubated patients, shedding of the virus and aspiration result in contamination of the lower respiratory tract, a finding not necessarily associated with pathogenicity. Daubin and colleagues (8) found that 31% of patients with VAP harbored HSV-1 in tracheobronchial secretions, but they could not ascribe a direct role to the virus as the pathogen causing VAP in the majority of patients. Indeed, frequent bacterial colonization/infection, as in Luyt and colleagues' study (1, 8–10), complicates any decision to identify HSV-1 as the primary cause of pulmonary abnormalities in critically ill, ventilated patients. Thus, a purely viral cause of VAP is probably rare. Similarly, HSV-1 was demonstrated histologically in only 1 of 37 open lung biopsies in patients with persistent acute respiratory distress syndrome (ARDS) on mechanical ventilation without bacterial pneumonia (11), even though a number of case studies suggest that HSV-1 pneumonia does exist and can result in ARDS (9, 12). Conversely, even open lung biopsies can be negative in case of autopsy-proven HSV-1 pneumonia (9). Obtaining evidence of infection and pulmonary injury directly attributable to HSV-1 is difficult in clinical practice and has, until Luyt and colleagues' study (1), not been pursued systematically in critically ill patients (7, 10). Demonstration of intracellular viral replication is the hallmark of viral pathogenicity and evidence for this can be obtained by studying intracellular Cowdry type A inclusions, tissue injury on histology, and viral DNA/RNA ratios (1, 7). Demonstration of HSV-1 DNA by PCR in the respiratory tract, as done in Luyt and coworkers' study (1), may be too sensitive and may not help to differentiate harmless shedding from invasive infection, unless, as suggested by the authors (1), viral load appears high, possibly as a result of active replication in the lower respiratory tract. Clear delineation of the route through which HSV-1 reaches the lungs could also help in attributing a pathogenic role to the virus versus harmless shedding. Hematogenic spread with viremia and positive HSV-1 PCR in blood would more likely be a cause of herpetic alveolitis and pneumonia. Local spread from thoracic dorsal root ganglia and nerve endings could contribute to herpetic tracheobronchitis, in the same way that this route can produce keratitis and herpes labialis. We are clearly in need of precise criteria for pathogenicity and diagnosis of HSV-1–associated pneumonia or tracheobronchitis. Indirect proof of pathogenicity associated with HSV-1 is demonstration of a treatment benefit. Although early acyclovir treatment in a small series of patients with ARDS prevented HSV-1 shedding and isolation (13), mortality was not affected. Indeed, there is no evidence yet, even in Luyt and colleagues' study (1), that antiviral therapy is beneficial in presumed HSV-1 pneumonia in the critically ill, as shown by decreased morbidity and mortality (10). In a retrospective study of a small number of critically ill patients in whom bronchoalveolar lavage fluid tested positive for HSV-1, van den Brink could not demonstrate a beneficial effect of 10 days of acyclovir treatment, even in patients who ultimately survived their stay in the ICU (10). In Luyt and colleagues' study (1), 19 of 42 patients with HSV-1 pneumonia were treated with acyclovir and a therapeutic benefit was not discernible. Prospective studies of acyclovir therapy in critically ill patients with presumed HSV-1 pneumonia are awaited. Luyt and coworkers' study (1) is a laudable effort to shed light on the role of HSV-1 in the critically ill. The lesson that may be drawn is that isolation of HSV-1 from the lower respiratory tract does not necessarily imply disease but may represent harmless shedding, and that evidence of disease should be actively sought for before deciding that a patient has viral pneumonia and should receive antiviral therapy. Obtaining evidence of HSV-1 pulmonary infection in clinical practice may include bronchoscopy and biopsy of lesions, as well as search for typical nuclear inclusion bodies and the cytopathic effect of the virus in alveolar or bronchial cells and macrophages obtained in washings or biopsies. This requires close collaboration among intensivists, pulmonologists, microbiologists, and pathologists to avoid delay of initiation of antiviral therapy in patients who may benefit from such treatment. However, a randomized, controlled clinical trial should be undertaken before such a strategy for treatment of presumed HSV-1 pneumonia can be recommended. FOOTNOTES Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
