key: cord-0855617-dnjr7be7
authors: Bartoletti, Michele; Pascale, Renato; Cricca, Monica; Rinaldi, Matteo; Maccaro, Angelo; Bussini, Linda; Fornaro, Giacomo; Tonetti, Tommaso; Pizzilli, Giacinto; Francalanci, Eugenia; Giuntoli, Lorenzo; Rubin, Arianna; Moroni, Alessandra; Ambretti, Simone; Trapani, Filippo; Vatamanu, Oana; Ranieri, Vito Marco; Castelli, Andrea; Baiocchi, Massimo; Lewis, Russell; Giannella, Maddalena; Viale, Pierluigi
title: Epidemiology of invasive pulmonary aspergillosis among COVID-19 intubated patients: a prospective study
date: 2020-07-28
journal: Clin Infect Dis
DOI: 10.1093/cid/ciaa1065
sha: 0933b720537f8a9eca93a4879af694e7129d2d3e
doc_id: 855617
cord_uid: dnjr7be7

BACKGROUND: In this study we evaluated the incidence of invasive pulmonary aspergillosis among intubated patients with critical coronavirus disease 2019 (COVID-19) and evaluated different case definitions of invasive aspergillosis. METHODS: Prospective, multicentre study on adult patients with microbiologically confirmed COVID-19 receiving mechanical ventilation. All included participants underwent screening protocol for invasive pulmonary aspergillosis with bronchoalveolar lavage galactomannan and cultures performed on admission at 7 days and in case of clinical deterioration. Cases were classified as coronavirus associated pulmonary aspergillosis (CAPA) according to previous consensus definitions. The new definition was compared with putative invasive pulmonary aspergillosis (PIPA). RESULTS: A total of 108 patients were enrolled. Probable CAPA was diagnosed in 30 (27.7%) of patients after a median of 4 (2-8) days from intensive care unit (ICU) admission. Kaplan-Meier curves showed a significant higher 30-day mortality rate from ICU admission among patients with either CAPA (44% vs 19%, p= 0.002) or PIPA (74% vs 26%, p<0.001) when compared with patients not fulfilling criteria for aspergillosis. The association between CAPA [OR 3.53 (95%CI 1.29-9.67), P=0.014] or PIPA [OR 11.60 (95%CI 3.24-41.29) p<0.001] with 30-day mortality from ICU admission was confirmed even after adjustment for confounders with a logistic regression model. Among patients with CAPA receiving voriconazole treatment (13 patients, 43%) A trend toward lower mortality (46% vs 59% p=0.30) and reduction of galactomannan index in consecutive samples was observed. CONCLUSION: We found a high incidence of CAPA among critically ill COVID-19 patients and that its occurrence seems to change the natural history of disease

The pandemic of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-associated coronavirus disease 2019 (COVID-19) is a major threat for global health. Approximately 14-30% of hospitalized patients diagnosed with COVID-19 develop a severe respiratory failure requiring intensive care [1] [2] [3] . Among ventilated patients with COVID-19, preliminary studies have reported a high incidence of invasive aspergillosis that may affect up to 30% of intubated patients [4] [5] [6] . These observations mirror previous reports of invasive pulmonary aspergillosis complicating severe influenza in patients admitted for intensive care unit (ICU) care [7, 8] .

Despite the high number of case reports of COVID-19 associated aspergillosis, a standardized case definition for this infection is lacking. Recently revised European Organisation for Research and Treatment of Cancer /Mycoses Study Group (EORTC/MSG) definitions of possible, probable and proven invasive aspergillosis [9] in immunocompromised patients, which rely on characteristic radiological features of invasive mould disease (i.e. nodular lesions with or without halo signs, cavitation) are difficult to apply in critically-ill COVID-19 patients who often have less-specific radiological signs of infection in the presence of acute respiratory distress syndrome (ARDS) [10] .

