SciELO - Scientific Electronic Library Online

vol.39 issue5Clinical findings in 16 patients with tomographic diagnosis of schizencephalyAssessment of cardiac function by magnetic resonance imaging: segmented x real-time steady-state free precession sequences author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Radiologia Brasileira

Print version ISSN 0100-3984On-line version ISSN 1678-7099

Radiol Bras vol.39 no.5 São Paulo Sept./Oct. 2006 



Computed tomography in the assessment of angioinvasive pulmonary aspergillosis in patients with acute leukemia*



Renata Carneiro LeãoI; Edson MarchioriII; Rosana RodriguesIII; Arthur Soares Souza Jr.IV; Emerson L. GasparettoV; Dante L. EscuissatoVI

IMD, Radiologist at Clínicas Luiz Felippe Mattoso, IRM and Centro de Mastologia do Rio de Janeiro
IITitular Professor of Radiology at Universidade Federal Fluminense, Adjunct Coordinator for the Course of Post-Graduation in Radiology at Universidade Federal do Rio de Janeiro
IIIMD, Radiologist at Hospital Universitário Clementino Fraga Filho Radiodiagnosis Service – Universidade Federal do Rio de Janeiro, and at Clínica Lab's
IVAdjunct Professor of Radiology at Faculdade de Medicina de São José do Rio Preto, SP
VAdjunct Professor of Radiology at Universidade Federal do Rio de Janeiro
VIAssistant Professor of Radiology at Universidade Federal do Paraná, Curitiba, PR

Mailing address




OBJECTIVE: The aim of this study was to evaluate the main findings of computed tomography in patients presenting acute leukemia complicated by angioinvasive aspergillosis.
MATERIALS AND METHODS: Computed tomography images of 19 patients were retrospectively studied for the presence of consolidations, nodules and masses, with or without presentation of halo sign, cavitation and air crescent sign.
RESULTS: Consolidation was the most frequent finding, occurring in 12 of the 19 cases, most of them presenting the halo sign; cavitation was found in 5 of 12 cases, one of them with air crescent sign. Nodules and masses occurred respectively in six and four cases, most of them with halo sign. Cavitation was found in only one case of mass. Other findings observed were: crazy-paving pattern (two cases), patchy areas of ground-glass attenuation opacity (three cases) and pleural involvement (seven cases) under the form of effusion or thickening.
CONCLUSION: Areas of consolidation, mass or nodule, even a solitary one, presenting halo sign on CT images evaluated in an appropriate clinical context are highly suggestive of angioinvasive aspergillosis.

Keywords: Angioinvasive pulmonary aspergillosis; Acute leukemia; Computed tomography.




Pulmonary aspergillosis is a mycotic infection caused by species of Aspergillus, usually A. fumigatus, whose main presentations are allergic bronchopulmonary as pergillosis and sensitivity pneumonitis, which are the most frequent forms of hypersensitivity reaction against the Aspergillus, aspergilloma, semi-invasive aspergillosis and invasive aspergillosis. The disease is characterized by a spectrum of clinical and radiological findings directly related to the immunologic condition of the host or to the presence of structural pulmonary disease(1–3).

The invasive form of pulmonary aspergillosis occurs primarily in deeply immunocompromised individuals, especially in patients presenting malignant hematological disease, most commonly acute leukemia. Invasive pulmonary aspergillosis may be bronchoinvasive or angioinvasive, clinically very similar, but presenting different radiological and histological aspects. The angioinvasive form is the most frequent one(4–6).

Computed tomography (CT), especially high resolution computed tomography (HRCT), has been systematically utilized to support the early diagnosis of angioinvasive aspergillosis, allowing an effective antifungal treatment and consequential improvement of the disease prognosis(7–9).

The most significant radiological finding is the halo sign, characterized by a halo of ground-glass attenuation surrounding the nodule, mass or consolidation, highly indicative of angioinvasive aspergillosis, detectable only early in the course of the disease, i.e. during the neutropenic period. Cavitation frequently develops later in the course of immunological recovery(4,10–13).

The present study objective was to evaluate the main findings on chest CT images of 19 patients with acute leukemia associated with angioinvasive pulmonary aspergillosis.



The retrospective analysis covered 19 chest CT of patients presenting acute leukemia with angioinvasive pulmonary aspergillosis diagnosis confirmed by anatomopathological studies, clinical radiological correlation or an appropriate therapeutic response. These cases were randomly gathered from files of several institutions in Rio de Janeiro, São Paulo and Salvador – Brazil.

