Multicenter Retrospective Study of Invasive Fusariosis in Intensive Care Units, France

Elevated SOFA scores at admission or history of allogeneic hematopoietic stem cell transplantation or hematologic malignancies were associated with death.

I nvasive fungal infections are common, and severe complications can occur in immunocompromised patients, especially in patients with hematologic malignancies who require intensive care unit (ICU) admission (1,2).Invasive fusariosis is a mycosis caused by infection with Fusarium spp.(3).Fusarium are ubiquitous filamentous fungi that can cause a range of infections, from localized lesions due to penetrating trauma in healthy persons, to acute invasive or disseminated infection in immunocompromised patients (3)(4)(5)(6).Most frequent clinical manifestations of invasive fusariosis are fever refractory to antimicrobial drugs, pneumonia, metastatic skin lesions of a disseminated infection, and sinusitis (3,4,6).
The European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the US National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) published definitions for proven and probable invasive fusariosis in immunocompromised patients (7).Although proven infection requires microscopic analysis or culture of a sterile material, probable infection is based on host factors, clinical features, and mycologic criteria (Appendix Table 1, https:// wwwnc.cdc.gov/EID/article/30/2/23-1221-App1.pdf).Despite progress in managing invasive fungal infections in recent decades, including the widespread use of antifungal prophylaxis in immunocompromised patients and improved treatment strategies, invasive fusariosis remains a serious and potentially life-threatening infection.Invasive fusariosis can lead to severe organ failure and has been associated with mortality rates ranging from 40% to 70% (8)(9)(10)(11).Even when amphotericin B and voriconazole are the first drugs of choice, sometimes in combination, the best antifungal treatment remains unclear (12).
Data focusing on fusariosis rely mainly on case reports (13)(14)(15), studies based on selected populations (9)(10)(11)16,17), or epidemiologic studies (18,19).None of those studies focused on critically ill patients with invasive fusariosis.We conducted a multicenter retrospective study to describe the characteristics and outcomes of invasive fusariosis in ICU patients in France and to identify the main risk factors associated with death and response to therapy.

Ethics
This observational study was based on anonymized hospitalization reports and was in strict compliance with the reference methodology MR-004 of France.The study was approved by the data protection authority, Commission Nationale de l'Informatique et des Libertés (registration no.2220799v0), and received a favorable opinion from the Comité Ethique de la Société de Réanimation de Langue Française institutional review board (approval no.20-95).The study was conducted in accordance with principles of the Declaration of Helsinki (World Medical Association, https://www.wma.net).

Study Population
We retrospectively included in the study adult ICU patients with a diagnosis of invasive fusariosis during  1).We identified patients by reviewing medical records, microbiologic databases, or both.To identify patients eligible for our study, we conducted a comprehensive screening of all ICUs and parasitology and mycology departments in France from which Fusarium species had been identified.Of 47 screened ICUs, only 16 had patients with a positive Fusarium microbiologic documentation during the inclusion period.We assessed a total of 120 patients for eligibility.We excluded 53 patients with Fusarium colonization and ultimately included 55 patients in the study (Figure 1).We collected data from anonymized hospitalization reports, including information on patient age, sex, underlying disease conditions, history of immunodeficiency, clinical and microbiologic characteristics of the Fusarium infection, any co-infections, antifungal treatment, need for organ support, and outcomes.For each patient, simplified acute physiology score (SAPS II) and sequential organ failure assessment (SOFA) scores were collected at admission, as previously defined (20,21).For response to therapy, we defined progression as clinical deterioration or death after antifungal treatment.We considered response complete if clinical improvement occurred, biological samples became sterile, and computed tomography (CT) features cleared.When all the complete response criteria were not met, we considered response to therapy to be partial.

Outcomes
Our primary objective was to identify factors associated with ICU mortality rates.Our secondary objective was to identify factors associated with response to therapy.

Statistical Analyses
We reported continuous variables as medians and interquartile ranges (IQRs) and categorical variables as numbers and percentages.We considered response to treatment, death in the ICU, death in the hospital, and death within 1 year as binary endpoints for the main analysis.We compared continuous variables by using Wilcoxon rank-sum test and compared categorical variables by using Fisher exact test.We performed adjusted analyses to evaluate factors associated with treatment response and death by using multivariable logistic regression models.We estimated cumulative incidence of death in the ICU by using standard methods for competing events and considered discharged alive as a competing outcome to ICU death.We performed all tests 2-sided at the 5% significance level.We performed all analyses on the R statistical platform (The R Foundation for Statistical Computing, https://www.r-project.org).
Results of serum galactomannan test were available for 50 (90.9%)patients and 15 (30%) of them had a positive serum galactomannan test on the day of invasive fusariosis diagnosis.Among those patients, 4 also had concomitant aspergillosis diagnosed.Other observed co-infections included bacterial co-infection in 58% (n = 32), viral co-infection in 35% (n = 19), and fungal co-infection in 34% (n = 19) of invasive fusariosis patients (Appendix Table 3).
Two thirds of invasive fusariosis patients received antifungal monotherapy.The 2 primary drugs used were voriconazole (62%, n = 34) and amphotericin B (60%, n = 33).Four (7%) patients died before invasive fusariosis diagnosis and did not receive treatment.Granulocyte colony-stimulating factor was administered to 55% (n = 12) of the neutropenic patients, and surgical debridement of localized lesions was performed in 13% (n = 7) of patients.Half of the patients experienced disease progression despite receiving adequate therapy.