Previous investigations have proposed a unique definition of putative invasive pulmonary aspergillosis (PIPA) for patients with Aspergillus-positive lower respiratory tract cultures (entry criterion) with compatible signs and symptoms of pneumonia, abnormal chest x-ray or computed tomography imaging, and either host risk immunosuppressive risk factors or lack of bacterial growth in lower respiratory cultures or the presence of a positive cytological smear demonstrating branching hyphae [11] . Recently an expert consensus proposed a case definition for influenza associated pulmonary aspergillosis (IAPA) based on galactomannan testing on serum or respiratory specimens.

A similar definition was also proposed for coronavirus associated pulmonary aspergillosis (CAPA) complicating severe COVID-19 cases [12] .

A c c e p t e d M a n u s c r i p t

In this study we aimed to describe incidence and outcome of CAPA in a larger cohort of COVID-19 ventilated patients. Additionally, we aimed to evaluate the prognostic impact of different aspergillosis case definition in this setting.

We performed a prospective multicentre cohort study of patients with laboratory-confirmed SARS-CoV2 virus infection, hospitalized from February 22 through April 20, 2020 in four intensive care units (ICUs) from 3 Hospitals in Bologna, Italy: 1 tertiary 1420-bed teaching and 2 tertiary hospitals with 870 and 320 beds, respectively. The number of maximal ICU beds per hospital available during the epidemic peak were 77, 22 and 13 respectively. Diagnostic testing for COVID-19 and hospitalization were performed according to local policy and clinical judgment and were not dictated by study protocol. Data were collected anonymously and managed using REDCap electronic data capture tools [13, 14] . The study was approved by the Ethic Committee (Comitato Etico Indipendente di Area Vasta Emilia Centro, n. 283/2020/Oss/AOUBo).

All consecutive adult (≥18 years) patients diagnosed with SARS-CoV-2 infection and requiring ICU admission for mechanical ventilation. Exclusion criteria were: i) early (<48h) ICU discharge ii) ICU admission for reason other than acute respiratory distress syndrome (ARDS).

A c c e p t e d M a n u s c r i p t

For all participants a screening protocol for invasive aspergillosis was proposed and consisted in bronchoalveolar lavage (BAL) performed on ICU admission (0-2 days), at day 7 (± 2 days) from the first day of mechanical ventilation and if the patient showed evidence of clinical disease progression, which was defined by either a) worsening of fever or b) increases in respiratory secretions or deterioration in respiratory status after a period of clinical stability. Samples were processed for galactomannan (GM) detection and cultures. Additionally, BAL samples that tested positive for galactomannan were store at -80°C and later analysed using a commercial quantitative real-time Aspergillus PCR assay (described below). Therefore, the result of PCR assay was not reported to clinicians. Direct cytological examination of BAL samples was deferred due to COVID-19 safety concerns.

Severe COVID-19 cases were treated with hydroxychloroquine, lopinavir-ritonavir or darunavircobicistat, intravenous tocilizumab (6 mg/kg in 1-2 doses within 12-24h) or subcutaneous tocilizumab administered in two simultaneous doses of 162 mg, methylprednisolone 1mg/kg for 5-7 days and low molecular weight heparin (LMWH) at daily dosage of 60-100 mg according to body weight.

The presence of SARS-Cov2 was detected by RT-PCR assay. Briefly, UTM-RT swab specimens (Copan, Italy) were immediately tested or stored at 4ºC until processed, no more than 48 hours. Total genomic DNA/RNA was extracted from 280 µl of the clinical swab sample by Nuclisens EasyMag (BioMerieux, Marcy l'Etoile, France) following manufacturer's instructions. Detection of SARS-CoV-2 virus was performed by real time RT-PCR following the WHO and/or CDC protocol in a QuantStudio S5 Real-time PCR system (ThermoFisher, USA).

