key: cord-0691886-sujsqx2j
authors: Amorim, Viviane Brandão; Rodrigues, Rosana Souza; Barreto, Miriam Menna; Zanetti, Gláucia; Hochhegger, Bruno; Marchiori, Edson
title: Influenza A (H1N1) pneumonia: HRCT findings
date: 2013
journal: J Bras Pneumol
DOI: 10.1590/s1806-37132013000300009
sha: 3e66fa6e5fa282546d206b4d2c55ad15f38f7563
doc_id: 691886
cord_uid: sujsqx2j

OBJECTIVE: To describe aspects found on HRCT scans of the chest in patients infected with the influenza A (H1N1) virus. METHODS: We retrospectively analyzed the HRCT scans of 71 patients (38 females and 33 males) with H1N1 infection, confirmed through laboratory tests, between July and September of 2009. The HRCT scans were interpreted by two thoracic radiologists independently, and in case of disagreement, the decisions were made by consensus. RESULTS: The most common HRCT findings were ground-glass opacities (85%), consolidation (64%), or a combination of ground-glass opacities and consolidation (58%). Other findings were airspace nodules (25%), bronchial wall thickening (25%), interlobular septal thickening (21%), crazy-paving pattern (15%), perilobular pattern (3%), and air trapping (3%). The findings were frequently bilateral (89%), with a random distribution (68%). Pleural effusion, when observed, was typically minimal. No lymphadenopathy was identified. CONCLUSIONS: The most common findings were ground-glass opacities and consolidations, or a combination of both. Involvement was commonly bilateral with no axial or craniocaudal predominance in the distribution. Although the major tomographic findings in H1N1 infection are nonspecific, it is important to recognize such findings in order to include infection with the H1N1 virus in the differential diagnosis of respiratory symptoms.

features of the disease becomes important in clinical practice.

The objective of the present study was to evaluate, by means of a retrospective analysis of HRCT scans of patients with confirmed H1N1 infection, the most common tomographic findings and the characteristics of their distribution in the lung parenchyma.

The study was approved by the Research Ethics Committee of the Clementino Fraga Filho University Hospital. This was a retrospective observational cross-sectional study in which we analyzed the HRCT scans of 71 patients from several hospitals in different states in Brazil between July and September of 2009.

The study population included patients presenting with flu-like symptoms and diagnosed with H1N1 infection, regardless of age or gender. The inclusion criterion was the diagnostic confirmation of H1N1 infection through laboratory testing (viral culture or real-time PCR) of aspiration material or of nasopharyngeal or oropharyngeal swab specimens. Another inclusion criterion was the required availability of complete chest HRCT scans of each study patient, performed during the acute phase of the disease. Patients with other pulmonary infections, confirmed through laboratory tests, were excluded, as were those with tomographic findings suggestive of pulmonary involvement resulting from chronic lung diseases. All HRCT scans were obtained 2-10 days after the onset of symptoms.

As multiple institutions were involved, imaging examination was performed with different tomography scanners, using the high-resolution technique, in accordance with the following protocol: patient in the supine position; 1to 2-mm slice thickness in increments of up to 10 mm, from the lung apices through the hemidiaphragm, at the end of a deep inhalation; and a high spatial resolution reconstruction algorithm. Parenchymal and mediastinal window settings (width = 1,400-1,600 HU; level = −600 to −800 HU; and width = 350-450 HU; level =

