Guidance on Imaging for Invasive Pulmonary Aspergillosis and Mucormycosis

  • Post category:Expert Opinion
  • Reading time:12 mins read


Clinical imaging has a role in early identification and helps guide additional testing in patients with suspected Invasive fungal diseases (IFD). While the presence of particular lesions may raise the risk of IFD, clinical imaging diagnosis of IFD lacks precision. Nodules, masses, segmental or subsegmental consolidations, atelectasis, ground-glass opacities, a tree-in-bud pattern, cavities, or pleural effusions are common radiologic symptoms of IFD. These radiologic abnormalities, on the other hand, can be seen in various viral or inflammatory pulmonary conditions.

Findings including the halo sign (HS), reversed HS, hypodense sign (HDS), and air crescent sign (ACS) may help distinguish between mould infections and non fungal pneumonias in particular clinical circumstances. Clinical studies that are frequently not comparable, as radiologic and other diagnostic methods, as well as medications and outcome evaluations, have changed dramatically over the previous two decades, make determining the real specificity of various radiologic lesions for IFD difficult.

In addition, findings on radiologic signs of IFD are based on hypothetical or suspected IFD, which may or may not be actual IFD. While imaging is still an important part of IFD diagnosis, the best test to use and the validity of the results to guarantee a confident and accurate diagnosis are still up for debate. The goal of this study was to evaluate the relevance of imaging in the diagnosis and management of pulmonary IFD in adults, with a particular focus on invasive pulmonary aspergillosis (IPA) and pulmonary mucormycosis (PM).

Radiologic testing for diagnosis of IFD


The CXR in immunocompromised hosts is less frequently used today as it suffers from low sensitivity, especially for early detection of pneumonia. The benefit of lower radiation compared with low-dose and ultra–low-dose CT techniques has dropped while availability of CT has increased. Post-processing of digital radiographs, including techniques to limit the visibility of overlying ribs, increase the conspicuity of focal opacities. However, these techniques cannot compensate for the overall lower contrast resolution of CXR compared with CT.

Many lesions, especially small nodules (<10 mm), will be indistinguishable from other parenchymal abnormalities; thus, the utility of CXR for the confident diagnosis of opportunistic lung infections is limited.


Volumetric thin-section (high-resolution) CT with a slice thickness of approximately 1 mm is the method of choice for lung imaging. The radiation dose required for a chest CT is highly dependent on the scanner and acquisition technique. As many patients with suspected IFD have underlying malignancy and have received chemotherapy or radiation therapy (which is 5000–60 000 times more radiation compared with a diagnostic CT scan), the impact of a possible and late negative side effect of diagnostic radiation seems less relevant.

CT imaging for the diagnosis of IFD can improve the early diagnosis and management of IPA in immunocompromised patients. Thin-section CT imaging can reveal distinct lesions early in infection that would not be detected or well characterized by conventional CXR imaging. Moreover, CT has a role in the follow-up of IFD to monitor lesion size, identify potential complications such as vessel erosion or bronchial compression, and detect typicals radiomorphologic findings of later stages of infection, which may be linked to therapeutic response and prognosis.


The principal benefits of MRI, in contrast with CT, are the absence of ionizing radiation and the superior resolution in solid organs. First, the relative paucity of protons in the lungs, which consist of 90% air, means that the signal-to-noise ratio is intrinsically low. Second, susceptibility artifacts related to air–tissue interfaces in the lungs cause a rapid diminution of the MR signal; and there is the effect of cardiac/respiratory motion on image quality. The most common contraindication is the presence of implanted devices (ie, pacemakers, certain intracranial aneurysm clips, or hemodynamic support devices) as well as claustrophobia.

High-speed gradient systems and sequences with short echo times have increased the role of MRI in diagnosis of IFD as an alternative option, with sensitivity and specificity comparable to those of CT. Nevertheless, the acquisition of quality MR images requires an elaborate optimization of technique and it requires that patients are able to sufficiently cooperate and hold their breath 20–50 times for 10–20 seconds during table time of 20–30 minutes.


