Clinical impact of altered gut microbiota and metabolite profiles on mortality in patients with candidemia: A prospective observation pilot study

doi:10.21203/rs.3.rs-6551378/v1

Download PDF

Research Article

Clinical impact of altered gut microbiota and metabolite profiles on mortality in patients with candidemia: A prospective observation pilot study

https://doi.org/10.21203/rs.3.rs-6551378/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The gut microbiota plays an important role in defense against infectious diseases. However, data on clinical effects of microbiome profiles in patients with candidemia are limited. This study investigated the intestinal microbiome and the effects on mortality in patients with candidemia.

Methods

In this prospective observational pilot study, fecal samples from adult patients with candidemia were subjected to 16S rRNA gene sequencing for microbiota analysis and gas chromatography–mass spectrometry metabolomics. Multivariate logistic regression was conducted to identify predictors of in-hospital mortality.

Results

Fifty-nine patients with candidemia were analyzed, with an in-hospital mortality rate of 40.7%. The median Shannon’s diversity index of the gut microbiota was significantly lower in survivors than in non-survivors (P = 0.009). Linear discriminant analysis Effect Size revealed 11 bacterial species that significantly differed between the two groups. Among 111 fecal metabolites, only 3-isopropoxy-hexamethyl-tetrasiloxane was differentially expressed between survivors and non-survivors (P = 0.007). Septic shock (adjusted odds ratio: 10.59; 95% confidence interval, 1.70–65.97), underlying malignancy (7.79 [1.41–43.10]), and Shannon’s diversity index (0.40 [0.19–0.84]) were significant predictors of in-hospital mortality.

Conclusion

Low gut bacterial diversity was independently associated with increased mortality in patients with candidemia. The intestinal microbiome could offer new perspectives for the prevention and the treatment of candidemia.

gastrointestinal microbiome

metabolome

candidemia

mortality

Candidemia is an important fungal disease caused by Candida species and, increasingly, non-albicans Candida pathogens, which are currently the fourth leading cause of nosocomial bloodstream infections. [1] Despite advancements in antifungal treatment, candidemia treatment is challenging, with mortality rates of up to 30%. [2] In immunity-impaired individuals, candidemia incidence is particularly elevated, primarily due to immunosuppressive agents, chemotherapy, and disruptions of the gut microbiome. [3]

Patients with bacteremia often exhibit bacterial pathobiont expansion in the intestines, resulting in bloodstream translocation and significant changes in the composition and diversity of the gut microbiota. [4] Candida species, a commensal fungus, exists as an endogenous human reservoir in several distinct anatomical sites, such as the gastrointestinal tract. [5] In cases of intra-abdominal bacterial dysbiosis, particularly loss of anaerobic bacteria, overgrown Candida species can translocate into the bloodstream and become opportunistic, causing lethal systemic infections. [6]

The composition and diversity of the gut microbiome in patients with invasive candidiasis has been highlighted using 16s rRNA gene sequencing. Gastrointestinal colonization by Candida species shapes immune responses, and antagonistic and synergistic Candida-bacteria interactions can influence microbial pathogenesis. [7, 8] Notably, commensal anaerobic bacteria, such as gut commensal Clostridia in the Firmicutes phylum, have also been suggested as factors that contribute to resistance to intestinal colonization by C. albicans. [9] Furthermore, invasive candidiasis is strongly associated with intestinal bacterial dysbiosis and extensive alteration in fecal metabolic profiles. [6, 10]

Although gut microbiota and metabolic profiling have been researched extensively, their clinical implications in patients with candidemia remain poorly understood, particularly from the perspective of the relationship between gut microbiota and metabolites.

The present study aimed to characterize intestinal microbiota and metabolites in relation to mortality in patients with candidemia, providing insights into the role of the gut microbiota in invasive fungal infections. Understanding these interactions may reveal novel therapeutic and diagnostic approaches for candidemia management.

Study design

This was a prospective observational case-control pilot study conducted at a 1,048-bed university-affiliated hospital in Korea between January 2022 and February 2024. The study included adult patients (≥ 19 years) with blood culture-confirmed candidemia, defined as at least one positive peripheral blood culture for Candida species with compatible clinical features. Non-survivors were defined as patients who died during hospitalization, whereas survivors were those that remained alive to be discharged.

Blood culture collection within 2 h after fever onset (≥ 38°C) and before antifungal therapy was prioritized to enhance diagnostic accuracy. Fecal specimens were collected within 5 days after candidemia diagnosis for the evaluation of gut microbiota and metabolite profiles. The fecal samples were aliquoted and stored frozen within 24 h of collection. Patients were excluded from this study if they refused to complete the consent form or if their stool sample was collected > 5 days after the date of candidemia diagnosis. Only the first episode of candidemia was analyzed for each patient.

