Introduction

Microbial co-infections in hospitalized patients with COVID-19 have been documented in many investigations from the onset of the global pandemic in Wuhan. Co-infections in SARS-CoV-2 infected individuals were caused by a variety of pathogens, and it was determined that these concomitant infections resulted in disease aggravation and poor outcomes [1,2,3].

In patients with SARS-CoV-2 infection, predisposing conditions such as mechanical ventilation and ICU admission in critically ill patients, the use of broad-spectrum antibiotics, immunosuppressive/anti-inflammatory treatments, catheterization, and underlying diseases all increase the risk of secondary infections [4, 5]. Among fungal pathogens, opportunistic fungi such as Aspergillus and Candida species cause fungal co-infections in COVID-19 patients [6, 7]. Since species of the genus Candida are normal inhabitants of such various sites and internal organs as skin, mucous membranes, respiratory tract, digestive system, and urinary tract, they can give rise to infections due to the presence of favorable conditions in COVID-19 patients [6, 8].

Iran has been one of the countries most afflicted by the COVID-19 pandemic in the Middle East, with 7,553,169 confirmed cases and 144,502 deaths as of October 16, 2022 [9]. Early diagnosis and management of concomitant fungal infections in COVID-19 patients with effective antifungal agents leads to improved clinical outcomes. As a result, clinicians can benefit from a better understanding of the epidemiological evidence of co-infections in COVID-19 patients, such as risk factors, species distribution, and susceptibility profiles of isolates, to manage and control super-infections. In view of these considerations, this study was conducted to assess the risk factors as well as clinical outcomes of Candida co-infections in patients with COVID-19 admitted to Abu-Ali Sina hospital in Shiraz, during the COVID-19 pandemic. Furthermore, the species distribution and antifungal susceptibility profiles of isolates recovered from COVID-19 patients were determined in this study.

Materials and Methods

Study Design and Participants

For this observational and single-center study, all patients with confirmed COVID-19 admitted to Abu-Ali Sina hospital in Shiraz, one of the largest hospitals in the south of Iran, were evaluated for Candida co-infections from September to November 2021. This center has 600 individual hospital beds, and the number of beds was increased during the COVID-19 pandemic. SARS-CoV-2 was diagnosed based on positive Real-time Polymerase Chain Reaction (PCR) tests for SARS-CoV-2 or according to the clinical guidelines definition for COVID-19 [10]. All yeast isolates were recovered from clinical samples of COVID-19 patients (blood, urine, tracheal aspirate) during their hospital stay. Moreover, no clinical intervention in antifungal treatment was included in this study.

The patient data was extracted from the electronic medical record system of the hospital considering several variables, such as demographic characteristics, comorbidities, COVID-19 severity, and management such as antivirals, corticosteroids, antibacterial and immunomodulators, bacterial co-infections, empirical/definitive antifungal therapy, length of hospitalization, and length of ICU stay. In addition, relative laboratory results for the patients who participated in the study were collected. Moreover, risk factors associated with yeast co-infections and also clinical outcomes were investigated. Informed consent was obtained from all individual participants included in the study, which was approved by the ethics committee of the Shiraz University of Medical Sciences (IR.SUMS.REC.1401.210).

Fungal Culture and Isolation

To detect fungal strains, all clinical samples obtained from confirmed COVID-19 patients (blood, urine, tracheal aspirate) were cultured on Sabouraud Dextrose Agar (SDA, HiMedia, India), and HiCrome Candida Differential Agar (HiMedia,India) in the Department of Microbiology of Abu-Ali Sina Medical Center.

Species Identification by Sequencing the rDNA Region

DNA extraction of pure colonies was performed according to the method as previously described [11]. Definitive species identification of isolates were performed by PCR-sequencing method following amplification of ITS region [12].The ITS region sequence for each isolate was assembled and used for BLAST explores (http://blast.ncbi.nlm.nih.gov/Blast.cgi?). Sequence data of ITS region were deposited in NCBI and GenBank, and accession numbers of sequences are available for all isolates as follows: ON 312540-69, ON312571-99, ON479767-77, ON514607.