The most common clinically used radionuclide in PET imaging is 18-fluoro-2-deoxyglucose (FDG), which is transported into cells and accumulates. One disadvantage of PET imaging alone is that images are of intrinsically poor resolution; however, this limitation has been overcome with the development of PET/CT machines. The principle role of FDG-PET imaging is that many tumor cells have a higher metabolic activity (compared with normal tissue) that, in turn, requires an increased expression of surface glucose transporters.

The accumulation in activated granulocytes and macrophages provides a possible role for PET in the search for a focus of inflammatory diseases. Infectious lesions that recruit inflammatory cells (that, like cancer cells, also have a high metabolic rate) may also demonstrate significant FDG avidity, resulting in PET/CT scans that are false-positive for cancer. A role of FDG-PET as a complement to CT imaging for the initial diagnosis and follow-up of IFD has been evaluated

CT findings associated with IPA

Well-circumscribed lesions (nodules) represent the main radiologic finding of IPD. Nonspecific and less common findings of IPA include consolidation, pleural effusions, ground-glass opacities, tree-in-bud-lesions, and atelectasis.


Nodules are defined as round opacities, at least moderately well marginated and no greater than 30 mm in maximum diameter In patients with hematologic malignancies, nodules are the most frequent lesions on CT and are more commonly observed in IFD compared with bacterial or viral infections. However, nodules are nonspecific for IPA and can be seen in pulmonary malignancies, lymphoma, secondary malignancies, and bacterial infections (Figure 1).

IPA is a common cause of pulmonary nodules in solid organ transplant recipients. Nodules (measuring 10–30 mm in diameter) were found to be prevalent and suggestive of IFD among liver transplant recipients. In lung transplant recipients, solitary macronodules (typically without a perifocal halo) may represent IPA or other causes, such as lung cancer or post-transplant lymphoproliferative disease. Centrilobular tree-in-bud nodular opacities may be present, indicating inflammatory disease in the small airways. The use of serial CT to monitor the evolution of nodular lesions has been shown to be useful in the follow-up of IPA and assessment of the outcome in patients with hematologic malignancy.

Halo Sign (HS)

The term “halo sign” refers to a pulmonary nodule or mass surrounded by a halo of ground-glass opacification on chest CT; it corresponds to an area of pulmonary infarction surrounded by alveolar hemorrhage (Figure 2A, 2B).

In a large cohort of 235 probable IPA cases, most with hematologic malignancy, the prevalence was 61%. The sensitivity of the HS for IPA is variable and related to the degree of the host’s immunosuppression and the stage of infection, but specificity is limited.

In general, in patients with hematologic malignancies and neutropenia, the HS is present on baseline CT (at the onset of symptoms) in >70% of IPA cases. The HS is less frequently observed in the pediatric population. The HS was reported to have good specificity (>90%) and positive predictive value (>60%) for the diagnosis of IPA in several studies.

Reversed HS

The reversed HS (RHS) is an area of ground-glass opacification with a peripheral ring of consolidation originally described in patients with cryptogenic organizing pneumonia (Figures 3, 4A).

On histopathologic exam in the setting of angioinvasive fungal disease, the reversed HS corresponds to peripheral hemorrhage around infarcted lung tissue. This CT sign is also called the “atoll sign” and can be seen in several diseases, including pulmonary infarction (secondary to thromboembolism), sarcoidosis, and tuberculosis. A recent study suggested it occurs more frequently in patients with PM than IPA (54% vs 6%; P < .001)

Air Crescent Sign (ACS)

The ACS is a peripheral crescentic-shaped collection of air that separates the wall of a cavity from an inner mass (Figure 4B).