The study protocol was approved by the Institutional Review Board of Korea University Anam Hospital (approval number: 2022AN0232), and written informed consent was provided by all participants or their surrogates prior to participation. This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational cohort studies.

Data collection

The following demographic and clinical data were prospectively obtained from the patients’ electronic medical records: age, sex, underlying diseases, Charlson’s Comorbidity Index [11], laboratory results, and risk factors identified within 1 month prior to candidemia diagnosis. The evaluated risk factors included neutropenia (absolute neutrophil count < 500 cells/µL), recent surgical history, and the use of immunosuppressive agents. The source of candidemia was determined based on clinical evidence of infection, regardless of whether causative organisms were isolated from the origin. [12–14] Clinical severity at the most severe stage of the disease was assessed using the following factors: presence of septic shock [15], the Pitt bacteremia scoring system [16], use of a central venous catheter, need for mechanical ventilation, admission to the intensive care unit, and in-hospital mortality.

Clinical microbiologic analysis

Identification of Candida spp. in blood culture and assessment of antifungal susceptibility were conducted using the BacT/ALERT 3D Microbial Detection System (bioMe´rieux, Inc., Durham, NC, USA) and the automated Vitek 2 Yeast Biochemical Card (bioMe´rieux, Inc., Durham, NC, USA), following a routine laboratory diagnostic procedure. The Candida strains were confirmed using matrix-assisted laser desorption/ionization-time of fight mass spectrometry (Bruker Daltonics, Bremen, Germany).

Stool specimen collection and sequencing

The V3–V4 region of the 16S rRNA gene was targeted for amplification to analyze the composition of the intestinal microbiota. The amplified products were purified using a magnetic bead-based purification process, and the appropriate concentration of the purified product was pooled together. Short fragments were removed using the ProNex® Size-Selective Purification System (Promega, Southampton, UK). The quality of the purified products was assessed using the PicoGreen assay (Molecular Probes, Invitrogen, USA). The pooled amplicons were sequenced using an Illumina MiSeq Sequencing System (Illumina, USA).

Low-quality sequence reads (Q < 25) were filtered out using Trimmomatic v0.32. Paired-end reads were merged using VSEARCH v2.13.4 with default parameters and trimmed at a similarity cutoff of 0.8 based on the Myers–Miller alignment algorithm. [17] Non-specific amplicons that did not encode the V3–V4 region of 16S rRNA were identified using HMMER v3.2.1. [18] Unique sequence reads were extracted, and duplicate reads were clustered using VSEARCH. [17] Taxonomic assignments were performed using the EzBioCloud 16S rRNA database, and chimeric sequences were removed using the UCHIME algorithm. [19, 20]

Metabolomic profiling

Fecal samples were homogenized using bead-beating and extracted with methanol for metabolomic analysis. After extraction, the samples were centrifuged, and the supernatant was filtered through a 0.45-µm syringe filter. All metabolites, including short chain fatty acids (SCFAs) such as acetic acid, butyric acid, valeric acid, and propionic acid, were analyzed using a gas chromatograph-mass spectrometer (gas chromatograph: Agilent 7890, mass spectrometer: LECO Pegasus HT TOFMS). All metabolites were analyzed using only significant peaks with a signal-to-noise ratio > 9. All procedures were performed in triplicate to ensure accuracy.

Statistical Analyses

Categorical variables were compared using Fisher’s exact test or Pearson’s chi-square test as appropriate and are expressed as numbers (proportions). Continuous variables were compared using either a two-sample Student’s t-test for normally distributed data or the Mann–Whitney U test for non-normally distributed data, and are summarized as median values (interquartile range [IQR]).

Potential prognostic factors associated with mortality were identified using univariate analyses. Variables with a P value < 0.2 in univariate analysis were included in the multivariate analysis. Backward stepwise selection was applied to refine the model and select the most relevant predictors. The goodness-of-fit of the final model was assessed using the Hosmer–Lemeshow test. Model discrimination was assessed by generating receiver operating characteristic curves, and predictive performance was validated using stratified 10-fold cross-validation.

Alpha diversity was evaluated using the Abundance-based Coverage Estimator and Shannon’s and Simpson’s diversity indices. Beta diversity was evaluated using the Bray–Curtis distance and visualized through principal-coordinate analysis. Taxonomic biomarkers were identified using linear discriminant analysis effect size (LEfSe) and the Kruskal–Wallis H Test. These analyses were conducted using the EzBioCloud 16s-based Microbiome Taxonomic Profiling platform.

Metabolomic data were normalized using the log10 fold-change for each metabolite. Differences between survivors and non-survivors were assessed using two-sample Student’s t-test. Correlations between stool metabolites and bacterial genera were analyzed using Spearman correlation. Partial least-squares discriminant analysis (PLS-DA) was performed to identify the stool metabolite signature of mortality, with a variable importance in projection (VIP) score threshold of 1.3.