In this study, an adequate standardization of protocols was unfeasible because CT imaging was performed in different institutions. The greatest part of the CT examinations applied the high-resolution technique, consisting of very narrow X-ray beam collimation (1-2 mm thick slices) at 10 mm intervals, from the pulmonary apex to the diaphragmatic cupula, and a so called "high spatial reconstruction algorithm, in a supine patient. Images were recorded on radiological films with windows ranging between 1,200 and 1,600 UH and center between –450 and –650 UH for evaluating pulmonary parenchyma, and windows ranging between 250 and 500 UH, and center between 30 and 50 UH for evaluating mediastinum, depending on the equipment utilized in each institution.

The images were independently analyzed by two radiologists and decisions were made by consensus.

The criteria for definition of tomographic patterns are in compliance with the Fleischner Society Definitions(14) . The terminology applied was that of the consensus elaborated by Colégio Brasileiro de Radiologia (Brazilian College of Radiology)(15).



The study covered 19 cases of angioinvasive pulmonary aspergillosis in patients with acute leukemia, 12 female and seven male, age range between two and 72 years, mean age 36 years.

Consolidation occurred in 12 of the 19 cases (63.2%) (Figure 1), three of them (25%) associated with nodules. In eight cases (66%), consolidations were multiple and in four (33.3%), solitary. As for localization, consolidations were peripheral in ten of 12 cases (83.3%) and central in five (41.6%) and, in three of these 12 cases, both (peripheral and central) were associated (25%). Air bronchogram was observed in six patients (50%). Halo sign was found in nine of the 12 cases of consolidation (75%), while cavitation occurred in six of these cases (50%), two with air crescent sign (16.7%).



Nodules were found in six of the 19 cases (31.6%) (Figure 2A), multiple in five (83.3%) and solitary in only one case (16.7%). In four (66.7%) of the six cases, nodules presented well-defined borders. Small nodules occurred in all of the six cases (100%), and in one of them small and large nodules were associated. Halo sign was observed in five of the six cases (83.3%), while in none of them cavitation was found.



Masses with irregular contour were found in four of the 19 cases (21.1%) (Figures 2B and 2C), three of them being solitary. Halo sign surrounding these masses was observed in three (75%) of the four cases, while cavitation occurred in only one of these cases (25%).

Areas of ground-glass attenuation non-associated with nodules and consolidations were found in three of the 19 cases (15.8%).

Superimposition of ground-glass opacities and thickened interlobular septa (crazing paving) was found in two of the 19 cases (10.5%), one of them associated with masses and another with consolidation (Figure 3).



Pleural involvement occurred in seven cases (36.8%), in three of them under the form of a very small effusion (42.9%) and in the other four, as pleural thickening (57.1%). The finding was unilateral in four (57.1%) and bilateral in three cases (42.9%).



All of the 19 patients with angioinvasive pulmonary aspergillosis were immunocompromised as a result of chemotherapy for treatment of acute leukemia which is in compliance with the literature since it describes extended granulocytopenia in patients undergoing treatment with chemotherapy as the most usual context where the angioinvasive pulmonary aspergillosis develops(4,7,10,16,17). There are studies demonstrating that about 50% of neutropenic patients undergoing treatment with chemotherapy, or submitted to bone marrow transplantation, present severe fungal infection at some phase of their evolution(16,18).

Clinical and radiographic presentations of angioinvasive pulmonary aspergillosis are similar to those of other infectious pneumonias. Additionally, by the time when the antifungal therapy could change the patient's survival, the sputum culture results positive in less than 10% of patients, and invasive procedures like biopsy usually are contraindicated because of severe thrombocytopenia and respiratory compromising, besides presenting a very low diagnostic sensitivity and high rates of false-negative results(4,10–12,16).

Computed tomography has produced a high impact on the conduct in relation to patients with angioinvasive pulmonary aspergillosis, since it is an important resource for the early diagnosis of this disease, especially when invasive procedures offer high risks. Compared to CT, HRCT provides a more clear definition of lesions, particularly ground-glass opacities present in the early phases of the disease(1,4,8,10–12,16–18).

Recent studies have reported the relevance of HRCT for the early diagnosis of angioinvasive pulmonary aspergillosis, through the detection of the typical halo sign. This sign is highly suggestive for the diagnosis of angioinvasive pulmonary aspergillosis in an early phase when radiographic findings are non-specific and the sputum culture presents negative results(4,9–11,16). Although non-pathognomonic and possibly associated with a range of infectious and noninfectious processes, the halo sign is highly suggestive of angioinvasive pulmonary aspergillosis when detected in neutropenic patients, and some authors consider it sufficiently characteristic to justify the commencement of the antifungal therapy(4,8,10,18,19).

