Abstract
Mold bloodstream infections are rare but highly fatal, especially with Lomentospora prolificans and Scedosporium spp. Among 84 episodes over 20 years, the 30-day mortality rate was 38%. Cancer, intensive care unit admission, and healthcare-onset infection were linked to increased mortality rates, highlighting the need for early detection and better management strategies.
Keywords: antifungal resistance, bloodstream infection, mold, mortality
Mold bloodstream infections (BSIs) are rare but life-threatening conditions associated with significant morbidity and mortality rates, particularly in immunocompromised patients, such as those with hematological cancer, organ transplant recipients, or patients receiving immunosuppressive therapy [1, 2]. Mold fungemia is often due to Fusarium spp, Lomentospora prolificans, Scedosporium spp, and, less frequently, Aspergillus spp or other filamentous fungi [3, 4]. These infections pose significant diagnostic and therapeutic challenges due to their intrinsic resistance to multiple antifungal agents and their ability to cause disseminated disease [5, 6]. Furthermore, they may cause confusion in posing the possibility of exogenous “contamination.”
Epidemiological studies of mold BSIs remain limited, with most data derived from small case series or single-center cohorts [7, 8]. A recent multicenter study from France reported that Fusarium spp accounted for most cases, whereas L prolificans was less frequent [3]. However, the distribution of causative pathogens may vary geographically, influenced by environmental exposures, at-risk patient populations, and local diagnostic practices [3, 5, 8]. Given the challenges in culturing filamentous fungi from blood, the true incidence and clinical burden of mold fungemia may be underestimated [6, 7].
Here we provide a comprehensive, population-based analysis of mold fungemia over a 20-year period in Queensland, Australia. We aim to describe the clinical characteristics, microbiological spectrum, antifungal susceptibility patterns, and outcomes associated with mold BSIs, with a focus on identifying mortality risk factors .
METHODS
This study included all patients who received care at public hospitals or health services (n = 70) in Queensland, Australia, from 1 January 2000 to 31 December 2019 and had a mold or filamentous fungus isolated from blood cultures. The BACT/ALERT 3D system (bioMérieux) was used throughout the study period, except that the BACT/ALERT VIRTUO system (bioMérieux) was implemented at the main central laboratory in 2018 that manages culture submissions from the Greater Brisbane area and several rural Queensland sites. Blood cultures were incubated for 5 days before being discarded if no growth was detected.
Mold isolates were identified to the species complex level based on morphological characteristics seen at culture [9]. The decision to report the presence of a mold or filamentous fungi in blood, including identification to the genus, species complex or species level, was at the discretion of the clinical microbiologist, as was the decision to perform antifungal susceptibility testing. Minimum inhibitory concentrations (MICs) to antifungal drugs were determined using the Sensititre YO10 method (Thermo Scientific), following the Clinical and Laboratory Standards Institute M38 document [10].
Data were analyzed using Stata 18.0 software (StataCorp). Univariate logistic regression was used to assess for predictors of the 30-day mortality status. Cox proportional hazards models were used for survival analysis, generating hazard ratios and a Kaplan-Meier curve for the 1-year all-cause mortality rate. The human research ethics committee at Royal Brisbane and Women’s Hospital approved this study and granted a waiver of individual consent (LNR/2020/QRBW/62494).
RESULTS
A total of 83 patients with mold or filamentous fungus fungemia (84 episodes) were identified. The episode breakdown of causative fungus according to genus was as follows: Fusarium spp (24 episodes, including Fusarium solani complex [n = 8] and Fusarium oxysporum complex [n = 1]), Scedosporium spp and L prolificans (25 episodes, including L prolificans [n = 21]), unidentified filamentous fungus (17 episodes), Aspergillus spp (8 episodes, including Aspergillus fumigatus complex and Aspergillus niger complex [n = 2], Aspergillus terreus complex, and Aspergillus flavus complex), Mucor spp (4 episodes), Curvularia spp (4 episodes), Penicillium spp (1 episode), and Purpureocillum spp (1 episode). One patient had recurrence of L prolificans BSI 1 year after the index episode. Of the episodes identified to at least the genus level (67 of 84 [79.8%]), only 18 (26.9%) and 22 (32.8%) were reported to species complex and species levels, respectively.