Functional profiles were predicted from normalized taxonomic data using PICRUSt and MinPath algorithms. Differentially abundant functional pathways were identified using the Kruskal–Wallis H test and LEfSe, with statistical significance set at P < 0.05.

IBM SPSS Statistics version 20.0 (IBM Corporation, Armonk, NY, USA), SAS 9.4 (SAS Institute Inc., Cary, NC, USA), and R 4.4.1 with RStudio (v2024.04.2 + 764) (Te R Foundation for Statistical Computing, Vienna, Austria) were used for all statistical analyses. Two-sided P values < 0.05 were considered significant.

Patient characteristics

Fifty-nine patients were diagnosed with clinically significant candidemia during the study period. All the patients were hospitalized in the general ward (n = 20, 33.9%) or intensive care unit (n = 39, 66.1%). The median patient age was 70 years (IQR, 60–80), and 52.5% of them were male. The most common sources of candidemia were central venous catheter (n = 28, 47.5%) and the gastrointestinal tract (n = 20, 33.9%).

Candida albicans (36.4%) was the most frequently isolated species, followed by Candida tropicalis (34.5%), Candida parapsilosis (18.2%), and Candida glabrata (7.3%). No case of infection by multiple Candida species was recorded. First-line antifungal therapy included echinocandins (n = 56, 94.9%), fluconazole (n = 2, 3.4%), and amphotericin B (n = 1, 1.7%).

Of the 59 patients, 57 (96.6%) received antibiotics before stool specimen collection, and 21 (35.6%) had concomitant bacteremia. The overall in-hospital mortality rate was 40.7% (24/59), and the median time to death was 29.5 days (IQR, 16.8–38.3 days) after candidemia onset.

Candidemia and gut microbiota

Alpha-diversity indices of gut bacterial communities differed significantly between the survivors and non-survivors (Table 1), with the non-survivors showing lower diversity metrics (P-value < 0.05; Fig. 1A). However, β-diversity analysis revealed no significant difference between the two groups (Fig. 1B).

Table 1
Comparison of demographic and clinical characteristics between the survivors and the non-survivors of candidemia
Characteristics	Total (N = 59)	Survivors (N = 35)	Non-survivors (N = 24)	P-value
Alpha-diversity index, median (IQR)
ACE^a	56.4 (28.0–97.8)	73.0 (38.2–103.2)	35.9 (16.4–60.6)	0.044
Shannon’s diversity index	1.7 (1.2–2.7)	2.3 (1.4–2.8)	1.6 (1.0–2.0)	0.009
Simpson’s diversity index	0.3 (0.1–0.4)	0.2 (0.1–0.4)	0.3 (0.2–0.6)	0.007
Demographic variable
Age (yrs), median (IQR)	70 (60.0–80.0)	72 (64.0–81.0)	68.5 (59.0–77.8)	0.261
Sex (male), N (%)	31 (62.9)	20 (57.1)	11 (52.5)	0.393
Identified isolates from culture results, N (%)
Candida albicans	21 (35.6)	11 (31.4)	10 (41.7)	0.596
Candida tropicalis	25 (42.4)	15 (42.9)	10 (41.7)	1.000
Candida glabrata	13 (22.0)	9 (25.7)	4 (16.7)	0.614
Source of candidemia, N (%)
Central venous catheter	28 (47.5)	16 (45.7)	12 (50.0)	0.746
Gastrointestinal tract	20 (33.9)	12 (34.3)	8 (33.3)	0.939
Unknown	11 (18.6)	7 (20.0)	4 (16.7)	1.000
Comorbidities, N (%)
Cardiovascular diseases	27 (45.8)	14 (40.0)	13 (54.2)	0.303
Cerebrovascular diseases	13 (37.1)	3 (12.5)	16 (27.1)	0.036
Diabetes mellitus	16 (27.1)	10 (28.6)	6 (25.0)	0.762
Chronic kidney diseases	16 (27.1)	7 (20.0)	9 (37.5)	0.137
Chronic liver diseases	5 (8.5)	3 (8.6)	2 (8.3)	1.000
Chronic pulmonary diseases	5 (8.5)	3 (8.6)	2 (8.3)	1.000
Malignancy	18 (30.5)	8 (22.9)	10 (41.7)	0.123
CCI ≥ 3	34 (57.6)	19 (54.3)	15 (62.5)	0.531
Risk factor within 1 month, N (%)
Corticosteroids	18 (30.5)	9 (25.7)	9 (37.5)	0.334
Neutropenia^b	4 (6.8)	0 (0.0)	4 (16.7)	0.023
Prior surgery	17 (28.8)	10 (28.6)	7 (29.2)	0.960
Prior antibiotic exposure	57 (96.6)	33 (94.3)	24 (100.0)	0.509
Use of immunosuppressive agents other than corticosteroids	4 (6.8)	2 (5.7)	2 (8.3)	1.000
Clinical severity at worst, N (%)
Concomitant bacteremia	21 (35.6)	11 (31.4)	10 (41.7)	0.420
Renal replacement therapy	10 (16.9)	5 (14.3)	5 (20.8)	0.376
ICU admission	39 (66.1)	24 (68.6)	15 (62.5)	0.628
Mechanical ventilation	25 (42.4)	13 (37.1)	12 (50.0)	0.326
Pitt bacteremia score ≥ 2	40 (67.8)	25 (71.4)	15 (62.5)	0.471
Septic shock	40 (67.8)	20 (57.1)	20 (83.3)	0.034
Laboratory findings at time of Candidemia diagnosis, N (%)
CRP ≥ 100 mg/L	46 (78.0)	25 (71.4)	21 (87.5)	0.143
Procalcitonin ≥ 1.00 ng/mL	35 (59.3)	17 (48.6)	18 (75.0)	0.042
First-line antifungal agent, N (%)
Amphotericin B	1 (1.7)	0 (0.0)	1 (4.2)	0.848
Echinocandins	56 (94.9)	34 (97.1)	22 (91.7)	0.736
Fluconazole	2 (3.4)	1 (2.9)	1 (4.2)	1.000
Footnotes. ^aAbundance-based Coverage Estimator, ^b Neutropenia is defined as an absolute neutrophil count (ANC) of < 500 cells/µL.
CCI, Charlson’s Comorbidity index; CRP, C-reactive protein; ICU, intensive care unit; IQR, interquartile range; yrs, years.

