key: cord-0755899-1tgowypx
authors: Busi Rizzi, E.; Schininà, V.; Ferraro, F.; Rovighi, L.; Cristoforo, M.; Chiappetta, D.; Lisena, F.; Lauria, F.; Bibbolino, C.
title: Radiological findings of pneumonia in patients with swine-origin influenza A virus (H1N1)
date: 2010-06-05
journal: Radiol Med
DOI: 10.1007/s11547-010-0553-9
sha: a2114cdfe6bde05777d1e833e8816387e6957859
doc_id: 755899
cord_uid: 1tgowypx

PURPOSE: During spring 2009, a pandemic swine-origin influenza A (H1N1) virus (S-OIV) emerged and spread globally. We describe the chest X-ray and computed tomography (CT) findings of 40 patients with pneumonia due to S-OIV observed in our institution. MATERIAL AND METHODS: Among 534 patients with S-OIV, according to the US Centers for Disease Control and Prevention case definition, seen between June and November 2009, 121 underwent chest X-ray and 40 (median age 44 years, range 16–79) had pneumonia. The initial chest radiographs were evaluated for pattern, distribution and extent of lung abnormalities. Unenhanced chest CT scans were performed in two patients and were reviewed for the same findings. Underlying medical conditions were present in 42% of patients (17/40). RESULTS: Our patients had predominantly mild illness, and pneumonia was observed in 40 individuals (40/121 patients who had chest X-rays, 33%; and 40/534 patients with S-OIV, 7.5%). However, S-OIV can cause severe illness requiring admission to the intensive care unit for advanced mechanical ventilation and extracorporeal life support, including adult respiratory distress syndrome (ARDS) and death. The major radiological abnormalities observed were interstitial changes (60.0%), with (22.0%) or without patchy ground-glass appearance, mostly bilateral, and located in the lower lung zones (7.5%). Extensive disease was seen in 37.5% (15/40), and ARDS was observed in three individuals (0.30%)with underlying medical conditions. Subtle pleural effusion was noted in four patients. CONCLUSIONS: In our series, the most frequent pneumonia patterns observed during S-OIV (H1N1) virus were interstitial changes and patchy ground-glass appearance, mostly bilateral, and located in the lower lung zones. CT, performed in severely ill patients, confirmed the ARDS identified with chest X-rays, better depicting the features and extent of lung abnormalities.

Swine-origin influenza A virus (S-OIV) (H1N1) was first reported in Mexico in late March 2009, and rapidly spread throughout the world. Epidemiological data suggest that the newly emerged H1N1 virus had relatively low virulence [1] . It is believed that the mechanism of transmission of S-OIV is predominantly by droplet secretions [2] . The clinical manifestations mostly consist of mild illness [3] , even though, as in the case of seasonal influenza [4, 5] , S-OIV can cause severe illness and death, above all in persons with underlying medical conditions. The role of radiological imaging is identifying pulmonary complications and depicting severe cases early in the course of the disease. In this paper, we review the chest radiographic and computed tomography (CT) findings observed in 40 individuals with pneumonia and S-OIV H1N1 infections.

Between June and November 2009, 534 patients fulfilling the US Centers for Disease Control and Prevention (CDC) clinical criteria for the diagnosis of S-OIV [6] were observed in our institution. Among 121 of these individuals who underwent chest radiography on admission, 40 had findings consistent with pneumonia. These patients formed our study population. S-OIV was subsequently confirmed by polymerase chain reaction (PCR) in all these individuals. The study population included 24 men, median age 44 years (range 16-79 years). All patients had temperature >38°C and one or more clinical findings of respiratory illness. Seventeen patients (42% 17/40) had one or more underlying medical conditions [obesity, HIV infection, chronic obstructive pulmonary disease (COPD), bronchial asthma, fibrosis or diabetes].

Chest radiographs were obtained on admission using conventional radiography in the two standard projections or performed at the bedside in the anteroposterior projection.

