Computed tomography findings of COVID-19-associated pulmonary mucormycosis: Data from a multicenter retrospective study (Mucovi2), India

Dear Editor, A large number of mucormycosis cases were reported in India following the second wave of COVID-19.[1] However, data on the imaging appearance of COVID-19-associated pulmonary mucormycosis (CAPM) are limited.[2] A few published case reports on CAPM and a systematic literature review suggested cavity and consolidation as the most common finding.[3-6] The primary objective of the current study was to describe the computed tomography (CT) thorax findings of CAPM at diagnosis. The secondary objectives were to compare the CT findings of (1) CAPM in patients with diabetes mellitus (DM) and those with no known traditional risk factors, and (2) survivors and non-survivors with CAPM. The MuCovi-2 study was a multicentre retrospective study conducted across 25 centres in India between 1 January 2021 and 30 June 2021. Data on COVID-19-associated mucormycosis (microbiologically or pathologically confirmed) cases and age-matched controls were collected in the original study.[7] COVID-19 was diagnosed by detecting SARS-CoV2 infection in nasopharyngeal or throat swabs by rapid antigen testing or reverse transcriptase polymerase chain reaction. Only subjects diagnosed with proven and probable CAPM were included in the present study, as defined previously.[2] The diagnosis and management of CAPM patients were per the prevalent guidelines.[2,8] The clinical presentation, risk factors, and outcome of the entire cohort of CAPM patients have been published recently.[9] For the current study, we excluded CAPM patients with inadequate CT data. We noted (as yes/no) the radiological findings on CT (cavity, consolidation, reversed halo sign (RHS), bird’s nest sign, pneumothorax, mycotic aneurysm, pleural effusion, nodules, and others) at diagnosis of CAPM. Data are presented as numbers (percentage) or mean with standard deviation (SD). We used the Mann–Whitney U and Fisher’s exact (or Chi-square) tests to analyse the difference between continuous and categorical variables. We assumed statistical significance as a P value <0.05. We included 84 subjects with CAPM [Table 1]. The participants’ mean (SD) age was 53.8 (12.5) years. DM was the most common predisposing factor in 58/84 (69%) subjects. Glucocorticoid treatment for COVID-19 was prescribed in 82.3% (65/79) subjects, and nearly 20% of these subjects received inappropriate glucocorticoid therapy (i.e., in the absence of COVID-19-related hypoxemia). Cavity was the most common finding (56/84; 77.8%). RHS or bird’s nest sign were observed in 14/84 (16.7%) subjects. Pleural effusion or >10 lung nodules were reported in 19/84 (22.6%) and 2/84 (2.4%) subjects, respectively. Tracheobronchial mucormycosis was noted in four subjects, one of whom had an isolated tracheal lesion with no other parenchymal or pleural abnormality. Of the 84 subjects, 43 (51.2%) and 41 (48.8%) had proven and probable CAPM, respectively. Bilateral disease was significantly more common in those with probable CAPM (20/41 [48.8%] versus 9/43 [20.9%] among proven cases), and pleural effusion was more frequent in proven CAPM (15/43 [34.9%] versus 4/41 [9.8%] in probable CAPM). None of the other findings were significantly different between proven or probable CAPM. The CT findings were similar in patients with COVID-19 as the only risk factor for CAPM versus those with DM, except for pleural effusion, which was significantly less common in DM [Table 2]. Survival at 6 weeks and 12 weeks was 43/79 (54.4%) and 37/73 (50.7%). We found no significant differences in CT findings among survivors versus non-survivors at 6-weeks [Table 3].Table 1: Baseline characteristics and imaging findings of CAPM (n=84) at diagnosisTable 2: Comparison of CT findings in CAPM patients with and without diabetes mellitusTable 3: Comparison of CT findings between survivors versus non-survivors at 6 weeks with CAPMTimely diagnosis improves survival in pulmonary mucormycosis (PM), a serious illness with a high mortality.[10,11] A CT scan provides one of the earliest clues for diagnosing PM [Figure 1].[12] However, most data on the CT findings of PM are from patients with risk factors such as haematological malignancies and organ transplant recipients.[13,14] Imaging appearances of CAPM are poorly studied. We found cavity (66.7%) the most common finding in CAPM, like in DM.[15,16] However, consolidation was less frequently reported in our study (13%) than in the previous studies (48%).[17] Interestingly, in the current study, consolidation was infrequent (12%), even in diabetic subjects with CAPM. The lesser frequency of consolidation and predominance of the cavities in our study could be related to the duration of illness before the CT scan.[15] The earliest imaging findings of PM include nodules, consolidation, and masses; cavitation occurs later when the lesions undergo central necrosis.[18] Neutropenic patients are evaluated early with chest CT and thus manifest nodules or masses. Contrarily, PM in DM or COVID-19 presents later and cavities are common. The current study participants (CAPM with and without DM) were possibly evaluated late during the illness due to severe COVID-19. Also, consolidation due to PM could have been attributed to COVID-19 or hospital-acquired pneumonia. Although insensitive, other notable findings in our study, like RHS and mycotic aneurysm, may suggest PM in those with predisposing factors (including COVID-19 and uncontrolled DM).Figure 1: Representative images showing a few typical signs of COVID-19-associated pulmonary mucormycosis: (a) consolidation in the right upper lobe along with diffuse ground glass opacities (GGOs; COVID-19 related) and pneumomediastinum, (b) reversed halo sign on the right side with patchy consolidation in bilateral lungs and a few GGOs, (c) cavity with air-fluid level in the right lower lobe and bilateral GGOs and (d) pulmonary artery aneurysm in a lesion in the right lower lobe; pneumothorax is also noted on the right sideOur study has a few limitations. The CT findings were retrospectively retrieved from patients’ records, and the imaging was not centrally reviewed. Despite a multicentre study, the number of PM patients was relatively small. Further, the CT protocol, the use of contrast, and the differences in the reporting could have influenced our study results. The chest CT findings may vary depending on the duration of PM, and we do not have this information. We do not have a comparator group (e.g., invasive pulmonary aspergillosis); therefore, the specificity of the imaging findings is unknown. We included only individuals with a confirmed diagnosis of CAPM, and we might have missed cases where microbiological or pathological confirmation was unavailable. Isolated tracheobronchial mucormycosis (also reported in COVID-19) without any parenchymal abnormalities could have been missed in our series.[19] In summary, we found similar imaging findings in CAPM as previously reported in PM complicating DM. Cavity was the most common finding in CAPM patients in India. Prospective studies, including serial imaging performed on a larger number of patients with varying risk factors, are required. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.