A total of 1,636 unique operational taxonomic units were identified from sequenced specimens, highlighting gut microbial diversity in patients with candidemia (Fig. 1C). No significant difference in 16s rRNA quantification was observed between survivors and non-survivors (read counts: 2,478,962 vs. 1,730,093; P = 0.077). Firmicutes was the most abundant phylum in both groups; however, no significant differences were observed in the abundances of Firmicutes (P = 0.253) and Actinobacteria between the two groups (P = 0.666) (Table 2).

Table 2
Characteristics of the fecal microbiome stratified according to mortality status in patients with candidemia
	Total (N = 59)	Survivors (N = 35)	Non-survivors (N = 24)	P-value
Relative abundance (%)
Phylum level
Firmicutes	51.51	55.60	45.53	0.254
Proteobacteria	24.52	23.96	25.34	0.793
Bacteroidetes	16.02	16.85	1.81	0.459
Actinobacteria	4.46	2.23	7.70	0.666
Verrucomicrobia	3.25	1.12	6.34	0.522
Others	0.25	0.23	0.27	0.260
Species level
Ruminococcus torques	0.79	1.29	0.05	0.04
Clostridium innocuum group	0.62	0.82	0.34	0.02
Streptococcus salivarius group	0.59	0.67	0.47	0.03
Clostridium symbiosum	0.24	0.34	0.10	0.04
Clostridium scindens	0.13	0.21	0.02	0.01
GL520168_s	0.09	0.15	0.00	0.00
Bacteroides cellulosilyticus	0.09	0.14	0.01	0.02
AJ315979_s	0.07	0.11	0.00	0.02
Clostridium leptum	0.05	0.08	0.01	0.04
PAC001643_s	0.02	0.03	0.00	0.04
PAC001597_s	0.04	0.06	0.01	0.01
Proteus mirabilis	0.04	0.00	0.09	0.04
Acinetobacter pittii group	0.18	0.00	0.44	0.04
Bacteroides caccae	0.05	0.44	0.60	0.05
Metabolites
3-isopropoxy-hexamethyl-tetrasiloxane^a	346556.12	441107.44	187000.75	0.047
KKLGZUJIFDXBCT-UHFFFAOYSA-N^b	64361.19	98935.56	6016.938	0.235
Phenylacetic acid	20706.44	32976.93	0	0.181
Functional metabolic pathway, median (IQR)
Endoglucanase	0.035 (0.020–0.047)	0.036 (0.026–0.052)	0.032 (0.016–0.041)	0.040
Footnotes. ^a3-Isopropoxy-1,1,1,7,7,7-hexamethyl-3,5,5-tris(trimethylsiloxy)tetrasiloxane, ^bPentacarbonyl(phosphasila-boracyclohexene) tungsten; IQR, interquartile range

LEfSe analysis identified 14 bacterial species that were significantly different (LDA score > 2.0) between the two groups (Fig. 1D). Non-survivors showed increased abundance of Bacteroides caccae (0.60%), Acinetobacter pittii group (0.44%), and Proteus mirabilis (0.09%). The remaining 11 species were more enriched in survivors than in non-survivors: Clostridium genera (Firmicutes), Bacteroides cellulosilyticus (Bacteroidetes), and Ruminococcus torques (Bacillota) (Table 2).