Another finding considered as suggestive of angioinvasive pulmonary aspergillosis is the air crescent sign when associated with an appropriate clinical presentation. However, contrarily to the halo sign, the air crescent sign appears later in the course of the disease and is seen in the phase of the infection resolution, coinciding with the improvement of neutropenia. This fact makes the clinical value of this finding limited in the early diagnosis of the disease, but is reported as an indication of good prognosis(4,8,10,11,16,18,19).

Won et al.(19) have found consolidation in 80% of leukemic patients with angioinvasive pulmonary aspergillosis, such consolidations being associated with nodules in 20% of the cases, a data that is similar to those found in this study, where consolidations occurred in 12 of 19 cases (63%), and associated with nodules in 25% of cases. Consolidations were predominantly peripheral in 83.3% in this casuistic, while in the great majority of studies in the literature these percentages have reached approximately 100%. In our study, consolidations were multiple in 66.6% of cases which is in agreement with authors like Kuhlman et al.(11) and Won et al.(19).

The halo sign in consolidations was found in nine of our 12 cases (75%), which coincides with the findings of Won et al.(19), where this percentage was 81%. Cavitation was observed in just six of our 12 cases (50%), with air crescent sign in only two of them. Franco et al.(16) have found consolidation in two of their seven cases (28.5%), both cavitary, one of them with air crescent sign. Kuhlman et al.(4) have found cavitary consolidations in five of the seven cases (71.4%).

In our study, nodules occurred in 31.6% of cases, 83.4% being multiple and 16.7% solitary, which is different from the study of Kuhlman et al.(9), whose percentages were respectively 55% and 45%.Well defined nodules were found in 66.7% of our cases, differently from studies like that of Franco et al.(16), where all the nodules presented regular contours.

A ground-glass halo was found in 83.4% of cases presenting nodules, which is in agreement with the studies of Kuhlman et al.(11) and Franco et al.(16), where the halo sign in nodules was found respectively in 88.8% and 80% of cases. On the other hand, Mori et al.(18) and Won et al.(19) have found this sign respectively in only 19% and 40% of cases. No cavitation was found in any case, differently from the findings of Kuhlman et al.(11), where cavitation was observed in five of seven cases (71.4%). In the histopathological study areas of necrosis were found with radially-disposed hyphae in the central region of the lesions, involved by inflammatory cells and blood-filled alveoli. This finding is compatible with those of the studies of Logan et al.(9) and Santos et al.(20).

Masses were found in 21% of our cases, the halo sign being present in 75% of them. This finding is different from those of Kuhlman et al.(11), whose percentages were respectively 55% and 22%. Cavitation was found in only one of our cases of mass (25%), while Kuhlman et al.(4) observed percentages of up to 71%.

Pleural involvement was observed in seven of our patients (35%), similarly to the majority of the referred authors(7,20,21). In five cases, the pleural involvement occurred in the form of an effusion, while in the other two cases only pleural thickening was observed, probably with no relationship with the fungal lesion.

The mosaic pattern of attenuation (crazy paving) (thickened interlobular septa associated with ground-glass opacity) occurred in two cases (10.5%), which is in agreement with the findings of Franco et al.(16), who have found this aspect in only one of the eight cases studied (12.5%). Data in the literature about this finding in patients with angioinvasive pulmonary aspergillosis are scarce, difficulting its analysis.

The clinical diagnosis of angioinvasive pulmonary aspergillosis is difficult and, generally, the prognosis is poor, with very high mortality rates. The role of computed tomography, especially HRCT, has become significant in the whole course of the disease, from its early detection, indicating the commencement of the antifungal therapy, to the monitoring of possible recurrences during the chemotherapy treatment(10,12,16). So, in patients with neutropenia under the risk of angioinvasive pulmonary aspergillosis, the systematic use of HRCT is justified since this is the best way to establish an early and probably specific diagnosis, once the halo sign is detected. Because of the short period during the course of the disease in which the halo sign is detectable by HRCT, the scan should be performed as soon as the clinical suspicion is raised, allowing an early diagnosis with a consequential effect on the morbidity and mortality rates. Additionally, HRCT is a rapid and non-invasive method, while invasive procedures like biopsy usually are contraindicated because of the poor condition of the patients and because sputum culture usually presents positive results in less than 10% of cases(4,8,11,12,18).

Serial CT studies also have played a significant role in the follow-up of such patients, since findings like cavitation usually represent a phase of immunological recovering with reabsorption of devitalized tissue(8,10).