The galactomannan antigen index was measured with a sandwich enzyme-linked immunosorbent assay (ELISA) (Platelia™ Aspergillus; Bio-Rad Laboratories) in BALs and serum M a n u s c r i p t specimens. BALs were further analysed by culture for filamentous fungi and quantitative real time-PCR for Aspergillus genus as follows A ten-microliter volume of BAL were cultured on Sabouraud Chloramphenicol agar tubes (Vakutainer Kima, Padova, Italy) at 30°C for up to five days. As soon as molds were visible they were subcultured on Sabouraud Dextrose Agar plates (Vakutainer Kima, Padova, Italy) for 2 to 3 days at 30 °C. Fungi identification was performed by microscopic examination of lactophenol cotton-blue stained slides and by MALDI-TOF Mass Spectrometry Instrument (Bruker, Italy), following manufacturer's instructions.

The residual volume was frozen at -20°C until used for PCR analysis. DNA extraction for PCR analysis was performed on ELITe InGenius automated platform as well as Real time-PCR (RT-PCR) using the Aspergillus spp. ELITe MGB kit (Elitgroup, Puteaux, France). The DNA was extracted from 1 ml volume of BAL fluid and was eluted in a 200 µL prior to DNA amplification in the same platform. RT-PCR for Aspergillus genus was performed by Aspergillus spp. ELITe MGB kit which was CE-IVD validated on diverse range of sample types. The target region was the rDNA18S gene and human beta-globin gene was used as an internal standard. The fungal DNA copy number was expressed as copies/ml in relation to a rDNA18s standard curve.

Microbiological diagnosis of SARS-CoV2 infection was defined as a positive RT-PCR test on respiratory specimens. These consisted of nasopharyngeal swabs or BAL in all cases.

Invasive pulmonary aspergillosis was defined according to the recently proposed CAPA definition consisting in COVID-19 positive patients admitted to the ICU with pulmonary infiltrates (entry criterion) who had at least one of the following: serum GM index >0.5 or BAL GM index >1.0 or positive Aspergillus BAL culture or cavitating infiltrate (not attributed to another cause) in the area of the pulmonary infiltrate [12] .

A c c e p t e d M a n u s c r i p t

To assess the prognostic performance of different cases definitions we compared cases of CAPA and cases of PIPA which was defined according with AspICU criteria [11] .

Exposure variables were assessed at hospital admission and included: age, sex, body mass index. Underlying conditions were recorded according to Charlson comorbidity index [15] .

Immunosuppression included neutropenia (neutrophil count <500/mm 3 ), solid organ transplantation, hematopoietic stem cell transplantation, corticosteroid therapy at a dosage higher then or equivalent to prednisone 16 mg/day ≥ 15 days, uncontrolled HIV infection (<200CD 4 /mm 3 ). Regarding the SARS-CoV2 infection, we collected symptoms vital signs, laboratory and radiological tests, at hospitalization and during follow-up, and treatments received. Clinical severity at hospitalization and ICU-admission was recorded according to sequential organ failure assessment(SOFA). Endpoint variables were assessed from hospital admission to discharge. We collected, duration of mechanical ventilation, in-hospital all-cause mortality and date of hospital discharge.

For descriptive analysis, categorical variables are presented as counts and percentages. Continuous variables as mean and standard deviation if normally distributed or as median and interquartile range (IQR) if non-normally distributed. Incidence rates of CAPA were calculated per 10,000 ICU patient-days, and 95% confidence intervals (CIs) for the incidence rates were estimated under the assumption of a Poisson distribution.

Survival was analysed by Kaplan Meier curves. The impact of PIPA and CAPA definitions on A c c e p t e d M a n u s c r i p t survival status of COVID-19 was assessed by the log-rank test after 30 days form ICU admission at univariate analysis. To assess the impact on mortality of CAPA and PIPA definitions we, first compared survivors and non-survivors after 30 days form ICU admission at univariate analysis. Age, sex, SOFA score at ICU admission and need for renal replacement therapy were included in a logistic regression model for 30-day mortality. Thereafter, the variables PIPA, CAPA and first galactomannan index were alternatively included to assess their effect on the mortality model. All statistical analysis was performed with Stata-IC 16 (College Station, Texas) and R version 3.5 (R Core team, Vienna, Austria).

