key: cord-0742222-gjpbne00
authors: Giusiano, Gustavo; Fernández, Norma B; Vitale, Roxana G; Alvarez, Christian; Ochiuzzi, María Eugenia; Santiso, Gabriela; Cabeza, Matías Sebastián; Tracogna, Fernanda; Farías, Luciana; Afeltra, Javier; Noblega, Luciana María; Giuliano, Carla Valeria; Garcia-Effron, Guillermo
title: Usefulness of Sōna Aspergillus Galactomannan LFA with digital readout as diagnostic and as screening tool of COVID-19 associated pulmonary aspergillosis in critically ill patients. Data from a multicenter prospective study performed in Argentina
date: 2022-04-08
journal: Med Mycol
DOI: 10.1093/mmy/myac026
sha: adc80735e80223371c4dc65e55cfa19a58022dbd
doc_id: 742222
cord_uid: gjpbne00

COVID-19 associated pulmonary aspergillosis (CAPA) incidence varies depending on the country. Serum galactomannan quantification is a promising diagnostic tool since samples are easy to obtain with low biosafety issues. A multicenter prospective study was performed to evaluate the CAPA incidence in Argentina and to assess the performance of the lateral flow assay with digital readout (Sōna Aspergillus LFA) as a CAPA diagnostic and screening tool. The correlation between the values obtained with Sōna Aspergillus LFA and Platelia® EIA was evaluated. In total, 578 serum samples were obtained from 185 critically ill COVID patients. CAPA screening was done weekly starting from the first week of ICU stay. Probable CAPA incidence in critically ill patients was 10.27% (19/185 patients when LFA was used as mycological criteria) and 9% (9/100 patients when EIA was used as mycological criteria). We found a very good correlation between the two evaluated galactomannan quantification methods (overall agreement of 92.16% with a Kappa statistic value of 0.721). CAPA diagnosis (>0.5 readouts in LFA) were done during the first week of ICU stay in 94.7% of the probable CAPA patients. The overall mortality was 36.21%. CAPA patients' mortality and length of ICU stay were not statistically different from for COVID (non-CAPA) patients (42.11% vs 33.13% and 29 vs 24 days, respectively). These indicators were lower than in other reports. LFA-IMMY with digital readout is a reliable tool for early diagnosis of CAPA using serum samples in critically ill COVID patients. It has a good agreement with Platelia® EIA.

The incidence of COVID associated pulmonary aspergillosis (CAPA) in critically-ill Argentinian patients was established (10.27%). Serum galactomannan quantification was useful as a screening tool for this mycosis. A good agreement between Platelia® EIA and Sōna Aspergillus LFA is reported.

As for October 20 th 2021, more than 240 million cases of coronavirus disease 2019 (COVID-19) were reported worldwide 1 .

COVID-19-associated pulmonary aspergillosis (CAPA) was reported as a new clinical complication in critically ill patients 2 with a reported incidence ranging from 3.5 to more than 26% with geographical variations [3] [4] [5] [6] [7] [8] [9] . In these patients, diagnosis is challenging. Radiological findings are nonspecific and bronchoalveolar lavages (BAL) for microbiology studies are usually not available.

In addition, as in any non-neutropenic patients, galactomannan (GM) quantification and interpretation is controversial. These facts raised questions about the used criteria for CAPA diagnostics 10-12 .

One recent improvement in invasive aspergillosis diagnosis is the commercialization of a specific GM-lateral flow assay (LFA). These devices demonstrated excellent performance, becoming a viable option when the wellestablished EIA GM quantification is not available 13, 14 . Moreover, these LFA devices reduced the turnaround time and cost 15 .

Although some reports were recently published 16-18 the incidence of CAPA in Argentina is barely known 19 . Data about this subject is important considering that Argentina is a big country, with important differences in access to the health system related to economic and demographic issues. A mixed health system (public, semi-public and private) has access to different techniques.