Forty-eight patients (57.1%) were male, and the median Charlson Comorbidity Index (CCI) was 2 (interquartile range, 0–4); 19 patients (22.9%) were children, of whom 5 had Fusarium spp fungemia. Forty-eight patients (57.1%) had cancer (Table 1). Fusarium spp (16 of 24 [66.7%]), Scedosporium spp (3 of 4 [75.0%]), and Lomentospora spp (20 of 21 [95.2%]) were most associated with cancer. Approximately half of episodes (41 of 84 [48.8%]) were associated with a documented hematological cancer, including acute myeloid leukemia (n = 16), acute lymphoid leukemia (n = 11), and non-Hodgkin lymphoma (n = 6). Of those with Aspergillus species fungemia (n = 8), only 1 patient had both underlying cancer and an indwelling vascular catheter; 5 (62.5%) had a CCI of 0. The median time to blood culture positivity was 71 hours (interquartile range, 48–83 hours), and all episodes of fungemia were detected in the aerobic blood culture bottle (BacT/ALERT FA plus; bioMérieux). Thirty-two patients (38.1%) had indwelling vascular catheters at the time of fungemia detection.
Table 1.
Clinical Characteristics by 30-Day Mortality Status
| Characteristic | Patients by 30-d Mortality Status, No. (%)a | P Value | |
|---|---|---|---|
| Death (n = 31) | Survival (n = 52) | ||
| Age, median (IQR), y | 50.8 (24.6–63.0) | 36.5 (16.7–59.0) | .23 |
| Sex | |||
| Female | 16 (50.0) | 20 (38.5) | .30 |
| Male | 16 (50.0) | 32 (61.5) | |
| CCI, median (IQR) | 3 (2–5) | 1.5 (0–2.5) | <.01 |
| Cancer | |||
| No | 3 (9.4) | 33 (63.5) | <.01 |
| Yes | 29 (90.6) | 19 (36.5) | |
| Diabetes mellitus | |||
| No | 26 (81.3) | 47 (90.4) | .23 |
| Yes | 6 (18.7) | 5 (9.6) | |
| Liver disease | |||
| No | 29 (90.6) | 48 (92.3) | .79 |
| Yes | 3 (9.4) | 4 (7.7) | |
| Pulmonary disease | |||
| No | 27 (84.4) | 44 (84.6) | .98 |
| Yes | 5 (15.6) | 8 (15.4) | |
| Kidney disease | |||
| No | 27 (84.4) | 48 (92.3) | .25 |
| Yes | 5 (15.6) | 4 (7.7) | |
| Rheumatological disease | |||
| No | 31 (96.9) | 52 (100.0) | .20 |
| Yes | 1 (3.1) | 0 (0.0) | |
| Species | |||
| Aspergillus spp | 1 (3.1) | 7 (13.5) | .15 |
| Fusarium spp | 8 (25.0) | 16 (30.8) | .57 |
| Scedosporium spp | 3 (9.4) | 1 (1.9) | .16 |
| Lomentospora spp | 16 (50.0) | 5 (9.6) | <.01 |
| Mucor spp | 0 (0) | 4 (7.7) | … |
| Curvularia spp | 0 (0) | 4 (7.7) | … |
| Unidentified fungus | 4 (12.5) | 15 (28.9) | .09 |
| Time to blood culture positivity, median (IQR), h | 72 (56–82) | 70 (39–84) | .37 |
| Admission to ICU | |||
| No | 24 (77.4) | 50 (96.1) | <.01 |
| Yes | 7 (22.6) | 2 (3.9) | |
| Acquisition and onset | … | … | |
| HAHO | 26 (83.8) | 20 (38.5) | <.01 |
| HACO | 3 (9.7) | 7 (13.5) | .57 |
| CACO | 2 (6.5) | 25 (48.1) | <.01 |
| Polymicrobial | 3 (9.4) | 5 (9.6) | .97 |
| Indwelling vascular catheter | |||
| No | 14 (45.2) | 37 (71.2) | .02 |
| Yes | 17 (54.8) | 15 (28.8) | |
Abbreviations: CACO, community-acquired community onset; CCI, Charlson Comorbidity Index; HACO, healthcare-associated community onset; HAHO, healthcare-associated healthcare onset; ICU, intensive care unit; IQR, interquartile range.
