Meeting Abstracts
» Imaging findings in acute invasive pulmonary aspergillosis: Utility and reliability of halo sign and air-crescent sign in the diagnosis and treatment of IPA in high-risk patients.

Reginald E. Greene, Paul Stark, Jörg-W. Oestmann, Christine Durand1

Objectives To characterize the radiological presentation of acute invasive pulmonary aspergillosis (IPA); to compare baseline imaging findings of IPA that was confirmed or unconfirmed by an independent data review committee (DRC); to evaluate reliability and utility of baseline halo sign (HS) and air crescent sign (ACS) in the diagnosis and treatment of IPA in high-risk patients.

Methods Among 391 high risk patients who participated in a recent comparative, multicenter aspergillosis treatment study2, the following baseline data were analyzed in a subgroup of 343 patients who had pulmonary involvement and imaging studies at baseline: prospectively-acquired imaging diagnoses and fungal findings, and global treatment response (TR) at 12 weeks. DRC-confirmed baseline HS or ACS satisfied criteria for the diagnosis of IPA in hematopoietic stem cell transplant (HSCT) recipients, and patients with a hematologic condition (HC) associated with neutropenia. Independent confirmation diagnoses of HS and ACS by DRC radiologists was based on standard definitions, and were compared with those of site investigators.

Results In 254 patients with DRC-confirmed baseline diagnoses of IPA, 141 had at least one nodular lesion with HS (56%), and 15 with ACS (6%). Both mycological and CT data were present at baseline in 138 patients (54%): 59 (43%) had either a HS (n=48) or ACS n= 11). Among the 79 patients in whom baseline diagnoses of IPA were based solely on fungal findings (none had a HS or ACS on CT), 66 had nodular lesions (84%) including 41 with unsharp margination (UM), and 13 had other lesions (7 with consolidations, 2 with centrilobular opacities, and 1 each with a cavitary lesion, non-nodular infarct, ground-glass opacity, or pleural lesion). TR was satisfactory in 79 of the 154 patients who had a HS or ACS on CT (51%), and in 51 of the 95 patients who had a HS or ACS on CT but lacked positive fungal findings (54%). Compared to the preceding two groups, the fraction of patients with satisfactory TR was smaller in patients in whom the diagnosis of IPA was based on fungal findings alone 30/98 (31%, p <= 0.001, x2). HS or ACS was confirmed by the DRC in 95/148 patients (64%) who were entered into the study based on imaging criteria alone. The major radiological findings in the 53 patients with un-confirmed IPA included nodular lesions in 48 (91%), 39 of which were associated with UM.

Conclusions Baseline imaging findings in IPA were dominated by nodular lesions, especially those associated with HS or ACS, or UM. The DRC was able to confirm a baseline HS or ACS in 2/3 of high-risk patients referred for entry based solely on imaging criteria. Non-reproducibility of HS may have resulted from difficulty in differentiation of HS from UM. A higher fraction of high risk patients with IPA had a satisfactory response to anti-fungal therapy when based on HS or ACS (either with or without supporting fungal findings) as compared to those in whom IPA was based on fungal findings alone.

1 In collaboration with the Global Aspergillus Study Group and the Invasive Fungal Infection Group of the EORTC. Supported by Pfizer.
2 Herbrecht R, Denning DW, Patterson TF, Bennett JE, Greene RE, Oestmann J-W, Kern WV, Marr KA, Ribaud P, Lortholary O, Sylvester R, Rubin RH, Wingard JR, Stark P, Durand C, Caillot D, Thiel E, Chandrasekar PH, Hodges MR, Schlamm HT, Troke PF, De Pauw B. Voriconazole Versus amphotericin for primary therapy of invasive aspergillosis. N Engl J Med 2002;347:408-15.

This page last modified on 3 February 2004 by Society Webmaster.