In April of 2009, there began an epidemic of acute febrile respiratory illness caused by a new virus: the influenza A (H1N1) virus. The first cases occurred in Mexico, and the infection rapidly spread worldwide. (1) By August of 2010, 214 countries had been hit by the infection and more than 18,000 deaths had been confirmed. (1, 2) Although the number of cases of H1N1 infection has decreased significantly since the 2009 pandemic, various studies have reported that the virus is still circulating together with other seasonal viruses, with different prevalences. In Brazil, by October of 2012, approximately 20,000 patients had been hospitalized for severe acute respiratory syndrome. Of those, approximately 2,600 cases were caused by the influenza A (H1N1) post-pandemic virus. (3) The most common clinical findings in the presentation of H1N1 infection are fever, cough, dyspnea, myalgia, and headache. Gastrointestinal symptoms, such as nausea, vomiting, and diarrhea, have also been reported. (4) (5) (6) In most cases, the symptoms are mild and run a self-limiting course; however, a small proportion of individuals develop a severe course, which can result in respiratory failure and death. (4, (7) (8) (9) (10) The most common laboratory test findings include increased serum lactate dehydrogenase levels (which can exceed 1,000 IU/L), bearing in mind that elevated lactate dehydrogenase levels are significantly associated with disease severity and ICU admission (6, 11) ; increased C-reactive protein levels; increased serum creatine kinase levels; lymphopenia; and thrombocytopenia. Elevated transaminase and D-dimer levels can occur in some patients. (4, 7, 12) Chest X-ray provides adequate information for defining the approach in most of the affected patients. (13) However, HRCT often becomes an important tool for determining the extent of pulmonary involvement, as well as being useful in the evaluation of complications and in the clarification of suspected mixed infections or failure to respond to therapy. Pulmonary abnormalities were identified on the HRCT scans of all patients. The most common tomographic findings, in decreasing order, were as follows: ground-glass opacities, in 60 patients (85%); consolidations, in 46 (64%; Figures 1 and 2); airspace nodules, in 18 (25%; Figure 3 ); bronchial wall thickening, in 18 (25%); interlobular septal thickening, with or without ground-glass opacities, in 15 (21%); crazy-paving pattern, in 11 (15%); perilobular pattern, in 2 (3%); and air trapping, 2 (3%; Figure 4 and Table 1 ).

The most common findings were ground-glass opacities and consolidations. A combination of ground-glass opacities and consolidations in the same patient was observed in 41 cases (58%). Only 6 patients had no ground-glass opacities, consolidations, or a combination of both.

Of the patients who had ground-glass opacities, 11 (15%) also had interlobular septal thickening, characterizing a crazy-paving pattern. Septal thickening, without ground-glass opacities, was a mild secondary finding in 4 patients (6%). Airspace nodules were found in 18 cases (25%), and bronchial wall thickening was found in 18 (25%). Bronchial wall thickening was observed in all 6 patients who had no ground-glass opacities, consolidations, or a combination of both.

Air trapping was an uncommon finding, being observed in 2 patients. Peribronchovascular interstitial thickening was observed in 8 cases (11%), and, in all of these cases, there were ground-glass opacities and consolidations. A perilobular pattern was found in only 2 patients (3%). Minimal pleural effusion was observed in 19 patients (27%), being bilateral in 10 (52%).

Parenchymal involvement was bilateral in 63 cases (89%) and unilateral in 8 (11%). In 55 patients (77%), the lesions were evenly distributed between the two lungs; in 9 patients (13%), the lesions predominantly affected the right lung; and, in 7 patients (10%), the lesions predominantly affected the left lung. In 7 of the 8 cases (87%) in which the involvement was unilateral, only one lobe was affected. An analysis of the craniocaudal distribution of the lesions by lung zone revealed the following: no preferential distribution, in 35 patients (49%); lower third predominance, in 26 (37%); middle third predominance, in 4 (6%); lower and middle third predominance, in 5 (7%); and upper third the inspiratory phase-and peribronchovascular interstitial thickening); and distribution of the lesions (central, peripheral, or random; unilateral or bilateral; upper, middle, or lower lung zone predominant, or any combination of the three). In addition, the presence of pleural effusion and lymphadenopathy was evaluated. The HRCT scans were independently interpreted by two experienced thoracic radiologists, and in case of disagreement, the decisions were made by consensus.