aData represent no. (%) of patients unless otherwise specified.
Mortality Rates
The mortality rates at 30, 90, and 365 days were 38.1%, 39.8%, and 47.0%, respectively. The highest rates were seen with L prolificans and Scedosporium spp, with Curvularia spp associated with no deaths and Aspergillus spp with only 1 death during 1 year of follow-up (Figure 1). On univariate analysis, CCI, cancer, admission to the intensive care unit, the presence of an indwelling vascular catheter, and healthcare-onset fungemia were all associated with elevated 30-day mortality rates. Conversely, community-onset fungemia was associated with a reduced risk of death at 30 days.
Figure 1.
Kaplan-Meier survival curve up to 1 year by mold species.
Sensitivity Analysis
To define mortality risk of mortality more accurately for true fungemia, a sensitivity analysis was performed that excluded patient-episodes (n = 23) with a CCI of 0 and those in which the filamentous fungus was not identified (n = 11). Significant baseline differences between groups based on CCI included age, fungus species, setting of fungemia, polymicrobial BSI, and mortality rates (Supplementary Table 1). No deaths occurred in patients with a CCI of 0. Mortality rates at 30, 90, and 365 days were 54.0%, 58.0%, and 66.0%, respectively. High mortality rates continued to be observed for L prolificans and Scedosporium spp. All patients with L prolificans BSI had died by 1 year (Supplementary Figure 1). The mortality rate for Aspergillus spp increased significantly, although patient numbers were low (Supplementary Table 2).
Antifungal Susceptibility
MIC range results were available for Fusarium spp isolates, as follows: Fusarium spp (n = 6; 0.25–4, 1–8, >8, and >16 mg/L, respectively, for amphotericin B, voriconazole, posaconazole, and itraconazole). F solani complex (n = 7; 0.5–2, 1–8, ≥8, and >16 mg/L, respectively), and F oxysporum complex (n = 1; 2, 4, >8, and >16). MIC range results were available for 12 L prolificans isolates: (>8, >8, >8, and >16 mg/L, respectively, for amphotericin B, voriconazole, posaconazole, and itraconazole).
DISCUSSION
Fungemia due to filamentous fungi is a rare but potentially rapidly lethal infection. Our study showed that Fusarium spp and L prolificans were the most common mold species identified in blood, in approximately equal numbers (in 28.6% and 25.0% episodes, respectively). Of patients who died within 1 year of follow-up, approximately half did so within 10 days. Recurrence was rare due to the high mortality rate associated with this infection. Cancer, indwelling vascular catheter, and hospital-onset fungemia were all associated with excess mortality rates. A sensitivity analysis confirmed very elevated and rapid mortality rates.
Epidemiology and species of mold fungemia varies by geographic region. In a cohort study from France (80 events), approximately two-thirds (67.5%) of cases were due to Fusarium spp and only 10% due to L prolificans [3]. Fusarium spp have been reported to be the leading agents of fungemia caused by molds in patients with hematological cancer [1, 12]. Fungemia also frequently complicates severe and disseminated L prolificans infection, occurring in up to 75.3% of cases [5]. It is also linked to hematological cancer [2]. Its geographic distribution is focused in Spain, Australia, the United States, and Germany. Definite fungemia due to Aspergillus spp is exceedingly rare, has been identified only in large surveillance studies, and almost exclusively affects patients with advanced hematological cancer and late in their infection [3, 4]. In fact, this has led to a recommendation in international guidelines to consider growth of Aspergillus spp from blood cultures to be a contaminant until proven otherwise [13].
