Candida lipolytica Bloodstream Infection in an Adult Patient with COVID-19 and Alcohol Use Disorder: A Unique Case and a Systematic Review of the Literature

Candida lipolytica is an uncommon Candida species causing invasive fungemia. This yeast is mainly associated with the colonisation of intravascular catheters, complicated intra-abdominal infections, and infections in the paediatric population. Here, we report a case of C. lipolytica bloodstream infection in a 53-year-old man. He was admitted for an alcohol withdrawal syndrome and mild COVID-19. Among the primary risk factors for candidemia, only the use of broad-spectrum antimicrobials was reported. The empiric treatment was commenced with caspofungin and then targeted with intravenous fluconazole. Infective endocarditis was ruled out using echocardiography, and PET/TC was negative for other deep-seated foci of fungal infection. The patient was discharged after blood culture clearance and clinical healing. To the best of our knowledge, this is the first case of C. lipolytica candidemia in a patient with COVID-19 and alcohol use disorder. We performed a systematic review of bloodstream infections caused by C. lipolytica. Clinicians should be aware of the possibility of C. lipolytica bloodstream infections in patients with alcohol use disorder, especially in a COVID-19 setting.


Introduction
Candida species are common pathogens causing nosocomial bloodstream infections (BSIs) worldwide and account for more than 90% of fungal BSIs [1]. The incidence of nosocomial Candida BSIs is quite variable, ranging between 0.3 and 5 per 1000 admissions [2]. There are several well-known risk factors for candidemia, including central venous catheterisation, parenteral hyperalimentation, broad-spectrum antibiotics, intensive care, malignancies, haematological aberrations, and immunocompromised conditions [3].
Historically, Candida albicans was the major pathogen of candidemia; nevertheless, in recent years, non-albicans Candida species have been responsible for up to 50% of all candidemia cases in some settings [4].
The incidence of Candida spp. BSIs in patients with COVID-19 syndrome is significantly higher than in patients without this syndrome. Various explanations have been suggested, such as poor central venous catheter (CVC) care, immunosuppression (e.g., the administration of tocilizumab or high doses of corticosteroids), and the increased use of antibiotics and their effects on the gut microbiome [5]. Furthermore, both viral infection and microthrombi formation can alter the gut-blood barrier, resulting in intestinal microbiota entering the blood [6].
Candida lipolytica (Yarrowia lipolytica) is a ubiquitous ascomycetous yeast growing in the environment, meat, and cheese products. It can occasionally colonise the gut and faeces, oropharynx, and the skin of asymptomatic persons [7]. C. lipolytica infection in humans was firstly reported by Wehrspann and Füllbrandt in 1985 [8]. Although this fungus was previously considered a low-virulence yeast, increasing episodes of nosocomial infections (i.e., catheter-related BSIs) in immunocompromised or critically ill patients have been recently reported. One hypothesis about the invasiveness of C. lipolytica, in addition to the production of proteases and lipases, is the ability to form biofilms [7]. Invasive C. lipolytica infections, except for catheter-related C. lipolytica infections, occurred in the context of traumatic ocular infection, an acute exacerbation of chronic sinusitis and acute pancreatitis [3,7]. Here, we present a unique case of C. lipolytica BSI in a patient affected by alcohol use disorder and concomitant COVID-19. Furthermore, we performed a systematic literature review of candidemia episodes due to C. lipolytica.

Materials and Methods
This systematic review was performed according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) [9]. This systematic review is registered with PROSPERO (CRD42023405326).
The One investigator (V.Z.) carried out the first selection of retrieved records by screening their titles and abstracts in order to establish eligibility for a full-text review. The second step (performed by V.Z., S.S. and L.C.) consisted of the further screening of full-text articles to define final inclusion in the systematic review according to the inclusion criteria. We included full texts (written in English, Spanish and German) of case reports, case series and systematic reviews about Candida lipolytica BSI. We excluded papers containing only microbiological data (e.g., susceptibility and genomics). Additional cases were sought from the reference list of included papers and reviews.
The following information was extracted from each article and entered into pilot-tested evidence tables: author, year, the country of diagnosis, age, gender, the origin of infection, underlying conditions and risk factors (immunodeficiency, parenteral nutrition and previous abdominal surgery), source control, clinical presentation (septic shock, coinfections and complications), susceptibility to main antifungals, antifungal therapy and outcomes.