Antifungal Susceptibility Testing (AFST)

Minimum Inhibitory concentration (MIC) values were determined using broth microdilution method according to the M27-A3/S4 protocol documented by Clinical and Laboratory Standards Institute (CLSI) [13]. All clinical isolates were evaluated for susceptibility to antifungals including Fluconazole (FLZ, Sigma,USA), Posaconazole (PSZ,Sigma,Germany), Voriconazole (VRZ, Pfizer, New York, USA), Itraconazole (ITR,Sigma,USA), Caspofungin (CSP,sigma,USA), Nystatin (NYS, Sigma, Germany), and also amphotericin B AMB,Sigma,Germany). RPMI 1640 (Sigma, St. Louis, Missouri,USA) was prepared as directed by the manufacturer, and buffered to pH 7.0 using 0.165 N-morpholino propanesulfonic acid (MOPS) (Sigma, USA). Pure isolates were grown on SDA by incubation for 24 h at 35 °C. Following the growth of isolates, the inoculum suspensions were made by suspending the colonies in NaCl, and the turbidity adjusted to 0.5 McFarland at 530 nm.Then, prepared suspension was diluted to 0.5–2.5 103 cells/ml in RPMI 1640 media. Two-fold serial dilutions of antifungals were made in 96-well plates, and 100 µlit of yeast inoculum was added to each well with in equal volume. Finally, the plates were incubated for 24 h at 35 °C.

Interpretation of data was performed using clinical breakpoints that described in former CLSI documents [14, 15], and new editions of defined breakpoints [16, 17].When the clinical breakpoint was not available, the epidemiological cutoff value (ECV) was used. MIC was defined as the lowest concentration of antifungal agents that inhibits the growth of yeast isolates by 50% for CSP, FLZ, VRZ, ITR, and PSZ in comparison to the controls (drug-free wells). Moreover, the lowest concentration that resulted in any visible growth of isolates (100% inhibition) was considered as the MIC for NYS and AMB. The final concentrations of the antifungal agents were 0.032–16 μg/ml for AMB, ITR, PSZ, VRZ, and NYS, 0.125–64 μg/ml for FLZ, and 0.015–8 μg/ml for CSP. C. krusei (ATCC 6258) was included as a reference strain for quality control.

Statistical Analysis

Data were analyzed using SPSS statistical software version 18.00 (SPSS Inc.IBM.USA). Continuous variables are shown as median and interquartile range (IQR). Also, continuous data with normal distribution were expressed either as means ± standard deviation. Categorical variables were reported as numbers and percentages. Statistical differences between groups were analyzed using the exact Chi-square and/or Fisher’s exact test, and the χ2 test or Mann–Whitney U test. p-value < 0.05 was considered statistically significant.

To highlight the risk factors associated with Candida co-infections, univariable and multivariable logistic regression models were used. Factors with a p-value < 0.2 were retained for multivariate analysis, and those demonstrating statistical significance (p-value < 0.05) on multivariate analysis were considered verifiable risk factors.

Results

A total number of 106 patients were included in this study (51 patients with Candida spp. positive culture and 55 patients without Candida spp. positive culture). All the statistical data analyzed in this study is shown in Table 1.

Table 1 Demographic characteristics, clinical data, laboratory findings, treatments and outcomes of COVID-19 patients (N = 106)
Full size table

Data are median (IQR) or n (%), *p- values were calculated by Mann–Whitney U test, χ2 test, or Fisher's exact test, as appropriate. ICU:Intensive Care Unit, MV:Mechanical Ventilation, CVC: Central Venous Catheter, DM:Diabetese mellitus, HTN:Hypertension, ACS: Acute coronary syndrome, COPD: Chronic obstructive pulmonary disease, LDH: lactate dehydrogenase.

The mean age of total patients was 61 ± 16 years, ranging from 18 to 99 years. Of the total patients, 64 (60.4%) were male and 42 (39.6%) were female. Our results showed about 48% of hospitalized COVID-19 patients during this study period had urinary tract, respiratory system, and bloodstream co-infections with Candida species. According to statistical analysis of the data, there was a significant difference between patients with and without Candida spp. in terms of age (66 vs. 57 years; p-value: 0.003),whereas no significant difference was found in sex distribution between the two groups of patients. In patients with Candida positive culture, the median time from hospital admission to positive fungal culture was 6 days, ranging from 2 to 36 days (IQR: 4–11). As far as the time of staying in hospital in patients is concerned, there was not found a statistical difference in length of hospitalization between two groups (p-value: 0.9).