Conclusioni. Il quadro di presentazione radiologica più frequente è stato l'impegno interstiziale, con o senza consolidamenti multifocali a vetro smerigliato, bilaterale e localizzato nei campi polmonari inferiori. La tomografia computerizzata, effettuata nei pazienti in condizioni cliniche gravi, ha confermato la ARDS, già identificata mediante la radiografia del torace, definendone meglio l'estensione e le caratteristiche semeiologiche.

Parole chiave Influenza A (H1N1) · Polmoniti virali · Radiografia del torace · TC Introduzione L'influenza suina A H1N1 (S-OIV) è stata identificata per la prima volta in Messico, nel marzo 2009, e si è successivamente diffusa in tutto il mondo con caratteristiche di una pandemia. Il virus H1N1 ha virulenza relativamente bassa [1] e la sua trasmissione interumana avviene prevalentemente attraverso droplet [2] . S-OIV si manifesta come malattia di media severità [3] , in persone con comorbilità può però decorrere in forma grave, anche mortale, analogamente a quanto avviene per l'influenza stagionale [4, 5] . Il ruolo della diagnostica per immagini è quello di rilevare la comparsa di complicazioni polmonari, identificando precocemente quelle con evoluzione sfavorevole. Scopo del presente lavoro è stato valutare i quadri radiologici rilevati in 40 soggetti con polmonite da influenza suina. The admission chest radiographs in the standard posteroanterior and lateral views were obtained with Opera G 650 rad-RT20 radiographic equipment (General Medical Merate, Seriate, Italy) at 110-140 kV, 12 mAs and 200-cm film-focus distance. The anteroposterior bedside radiographic examinations were performed with portable equipment TMXR plus (General Electric Medical System, Milwaukee, WI, USA) using 90 kV, 5 mAs and 150-cm film-focus distance. Images were evaluated on high-resolution monitors (2,048×2,560 pixels, display gradation 1,021 (10-bit), maximum brightness 750 cd/m 2 , LCD display device 54 cm) of the Picture Archiving and Communications System (PACS 5.1, Kodak Carestream, Rochester, NY, USA).

Two patients underwent unenhanced helical CT with a HiSpeed scanner (General Electric Medical System). CT was performed at a table speed of 15 mm/s, 5-mm collimation and a 5-mm reconstruction interval. CT scans were displayed on monitors with window setting appropriate for lung parenchyma and mediastinum (2,048×2,560 pixels, display gradation 1,021, 10-bit), maximum brightness (750 cd/m 2 , LCD display device 54 cm) of the PACS system. Three experienced, board-certified radiologists (LR, EBR, VS) independently reviewed the radiographs obtained on admission, and consensus was reached in all cases.

The chest radiographs and CT scan were assessed for the presence and distribution of the following findings: (1) reticular opacities, (2) ground-glass opacities, (3) nodular opacities, (4) parenchymal consolidation, (5) pleural effusion, and (6) adenopathies. In accordance with the Fleischner Society Glossary [7, 8] , reticular opacities were defined as linear opacities forming a mesh-like pattern, which may be thin or thick and coarse. Ground-glass opacities were defined as hazy areas of increased attenuation with preservation of bronchial/vascular markings and without obscuration of the ribs. Nodular opacities were defined as focal round opacities, with a diameter no greater than 7 mm. Consolidation was an opacification of the parenchyma with obscuration of the underlying anatomical structures. A mixed pattern denoted a combination of ground-glass opacities, dense consolidations and reticular opacities. Distribution was categorised as bilateral or unilateral, focal, patchy or diffuse. Focal was defined a single focus of abnormality, patchy as more than one focus and diffuse when abnormalities involved three or more lung zones. The predominant distribution was also assessed as being in the upper (above the level of the anterior end of the third rib), middle (between the third and fifth anterior ribs) or lower (below the level of the anterior end of the fifth rib) lung zones. Le radiografie del torace sono state valutate analizzando la presenza e la distribuzione dei seguenti rilievi: 1) opacità reticolari, 2) vetro smerigliato, 3) opacità nodulari, 4) consolidamenti parenchimali, 5) versamento pleurico, 6) adenopatie. In accordo con il glossario della Fleischner Society [7, 8] , reticolari sono state definite le opacità lineari disposte a formare un reticolo, con setti interlobulari sottili o spessi; a vetro smerigliato è stata definita l'opacità nel cui contesto è stato possibile valutare i vasi, i bronchi sottostanti e le coste regionali sovrapposte; opacità nodulari sono state definite quelle focali rotondeggianti, con diametro non eccedente i 7 mm; consolidamento è stato definito l'opacità parenchimale densa, con cancellazione delle strutture anatomiche sottostanti/locoregionali. Il quadro di presentazione misto è quello nel quale le opacità a vetro smerigliato coesistono con consolidamenti parenchimali densi e impegno interstiziale. La distribuzione delle alterazioni è stata classificata come mono-o bilaterale, focale, multifocale e diffusa. In questo caso l'impegno parenchimale si estendeva a 3 o più regioni polmonari. La distribuzione predominante è stata inoltre valutata relativamente alla sede polmonare coinvolta, come superiore (sotto il bordo anteriore della 3 a costa), media (tra il bordo anteriore della 3 a e della 5 a costa) e inferiore (al di sotto del bordo anteriore della 5 a costa).