Consolidation was defined as increased attenuation of the lung parenchyma, resulting in the obscuration of the vascular outlines and adjacent airway walls; ground-glass opacity was defined as slightly increased attenuation of the lung parenchyma that is unrelated to the obscuration of the vessels and adjacent airway walls; interlobular septal thickening was defined as thin linear opacities, which correspond to the thickened peripheral connective septa; a crazypaving pattern was defined as interlobular septal thickening superimposed on ground-glass opacities; airspace nodules were defined as ill-defined nodules smaller than 1 cm and tending to confluence; peribronchovascular thickening was defined as an increase in connective tissue around the bronchi, pulmonary arteries, and lymphatic vessels; a perilobular pattern was defined as thick and irregular polygonal opacities in the periphery of the secondary pulmonary lobule; and air trapping was defined as decreased attenuation of the lung parenchyma, revealed by a lowerthan-usual density. (14, 15) The bronchial walls were considered thickened if the bronchial lumen internal diameter was equal to or greater than 80% of its external diameter. (16) The distribution of the lesions in the lung parenchyma was evaluated along the axial and craniocaudal axes. The craniocaudal distribution of the lesions was classified as upper, middle, or lower lung zone predominant, or any combination of the three. The axial distribution of the findings was classified as central or peripheral. In addition, the predominance of the findings in one lung or the lack of predominance along the two axes was recorded.

Of the 71 patients, 38 (53.53%) and 33 (46.47%) were female and male, respectively. Airspace nodules were present in 25% of the cases, and this was the dominant finding in only 1. This finding has not commonly been reported in the literature, and few studies have described the presence of airspace nodules in patients with H1N1 infection. (21, 24) This can be attributed to an alternative interpretation, by some authors, of airspace nodules being focal areas of consolidation, or it can even be attributed to the fact that airspace nodules can be obscured by predominance, in only 1 (1.5%). An analysis of the axial distribution of the lesions revealed the following: no specific distribution, in 48 patients (68%); peripheral lung zone predominance, in 21 (29%); and central lung zone predominance, in 2 (3%). In 14 of the 21 patients (67%) with peripheral distribution, there were also lesions in the peribronchovascular interstitial space.

In the present study, we retrospectively evaluated the HRCT scans of 71 patients with confirmed H1N1 infection. The most common tomographic findings were ground-glass opacities, consolidations, and a combination of ground-glass opacities and consolidations in the same patient. These data are similar to literature data showing that these tomographic findings predominate in patients with H1N1 infection. (4, (17) (18) (19) (20) (21) A crazy-paving pattern, in which ground-glass opacities are associated with septal thickening, has been reported in few cases in the literature. (22, 23) In our study, 15% of the patients showed in few studies and in varying frequencies. (21, 24) One group of authors (24) found bronchial wall thickening in all of their patients. Other authors (25) suggested that these changes occur early in the disease course, a period when patients have not yet sought medical care, and is therefore poorly reported. Tanaka et al. (25) related the high frequency of these findings in their study (68%) to the fact that their patients had quick access to medical treatment. Likewise, the immunocompromised patients in another study (24) are probably patients who received prompt medical attention, given their comorbidities. In contrast, other authors (20) reported air trapping in a female patient in a late phase, three months after the onset of the disease.

In our sample, a mosaic attenuation pattern was observed in 3% of the cases and was interpreted as being air trapping because it was associated with bronchial wall thickening. As occurred in our sample, air trapping related to H1N1 infection has rarely been described in the literature. (21, 24) However, it should be emphasized that, in those studies, the expiratory phase, which increases sensitivity in the characterization of this pattern, was not performed. Li et al. (21) found a mosaic attenuation pattern due to air trapping in 13% of their patients and related it to the use of mechanical ventilation.

A perilobular pattern is a finding that is commonly associated with organizing pneumonia. (15) This finding was seen in 2 of our patients (3%), and it had not been described in H1N1 infection. Because patients affected by the pandemic virus can develop organizing pneumonia during the convalescence phase of the disease, (18, 19, 26) a perilobular pattern is an expected finding in these cases.

other findings when there is extensive involvement by the disease.