Compared to patients without Candida positive culture, a higher percentage of patients with Candida co-infections were admitted in the ICU ( 47.1% vs 27.3%; p-value:0.03), used mechanical ventilation (47.1% vs 25.5; p-value:0.02) and used urinary catheter (54.9% vs 30.9%; p-value:0.013). Also, the median time of urinary catheterization in patients with Candida spp. positive culture was 5 days, with a significant difference in comparison to patients without Candida spp. positive culture (p-value: 0.036).Although hypertension (HTN) and diabetes mellitus (DM) were the most common comorbidities among the study population, there was not a statistically significant association between the presence of Candida co-infection and HTN/DM in COVID-patients. In regards to other underlying conditions evaluated in this study, significant differences were observed in the presence of chronic obstructive pulmonary disease (COPD), heart failure, and solid organ transplantation. Among patients with COVID-19, those with Candida positive cultures showed a higher proportion of bacterial co-infection (60.8% vs. 32.7; p-value: 0.004). Moreover, in 24 of 45 patients with positive culture of Candida spp. from respiratory samples, isolation of bacterial species was also reported. Enterococcus and Acinetobacter spp. were the most commonly identified bacteria.

Regarding treatments, 93.4% of patients received antibiotics, 96.2% received high doses of corticosteroids, 82.1% received antiviral treatments, and 47.2% received antifungals as empirical therapy. Toclizumab, a recombinant anti-interleukin-6 receptor (IL-6R) monoclonal antibody, was prescribed for 19.8% of patients. Moreover, fluconazole was the most common antifungal agent used as empirical therapy in hospitalized patients with COVID-19 (31/50).

Antifungal treatment was administered to 34/51 patients with a Candida positive culture. The most commonly prescribed antifungal agents were as follows: fluconazole(12/34), combination therapy with caspofungin and fluconazole (6/34), liposomal amphotericin B (4/34), and nystatin (3/34). Although a higher proportion of in-hospital mortality was observed in patients with Candida positive cultures than in patients without Candida positive cultures, no significant difference in mortality rate was observed between the two groups (35.3% vs 20%; p-value = 0.07).

Regression analysis demonstrates that odds for Candida co-infection were higher in patients who admitted in ICU, had mechanical ventilatory supports, urinary catheterization, bacterial infection and patients with severe /critical diseases (Table 2). With regards to underlying conditions, HTN, COPD, and heart failure are associated with an increased risk of the development of Candida co-infection among patients with COVID-19. Moreover, patients who had severe /critical SARS-CoV-2 infection were at a greater risk for Candida co-infections.

Table 2 Risk factors associated with isolation of Candida spp. in hospitalized COVID-19 patients (N = 106)
Full size table

In univariable analysis, the odds of in-hospital mortality was higher in patients with positive Candida culture (RR: 2.18, 95% confidence interval: 0.9, 5.2). Moreover, a multiple logistical regression showed that heart failure, bacterial co-infection, and prescribed antifungals as empirical treatment were significant risk factors for Candida co-infection among hospitalized COVID-19 patients.

Overall, 67 episodes of Candida positive cultures were documented from different clinical samples (45 teracheal aspirate samples, 20 urine samples, and 2 blood samples) collected from 106 COVID-19 patients during the study period. With regards to the categorization of fungal infections according to European Organisation for Research and Treatment of Cancer and Mycoses Study Group (EORTC-MSG), two patients had proven invasive candidiasis [18]. Moreover, 45 patients were identified with Candida airway colonization and 20 patients with candiduria. Also, 4 samples (2 urine and 2 teracheal aspirate) presented mixed infection. Of the patients with Candida positive culture, 11 patients showed multiple episodes of Candida infections during their hospitalization. Totally, 71 Candida spp. were isolated from 51 COVID-19 patients with Candida positive culture as follows: C. albicans (n:49, 69%), C.glabrata (n:14, 19.7%), C. tropicalis (n: 7, 9.9%), and C. dubliniensis (n:1, 1.4%) (Table 3).

Table 3 Species distribution of Candida isolates from different clinical specimens
Full size table

The categorization of all Candida isolates in this study was based on obtained MICs according to recommended breakpoints by CLSI as shown in Table 4. Moreover, in-vitro antifungal susceptibility patterns of all isolates are shown in Table 5 in detail.