All chest radiographs obtained on admission were interpreted prospectively and retrospectively as abnormal. The imaging findings are summarised in Table 1 . The predominant chest radiographic findings were interstitial changes ( Fig. 1) , which constituted the sole finding in 37.5% (15/40)

Tutti i radiogrammi valutati sono stati effettuati al momento del ricovero presso la nostra struttura e interpretati, sia prospetticamente che retrospettivamente, come patologici. Le caratteristiche semeiologiche rilevate nel corso dello studio sono rappresentate nella Tabella 1. Le alterazioni of the patients. Ground-glass opacities (Fig. 2 ) associated with interstitial changes were the second most common predominant finding, found in 22.5% (9/40). A mixed pattern (Fig. 3) , a combination of ground-glass opacities, dense consolidations and reticular opacities, was observed in 15% (6/40). The lung abnormalities were bilateral in 70% (28/40) and most commonly found in the lower lung zones 70% (28/40) (Fig. 4 ). Extensive disease (Fig. 4) was seen in 37.5% (15/40), and acute respiratory distress syndrome (ARDS) (Figs. 5, 6) was observed in three individuals (7.5%). Significant lymph node enlargement was not identified in any chest X-ray. Subtle pleural effusions were seen in four patients. Two patients with severe illness and ARDS identified on chest radiography were also studied with CT, which confirmed the ARDS pattern, better depicting the features and extent of lung abnormalities. One of the patients, despite showing improvement of lung disease at subsequent CT, unfortunately died from cerebral ischaemia confirmed at the postmortem examination. Follow-up radiographs (F-Ur) were never performed in individuals with rapid clinical recovery. Among the F-Ur available, advanced improvement of the abnormalities were always depicted. In our series, the influenza virus infection did not lead to secondary bacterial pneumonia in any individual. interstiziali (Fig. 1) presenti nel 37,5% dei pazienti (15/40) , sono state il quadro di presentazione radiologico prevalente. Vetro smerigliato (Fig. 2) associato ad alterazioni interstiziali è stato identificato nel 22,5% (9/40) dei casi. Il quadro di presentazione misto (Fig. 3) , costituito d una combinazione di vetro smerigliato, consolidamenti parenchimali densi e opacità reticolari, è stato osservato nel 15% dei pazienti (6/40) . L'impegno parenchimale è stato bilaterale nel 70% dei casi (28/40), più comunemente i campi polmonari interessati sono stati quelli inferiori 70% (28/40) (Fig. 4) . L'impegno parenchimale è stato diffuso in 15 dei 40 pazienti (37,5%), con evoluzione in acute respiratory distress syndrome (ARDS) (Figg. 5, 6) in 3 casi (7,5%). Non abbiamo mai rilevato la presenza di adenopatia. In 4 pazienti abbiamo identificato un versamento pleurico di modesta entità. Due pazienti con malattia clinicamente grave e ARDS identificato con la radiografia del torace, sono stati studiati anche con TC, che ha meglio rilevato la presenza e l'estensione delle anormalità polmonari già rilevate. Uno di questi soggetti, pur mostrando un miglioramento della patologia polmonare nella TC di controllo, è morto per ischemia cerebrale confermata dall'autopsia. Non abbiamo mai effettuato controlli radiografici nei pazienti con rapido miglioramento clinico e nei pazienti sottoposti a monitoraggio la guarigione radiologica è sempre stata rapida. Nessuna Fig. 1 A 32- year-old man with swine-origin influenza virus. Chest X-ray shows parenchymal band consistent with atelectasis, and interstitial changes. Abnormalities were bilateral and involved the mid and lower lung zones. Fig. 3 A 16 -year-old man with swine-origin influenza virus. Chest X-ray shows patchy ground-glass opacity, interstitial changes and confluent centrilobular nodules. The lung abnormalities were unilateral and extensive, involving three lung zones.

