Context.—Hantavirus pulmonary syndrome (HPS) and acute interstitial pneumonia (AIP) can share similar clinical presentations. AIP is an acute, diffuse lung disease that has some clinical features suggesting a viral infection, although causative agent(s) have not been identified.
Objective.—To clinically, histologically, and immunohistochemically compare cases of HPS to cases of AIP and to determine if any cases of AIP were actually examples of HPS.
Design.—Seven cases of HPS and 9 cases of AIP were compared clinically and histologically by semiquantitative grading of features in lung tissue. The cases were also evaluated immunohistochemically for the presence of hantaviral antigens.
Results.—Hantavirus pulmonary syndrome had a shorter clinical duration and more acute changes histopathologically; AIP was of longer clinical duration and was usually accompanied by histologic evidence of organization. Hantavirus pulmonary syndrome was distinguished by the presence of immature leukocytes in the pulmonary vasculature. No hantaviral antigens were identified immunohistochemically in the 9 case of AIP. Hantaviral antigens were identified in all 7 cases of HPS.
Conclusion.—Cases of AIP and fatal cases of HPS can generally be distinguished on clinical and histologic grounds, and this distinction can be further confirmed immunohistochemically.
In the spring and summer of 1993, an outbreak of acute respiratory disease in the southwestern United States was attributed to a novel hantavirus; the disease was subsequently named hantavirus pulmonary syndrome (HPS).1–9 Hantavirus pulmonary syndrome is a severe respiratory illness with a case fatality rate of 50% to 75%.3 The case definition for HPS evolved over time and facilitated identification of cases (Table 1).6–8 Cases fulfilling clinical or autopsy criteria and cases in which no other cause was found were confirmed as HPS by serologic, immunohistochemical, or polymerase chain reaction procedures.8,9
On autopsy, HPS is marked by a constellation of findings, including large pleural effusions, severe fibrinous pulmonary edema with hyaline membranes, and an immunoblastic proliferation involving the reticuloendothelial system.5,9 These findings occur in concert with a characteristic hematologic tetrad that includes left-shifted neutrophilic leukocytosis, thrombocytopenia, an increased hematocrit reflective of hemoconcentration, and circulating immunoblastic cells.5
Acute interstitial pneumonia (AIP) is an idiopathic, acute lung disease associated with the rapid development of the adult respiratory distress syndrome.10–13 Most patients with this condition have a prodrome suggestive of a viral infection that rapidly evolves into respiratory failure requiring assisted ventilation. The pathologic findings of AIP are characterized by acute and organizing diffuse alveolar damage. The fatality rate for AIP approaches 50%.11 Despite the clinical features suggesting viral infection, the cause(s) of AIP have (by definition) not been identified. In the initial clinical evaluation of patients with AIP, HPS is often included in the differential diagnosis. In fact, cases of AIP fulfill the case definition criteria for HPS. Because of this similarity to HPS and because sporadic cases of HPS have been retrospectively identified,14,15 we have studied 9 cases of AIP, 8 of which occurred prior to the recognition of HPS. We compared these cases with 7 cases of HPS and evaluated the AIP cases for hantaviral antigens by immunohistochemistry.
MATERIALS AND METHODS
Paraffin blocks from 9 cases of AIP retrieved from the files of one of the authors (T.V.C.) were selected for study. Eight of these were previously reported as examples of AIP.11 Clinical findings were obtained from referring physicians and the medical records. Seven fatal cases of HPS, comprising part of the original series of cases,5 were selected for comparison, and autopsy reports were reviewed.
The following parameters were assessed semiquantitatively (0–4+) on H&E-stained slides from both groups: interstitial edema; airspace edema and fibrinous exudates; hyaline membranes; hemorrhage; acute inflammation in airspaces; interstitial organization (fibroblast proliferation); airspace organization (intraluminal polyps of fibroblastic proliferation); type 2–cell hyperplasia; alveolar macrophages; interstitial chronic inflammation; alveolar wall atelectasis; cellular inflammation of bronchioles (cellular bronchiolitis); bronchiolar mucosal necrosis and metaplasia; and pleuritis.
Immunohistochemical assays for hantaviral antigens were performed as previously described.9,11 Briefly, tissue sections were predigested with proteinase-K and then incubated with a cross-reactive hantaviral monoclonal antibody (GB04-BF07). This step was followed by serial application of rabbit anti-mouse link antibody and an alkaline phosphatase–anti-alkaline phosphatase complex (Dako Corporation, Carpinteria, Calif). Hantaviral antigens were then detected by using a naphthol/fast red substrate. Isotype-identical monoclonal antibodies and nonimmune sera were used as negative controls.
Statistical comparison between the 2 groups was performed with the Wilcoxon rank sum test.
The clinical features of the 2 groups are compared in Table 2. Differences in the length of prodrome and the time to tissue evaluation were statistically significant between the 2 groups (P = .002 and P = .008, respectively).
In all 7 cases of HPS, viral antigens were detected immunohistochemically in lung tissue. There was widespread staining of hantaviral antigens primarily within endothelial cells of the pulmonary microvasculature. None of the cases of AIP showed positivity for hantaviral antigens. Representative histologic findings of both entities are shown in Figures 1 and 2. Comparison of the histologic parameters as assessed by 3 observers (T.V.C., R.M.F., and K.B.N.) is shown in Table 3.
The clinical characteristics and pulmonary histologic findings in our cases of HPS and AIP are distinct. Not surprisingly, none of the cases of AIP studied showed any hantaviral antigens by immunohistochemistry. Hantavirus pulmonary syndrome tends to be of shorter duration and shows disproportionately more pulmonary edema than AIP. Organization and type 2–cell proliferation are characteristic of AIP and are rarely seen in HPS cases. Intravascular immature lymphoid cells (that correlate with the distinctive hematologic tetrad described previously) were common in HPS and unusual in AIP; furthermore, microatelectasis, one of the most characteristic features of AIP, was rarely seen in HPS cases. In addition, the pleural effusions, immunoblastic proliferation in the reticuloendothelial system, and hematologic tetrad seen in HPS are not present in cases of AIP.5,11 These differences, combined with immunohistochemical staining for hantaviral antigens, should allow pathologic distinction between the 2 conditions.
All HPS cases were fatal and autopsy tissue was evaluated, whereas open lung biopsy tissue was evaluated in all of the AIP cases. Ideally, either biopsies in both conditions or autopsies in both conditions would be compared, but as yet, this has not been possible. The extremely rapid evolution and early mortality of HPS has precluded biopsy in most cases. This clinical pattern in itself is one of the most important differences between HPS and AIP. It is possible that HPS may manifest a very early phase of severe diffuse alveolar damage and in the few cases when patients survived in the hospital for 2 days or more, the histologic changes may more closely resemble those of AIP.9 No such cases were available for inclusion in this study. While HPS may be a very early phase of diffuse alveolar damage, there is insufficient time for most cases to develop the reparative changes (such as organization and type 2–cell metaplasia) that are seen in cases of AIP.
The authors gratefully acknowledge Jane Sweeney for technical help and V. Shane Pankratz, PhD, for statistical help.