Table 4 MIC interpretation of tested antifungal drugs for Candida spp. recovered from COVID-19 patients
Full size table
Table 5 In-vitro antifungal susceptibility patterns of Candida spp. recovered from COVID-19 patients using micro broth dilution method
Full size table

All Candida isolates presented low MIC values for CSP (MIC range: 0.03–0.12) that classified as susceptible according to M-60 CLSI breakpoints. Resistant to itraconazole were observed with a high proportion among Candida species regarding breakpoints documented by M27-S3. According to the M59 CLSI document, all C. albicans isolates were wild-type (WT) for AMB. Among non-albicans Candida species, one C.glabrata isolate had a non-wild-type (NWT) phenotype for AMB. In regards to another polyene drug, nystatin, all Candida isolates were categorized as susceptible. Although no resistant isolate was found among all Candida species against posaconazole, few isolates were resistant to other azoles tested in this study. One C. albicans isolate was cross-resistant to VRZ (≥ 1 μg/ml) and ITZ (≥ 1 μg/ml), and also two C.albicans were cross-resistant toFLZ (≥ 8 μg/ml), VRZ (≥ 1 μg/ml) and ITZ (≥ 1 μg/ml). Cross-resistant species against FLZ (≥ 8 g/ml) and ITZ (≥ 1 g/ml) were also found in one C.albicans and one C.tropicalis isolate.

Discussion

The presence of fungal co-infections is considered an important complication in COVID-19 hospitalized patients by association with increased mortality [19,20,21]. COVID-19-Associated Candidiasis has been studied in different parts of the world with a range of 0.7–23.5% [4]. According to obtained data, almost half of hospitalized patients with COVID-19 have shown at least one positive culture of urinary, respiratory, or blood specimens for Candida species during their stay in hospital. To the best of our knowledge, this is the first comprehensive study on COVID-19-associated candidiasis (CAC) that included all clinical samples collected from patients with COVID-19 in Iran. It should be noted that the inclusion of both invasive and non-invasive candidiasis cases in this study led to a considerable difference in our reported rates compared to earlier studies.

As we know, Candida spp. are a part of the human microbiome that resides in different sites. So, dysregulation of the immune system following infection with SARS-CoV-2 leaves patients vulnerable to Candida superinfections. In general, previous evidence supports our findings that ICU-admission, mechanical ventilation, and urinary catheterization represent important risks for the development of candidiasis [4, 22,23,24,25,26,27,28].

Moreover, underlying conditions in COVID-19 patients, including DM, hypertension, and organ failure, in addition to a number of various antimicrobial and immunosuppressive medications, may be related to fungal co-infections [21, 29,30,31,32]. However, in the present work we identified HTN, COPD, hepatic and heart failures as risk factors for CAC that are in consistent with other previous studies in Iran and other parts of the world [25, 27, 28, 33].

Similar to other investigations, our results revealed that more severe conditions of pneumonia associated with SARS-Cov-2 infection increased the probability of fungal co-infections [25, 34]. Certainly, such therapeutic approaches in severe cases of COVID-19 patients as steroids, mechanical ventilation, broad-spectrum antimicrobial therapy, and ICU admission make them more susceptible to candidiasis, which has been previously reported in the Iranian population [27, 28, 35]. However, further studies are needed to investigate the possibility that Candida infection may increase the severity of the disease in patients with COVID-19.

Despite the fact that isolation of Candida spp. from respiratory samples occurs particularly in patients on mechanical ventilation, the link between this colonization and pneumonia is still controversial. Because the definitive diagnosis of Candida pneumonia is confirmed by examining the invasive form of the fungus in biopsy tissue, that is a limited diagnostic procedure. Since the lung is the main affected organ in COVID-19 patients, Candida colonization following immunosuppressive treatments and antibacterial medications can lead to overgrowth and morphogenesis of Candida spp. due to induced defects in cell-mediated immunity of the respiratory epithelium [36,37,38,39].

According to our research findings, patients with bacterial co-infection were at higher risk for Candida infections. Also, a high proportion of COVID-19 patients received antibacterial treatment, which is in agreement with the previous literature [5, 40]. Aligned to these results, it could be explained that overprescribing of broadspectrum antibiotics in COVID‐19 patients causes an imbalance of normal flora and overgrowth of Candida, which could result in Candida co-infections [41]. Furthermore, the coexistence of bacterial and fungal infections could contribute to increased pathogenicity, host damage, and inflammation [42]. Our findings showed that mortality among COVID-19 patients with Candida co-infection was higher in comparison to COVID-19 alone, which is in agreement with many previous studies [25, 34, 43,44,45].