A 54-year-old man with swine-origin influenza virus. Chest X-ray shows patchy, bilateral areas of ground-glass opacity in both lower zones, associated with interstitial changes.

. Chest X-ray shows bilateral areas of dense consolidation and ground-glass opacities. Lung involvement was extensive, consistent with ARDS. 

Influenza virus belongs to the orthomyxovirus family of RNA viruses, and human disease is predominantly caused by type A, the most virulent, which can easily mutate [9] . Many subtypes of type A have been identified on the basis of the occurrence of surface glycoproteins, haemagglutinin (H) and neuraminidase (N). The 2009 H1N1 virus contained a unique combination of gene segments that had not previously been identified in humans or animals [10, 11] .

Whereas the elderly and young children are at high risk for seasonal influenza, H1N1-related illness predominantly affects young individuals. The possible explanation for this phenomenon includes the fact that the young have a greater susceptibility to the virus, as proven on the basis of serological studies [12, 13] . Another reason might be that elderly people have some level of cross-reactive neutralising antibodies to the H1N1 virus [14] . However, it is also possible that there is a case-ascertainment bias because more young people are tested as part of outbreak investigations in schools [15] The majority of S-OIV infections reported has been mild illness. However, the pandemic strain of H1N1 virus can cause severe illness, including pneumonia and ARDS [16] [17] [18] [19] . In a pattern similar to patients with seasonal influenza, patients with severe illness had a high prevalence of underlying medical conditions [20, 21] . Risk factors for severe S-OIV illness are obesity, pulmonary diseases, cardiac diseases, neurological diseases, immunosuppression and pregnancy [16] [17] [18] [19] . One contributing factor for death during H1N1 infection may be delayed admission and/or delayed initiation of therapy [16, 19] . Concurrent bacterial infection does not appear to be a contributing factor to the severity of illness, and lung damage was mostly due to the primary effect of infection with influenza virus [16] [17] [18] [19] . Possible mechanisms of damage include direct injury to the respiratory epithelium with a secondary cytokine storm [22] .

The radiographic findings in viral pneumonia consist of interstitial changes, ground-glass opacities, centrilobular nodules and consolidation. These findings are variable and overlapping [23] and reflect the variable extent of the underlying histopathological features. The most prominent histological changes are seen in the bronchial epithelium and in the adjacent interstitial tissue. Necrosis and desquamation of the epithelial cells in the bronchial lumen, associated with endoluminal exudates, may produce obstructions. The mononuclear cells infiltrate the bronchial wall, and the inflammation may extend to the adjacent interstitial tissue. The radiological features of those changes are dystelectasis, with consolidations or air trapping [24, 25] and/or centrilobular nodules. superinfezione polmonare batterica è stata identificata nella popolazione in studio.

Il virus dell'influenza appartiene alla famiglia degli virus a RNA Orthomixovirus. La malattia umana è generalmente provocata dal tipo A, il più virulento, che può facilmente mutare [9] . Esistono una grande quantità di sottotipi del virus A, identificati per la presenza di glicoproteine, emoagglutinine (H) e neuraminidasi (N), sulla membrana di superficie. Il virus H1N1 identificato nel 2009 contiene una combinazione genetica mai identificata precedentemente nell'uomo e negli animali [10, 11] .