During the study period 822 patients with diagnosis of COVID-19 were admitted in the 3 centres. Of these, 185 (22%) were admitted to ICU and 163 (20%) were intubated. Screening for aspergillosis was performed in 108 patients, this group was selected as study cohort. Main reason for protocol exclusion were early (<48h) extubation (12 cases), ICU admission and intubation for reason other than ARDS (13 cases), incompliance with protocol (30 cases). In these latter cases bronchoscopies were not performed for safety concerns (16 cases) or GM was not tested because insufficient BAL quantity (8) or by mistake (5) (Figure 1 ).

Overall, median (IQR) age was 64 (57-70) years and 83 (78%) were male. The median ageadjusted Charlson comorbidity index was 2.5 (1) (2) (3) (4) . At ICU admission the median (IQR) SOFA score was 4 (3) (4) (5) . A total of 189 samples from BAL were obtained and analysed (Table 1) Table 2 . Briefly, the only factor associated to CAPA was chronic steroid therapy (p=0.02) at dosage higher then or equivalent to prednisone 16 mg/day for at least 15 days.

Simultaneous serum and BAL galactomannan were available only in 59 patients. Among Table 1 .

The last evaluable follow-up date was May 19,2020. The median patient follow-up time was 31 (20-43) days. At this time, 54 (50%) patients were discharged, 44 (41%) died and, in 9 patients the followup was ongoing. Differences between survivors and non-survivors are reported in Table 3 . Kaplan-Meier curves showed a significantly higher 30-day mortality rate from ICU admission among patients with either probable CAPA when compared with patients without CAPA (44% vs 19%, p= 0.002; 

The relationship between initial BAL galactomannan index and 30-day survival is show in Figure 3 .

The odds of death within 30 days of ICU admission increased 1.41-fold (1.10-1.81; p=0.007) for each point increase in the initial BAL galactomannan index. When adjusted for age, need for renal replacement therapy and SOFA score at ICU admission the initial BAL galactomannan index was still independently associated with increased odds of death within 30-days of ICU admission (OR 1.44; 95% CI 1.08-1.94; p=0.014).

Among patients fulfilling probable CAPA definition a total of 16 (53%) of patients received antifungal of which 13 (43%) were treated with voriconazole. Reason for non-treatment was postmortem diagnosis or clinical decision not to start antifungal therapy in 7 cases, each.

Among patients with probable CAPA voriconazole treatment was associated to a trend toward lower mortality (Figure 3 To overcome all these limitations, we tried to validate a novel proposed definition of CAPA based on recently published expert consensus definitions [12] . When we evaluated the survival impact of PIPA and probable CAPA criteria, both were independent predictors of mortality. We believe that the use of CAPA criteria could be more useful in clinical practice for guiding clinical decisions. PIPA criteria do not consider non-culture-based methods such as BAL GM or BAL/serum PCR. These latter may allow prompt earlier diagnosis and treatment of aspergillosis potentially leading to improved patient survival.

According to this new definition, we were able to identify 30/108 (28%) patients fulfilling probable CAPA criteria. The incidence rate ratio of positive BAL galactomannan with CT findings during the COVID period (38.83 per 10,000 unit days) vs. the same months (February to April ) of the previous year in the same ICUs (9.69 per 10,000 unit days) was 4.04 (95% CI 1.77-9.91); p<0.0001.

However, these estimates are likely biased towards higher number of cases during the COVID period A c c e p t e d M a n u s c r i p t because serial BAL testing was performed in the COVID group but not the non-COVID historical controls. These results are slightly higher to prevalence of aspergillosis complicating severe influenza cases found in previous studies (7-19%) [7, 8] . Similarity between severe IAPA and CAPA could be expected given the extensive damage to the respiratory epithelium associated with both infections, which is considered a key predisposing event for semi-invasive and invasive pulmonary aspergillosis in these populations [12] . In our study the median number of days between COVID-19 symptoms onset and CAPA diagnosis was 14 days. Other studies performed on IAPA found comparable timing between influenza onset and aspergillosis [16, 17] . Likewise influenza, severe COVID-19 is characterized by lymphopenia. This factor was previously associated to development of invasive aspergillosis [18] .