In contrast to the HS, the ACS is observed at later stages of IPA, typically appearing on CT imaging about 2 weeks after the initial diagnosis of IPA and usually preceding the appearance of more complete cavitation. It may be observed in as many as one-half to two-thirds of cases during follow-up imaging of IPA but is usually not present at initial presentation; thus, its diagnostic value is limited. The ACS has been described in other fungal infections (notably PM), bacterial infections, and noninfectious conditions. Moreover, the ACS is rarely observed in nonhematologic patients or in the pediatric population.

Hypodense Sign (HDS)

The ACS has been described in other fungal infections (notably PM), bacterial infections, and noninfectious conditions. The HDS on nonenhanced CT is a discrete central area of lower attenuation that becomes detectable within a focus of consolidation or a mass (Figure 5).

In contrast to this is the central ground-glass component characteristic of the RHS, best seen in lung window settings. The central necrosis may be associated with a sudden stop of an air-filled bronchus (bronchus cutoff), indicating the underlying parenchymal distortion. The HDS, appearing about 1 week after the initial nodular presentation and 1 week before the development of the ACS, may represent an intermediate radiologic stage of IPA and an adjunct in the diagnosis of IPA.

Other Findings of IPA

The ACS has been described in other fungal infections (notably PM), bacterial infections, and noninfectious conditions. Concomitant pulmonary infections and noninfectious lung diseases in patients with IPA are relatively common and may influence the radiographic findings. Bronchial wall thickening associated with the tree-in-bud pattern on CT was a common finding among lung transplant recipients with airway-invasive aspergillosis but is less common in IPA. These findings may be the only radiologic signs of IPA in lung transplant recipients as the typical macronodular lesion may be absent. The recent observation of the increasing incidence of IPA among intensive care unit patients with severe influenza also poses a diagnostic challenge because of the lack of specificity of lung lesions, which may mimic bacterial pneumonia on chest CT.

Radiologic CT changes associated with PM

The ACS has been described in other fungal infections (notably PM), bacterial infections, and noninfectious conditions. In one study that compared CT findings of PM and IPA in patients with hematologic malignancies, PM patients were more likely to have multiple nodules (≥ 10) and pleural effusions. In a recent study, there was an increase in the frequency of the RHS in patients with PM compared with patients with IPA (54% vs 6%; P < .001).

Another study, including 37% of patients with hematologic malignancy and 63% with diabetes mellitus and/or Solid organ transplant (SOT), found that consolidation was the most frequent presentation of PM (65% of cases), followed by cavitation (40%) and masses (25%); nodules were observed in only 16% of cases. The RHS appears to be more specific of the early phase of PM in patients with hematologic malignancies


Radiologic assessment of the lung is an important component of the diagnostic work-up and management of IFD, and CT imaging is recommended. One finding, the HS, is highly suggestive of IPA and associated with specific stages of the disease. In other populations, both IPA and PM are more frequently associated with “atypical” non nodular presentations, with consolidation, ground-glass opacities, or tree-in-bud patterns associated with bronchial wall thickening.

The group acknowledged this and broadened the radiologic criteria of IFD in the updated definitions, also including wedge-shaped and segmental or lobar consolidation as a fourth criterion of IPA and the RHS for PM, in addition to the 3 criteria of the previous definitions. While IFD criteria can be restricted to these radiologic features for defining IFD in clinical trials, the aforementioned CT lung abnormalities can be suggestive of IFD and should be interpreted in an appropriate context of host risk for guiding therapeutic decisions.

Source: Barbara D Alexander, Frédéric Lamoth, Claus Peter Heussel, Cornelia Schaefer Prokop, Sujal R Desai, C Orla Morrissey, John W Baddley, Guidance on Imaging for Invasive Pulmonary Aspergillosis and Mucormycosis: From the Imaging Working Group for the Revision and Update of the Consensus Definitions of Fungal Disease from the EORTC/MSGERC, Clinical Infectious Diseases, Volume 72, Issue Supplement_2, 15 March 2021, Pages S79–S88,