L'influenza H1N1 colpisce prevalentemente soggetti giovani, al contrario dell'influenza stagionale per la quale i bambini e gli anziani sono le popolazioni a rischio più elevato di malattia. Una delle possibili spiegazioni di questo fenomeno, comprovata da studi sierologici, è la grande suscettibilità dei giovani al virus [12, 13] . Un'altra ragione potrebbe essere la presenza di livelli di anticorpi adeguati al controllo del virus prodotti dalle persone anziane [14] . I dati statistici potrebbero comunque essere inquinati per un errore di campionamento, che viene effettuato prevalentemente nei giovani, in corrispondenza dell'insorgenza di epidemie scolastiche [15] .

La maggior parte delle influenze suine segnalate hanno presentato un decorso clinico di media gravità, ma il virus H1N1 è in grado di provocare malattia grave con polmonite e ARDS [16] [17] [18] [19] . I pazienti con malattia grave hanno presentano un'alta prevalenza di comorbilità [20, 21] , come accade in corso di influenza stagionale. I fattori di rischio correlati all'insorgenza di malattia grave sono: l'obesità, le malattie cardiache, neurologiche e polmonari pre-esistenti, l'immunosopressione, la gravidanza [16] [17] [18] [19] . Un ritardo nel ricovero e/o nell'inizio della terapia rappresentano un fattore prognostico negativo fino al decesso [16, 19] . In questi casi fatali le superinfezioni batteriche non hanno contribuito alla severità della malattia e il danno polmonare è stato essenzialmente provocato proprio dal virus. I possibili meccanismi di danno includono infatti l'azione del virus sull'epitelio respiratorio, diretta e indiretta per la successiva liberazione di citochine nell'ambito di complessi meccanismi flogistici [22] . I segni radiologici della polmonite virale consistono di alterazioni interstiziali, opacità a vetro smerigliato, noduli centrolobulari, consolidamenti, variamente rappresentati, e riflettono la presenza e l'estensione dei fenomeni istopatologici sottostanti [23] . Le principali alterazioni si osservano a livello dell'epitelio bronchiale e dell'adiacente tessuto interstiziale. Necrosi e desquamazione dell'epitelio nel lume bronchiale, in concomitanza con l'essudazione endoluminale, Interstitial thickening may produce reticular or linear opacities. Parenchymal inflammation initially involves the air space adjacent to the terminal and respiratory bronchioles. Variable extension and amount of exudate occur. Imaging findings are ground-glass opacities or dense consolidations.

Rapidly progressive pneumonia may be seen, particularly in patients with underlying medical conditions; in these cases, the lung histologically shows diffuse alveolar damage, comprising interstitial lymphocyte infiltration, airspace haemorrhage, oedema and hyaline membrane formation [26, 27] . In these cases chest X-ray and CT show the ARDS pattern [25, 28] . Pleural effusion is rare.

The literature data [16] [17] [18] [19] show that also in S-OIVrelated pneumonia, the radiographic pattern is consistent with the nonspecific pattern of viral pneumonia. In our series, the major radiological abnormalities were interstitial changes, with or without patchy ground-glass opacities and centrilobular nodules, mostly bilateral and located in the lower lung zones. These data differ considerably from the literature data [16] [17] [18] [19] in which the most prevalent pattern is patchy ground-glass opacities, more frequently bilateral and located in lower zones.

In our series, extensive lung abnormalities with involvement of three or more lung zones were observed in 37.5% of patients, and ARDS was observed in three patients, all with underlying medical conditions. CT was performed only three times in our series, as it is known that it does not add information in patients with evident radiographic abnormalities, leading to only a modest increase in diagnostic accuracy [29, 30] . On the contrary, incipient and/or overt ARDS should be further investigated with CT, which is more sensitive and specific than radiography for detecting thoracic abnormalities and may be useful for guiding treatment decisions in critical patients, providing a more accurate evaluation of the extent and distribution of the disease.