The aims of this work were to evaluate the CAPA incidence in Argentina and to assess the performance of the GM lateral flow assay (Sōna Aspergillus LFA-IMMY®) as a CAPA diagnostic and screening tool. A prospective multicenter study, including Hospitals and referral centers from diverse Argentinian regions

with different economic and demographic characteristics, was performed for this purpose.

Participating center, inclusion criteria and patient data

In this work, we studied samples obtained in three public hospitals from Buenos Aires city (Clínicas-UBA Hospital, Durand Hospital and Ramos Mejía Hospital) and in three referral regional centers, one from the northwest Argentina (Tucumán), one from the northeast region (Chaco-Corrientes) and one from central Argentina (Santa Fe-Entre Ríos). These three referral regional centers received samples from 8 different regional Hospitals (public, semi-public and private Hospitals). Moreover, Muñiz Hospital participated in processing samples obtained at the Durand Hospital (no patients were enrolled in Muñiz Hospital). All the participating centers were managed and financed by state governments with the exception of three centers, semi-public (n = 1) and private (n = 2).

The study included critically ill adult patients (>18 years-old) with a COVID-19 positive RT-PCR SARS-CoV-2 test admitted to ICU with respiratory support. These patients were admitted between March 15 th 2020 and October 15 th 2020 during the so-called first wave of SARS-CoV-2 infections in Argentina.

Patient´s clinical and demographic data were collected including: age, genre, underlying disease (if any), ICU admission and discharge (or death) dates, microbiology laboratory results (mycology and bacteriology studies), corticosteroid and antibiotic treatments received, type of respiratory support (tracheotomy, intubation or other assisted ventilation) and radiological findings.

Serum samples were obtained as follows: during the first two weeks of ICU stay, one sample per week (days 2 or 3 and 9 or 10 of ICU stay) were taken. In the following weeks, two serum samples were obtained. All serum samples (and BAL if received) were subjected to GM quantifications using IMMY´s SŌNA LFA (from now on: GM-LFA) (IMMY Diagnostics, OK, USA). Moreover, BAL samples (when available) were subjected to the routine diagnostic procedures including microscopic examination and culture. BAL pellets were cultured in two Sabouraud-chloramphenicol agar slants (incubated for 30 days at 28°C and 37°C and examined every day). The rest of the resuspended pellet was used for microscopic examination (Giemsa stain and direct examination with and without

calcofluor-white and with and without 20% KOH) 20 . Furthermore, four of the participating centers performed GM quantification by EIA (Platelia®, Biorad, from now on: GM-EIA) in parallel (see below for further details). GM-LFA and GM-EIA were performed strictly following the manufacturer instructions. Cutoff values for GM-LFA and GM-EIA positivity were an index value of 0.5 and 1.0 for serum and BAL, respectively 21, 22 . SŌNA cube reader (IMMY diagnostics, OK, USA) was used to accurately obtain the readout results of the GM-LFA.

Used definitions of CAPA and study characteristics.

This was a prospective multicenter study. The 2020 ECMM/ISHAM consensus criteria for research and clinical guidance definitions of CAPA were followed 23 . Briefly, the entry criterion of the consensus is positive SARS-CoV-2 RT-PCR anytime during two weeks between hospital admission and ICU admission or within 72-96 h after ICU admission and acute respiratory distress syndrome. We also used the proposed grades (possible, probable and proven CAPA). A proven CAPA diagnosis requires normally sterile pulmonary samples (e.g., pulmonary biopsies). Due to the known difficulties to obtain these samples in severely ill COVID-19 patients, none of the described cases of this study could be categorized as proven CAPA. Probable CAPA diagnosis requires the demonstration of pulmonary nodules or infiltrates and/or cavitating infiltrates (by a chest CT scan) with no other attributable cause than SARS-CoV-2 infection together with one or more mycological evidence. Radiological series were analyzed locally (each hospital analyzed their own radiological reports as each hospital usually does). The followed consensus criteria included at least one of the following positive mycological tests: observation of filamentous fungal elements in BAL by microscopy, Aspergillus spp. isolated from BAL culture, GM ratio >0.5 in serum and/or ≥1.0 in BAL, PCR in BAL, 2 PCR in plasma, serum or whole blood 23 . Therapeutic data was retrospectively collected and analyzed. 

and by determining a Spearman correlation. A P value <0.05 was considered significant. The participating centers ethics committees approved this study. Technical comparison of GM-LFA with GM-EIA.