Candidemia and host stool metabolomic profiles

A total of 111 compounds were included in the analysis. Metabolites within each group, including SCFAs, amino acids, cholic acids, and vitamins, were identified (Fig. 2A). The heatmaps of fecal metabolites derived from the survivors and non-survivors are depicted in Fig. 2A. PLS-DA plots of all metabolites did not completely distinguish the two groups (Supplementary Fig. 1). However, three metabolites with VIP scores > 1.3 were highlighted for their significant role in distinguishing survivors from non-survivors (Fig. 2B). Particularly, only one metabolite of 3-isopropoxy-hexamethyl-tetrasiloxane was significantly different between the two groups (P < 0.05) (Table 2).

Based on the taxonomic biomarkers, increased endoglucanase activity in the functional metabolic pathway was identified as a key metabolic pathway in survivors compared with non-survivors (Fig. 2C). Notably, Clostridium genera, one of the significant taxonomic biomarkers, exhibited significant positive correlations with the metabolic pathway of endoglucanase. However, none of the three metabolites with VIP scores > 1.3 were significantly associated with the significant taxonomic biomarkers (Fig. 3).

Predictors of in-hospital mortality in patients with candidemia

The demographic and clinical characteristics of the survivors and non-survivors are presented in Table 1. Cerebrovascular disease was more prevalent in non-survivors, whereas concomitant bacteremia and septic shock occurred at similar rates in the two groups. Although C-reactive protein level (≥ 100 mg/L) did not differ significantly between the two groups, procalcitonin level ≥ 1.00 ng/mL was more common in non-survivors than in survivors (Table 1).

Univariate analysis identified potential risk factors for in-hospital mortality (Supplementary Table 1). Despite its significance in the univariate analysis, Simpson’s diversity index was excluded from the multivariate analysis due to its high skewness and collinearity with Shannon’s diversity index (variance inflation factor = 9.831) (Supplementary Fig. 2). Backward stepwise selection was applied to refine the model and select the most relevant predictors. Variables with P values < 0.2 in the univariate analysis were included in the multivariate model, which identified three significant predictors of mortality: lower gut microbiota diversity index (Shannon’s diversity index; odds ratio [OR], 0.401; 95% confidence interval [CI], 0.191–0.844), underlying malignancy (OR, 7.794; 95% CI, 1.410–43.10), and septic shock (OR, 10.59; 95% CI, 1.701–65.97) (Table 3).

Table 3
Multivariate logistic regression analysis of prognostic factors associated with in-hospital mortality in patients with candidemia
	Multivariable logistic regression model			Multivariable logistic regression model refined using backward stepwise selection based on the Wald statistic
Variable	OR	95% CI	P-value	OR	95% CI	P-value
ACE^a	0.998	0.983–1.013	0.815
Shannon’s diversity index	0.469	0.181–1.218	0.120	0.401	0.191–0.844	0.016
Cerebrovascular diseases	0.329	0.062–1.755	0.193
Malignancy	4.769	0.729–31.21	0.103	7.794	1.410–43.10	0.019
Chronic kidney diseases	1.697	0.369–7.806	0.497
Septic shock	9.934	1.351–73.03	0.024	10.59	1.701–65.97	0.011
CRP ≥ 100 mg/L	2.161	0.369–12.66	0.393
Procalcitonin ≥ 1.000 ng/mL	2.234	0.528–9.451	0.275
Footnotes. The abovementioned covariates had P values < 0.2 in univariate logistic regression model and were included in the multivariable logistic regression model.
^aAbundance-based Coverage Estimator; CRP, C-reactive protein

The final model demonstrated a good fit (Hosmer–Lemeshow test: χ² = 7.013, P = 0.535) and strong discrimination (area under curve [AUC] = 0.805; 95% CI, 0.67–0.90) (Supplementary Fig. 3). Cross-validation supported model accuracy (10-fold log-loss = 0.606; AUC = 0.773), indicating robust calibration and discrimination.

In this study, the gut microbiota in non-survivors of candidemia was associated with lower α-diversity as indicated by Shannon’s diversity index. Notably, no significant differences in ß-diversity was observed between survivors and non-survivors. Furthermore, analysis of alterations in gut microbiota revealed significant differences in the composition of 11 species between the two groups. Specifically, Clostridium genera involved in endoglucanase activity in the functional metabolic pathway and the metabolite of 3-isopropoxy-hexamethyl-tetrasiloxane were significantly enriched in the survivors compared with the non-survivors. Clinical applications of gut microbiota have shown promise in predicting the prognoses of patients with candidemia.