The clinical importance of fungemia due to specific molds (eg, Aspergillus spp and Mucor spp) can be difficult to discern. Pseudofungemia due to laboratory or environmental contamination has been reported infrequently and should be considered in the appropriate context [11]. However, underlying disease status (eg, hematological cancer), immunosuppression, and the presence of an indwelling vascular catheter dramatically increases the pretest probability of true fungemia [14]. In our study we attempted to distinguish true fungemia from pseudofungemia by looking at a sensitivity analysis subcohort that excluded patients with no comorbid conditions and in which the fungal isolate was not identified. Mortality rates increased at all time points during follow-up (especially for Aspergillus spp), which adds weight to their significance as true pathogens.
Filamentous fungi typically do not grow well and are not detected in modern automated blood culture systems [7]. This can hinder early diagnosis and treatment. There remains a strong clinical need for rapid and accurate molecular diagnostics for invasive fungal disease [6]. Moreover, mold species identification from blood can also be problematic. Nineteen patient-episodes (22.6%) involved an unidentified filamentous fungus. There are many possible reasons why this situation may have occurred: (1) mold determined to be a contaminant, (2) isolate identification deemed not likely to change outcome (eg, patient death), (3) hyphal fragments identified with Gram stain but fungal isolate unable to be grown, or (4) insufficient laboratory resources to accurately identify the filamentous fungus.
True mold fungemia, especially when due to certain fungal species (eg, L prolificans), often heralds a terminal event. It is frequently uncertain as to whether this reflects host immune failure (eg, iatrogenic or disease related) or fungal virulence. Pooled analysis of mold fungemia episodes from French hospitals identified a 90-day overall mortality rate of 53% [3], higher than the approximately 40% we identified in our study. This likely reflects their strict case definitions and exclusion of pseudofungemia. The 90-day mortality rates for both Fusarium spp (33%) and L prolificans (86%) fungemia in our study were slightly lower than in other studies. Of note, 0% of those with Mucor spp in blood in our study and 12% of those with Aspergillus spp died within 90 days (compared with 75% and 100% in the French study [3]), with only 1 patient in each group having an underlying cancer.
The strengths of our study included the relative cohort size of patients with mold fungemia (largest recorded cohort), the study duration (>20 years), the large population size covered (state wide), and complete follow-up mortality data. The limitations included a small sample size, which hindered statistical analyses, the likely inclusion of patients with pseudofungemia, and incomplete or absent clinical (eg, other sites of infection and antifungal therapy), microbiological (eg, species identification and antifungal susceptibility results), and radiological data.
In conclusion, fungemia due to mold pathogens is a rare but often fatal condition. Our study, which represents the largest reported cohort of patients with mold fungemia, highlights the predominance of Fusarium spp and L prolificans as causative pathogens. The high mortality rate associated with these infections—especially within the first 30 days—underscores the urgent need for improved diagnostic strategies and more effective antifungal therapies.
Supplementary Material
Contributor Information
Adam G Stewart, Transplant and Immunocompromised Host Infectious Diseases, Infectious Diseases Division, Massachusetts General Hospital, Boston, Massachusetts, USA; Harvard Medical School, Boston, Massachusetts, USA; Centre for Clinical Research, Faculty of Medicine, The University of Queensland, Royal Brisbane and Women's Hospital Campus, Brisbane, Queensland, Australia.
Kevin B Laupland, Department of Intensive Care Services, Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia; Queensland University of Technology, Brisbane, Queensland, Australia.