Galactomannan quantification by GM-LFA were performed in 578 serum samples obtained from the 185 patients included in the study (averaging 3.82 sera per patient ranging from 2 to 10 sera/patient). Moreover, 35 BAL specimens from 18 patients were analyzed (1.94 per patient).

Quantification of GM was also carried out in parallel by GM-EIA in 258/578 serum (obtained from 100 individual patients) and in 19/35 BAL samples (taken from 9 patients). Despite the used method, most of the serum samples showed low (< 0.5) GM ratios, 497/578 (85.99%) and 211/258 (81.78%) for GM-LFA and GM-EIA, respectively.

We received 10 hemolyzed sera samples that were considered invalid by IMMY´s CUBE reader (n=3) or showed result discrepancies between methods (n=7) (>0.5 and <0.5 for GM-LFA and GM-EIA, respectively). After excluding these hemolyzed samples, a good agreement between GM-LFA and GM-EIA was statistically confirmed (248 samples were analyzed). The overall observed agreement was 92.16% (95% confidence interval (CI) from 0.606 to 0.836) with a Kappa statistic value of 0.721 (95% CI from 0.390 to 0.747) that represent a substantial agreement following Landis et al. criteria (Kappa strata 0.61-0.80) 24 and a moderate agreement by Spearman´s coefficient (P<0.0001) ( Figure 1B) . Briefly, the results obtained from 235 serum samples were interpreted equally by using both methods (202 negative and 33 positives by both). The rest (n=13) showed discrepant results. The most common discrepancies were detected when samples were positive by GM-LFA and negative by GM-EIA. Seven of these discrepancies were seen in samples showing borderline GM values (between 0.5 and 0.70 for GM-LFA and between 0.30 and 0.50 for GM-EIA). As an example, we can state the second serum sample of the 3 rd week of ICU stay of patients 1 and 4, depicted in Table 2 Out of these 19 probable CAPA patients, eight died (42.11%) ( Table 2) . This mortality percentage showed no statistical differences with the mortality of the non-CAPA patients (33.13%, P=0.3237). The average length of ICU stay of these CAPA-patients was 29.32 ± 19.24 days (ranging from 7 to 83 days) and there were no statistical differences with the other COVID (non-CAPA) enrolled patients (P=0.620).

Aspergillus spp. was recovered in culture from eight BAL samples. All but one positive Aspergillus cultures were obtained from patients with at least one positive GM-LFA ( Table 2) . On the other hand, five of the eleven patients diagnosed as probable CAPA by means of GM-LFA alone did not receive antifungal treatment and six died. Most of the patients that were treated received voriconazole (63.3%, n=7) followed by amphotericin B (n=4, 3 liposomal presentation), isavuconazole (n=1) and itraconazole (n=1). Some patients received multiple antifungal treatments. In one case, fluconazole preemptive treatment was changed to voriconazole after the high GM-LFA report. In some cases, treatment was initiated with itraconazole or amphotericin B until voriconazole was received (some centers had no voriconazole in their pharmacies on a regular basis). In one of the CAPA patients, voriconazole was replaced by isavuconazole when voriconazole serum level was considered not adequate (low concentration in serum) ( Table 2 ).

We present the results of a prospective multicenter study from different Argentinian regions with differences in climate, economic and demographic characteristics. The selection of centers was done in order to know CAPA incidence in Argentina including these heterogeneities. In addition, the evaluation of the IMMY®´s SŌNA Aspergillus LFA with cube reader lecture GM quantification method was done.

We studied 185 critically ill COVID-19 patients (ICU-with respiratory assistance). Age ranges and underlying diseases of our patients were similar to other worldwide reports 19, 25, 26 . It has to be highlighted that four (2.16%) of our patients had a diagnosis of Chagas disease (all with other comorbidities).

