The incidence of Candida bloodstream infections (CBI) has increased dramatically in recent decades and Candida is now the most common etiology of fungal sepsis (Lamoth et al., 2018; Arastehfar et al., 2021). In the United States, Candida is the fourth most common cause of nosocomial bloodstream infections (BSI; Wisplinghoff et al., 2004). In Europe, Candida is the third most common etiology of fungal sepsis and is associated with a 37% 30-day mortality rate (Koehler et al., 2019). The Chinese Consensus on the Diagnosis and Management of Adult Candidiasis has observed a substantial increase in the incidence of invasive fungal disease (IFD), of which Candida is the primary cause (Gui-Qiang, 2020). Candida sepsis is associated with prolonged hospitalization times and increased costs, burdening both patients and the healthcare system (Hassan et al., 2009; Bloos et al., 2013; Heimann et al., 2015).

In Asian countries, non-albicans Candida spp. (NAC) is more prevalent than Candida albicans (Lortholary et al., 2017; Zhang et al., 2020; El Zakhem et al., 2021). Hematologic disorders are significant risk factors for Candida infection due to several factors including neutropenia, frequent use of immunomodulatory agents, prolonged use of broad-spectrum antimicrobials, central vascular catheters (CVC), chemotherapy and hematopoietic stem cell transplantation (HSCT; Kashefi et al., 2021). The highest percentage of Candida tropicalis among blood isolates was found in haemato-oncology wards (Tan et al., 2015). This fungus is associated with the highest rates of sepsis and the poorest prognosis of any Candida spp. (Ko et al., 2019). The emergence of antifungal-resistant C. tropicalis has also been widely reported in recent years (Zuza-Alves et al., 2017). Multiple studies have shown that almost half of Candidemia patients infected with C. tropicalis are azole-resistant (Chong et al., 2012; Chen et al., 2019; Arastehfar et al., 2020a,b).

The study of C. tropicalis bloodstream infections (CTBI) with hematological diseases still needs further exploration. This study describes the clinical features of CTBI in a hematology ward, evaluates the risk factors associated with mortality, and analyzes the antifungal susceptibility of this pathogen. Understanding CTBI-associated hematologic diseases are critical for the development of effective prevention and treatment options.

C. albicans is the primary cause of CBI, however, C. tropicalis is becoming increasingly more common in patients with hematologic disorders and HSCT (Tan et al., 2015). This species is shown to have higher virulence than C. albicans at the time of neutropenia or mucosal infection (Deorukhkar et al., 2014). The current study found that all C. tropicalis-infected patients were neutropenic and more than half had skin or mucosal infections (57.7%), indicating a likely association between these factors and CTBI.

A prior study found that infection with C. tropicalis was more severe and had worse clinical outcomes, including a 30-day patient mortality rate of 44.1%, than non-albicans Candidemia (Ko et al., 2019). The current study identified a CTBI-attributable mortality rate of 34.6%, with clinical symptoms that included fever, respiratory symptoms, digestive tract symptoms, skin or mucosal infections, and septic shock. Septic shock, one of the most serious BSI complications, is more often associated with C. tropicalis than C. albicans infection (Zuza-Alves et al., 2017). In the current study, septic shock occurred in 38.5% of patients. This symptom was an independent risk factor for CTBI-attributable death (p = 0.006). These findings highlight how critical it is to actively prevent and control shock among patients with neutropenia and fever.

CTBI has no characteristic clinical manifestations and diagnosis is primarily dependent on blood culture results. However, blood culture positivity rates are low, and early or bacterial bloodstream infections can be difficult to identify, contributing to high mortality rates. Previous studies have shown that a PCT value ≤2 ng/mL in a critically ill septic patient is more likely the result of Candidemia than bacteremia (Martini et al., 2010). Indeed, the current study found that the PCT value of most patients (73.7%) was ≤2 ng/mL. The PCT value ≥4 ng/mL is a risk factor for CTBT-associated death among patients with hematologic malignancies (Chao Zhi et al., 2016). The present findings showed that a PCT value ≥10 ng/mL was linked to a poor prognosis (p = 0.042). These data indicate that a low PCT value is a more likely indicator of Candidemia; however, poor prognosis is most often associated with a significant rise in PCT values among patients with hemopathy and CTBI.

A large-scale, multicenter clinical study (Sun et al., 2015) found that risk factors of IFD among chemotherapy patients with malignant hematologic disease included neutropenia, AML or MDS, non-CR patients who receipt of induction chemotherapy or repeat induction chemotherapy, decreased serum albumin levels, and CVC. All patients in the current study were neutropenic, 61.5% had AML, and 42.3% had albumin levels <30.0 g/L, which correlated with a poor prognosis (p = 0.006). Thus, actively correcting hypoproteinemia can both prevent IFD and improve CTBI-associated outcomes.