It was recently hypothesized that the uncontrolled cytokine storm may play role in determining disease progression for COVID-19 patients [19] . Consistently, most patients received corticosteroids and tocilizumab in our cohort. This heavy use of immunomodulant drugs may had contributed to the high prevalence of CAPA in our study. Similar findings were observed in a large number of studies including those focused on patients with severe influenza cases [8, 20] . Conversely, cytokine storm may add challenges to diagnosis of bacterial and fungal infection as symptoms and radiological findings may overlap. Therefore, a screening algorithm for CAPA as performed in our study may provide prompt diagnosis and treatment.

Studies on best antifungal treatment for CAPA are lacking. Most common recommendation of use voriconazole or isavuconazole in the setting of aspergillosis complicating severe influenza cases are based on studies performed in immunocompromised [10] . In this report most patients received voriconazole. Although this study was not designed to address this point in explorative analysis an interesting trend toward higher survival (figure 4) or reduced BAL galactomannan index was observed. Unfortunately, the relatively low sample size prevents any firm conclusion about antifungal treatment.

A c c e p t e d M a n u s c r i p t Another interesting finding in our study is the correlation between the magnitude of the BAL GM index and 30-day patient mortality. Similar results were also found in studies on serum GM in hematological patients and may be the expression of higher fungal burden [21] . If confirmed by further studies BAL GM may be useful to prioritize antifungal treatment to patients considered at higher risk of mortality.

In conclusion we found a high incidence of invasive aspergillosis among critically ill COVID-19 patients and that its occurrence seems to change the natural history of disease. The use of CAPA criteria for diagnosis of invasive aspergillosis may provide earlier diagnosis than AspICU criteria and might prioritize prompt antifungal treatment.

M a n u s c r i p t A c c e p t e d M a n u s c r i p t M a n u s c r i p t 

Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the

Critical Care Utilization for the COVID-19 Outbreak in Lombardy, Italy: Early Experience and Forecast During an Emergency Response

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Prevalence of putative invasive pulmonary aspergillosis in critically ill patients with COVID-19. The Lancet Respiratory

COVID-19 Associated Pulmonary Aspergillosis. American journal of respiratory and critical care medicine

COVID-19 associated pulmonary aspergillosis

High rates of influenza-associated invasive pulmonary aspergillosis may not be universal: a retrospective cohort study from Alberta, Canada. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America

Invasive aspergillosis in patients admitted to the intensive care unit with severe influenza: a retrospective cohort study

Revision and Update of the Consensus Definitions of Invasive Fungal Disease From the European Organization for Research and Treatment of Cancer and the Mycoses Study Group Education and Research Consortium

Intensive care management of influenza-associated pulmonary aspergillosis. Clinical microbiology and infection : the official publication of the

A clinical algorithm to diagnose invasive pulmonary aspergillosis in critically ill patients

Review of influenza-associated pulmonary aspergillosis in ICU patients and proposal for a case definition: an expert opinion

Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support

The REDCap consortium: Building an international community of software platform partners

A new method of classifying prognostic comorbidity in longitudinal studies: development and validation

Influenza-associated aspergillosis in critically-ill patients-a retrospective bicentric cohort study. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 2020. 17. van de Veerdonk FL

Development and internal validation of a model for predicting 60-day risk of invasive mould disease in patients with haematological malignancies

COVID-19: consider cytokine storm syndromes and immunosuppression

Invasive aspergillosis in glucocorticoid-treated patients

The serum galactomannan index predicts mortality in hematopoietic stem cell transplant recipients with invasive aspergillosis

A c c e p t e d M a n u s c r i p t M a n u s c r i p t