The overall mortality rate in our cohort was 36.21%. It was higher than some US centers (26.5%) 27 but similar to most reported mortality rates in COVID-19 ICUmechanical ventilated patients [27] [28] [29] [30] . In contrast to other series, age was not related with mortality in our group of patients 28 and the average length of ICU stay (24.43 days) was three-times longer than some reports from China (8 days 31 to 9 days 32 ) and similar to others (outside China) 33,34 .

Following the described ISHAM/ECMM consensus criteria, the incidence of CAPA in critically ill patients (ICU-with respiratory assistance) during the first COVID-19 wave across Argentina was 10.27% (if GM-LFA was used as mycological criterion) and 9.0% (if GM-EIA was used as the unique biomarker). This incidence is similar to the overall incidence of CAPA in COVID-19 patients on mechanical ventilation published by Fungiscope 19 and comparable with data from some European centers (11.43 ± 1.68 % in UK, Germany and Belgium one center each). On the other hand, CAPA incidence in Argentina was lower than in some European countries (>20%) 19 . These variations on incidence would be due to differential exposure to Aspergillus, different CAPA definition usage, regional differences in diagnostic capabilities, different clinical approaches and differences in awareness of CAPA (as suggested before for influenza-associated pulmonary aspergillosis) 12 .

Using the classical mycology techniques (no biomarkers), we were able to diagnose eight probable CAPA patients. These patients showed positive BAL cultures and hyphae were observed in five samples. Aspergillus section Fumigati was the most commonly isolated agent (6 out of 8 isolates). Thus, the performance of these techniques in terms of the CAPA diagnosis was similar to other reports (lower but close to 50% of the cases) 19, [35] [36] [37] [38] [39] [40] . The rest of the CAPA patients were diagnosed by GM detection in serum using LFA and EIA. Our data shows a good correlation of results between IMMY´s LFA with cube reader and BIO-RAD`s EIA results, replicating previous reports 15 Table 2 ), GM-LFA and GM-EIA in BAL samples were below the cut-off value and cultures and microscopy were negative confirming that serum GM were false positives. This example also confirms what was reported previously: respiratory samples would be better for probable CAPA diagnostics 13, 14, 41 .

In our series, CAPA diagnosis was mostly done during the first week of ICU stay. This quick diagnosis was possible by the implementation of the GM-LFA quantification as a screening tool in all the patients. Similarly, early diagnosis was described by Alanio et al. 35 4, 5, 37 . These differences on diagnosis speed seems to be related with the use of aspergillosis diagnostic tests on ICU admission (in some cases in the first 48 hs 43 ) or using these analyses when patients showed a deterioration on respiratory status.

It was reported that pulmonary aspergillosis increases the mortality of COVID-19 patients (e.g., 66.7% vs 32% 5 and 71.4% vs 36.8% 4 for CAPA and non-CAPA, respectively). Similarly, the mortality rate in our cohort was higher for CAPA patients (42.11% vs 33.13% for CAPA and non-CAPA, respectively). However, this difference was not-statistically significant and it was lower than in the majority of previous reports.

As in most reports, there was no postmortem histological confirmation of pulmonary aspergillosis in our CAPA patients 44 . In others, it was demonstrated that a combination of tools is needed to support a CAPA diagnosis (conventional mycology, PCR and biomarkers) 40, 45 . It is clear that further studies and perhaps a more rigorous criterion for CAPA diagnosis are needed. Until then, we would support that serial and routine (weekly) GM quantification in critically ill COVID-19 patients would be useful for CAPA diagnosis and mortality reduction 45, 46 . 