Felicity Edwards, Department of Intensive Care Services, Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia; Queensland University of Technology, Brisbane, Queensland, Australia.
Sharon C A Chen, Sydney Infectious Diseases Institute, The University of Sydney, Sydney, New South Wales, Australia; Department of Infectious Diseases, Westmead Hospital, Sydney, New South Wales, Australia; Centre for Infectious Diseases and Microbiology Laboratory Services, New South Wales Health Pathology, Westmead Hospital, Sydney, New South Wales, Australia.
Monica A Slavin, National Centre for Infections in Cancer, Peter MacCallum Cancer Centre, Melbourne, New South Wales, Australia; Department of Infectious Diseases, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia; Sir Peter MacCallum Department of Oncology, The University of Melbourne, Melbourne, Victoria, Australia.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
References
- 1. Nucci M, Anaissie E. Invasive fusariosis. Clin Microbiol Rev 2023; 36:e00159-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Neoh CF, Chen SC, Lanternier F, et al. Scedosporiosis and lomentosporiosis: modern perspectives on these difficult-to-treat rare mold infections. Clin Microbiol Rev 2024; 37:e0000423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Tala-Ighil T, Garcia-Hermoso D, Dalle F, et al. Epidemiology and prognostic factors associated with mold-positive blood cultures: 10-year data from a French prospective surveillance program (2012–2022). Clin Infect Dis 2025; 80:529–39. [DOI] [PubMed] [Google Scholar]
- 4. Kontoyiannis DP, Sumoza D, Tarrand J, Bodey GP, Storey R, Raad II. Significance of aspergillemia in patients with cancer: a 10-year study. Clin Infect Dis 2000; 31:188–9. [DOI] [PubMed] [Google Scholar]
- 5. Konsoula A, Agouridis AP, Markaki L, Tsioutis C, Spernovasilis N. Lomentospora prolificans disseminated infections: a systematic review of reported cases. Pathogens 2022; 12:67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kidd SE, Chen SC, Meyer W, Halliday CL. A new age in molecular diagnostics for invasive fungal disease: are we ready? Front Microbiol 2019; 10:2903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Herrera LN, Khodadadi R, Leal S, Kulkarni P, Pappas P, McCarty T. Clinical utility of routine use of fungal blood cultures. Am J Med 2023; 136:514–7. [DOI] [PubMed] [Google Scholar]
- 8. Carlesse F, Amaral AC, Gonçalves SS, et al. Outbreak of Fusarium oxysporum infections in children with cancer: an experience with 7 episodes of catheter-related fungemia. Antimicrob Resist Infect Control 2017; 6:93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ciardo DE, Schär G, Altwegg M, Böttger EC, Bosshard PP. Identification of moulds in the diagnostic laboratory—an algorithm implementing molecular and phenotypic methods. Diagn Microbiol Infect Dis 2007; 59:49–60. [DOI] [PubMed] [Google Scholar]
- 10. CLSI. Epidemiological Cutoff Values for Antifungal Susceptibility Testing. 3rd ed. CLSI supplement M59. Wayne, PA: Clinical and Laboratory Standards Institute; 2020.
- 11. Hruszkewycz V, Ruben B, Hypes CM, Bostic GD, Staszkiewicz J, Band JD. A cluster of pseudofungemia associated with hospital renovation adjacent to the microbiology laboratory. Infect Control Hosp Epidemiol 1992; 13:147–50. [DOI] [PubMed] [Google Scholar]
- 12. Nucci M, Anaissie E. Emerging fungi. Infect Dis Clin North Am 2006; 20:563–79. [DOI] [PubMed] [Google Scholar]
- 13. Donnelly JP, Chen SC, Kauffman CA, et al. 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. Clin Infect Dis 2020; 71:1367–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Duthie R, Denning DW. Aspergillus fungemia: report of two cases and review. Clin Infect Dis 1995; 20:598–605. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.