Our results demonstrated that lower microbial α-diversity in patients with candidemia is associated with a higher risk of mortality, which was consistent with the results of previous in vivo studies. [10, 21] Decreased gut microbiota diversity and changes in metabolites are well documented as risk factors for several diseases, such as type 2 diabetes, inflammatory bowel disease, and metabolic syndrome. [22–24] Notably, gut microbiome disruption appears to be a known risk factor for sepsis and subsequent organ dysfunction, leading to worse outcomes. [25] Considering the critical roles of the microbiome in the training and development of the host immune system, these changes may compromise the mucosal barrier and immune function, worsening the prognoses of patients with candidemia. [21, 26] Consistent with previous findings, our study showed that underlying malignancy and septic shock are independent predictive factors associated with mortality in patients with candidemia. [27, 28]

Our analysis identified 14 causal bacterial taxa associated with candidemia, with increased abundance of three taxa that showed positive causal relationships with mortality. A previous study indicated that Acinetobacter pittii can transform deoxynivalenol, which is associated with inflammation and apoptosis, into less toxic or even non-toxic metabolites. [29] Over-abundance of Bacteroides caccae can lead to lysis of mucus, resulting in thinning of the mucosal barrier and enabling pathogen translocation. [30] Furthermore, lipopolysaccharides, a virulence factor of Proteus mirabilis, may be associated with pathological changes via gut leakage and inflammatory actions. [31] These results suggest the roles of these species in the pathogenesis of candidemia or worsening prognosis.

The remaining 11 taxa were significantly abundant in survivors compared with non-survivors. This finding may be because Ruminococcus torques degrades intestinal mucin, helping to regulate pH and maintain gut microbial diversity and stability. [32] In addition, Clostridium innocuum contributes to intestinal nutrient metabolism by breaking down glucose ureide and is detected in approximately 80% of healthy adults. [33] Furthermore, Streptococcus salivarius suppresses intestinal inflammation and regulates immune response. [34] In addition to the diverse roles of gut flora in patients with candidemia, these findings suggest the possibility of improving patient outcomes using microbiota-modulating therapy.

Interestingly, several Clostridium species and the metabolic pathway of endoglucanase, which showed relatively high abundance and activity, respectively, in survivors compared with non-survivors, showed a significant positive correlation in this study. The endoglucanases, which functionally hydrolyze glycosidic bonds, are enzymes found in microbiomes that break down cellulose. In addition to maintaining the balance of intestinal microbiota, the gut flora associated with the endoglucanase of the functional metabolic pathway contribute to the nutrition and health of the host through the production of SCFAs. [35] Considering this interplay between the gut flora and endoglucanase, strengthening important functional metabolic pathways using beneficial taxa of the gut microbiota may be useful in the prevention and treatment of candidemia.

Our current gut microbiome data provides further evidence that supports a potential contribution of gut microbiota to host fecal metabolic profiles. Although none of the three metabolites with VIP scores > 1.3 was significantly associated with gut taxonomic biomarkers, the levels of 3-isopropoxy-hexamethyl-tetrasiloxane detected in the feces of the survivors were significantly higher than those in the fecal samples of non-survivors. Previous studies have shown that 3-isopropoxy-hexamethyl-tetrasiloxane, a siloxane derivative, exhibits antifungal activity against Candida albicans, Candida krusei, and Candida glabrata. [36, 37] These findings suggest that 3-isopropoxy-hexamethyl-tetrasiloxane may play a role in suppressing pathogenic microorganisms and restoring the gut ecosystem.

Phenyl acetic acid and KKLGZUJIFDXBCT-UHFFFAOYSA-N (tungsten) with VIP scores > 1.3 were observed in higher levels in the stool of the survivors than in those of non-survivors; however, this finding was not statistically significant. Phenyl acetic acid exhibits antifungal activity against Candida albicans, and the metabolism of Clostridium species is involved in this process. [38, 39] Based on a recent study, the calcium tungstate microgel of a heavy metal may act as a microbe-based therapy against inflammatory bowel diseases by restoring gut microbiota homeostasis and improving the intestinal mucosal immune barrier. [40]

This study has some limitations. First, bias in data collection is possible. In addition, the statistical power of the study was possibly limited by its single-center design and small sample size. Moreover, the small sample size may have also affected the detection of significant differences in β-diversity. Larger multicenter cohort studies are needed to confirm these findings and provide more generalizable conclusions. Second, the cause–effect relationship of the changes observed in this study could not be determined. Third, as combined antimicrobial therapy plays a role in gut dysbiosis, co-infections with other pathogens or other treatments may have interfered with our findings. Future large randomized controlled trials are needed to comprehensively investigate the effect of gut microbiome profiling on the prevention and treatment of candidemia. In addition, in vivo models are warranted to decipher its pathogenesis and validate the findings of this study.