1. Klein DL, Gamsu G. Thoracic manifestations of aspergillosis. AJR Am J Roentgenol 1980;134: 543–552.         [ Links ]

2. Pozes AS, Maranhão B, Marchiori E, Zanardi PN, Martins EML, Vabo KA. Aspergilose pulmonar semi-invasiva: relato de caso. Rev Imagem 2002; 24:197–201.         [ Links ]

3. Yousem SA. The histological spectrum of chronic necrotizing forms of pulmonary aspergillosis. Hum Pathol 1997;28:650–656.         [ Links ]

4. Kuhlman JE, Fishman EK, Burch PA, Karp JE, Zerhouni EA, Siegelman SS. Invasive pulmonary aspergillosis in acute leukemia. The contribution of CT to early diagnosis and aggressive management. Chest 1987;92:95–99.        [ Links ]

5. Nuño CG, Alfonso PP, Vazquez JCR, Gomez MMR, Prats IP, Gonzáles JG. Aspergilosis pulmonar: un nuevo enfoque en la reemergencia. Acta Médica 2000;9:67–72.         [ Links ]

6. Webb WR, Müller NL, Naidich DP. High-resolution CT of the lung. 3rd ed. New York: Raven, 2001.         [ Links ]

7. Aquino SL, Kee ST, Warnock ML, Gamsu G. Pulmonary aspergillosis: imaging findings with pathologic correlation. AJR Am J Roentgenol 1994;163:811–815.         [ Links ]

8. Caillot D, Couaillier JF, Bernard A, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19:253–259.         [ Links ]

9. Logan PM, Primack SL, Miller RR, Müller NL. Invasive aspergillosis of the airways: radiographic, CT, and pathologic findings. Radiology 1994;193:383–388.         [ Links ]

10. Kuhlman JE, Fishman EK, Burch PA, Karp JE, Zerhouni EA, Siegelman SS. CT of invasive pulmonary aspergillosis. AJR Am J Roentgenol 1998;150:1015–1020.         [ Links ]

11. Kuhlman JE, Fishman EK, Siegelman SS. Invasive pulmonary aspergillosis in acute leukemia: characteristic findings on CT, the CT halo sign, and the role of CT in early diagnosis. Radiology 1985;157:611–614.        [ Links ]

12. Marchiori E, Valiante PM, Souza Jr AS. Nódulos com sinal do halo na aspergilose pulmonar angioinvasiva: correlação da tomografia computadorizada de alta resolução com a anatomopatologia. Radiol Bras 2002;35:195–198.         [ Links ]

13. Persegani MK, Marchiori E, Rodrigues R, et al. O "sinal do halo" na tomografia computadorizada de alta resolução do tórax. Rev Imagem 2001;23: 225–231.         [ Links ]

14. Austin JHM, Müller NL, Friedman PJ, et al. Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology 1996;200:327–331.         [ Links ]

15. Souza Jr AS, Araújo Neto C, Jasinovodolinsky D, et al. Terminologia para a descrição de tomografia computadorizada do tórax. Radiol Bras 2002; 35:125–128.         [ Links ]

16. Franco IO, Marchiori E, Souza Jr AS, Melo AVF, Crespo SJV, Irion K. Aspergilose pulmonar angioinvasiva: aspectos na tomografia computadorizada. Rev Imagem 2002;24:77–82.         [ Links ]

17. Kenney HH, Agrons GA, Shin JS; Armed Forces Institute of Pathology. Best cases of the AFIP. Invasive pulmonary aspergillosis: radiologic and pathologic findings. RadioGraphics 2002;22: 1507–1510.         [ Links ]

18. Mori M, Galvin JR, Barloon TJ, Gingrich RD, Stanford W. Fungal pulmonary infections after bone marrow transplantation: evaluation with radiography and CT. Radiology 1991;178:721–726.         [ Links ]

19. Won HJ, Lee KS, Cheon JE, et al. Invasive pulmonary aspergillosis: prediction at thin-section CT in patients with neutropenia – a prospective study. Radiology 1998;208:777–782.         [ Links ]

20. Santos MLO, Marchiori E, Vianna AD, Souza AS, Moraes HP. Opacidades em vidro fosco nas doenças pulmonares difusas: correlação da tomografia computadorizada de alta resolução com a anatomopatologia. Radiol Bras 2003;36:329–338.         [ Links ]

21. Franquet T, Müller NL, Giménez A, Guembe P, de LaTorre J, Bagué S. Spectrum of pulmonary aspergillosis: histologic, clinical, and radiologic findings. RadioGraphics 2001;21:825–837.        [ Links ]



Mailing address:
Prof. Dr. Edson Marchiori
Rua Thomaz Cameron, 438, Valparaíso
Petrópolis, RJ, Brazil 25685-120

Received January 11, 2006.
Accepted after revision January 30, 2006.



* Study developed at Hospital Universitário Clementino Fraga Filho Radiodiagnosis Service – Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License