The Pitt Bacteremia Score is used to measure acute illness severity and predict the mortality of patients with BSI (Al-Hasan and Baddour, 2020). This score was used as a stratification tool in pivotal multicenter studies of Candida spp.-related BSI (Nguyen, 1995). Patients with a Pitt Bacteremia Score < 4 were considered noncritically ill, while those with a Pitt Bacteremia Score ≥ 4 have a higher mortality risk and are classified as critically ill (Al-Hasan and Baddour, 2020). The current study determined the Pitt Bacteremia Score using the clinical symptoms of patients, identifying eight with a Pitt Bacteremia Score ≥ 4. Univariate analysis linked Pitt Bacteremia Score ≥ 4 to a poor prognosis (p = 0.001).

It is rarely study about the infection of Candida spp. with other pathogens. This study (Medina et al., 2020) documented the presence of mixed yeast infections. It showed that 6.5% of C. tropicalis/C. glabrata were infected in 122 cases, and All Candida isolates were susceptible to amphotericin B. In our study, mixed infection was no significant correlation with prognosis. We find a patient had C. tropicalis co-infected with C. glabrata and another had C. tropicalis co-infected with C. albicans, and they had a good prognosis in antifungal.

Azole-resistant C. tropicalis infections have markedly increased in recent years. Multiple studies have shown that almost half of Candidemia patients infected with C. tropicalis are azole-resistant (Chong et al., 2012; Chen et al., 2019; Arastehfar et al., 2020a,b). The current study found that 41.7, 50, and 41.7% of the 24 C. tropicalis isolates were resistant to fluconazole, itraconazole, or voriconazole, respectively. Our finding supports those studies. The seven C. tropicalis-infected patients changed treatments or added echinocandin, amphotericin B, and flucytosine to their current treatment because the original regimens, azole alone, were not effective. These results indicated that a therapeutic schedule containing amphotericin B and drug combinations protected hemopathy patients from CTBI-associated death (p = 0.041). Drug sensitivity testing also showed that C. tropicalis was sensitive to amphotericin B. In addition, echinocandin antifungal drugs have a strong bactericidal effect on Candida spp., and there are only a few reports of C. tropicalis resistance. Both Chinese and American guidelines (Pappas et al., 2015; Xiaojun, 2020) indicate that echinocandins are the preferred treatment for Candidemia in neutropenic patients, followed by amphotericin B and lipid formulation. In the current study, echinocandins had no significant effect on prognosis. This may be explained by the high cost of these drugs, which caused some patients to delay treatment initiation. Development of azole resistance in C. tropicalis may occur through increased levels of the cellular target, upregulation of genes controlling drug efflux, alterations in sterol synthesis and decreased affinity of azoles for the cellular target (Lupetti et al., 2002). The resistance may be related to the ERG11, ERG3, MDR1, and CDR1 genes (Gaur et al., 2008; Prasad and Goffeau, 2012; Forastiero et al., 2013; Branco et al., 2017; Zhou et al., 2018). It is likely that more mechanisms for Candida-specific drug resistance will be identified. Guidelines (Pappas et al., 2015; Xiaojun, 2020) emphasizes that these are important for the development of antifungal prophylaxis and empiric treatment. The current study confirmed that the timing of antifungal treatment initiation is equally important. A time between fever and antifungal treatment of ≥5 days and a time between agranulocytosis and antifungal treatment of ≥10 days correlated with a poor prognosis (p = 0.011 and p = 0004, respectively).

This study has several limitations. It is a single-centered retrospective analysis and the number of cases is relatively small, making it difficult to apply the findings to other settings. Multicenter and large-scale clinical studies are required to further understand CTBI-associated hematologic disorders. In addition, ongoing studies of C. tropicalis-specific virulence factors and resistance genes will help to inform new strategies for clinical diagnosis and treatment.

This study identified a high mortality rate among patients with CTBI-associated hematologic disorders, which the attributable mortality rate was 34.6%. Septic shock is an independent risk factor for death attributed to CTBI. There were poor prognoses in the time from fever to antifungal treatment initiation ≥5 days and the time between neutropenia and antifungal treatment ≥10 days, so early antifungal therapy can significantly reduce the risk of death. The intermediate or resistance of C. tropicalis to azole drugs was close to 50%, and it was sensitive to flucytosine and amphotericin B. According to our study, the preferred agent is amphotericin B and drug combinations should be considered for severe infections.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Written informed consent was obtained from the individual(s), and minor(s)’ legal guardian/next of kin, for the publication of any potentially identifiable images or data included in this article.

WZ coordinated the study, initiated the project, and supervised. BY, ZW, and MW developed the protocol. BY and ZW analyzed the data and prepared the manuscript. WZ, MW, and YL critically revised the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.

This work was supported by the China Postdoctoral Science Foundation (2020M673097).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.