Aspergillosis complicating severe coronavirus disease

Incidence of invasive pulmonary aspergillosis among critically ill COVID-19 patients

Risk factors associated with COVID-19-associated pulmonary aspergillosis in ICU patients: a French multicentric retrospective cohort

COVID-19-associated pulmonary aspergillosis

COVID-19 associated pulmonary aspergillosis

Accuracy of galactomannan testing on tracheal aspirates in COVID-19-associated pulmonary aspergillosis

Prevalence of opportunistic invasive aspergillosis in COVID-19 patients with severe pneumonia

Frequency of Positive Aspergillus Tests in COVID-19 Patients in Comparison to Other Patients with Pulmonary Infections Admitted to the ICU

American Association for Bronchology and Interventional Pulmonology (AABIP) Statement on the Use of Bronchoscopy and Respiratory Specimen Collection in Patients With Suspected or Confirmed COVID-19 Infection

COVID-19 Associated Pulmonary Aspergillosis (CAPA)-From Immunology to Treatment

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

Point of care Aspergillus testing in intensive care patients

Point-of-care diagnostics for invasive aspergillosis: nearing the finish line

Evaluation of the performance of the IMMY sona Aspergillus galactomannan lateral flow assay when testing serum to aid in diagnosis of invasive aspergillosis

Ventilator-associated pneumonia involving Aspergillus flavus in a patient with coronavirus disease 2019 (COVID-19) from Argentina

COVID-19 associated pulmonary aspergillosis in ICU patients: Report of five cases from Argentina

SARS-CoV-2 and Aspergillus section Fumigati coinfection in an immunocompetent patient treated with corticosteroids

Clinical and Laboratory Strandards Institute CLSI. Principles and Procedures for Detection and Culture of Fungi in Clinical Specimens -M54

Galactomannan testing and Aspergillus PCR in same-day bronchoalveolar lavage and blood samples for diagnosis of invasive aspergillosis

A prospective study of fungal biomarkers to improve management of invasive fungal diseases in a mixed specialty critical care unit

Defining and managing COVID-19-associated pulmonary aspergillosis: the 2020 ECMM/ISHAM consensus criteria for research and clinical guidance

The measurement of observer agreement for categorical data

Clinical features and outcomes of critically ill patients with coronavirus disease 2019 (COVID-19): A multicenter cohort study

Clinical features and prognostic factors of critically ill patients with COVID-19 in Daegu, South Korea: A multi-center retrospective study

ICU outcomes and survival in patients with severe COVID-19 in the largest health care system in central Florida

Clinical characteristics and day-90 outcomes of 4244 critically ill adults with COVID-19: a prospective cohort study

ICU and Ventilator Mortality among Critically Ill Adults with Coronavirus Disease

Mortality rates of patients with COVID-19 in the intensive care unit: A systematic review of the emerging literature

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Baseline Characteristics and Outcomes of 1591 Patients Infected with SARS-CoV-2 Admitted to ICUs of the Lombardy Region, Italy

Prevalence of putative invasive pulmonary aspergillosis in critically ill patients with COVID-19

Invasive pulmonary aspergillosis in critically ill patients with severe COVID-19 pneumonia: Results from the prospective AspCOVID-19 study

Invasive pulmonary aspergillosis in the COVID-19 era: An expected new entity

Evaluation of Real-Time PCR, Galactomannan Enzyme-Linked Immunosorbent Assay (ELISA), and a Novel Lateral-Flow Device for Diagnosis of Invasive Aspergillosis

Detection of invasive pulmonary aspergillosis in critically ill patients by combined use of conventional culture, galactomannan, 1-3-beta-D-glucan and Aspergillus specific nested polymerase chain reaction in a prospective pilot study

Aspergillus Lateral Flow Assay with Digital Reader for the Diagnosis of COVID-19-Associated Pulmonary Aspergillosis (CAPA): a Multicenter Study

Invasive aspergillosis in patients with severe COVID-19 pneumonia

Epidemiology of Invasive Pulmonary Aspergillosis Among Intubated Patients With COVID-19: A Prospective Study

Late histopathologic characteristics of critically ill COVID-19 patients: Different phenotypes without evidence of invasive aspergillosis, a case series

COVID-19-associated invasive aspergillosis: Data from the UK national mycology reference laboratory

Confronting and mitigating the risk of COVID-19 associated pulmonary aspergillosis

HIV/AIDS, Hepatitis B infection, healthcare worker, pregnancy, etc