Case Presentation
On 3 June 2022, a 53-year-old man entered the emergency room for alcohol withdrawal syndrome associated with chronic alcoholism. He was a regular smoker, and his past medical history was unremarkable, except for reported episodes of unspecified haematemesis. He was further hospitalised for acute care because of withdrawal symptoms.
On admission to a medical ward, he was found to have a right submandibular tumour. Routine laboratory tests were unremarkable, and a screening nasal swab was negative for SARS-CoV-2. A computed tomography (CT) scan of the neck and head revealed a productive submandibular lesion with no homogeneous enhancement. The patient underwent an ultrasound-guided needle biopsy of the right parotid gland, and its histology was consistent with possible sialocele or adenoma.
The above-mentioned CT examination also described hypodense material in all mastoid cells and in the right tympanic cavity with a retracted tympanic membrane, as well as a suspected focal brain lesion. Magnetic resonance imaging (MRI) of the brain demonstrated both cholesteatomatous otitis in the right tympanic cavity and a cerebral cavernoma. The patient received empirical antibiotic therapy with piperacillin/tazobactam from 13 to 22 June. On 30 June, the patient was found to be positive for SARS-CoV-2 at a screening using a nasal swab and then transferred to the COVID-19 department. Then, mild febrile COVID-19 was diagnosed, and another course of piperacillin/tazobactam was prescribed.
Blood cultures were found to be positive for C. lipolytica using Matrix-Assisted Laser Desorption Ionization Time-of-Flight (MALDI-TOF) technology (Figure 1). At that time and following patient admission, no CVC was present. The abnormal blood test results were as follows: C-reactive protein 98.1 mg/L (reference value < 5 mg/L) and procalcitonin 0.73 µg/L (reference value < 0.5 µg/L). Caspofungin was started pending sensitivity tests and the BSI clearance. A transthoracic echocardiography was negative for endocarditis, and a funduscopic examination ruled out retinal embolisms. In addition, positron emission tomography (PET/CT) ruled out other deep-seated foci of fungal infection.

Review
The literature search identified 17 articles about cases of Candida lipolytica bloodstream infections ( Figure 2). After 72 h of echinocandin treatment, the patient was afebrile, and three blood culture sets were negative. Thus, the patient received caspofungin up to 15 July and subsequently concluded a total of 14 days of targeted therapy with fluconazole since bloodstream clearance.
In summary, we report a BSI episode due to C. lipolytica of unknown origin in a patient with alcohol use disorder and concomitant mild COVID-19.
The mean age of the affected patients was 41.48 years old (+/−23.86). A total of 76% of the patients were males (n = 68). Candidemia was catheter-related in 94% of cases (84 out of 89 subjects). Regarding other predisposing risk factors, parenteral nutrition was described in 25 patients (28%), while previous abdominal surgery and immunodeficiency were described in 20 and 14 patients (23% and 16%), respectively.
Candidemia occurred with septic shock in 16 patients (18%). Sixty-nine patients (78%) received antifungal therapy. In 15 cases (17%), antifungal combination therapy was prescribed. Of these, 87% of patients received targeted therapy with amphotericin B and fluconazole, while 13% received amphotericin B and caspofungin [12,23]. In total, 5 out of 15 (67 %) of the patients who received antifungal combination therapy died. Amphotericin B and fluconazole were the most prescribed drugs. The resolution of the candidemia was reported in 53 patients (60%). For 11 surviving patients, the treatment only consisted of source control with CVC removal. Death occurred in 34 cases (38%). Source control was not carried out in almost one-third of the patients who died (12 patients out of 34).