Factors Associated with Death
By univariate analysis, signs and symptoms significantly associated with death in the ICU included history of hematologic malignancies and allo-HSCT (p = 0.017), immunosuppressive therapy other than corticosteroids (p = 0.036), elevated SAPS II (p = 0.007) or SOFA (p = 0.001) score at admission, and neutropenia (neutrophils <0.5 G/L) (p = 0.05) (Appendix Table 5).Among patients with organ failure, only vasopressors were associated with death (p = 0.006).Conversely, surgical debridement of localized lesion was associated with ICU survival (p = 0.014).

Discussion
We conducted a large retrospective study to describe the clinical characteristics and outcomes of patients   with invasive fusariosis admitted to ICUs in France.We found that invasive fusariosis can be life-threatening; often is associated with bacterial, viral, and fungal co-infections; and occurs mainly in immunocompromised patients or patients enduring extended ICU stays.Pneumonia is the prevailing clinical manifestation in ICU patients.Despite ICU hospitalization, organ support, and adequate treatment, the fusariosis mortality rate remains high.SOFA score and a history of allo-HSCT or hematologic malignancies, or both, are significantly associated with death in the ICU.
The available medical literature on invasive fusariosis remains scarce, and a paucity of studies specifically focused on invasive fusariosis in the ICU setting.The few studies dedicated to invasive fungal infections in ICU patients included <5 patients with invasive fusariosis (2,22).Our study provides a comprehensive description of clinical, biologic, and microbiologic characteristics of critically ill invasive fusariosis patients.
Previous studies indicate that most invasive fusariosis patients have immunocompromising conditions, including hematologic malignancies, recent allo-HSCT, or solid organ transplantation (9,11,17,19,23).However, we found that one fifth of patients with invasive fusariosis in the ICU are considered immunocompetent at ICU admission but experienced a prolonged ICU hospitalization, mainly because of septic shock.That finding supports the concept of sepsisinduced immunosuppression, wherein an imbalanced inflammatory state contributes to immunoparalysis and increases the risk for nosocomial infections (24).Therefore, physicians should investigate the possibility of invasive fusariosis in patients with prolonged ICU hospitalization, especially in cases of a secondary sepsis unresponsive to antimicrobial agents and clinical manifestations consistent with invasive fusariosis.
The clinical manifestations we observed in our cohort align with those from previous reports (3,4,6).However, a notable finding in our study was identification of Fusarium in mycologic culture (semiquantitative results showing numerous Fusarium colonies) from pancreatic fluid collected in a case of suspected infection from a site that has not been previously described.That novel observation highlights the importance of considering Fusarium as a potential pathogen in unusual infection sites and expands our understanding of clinical manifestations of fusariosis.In addition, our study revealed a lower prevalence of disseminated infection, affecting only one fifth of patients, in contrast to reports from previous publications focusing on non-ICU patients (8,11,17).However, patients with hematologic malignancies in our study exhibited much higher rates of disseminated infections.That finding aligns with a hypothesis proposed by others that suggests the larger proportion of neutropenia in fusariosis patients might contribute to the increased susceptibility to disseminated infection (17).
We noted a marked predominant prevalence of pneumonia (76%) among our study population.That finding highlights that fusariosis-related pneumonia can lead to acute respiratory failure, often necessitating invasive mechanical ventilation.Thoracic CT patterns of fusariosis-related pneumonia included pulmonary consolidations, micronodules and nodules, and ground-glass opacities.Excavations and pleural effusions have also been observed, but proportions of those CT patterns vary across different publications, mainly due to the small number of patients included (25)(26)(27).In addition, the timing of imaging and presence of neutropenia or co-infections might influence those patterns.Many patients in our study had coinfections; thus, we cannot attribute their CT patterns solely to invasive fusariosis.
All patients in our study who had solid organ transplants also had acute kidney injury.That difference varied from previous reports and could be explained mainly by the presence of calcineurin inhibitor, well known for its nephrotoxicity (28).In addition, one third of solid organ transplant patients in our study had undergone a kidney transplant, which might have contributed to the increased susceptibility to acute kidney injury in this subgroup.
Despite identifying various factors associated with treatment response in the univariate analysis, the multivariate analysis in our study did not reveal any independent risk factors.However, the small number of patients included in our study might have limited the statistical power of the analysis.Moreover, we considered all non-ICU survivors to be nonresponders and 4 patients died before receiving treatment, findings others should consider when interpreting our results but that further emphasize the need for larger studies among more extensive patient populations to better elucidate the independent risk factors associated with treatment response in invasive fusariosis.
The optimal antifungal treatment for invasive fusariosis remains uncertain (29).The heterogeneity of treatments administered to patients across different centers in our study further complicates the interpretation of our results.However, our analysis was underpowered to detect a favorable response with voriconazole.Conversely, a previous study reported a 90-day survival rate of 60% with voriconazole monotherapy (8), surpassing the outcomes associated with liposomal or deoxycholate amphotericin B. Another study demonstrated an overall response rate of 47% with voriconazole (10).Nevertheless, because the current literature primarily consists of case reports and small retrospective studies, determining the optimal antifungal regimen for such patients remains challenging.
The mortality rate observed for ICU patients with invasive fusariosis in our study was notably high, reaching 56%.That finding is consistent with previous studies reporting mortality rates ranging from 40% to 70% in patients with invasive fusariosis, although those studies did not specifically focus on ICU patients (8)(9)(10)(11).In univariate analysis, the observed association between surgical debridement and survival could be attributed to the fact that patients were well enough, and possibly had less severe and fewer disseminated infections, to undergo debridement.By multivariate analysis, we identified history of hematologic disease, including active hematologic malignancy or recent allo-HSCT, as an independent risk factor for death.Patients with hematologic malignancies and those who have undergone allo-HSCT are more likely to experience neutropenia.Persistent neutropenia has been identified as a factor associated with increased mortality rates among invasive fusariosis patients in several previous studies (8,16,17,30).
One limitation in our study is the lack of assessment of persistent neutropenia during hospitalization because of missing data on hospitalization reports; those missing data prevented a comprehensive analysis of the effects of persistent neutropenia on patient outcomes in our study population.Also, because we did not have access to the total number of immunocompromised patients admitted to ICUs during the entire study period, we were unable to estimate the prevalence of invasive fusariosis in this population.Finally, the variability in ICU admitting policies across different centers might have influenced our study results.Some patients with invasive fusariosis and underlying conditions or poor prognosis related to hematologic malignancies might have been denied ICU admission.That potential selection bias could affect the generalizability of our findings and should be considered when interpreting the results.
In conclusion, invasive fusariosis is a severe condition that can lead to multiorgan failure and is associated with high mortality rates in the ICU.Clinicians should consider invasive fusariosis as a potential diagnosis in immunocompromised patients who have pneumonia or persistent fever unresponsive to antimicrobial agents.Treatment for invasive fusariosis includes antifungal therapy, rapid reversal of neutropenia, and surgical debridement for localized lesions.Further research is warranted to optimize diagnostic strategies and treatment approaches for this challenging and life-threatening infection.However, clinicians should closely monitor ICU patients with a history of hematologic malignancies or allo-HSCT because of significantly higher invasive fusariosis ICU mortality rates among those patients.
L.Z. reports receiving fees for lectures for MSD and Sanofi and her institution received a research grant from Jazz Pharmaceuticals.E.C. received fees for lectures and conference talks and had travel and accommodation expenses related to attending scientific meetings covered by Gilead, Shionogi B.V., and Sanofi-Genzyme.N.B. reports fees and reimbursements for national congresses from Sanofi.F.P. report receiving fees for lectures and consulting from Gilead and an institutional grant from ALEXION Pharma.A.D.J. reports receiving remuneration for presentations from Medtronic, Drager, Viatris, and Fisher & Paykel.Other authors declare no conflict of interest.