Efficacy of empirical or preemptive antifungal therapy in high-risk patients has been studied in a number of researches [46, 47]. In nearly half of our research population, antifungal agents were provided as empirical therapy, despite the fact that there is no treatment guideline for fungal co-infections in COVID-19-positive individuals. Excessive use of empiric antifungal regimens not only has not been proven to be safe or beneficial but also could lead to resistance and therapeutic failures in patients who have previously been exposed, in addition to imposing an economic burden. Obviously, drug-drug interactions, toxicity, and adverse side effects of antifungals used in COVID-19 patients should all be considered [38].

Based on recent studies, C.albicans is known as the most prevalent species responsible for candidiasis in COVID-19 patients [48,49,50,51]. However, the identification of C. auris, a multi-drug resistant species, has been reported in previous studies with a high mortality rate [25]. In accordance with the findings presented in previous reports, C. albicans was the major species for candidiasis in COVID-19 patients in this study. Moreover, the species distribution of isolates from CAC with the highest proportion of C.albicans and C.glabrata is in accordance with previous reports from different countries[31, 34, 49, 51,52,53,54]. Although it is known that C.albicans accounts for half the isolates recovered from different types of Candida infections, it should be noted that non-albicans Candida species have emerged as prominent species in recent years [55]. Importantly, outbreaks of C.auris infections have been reported among COVID-19 patients in some regions of the world [25, 56, 57].

Although many reports have been published, there are a few studies about the susceptibility of fungal isolates from COVID-19 patients. Also, there have been few studies about the susceptibility patterns of Candida species isolated from non-invasive candidiasis. In total, resistance to azole drugs was observed among Candida species recovered from different samples in our study in spite of their high level of susceptibility to all tested antifungals. As reported from previous studies in Shiraz, the current study found azole-resistant Candida species among clinical isolates [58, 59]. Moreover, assessment of the azole susceptibility profiles of Candida isolates showed that C. albicans had cross-resistant profile to azoles. [38]Also, the resistant rate among C.glabrata was low in agreement with previous reports from Iran [60, 61]. Of note, itraconazole resistance among Candida species has been reported in previous studies[62,63,64,65]. Collectively, various degrees of resistance to azole antifungals have been declared previously in many reports from Iran [66, 67]. As a consequence, the presence of azole-resistant species could be a result of high background usage of azoles as empirical or first-line choice treatments and repeated exposure in our reviewed medical center.

Although posaconazole has no defined breakpoint in CLSI documents, it totally showed low MIC values for Candida species in the current study, which is consistent with the other studies [67, 68]. Considering the potent activity of posaconazole against Candida species, administration of this agent is proposed for the treatment of Candida infections. Additionally, the efficacy of posaconazole for treatment of mucormycosis, as a post-COVID-19 fungal infection, makes this azole antifungal an efficacious choice for treatment of other mycoses than yeast infections. In this survey, the most potent antifungal agent was caspofungin, with low MIC values for all species in accordance with prior research [32, 65, 69]. From the perspective of availability and cost-effectiveness, echinocandins appear to be an appropriate drug of first choice to define the most effective approach to the management of candidiasis in our country.

Some limitations must be acknowledged. In this study,we analyzed the available information for a small research population in one center that follow-up of patients after discharge was not included. The unavailability of new antifungals such as Isavuconazole in our country precludes the possibility of evaluating their efficacy in therapeutic procedures.

In conclusion, the results of our study declared some fundamental issues for patients infected with SARS-CoV-2 that had Candida co-infections. The main finding of this study reveals a high rate of Candida co-infections among hospitalized COVID-19 patients. In addition, comorbidities such as heart failure, HTN, COPD, bacterial infections as well as therapeutic interventions including catheterization, mechanical ventilation, and ICU admission increased the risk of infection by Candida spp. in COVID-19 patients. The isolation of Candida species from urinary/respiratory tracts and the bloodstream led to poor clinical outcomes, which presented as a higher in-hospital mortality rate. Also, obtained data clarified that empiric antifungal therapy did not achieve expected effectiveness. Finally, our results could assist infectious disease specialists with a better understanding of COVID-19-associated candidiasis. Definitely, selection of the appropriate antifungal therapies considering the species distribution and susceptibility patterns of causative agents can result in more effective therapeutic interventions and desired outcomes.