Discussion
The majority of patients with C. lipolytica BSI described in the literature are represented by adult males, characteristics consistent with our experience. Furthermore, the presentation of symptoms described herein was mild, in line with previous reports where septic shock was rare. Interestingly, the case we described was a C. lipolytica candidemia not related to the presence of an intravascular device. As a matter of fact, the source of the candidemia was never detected. This is in contrast with the results of our literature review, which shows a high burden of C. lipolytica fungemia related to CVC. This yeast is ubiquitous both in hospital environments and at the community level. Nevertheless, because of its selective ability to adhere and form a biofilm on medical devices, the majority of invasive infections encountered are of nosocomial origin [7,24].
Our case is a unique report of C. lipolytica BSI in a patient suffering from COVID-19 without indwelling catheters as a risk factor for candidemia; however, the patient had a history of alcohol use disorder (AUD). Such behaviour profoundly alters the gut microbiome, increases intestinal permeability, causes gut dysfunction, induces bacterial translocation and exacerbates the process of alcohol-associated liver disease (ALD). Furthermore, ethanol abuse decreases the prevalence of Epicoccum, Galactomyces and Debaryomyces in the gut, while Candida spp. burden increases significantly [25]. Indeed, AUD has been previously described to be a risk factor for C. lipolytica BSI, as shown by two reports included in our analysis [8,22].
In addition, our patient underwent two courses of therapy with a broad-spectrum antibiotic (piperacillin/tazobactam). Antibacterial drugs have a long-term effect on the microbiome of the human gut by shifting fungal communities from mutualism to competition and reducing the abundance of bacteria that actively suppress the pathogenicity of opportunistic fungi, such as Candida spp. Nevertheless, piperacillin/tazobactam does not seem to be associated with invasive fungal infections like other classes of antibiotics, such as fluoroquinolones [26,27].
The risk factors and incidence of invasive candidiasis in patients with COVID-19 are in the progress of being defined. Indeed, the frequency of this fungal infection ranges from 0.03 to 14% because of the heterogeneity of patients and cohorts of study [28]. The faecal microbiome was also studied in patients hospitalised for COVID-19, and an intestinal accumulation of fungal pathogens belonging to the genera Candida and Aspergillus was found compared to controls [29]. Other authors showed that SARS-CoV-2 intestinal mu-cosa damage and malnutrition correlate with secondary infection, such as bloodstream infections [6]. Thus, our patient, who appeared to have no primary risk factor for invasive candidemia, had really three underlying conditions favouring intestinal C. lipolytica translocation, namely, AUD, COVID-19, and broad-spectrum antibiotic therapy.
There is a scarcity of data on C. lipolytica BSI outcomes. Our analysis revealed a high mortality rate in affected patients. Nevertheless, up to 12% of the included patients survived without any antimycotic therapy. As the management of this rare infection is not standardised, we decided to follow the available guidelines and prescribe a minimum duration of therapy of 2 weeks after the documented clearance of Candida from the bloodstream [1]. International guidelines for the treatment of candidiasis generally do not include combination therapy, except in certain clinical cases, such as endocarditis. Nevertheless, some combination therapies seem to have a synergistic effect against difficult-to-treat Candida species by preventing or reducing biofilm formation [30]. Although C. albicans remains the most pathogenic yeast, the selective abilities of C. lipolytica to form biofilms on devices and produce haemolytic enzymes are of particular interest in a nosocomial setting [24]. The role of antimycotic combination therapy needs to be further studied. The patient here described recovered after prolonged antifungal monotherapy and was discharged from our institution in good clinical condition.

Conclusions
We described a unique case of C. lipolytica BSI in a patient with AUD, COVID-19, and antibiotic therapy, which all represent conditions favouring the intestinal translocation of Candida spp. We also conducted a systematic review of previously published cases of C. lipolytica candidemia. This study reinforces the available data on the specific risk factors for such an invasive fungal infection and contributes with a personal perspective on its management.