In conclusion, S-OIV-related pneumonia does not differ in radiographic presentation from the other viral pneumonias, and it may evolve into ARDS. Therefore, during S-OIV, recognition of the radiological findings of pneumonia is imperative for determining appropriate management. Chest radiograph is an effective and adequate tool for identifying S-OIV-related pneumonia, whereas CT is helpful in individuals with severe illness to depict incipient ARDS and to orient prompt therapeutic management.

possono determinare l'ostruzione bronchiale, e l'infiltrazione flogistica si può propagare all'interstizio peribronchiale. Questo substrato anatomopatologico si traduce radiologicamente nella presenza di distelectasie, con consolidamenti o intrappolamento aereo [24, 25] , e/o noduli centrolobulari. L'impegno interstiziale produce opacità reticolari o lineari. L'essudazione endoalveolare è responsabile, relativamente alla sua entità, della comparsa di consolidamenti parenchimali, densi o vetro smerigliato.

Nei soggetti con comorbilità l'evoluzione della polmonite può essere rapidamente ingravescente per la comparsa di un danno alveolare con aspetti anche emorragici, formazione di membrane ialine e concomitante impegno interstiziale [26, 27] . In questi casi l'ARDS sarà la manifestazione identificabile radiologicamente [25, 28] . Il versamento pleurico non è frequente.

I dati desumibili dalla letteratura [16] [17] [18] [19] mostrano che anche nella polmonite correlata all'influenza suina, la presentazione radiologica è quella aspecifica delle polmoniti virali, ma mentre nella nostra casistica le alterazioni identificate con maggiore frequenza sono state quelle interstiziali, più spesso bilaterali e con prevalente impegno dei campi polmonari inferiori, il quadro di presentazione più frequentemente rilevato dalla letteratura [16] [17] [18] [19] , è stato il consolidamento multifocale a vetro smerigliato, più spesso bilaterale e localizzato nei campi polmonari inferiori.

Nel 37,5% dei nostri casi la malattia è stata diffusa, coinvolgendo 3 o più campi polmonari e in 3 pazienti la polmonite si è complicata per la comparsa di ARDS. È noto che la TC, nei casi in cui la presenza di patologia polmonare sia già stata rilevata mediante radiografia del torace, non aggiunge un significativo incremento di accuratezza diagnostica [29, 30] . Nella nostra esperienza la TC è stata effettuata solo 3 volte, in 2 soggetti con clinica rapidamente ingravescente, per i quali la precoce identificazione del tipo e dell'entità delle alterazioni polmonari ha comportato una modificazione nelle scelte terapeutiche.

In conclusione, la presentazione radiologica della polmonite da virus dell'influenza suina non differisce dalle altre polmoniti virali e l'ARDS può rappresentarne la temibile evoluzione. Per questo motivo in casi clinicamente selezionati è indispensabile uno studio radiologico del paziente, al fine di orientare le successive scelte terapeutiche. La radiografia del torace si conferma strumento efficiente ed efficace, generalmente sufficiente, la TC deve essere riservata ai casi critici che richiedono aggiustamenti e/o modificazioni dei regimi terapeutici. 

Update: Novel influenza A (H1N1) virus infectionsworld wide

Lack of airborne transmission during outbreak of pandemic (H1N1) 2009 among tour group members, China

H1N1 flu situation update

Excess morbidity and mortality associated with influenza in England and Wales

Influenza-associated hospitalizations in the United States

Interim guidance on case definitions for swine influenza A (H1N1) human case investigations

Glossary of terms for thoracic radiology: recommendations of the Nomenclature Committee of the Fleischner Society

Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society

Influenza, an existing public health problem

Emergence of a novel swineorigin influenza A (H1N1) virus in humans

Antigenic and genetic characteristics of swine-origin 2009 A (H1N1) influenza viruses circulating in humans

Cross-reactive antibody responses to the 2009 pandemic H1N1 influenza virus

Clinical trials of bivalent influenza A/New Jersey/76-A/Victoria/75 vaccines in the elderly

Immunogenicity and reactogenicity of influenza A/New Jersey/76 virus vaccines in normal adults

Swine-origin influenza A (H1N1) virus infections in a school -New York city

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

Imaging findings in a fatal case of pandemic swine-origin influenza A (H1N1)

Chest radiographic and CT findings in novel swine-origin influenza A (H1N1) virus (S-OIV) infection

Hospitalized patients with 2009 H1N1 influenza in the United States

Clinical features of influenza A in older hospitalized persons