This study investigates the potential dysregulation function of the gut microbiota in patients with candidemia. In addition, it highlights the link between reduced gut microbiota diversity and higher mortality in patients with candidemia. Our findings suggest the alteration of gut microbiota in candidemia and its potential effects on host fecal metabolic profiles. Our study supports the potential therapeutic role of addressing disruptions in the gut microbiota in patients with candidemia.

Acknowledgments. None

Author Contributions. Conceptualization and design: Y.K.Y. Acquisition, analysis, and interpretation of data: Y.K.Y., S.H.P., and S.M.P. Statistical analysis: Y.K.Y., S.H.P., and S.M.P. Drafting of the manuscript: Y.K.Y., S.H.P., and S.M.P. Critical review and revision of the manuscript: All authors.

Data Availability Statement. Data supporting the findings of this study are available from the corresponding author upon reasonable request.

Funding. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2022R1A2C1010808). The Funding sources had no role in the study design, data collection, data analysis, decision to publish, or manuscript preparation.

Institutional Review Board Statement. This study was approved by the Institutional Review Board of Korea University Anam Hospital (IRB No. 2024AN0385).

Informed Consent Statement. The need for patient consent was waived owing to the nature of the study.

Conflicts of Interests. The authors declare no competing interests.

Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis. 2004;39:309–17.
Salmanton-García J, Cornely OA, Stemler J, et al. Attributable mortality of candidemia - Results from the ECMM Candida III multinational European Observational Cohort Study. J Infect. 2024;89:106229.
Thomas-Rüddel DO, Schlattmann P, Pletz M, Kurzai O, Bloos F. Risk Factors for Invasive Candida Infection in Critically Ill Patients: A Systematic Review and Meta-analysis. Chest. 2022;161:345–55.
Tamburini FB, Andermann TM, Tkachenko E, Senchyna F, Banaei N, Bhatt AS. Precision identification of diverse bloodstream pathogens in the gut microbiome. Nat Med. 2018;24:1809–14.
Mba IE, Nweze EI. Mechanism of Candida pathogenesis: revisiting the vital drivers. Eur J Clin Microbiol Infect Dis. 2020;39:1797–819.
Zhai B, Ola M, Rolling T, et al. High-resolution mycobiota analysis reveals dynamic intestinal translocation preceding invasive candidiasis. Nat Med. 2020;26:59–64.
Eichelberger KR, Paul S, Peters BM, Cassat JE. Candida-bacterial cross-kingdom interactions. Trends Microbiol. 2023;31:1287–99.
Eckstein MT, Moreno-Velásquez SD, Pérez JC. Gut Bacteria Shape Intestinal Microhabitats Occupied by the Fungus Candida albicans. Curr Biol 2020; 30:4799 – 807.e4.
Fan D, Coughlin LA, Neubauer MM, et al. Activation of HIF-1α and LL-37 by commensal bacteria inhibits Candida albicans colonization. Nat Med. 2015;21:808–14.
Hu W, Xu D, Zhou Z, Zhu J, Wang D, Tang J. Alterations in the gut microbiota and metabolic profiles coincide with intestinal damage in mice with a bloodborne Candida albicans infection. Microb Pathog. 2021;154:104826.
Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47:1245–51.
Kim SH, Yoon YK, Kim MJ, Sohn JW. Clinical impact of time to positivity for Candida species on mortality in patients with candidaemia. J Antimicrob Chemother. 2013;68:2890–7.
Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1–45.
Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control. 1988;16:128–40.
Mahapatra SHA. Septic Shock. StatPearls [Internet]: Treasure Island (FL): StatPearls Publishing; 2024 Jan. Available from: https://www.ncbi.nlm.nih.gov/books/NBK430939/
Nakada-Motokawa N, Miyazaki T, Ueda T, et al. Modified Pitt bacteremia score for predicting mortality in patients with candidaemia: A multicentre seven-year retrospective study conducted in Japan. Mycoses. 2021;64:1498–507.
Geng J, Sui Z, Dou W, et al. 16S rRNA Gene Sequencing Reveals Specific Gut Microbes Common to Medicinal Insects. Front Microbiol. 2022;13:892767.
Wheeler TJ, Eddy SR. nhmmer: DNA homology search with profile HMMs. Bioinformatics. 2013;29:2487–9.
Yoon SH, Ha SM, Kwon S, et al. Introducing EzBioCloud: a taxonomically united database of 16S rRNA gene sequences and whole-genome assemblies. Int J Syst Evol Microbiol. 2017;67:1613–7.
Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27:2194–200.