About the Author
Dr. Demonchy is a specialist in intensive care at the Medical Intensive Care Unit, Critical Care Center, University Hospital of Lille, France.Her primary interest is in infectious diseases and immunocompromised patients, especially patients with hematologic malignancies.In this EID podcast, Dr. Niaz Banaei, a professor of pathology and medicine at Stanford University in California, discusses Mycobacterium marinum infection after an iguana bite in Costa Rica.

Figure 2 .
Figure 2. Main clinical manifestations among patients in a multicenter retrospective study of invasive fusariosis in intensive care units, France.The figure was partly generated using Servier Medical Art (https://smart.servier.com),licensed under a Creative Commons Attribution 3.0 unported license.

Figure 3 .
Figure 3. Thoracic computed tomography scans from patients included in a multicenter retrospective study of invasive fusariosis in intensive care units, France, showing findings of fusariosis-related pneumonia.A) Unilateral consolidations (asterisk); B) ground glass opacities (triangle); C) 7-mm nodule (arrow) and bilateral pleural effusion.

Figure 4 .
Figure 4. Cumulative incidence of death among patients with and without history of hematologic malignancies and allo-HSCT in a multicenter retrospective study of invasive fusariosis in ICUs, France.Calculations used Fisher exact test (p = 0.017).Chart shows 14-day and 28day death rates.Allo-HSCT, allogeneic hematopoietic stem cell transplant; ICU, intensive care unit.

®
Zoonotic infections associated with animal bite injuries are common and can result in severe illness.Approximately 5 million animal bites occur annually in North America, and 10 million injuries occur globally from dog bites alone.Pathogens causing infections after dog or cat bites are well described; pathogens from other animal bites are less well defined, although their oral microbiota are known.

Table 2 .
Organ failure and outcomes among 55 patients in a multicenter retrospective study of invasive fusariosis in intensive care units, France* Values are no.(%) except as indicated.IQR, interquartile range.With orotracheal intubation cannula or tracheotomy. *