Bronchial wall thickening was observed in one fourth of the patients in our sample. In the literature, these findings have been described In the literature, the most common distribution of HRCT findings is bilateral multifocal involvement predominantly in the lower lobes. (4, 20, 21, 24) However, in some studies, the distribution is diffuse, with no zonal predominance. (27) In our sample, the HRCT scans showed bilateral involvement in the vast majority of the cases (89%), and both lungs were similarly affected in 77% of them. The craniocaudal distribution of the lesions was random, with no zonal predominance in approximately half of the cases and with lower third predominance in 37%. The axial distribution of the findings showed no zonal predominance in most cases (68%), but, in approximately one third of the cases, it showed peripheral lung zone predominance. A central distribution was rarely observed. In the literature, data regarding the axial distribution of HRCT findings are varied, there being reports of central or peripheral predominance. (4, 20, 24) Regarding the craniocaudal distribution, the literature reports lower lung zone predominance.

Pleural effusion, either unilateral or bilateral and typically minimal, has been described in some cases. (4, 17, 21) In our sample, pleural effusion was observed in 27% of the cases, being bilateral in 52% and typically minimal. No mediastinal lymphadenopathy was identified in any of the patients studied.

Our study had some limitations. First, the study was retrospective, which made it impossible to establish a proper correlation between clinical and radiological findings. Second, there were some differences in relation to the technical parameters in tomography image acquisition, since we evaluated HRCT scans performed at various institutions.

In conclusion, the most common tomographic findings were ground-glass opacities and airspace consolidations, or a combination of both. The findings were predominantly bilateral, with no axial or craniocaudal predominance in the distribution in most cases. When zonal predominance was present, it was more common in the lower thirds and peripheral regions of the lungs. Pleural effusion, when observed, was typically minimal. No lymphadenopathy was identified. Although the major tomographic findings in H1N1 infection are nonspecific, it is important to recognize such findings in order to include infection with the H1N1 virus in the differential diagnosis of respiratory symptoms

World Health Organization

Geneva: World Health Organization

Pandemic Influenza A (H1N1)

Clinical features of the initial cases of 2009 pandemic influenza A (H1N1) virus infection in China

Brasília: Ministério da Saúde

Boletim Informativo -Secretaria de Vigilância em Saúde -Influenza (gripe) -Semana Epidemiológica (SE) 44

High-resolution computed tomography findings from adult patients with Influenza A (H1N1) virus-associated pneumonia

Pandemic influenza A (H1N1) 2009: risk factors for hospitalization

Influenza A (H1N1)-associated pneumonia

Pneumonia and respiratory failure from swine-origin influenza A (H1N1) in Mexico

Swine-origin influenza A (H1N1) viral infection: thoracic findings on CT

A successfully treated case of parainfluenza virus 3 pneumonia mimicking influenza pneumonia

Lung pathology in fatal novel human influenza A (H1N1) infection

Factors associated with death in hospitalized pneumonia patients with 2009 H1N1 influenza in Shenyang, China

High-resolution CT findings in a patient with influenza A (H1N1) virus-associated pneumonia

Swine-origin influenza A (H1N1) viral infection: small airways disease

Chest computed tomography findings in severe influenza pneumonia occurring in neutropenic cancer patients

Crazy-paving pattern on HRCT of patients with H1N1 pneumonia

Image findings of patients with H1N1 virus pneumonia and acute respiratory failure

Thoracic CT findings of novel influenza A (H1N1) infection in immunocompromised patients

High-resolution computed tomography findings of influenza virus pneumonia: a comparative study between seasonal and novel (H1N1) influenza virus pneumonia

Organising pneumonia as a late abnormality in influenza A (H1N1) virus infection

Follow-up after acute respiratory distress syndrome caused by influenza a (H1N1) virus infection

H1N1 influenza: initial chest radiographic findings in helping predict patient outcome

Fleischner Society: glossary of terms for thoracic imaging

Comissão de Imagem da Sociedade Brasileira de Pneumologia e Tisiologia. Illustrated Brazilian consensus of terms and fundamental patterns in chest CT scans

Bronchial abnormalities

Serial evaluation of high-resolution CT findings in patients with pneumonia in novel swine-origin influenza A (H1N1) virus infection

Influenza A (H1N1) virus-associated pneumonia: high-resolution computed tomographypathologic correlation

Follow-up aspects of influenza A (H1N1) virus-associated pneumonia: the role of high-resolution