Bertolini M, Ranjan A, Thompson A, et al. Candida albicans induces mucosal bacterial dysbiosis that promotes invasive infection. PLoS Pathog. 2019;15:e1007717.
Ning L, Zhou YL, Sun H, et al. Microbiome and metabolome features in inflammatory bowel disease via multi-omics integration analyses across cohorts. Nat Commun. 2023;14:7135.
Menni C, Zhu J, Le Roy CI, et al. Serum metabolites reflecting gut microbiome alpha diversity predict type 2 diabetes. Gut Microbes. 2020;11:1632–42.
Pallister T, Jackson MA, Martin TC, et al. Hippurate as a metabolomic marker of gut microbiome diversity: Modulation by diet and relationship to metabolic syndrome. Sci Rep. 2017;7:13670.
Adelman MW, Woodworth MH, Langelier C, et al. The gut microbiome's role in the development, maintenance, and outcomes of sepsis. Crit Care. 2020;24:278.
Iliev ID, Leonardi I. Fungal dysbiosis: immunity and interactions at mucosal barriers. Nat Rev Immunol. 2017;17:635–46.
Haroun E, Kumar PA, Saba L, et al. Intestinal barrier functions in hematologic and oncologic diseases. J Transl Med. 2023;21:233.
Guzman JA, Tchokonte R, Sobel JD. Septic shock due to candidemia: outcomes and predictors of shock development. J Clin Med Res. 2011;3:65–71.
Liu Y, Xu L, Shi Z, et al. Identification of an Acinetobacter pittii acyltransferase involved in transformation of deoxynivalenol to 3-acetyl-deoxynivalenol by transcriptomic analysis. Ecotoxicol Environ Saf. 2023;263:115395.
Zafar H, Saier MH. Jr. Gut Bacteroides species in health and disease. Gut Microbes. 2021;13:1–20.
Choi JG, Kim N, Ju IG, et al. Oral administration of Proteus mirabilis damages dopaminergic neurons and motor functions in mice. Sci Rep. 2018;8:1275.
Tran NTD, Chaidee A, Surapinit A, et al. Chronic Strongyloides stercoralis infection increases presence of the Ruminococcus torques group in the gut and alters the microbial proteome. Sci Rep. 2023;13:4216.
Bhattacharjee D, Flores C, Woelfel-Monsivais C, Seekatz AM. Diversity and Prevalence of Clostridium innocuum in the Human Gut Microbiota. mSphere. 2023;8:e0056922.
Kaci G, Goudercourt D, Dennin V, et al. Anti-inflammatory properties of Streptococcus salivarius, a commensal bacterium of the oral cavity and digestive tract. Appl Environ Microbiol. 2014;80:928–34.
Froidurot A, Julliand V. Cellulolytic bacteria in the large intestine of mammals. Gut Microbes. 2022;14:2031694.
Omoruyi BE, Afolayan AJ, Bradley G. The inhibitory effect of Mesembryanthemum edule (L.) bolus essential oil on some pathogenic fungal isolates. BMC Complement Altern Med. 2014;14:168.
Barani K, Manipal S, Prabu D, Ahmed A, Adusumilli P, Jeevika C. Anti-fungal activity of Morinda citrifolia (noni) extracts against Candida albicans: an in vitro study. Indian J Dent Res. 2014;25:188–90.
Jiao M, He W, Ouyang Z, Shi Q, Wen Y. Progress in structural and functional study of the bacterial phenylacetic acid catabolic pathway, its role in pathogenicity and antibiotic resistance. Front Microbiol. 2022;13:964019.
Kim Y, Cho JY, Kuk JH, et al. Identification and antimicrobial activity of phenylacetic acid produced by Bacillus licheniformis isolated from fermented soybean, Chungkook-Jang. Curr Microbiol. 2004;48:312–7.
Yang J, Peng M, Tan S, et al. Calcium Tungstate Microgel Enhances the Delivery and Colonization of Probiotics during Colitis via Intestinal Ecological Niche Occupancy. ACS Cent Sci. 2023;9:1327–41.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
26 Sep, 2025
Reviews received at journal
28 Aug, 2025
Reviewers agreed at journal
24 Aug, 2025
Reviewers agreed at journal
21 Aug, 2025
Reviewers invited by journal
21 Aug, 2025
Editor assigned by journal
15 May, 2025
Submission checks completed at journal
30 Apr, 2025
First submitted to journal
28 Apr, 2025

You are reading this latest preprint version

Clinical impact of altered gut microbiota and metabolite profiles on mortality in patients with candidemia: A prospective observation pilot study

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials and Methods

Study design

Data collection

Clinical microbiologic analysis

Stool specimen collection and sequencing

Metabolomic profiling

Statistical Analyses

Results

Patient characteristics

Candidemia and gut microbiota

Candidemia and host stool metabolomic profiles

Predictors of in-hospital mortality in patients with candidemia

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1