Fatal Lodderomyces elongisporus Fungemia in a Premature, Extremely Low-Birth-Weight Neonate

Many rare yeasts are emerging as pathogens, causing invasive infections in susceptible hosts that are associated with poor clinical outcome. Here, we describe the first and fatal case of Lodderomyces elongisporus fungemia in a premature, extremely low-birth-weight neonate after spontaneous vaginal delivery. The bloodstream isolate was identified as C. parapsilosis by the VITEK 2 yeast identification system and as L. elongisporus by PCR-sequencing of the internal transcribed spacer (ITS) region of ribosomal DNA. Antifungal susceptibility testing data for the isolate, performed by the broth microdilution-based MICRONAUT-AM assay, showed susceptibility to all nine antifungal drugs tested. Despite the initiation of treatment with liposomal amphotericin B, the patient died on the same day that the blood culture yielded yeast growth. This is the first report of L. elongisporus bloodstream infection in a neonate as the previous nine cases reported in the literature occurred in adult patients. The crude mortality rate for invasive L. elongisporus infection is 50%, as only 5 of 10 patients survived.

Lodderomyces elongisporus, an ascomycetous yeast, shares many phenotypic and physiological characteristics with the three species (viz. Candida parapsilosis sensu stricto, Candida orthopsilosis, and Candida metapsilosis) of the C. parapsilosis complex [16,17]. L. elongisporus forms turquoise color colonies, while C. parapsilosis complex members form pink or lavender colonies on chromogenic CHROMagar Candida medium [16,18]. Biochemical identification systems for yeasts such as VITEK 2 (bioMérieux) and API 20C (bioMérieux) also often misidentify them as C. parapsilosis as their databases do not include L. elongisporus. Specific identification is achieved only by PCR amplification or PCR amplification followed by DNA sequencing (PCR-sequencing) of the internal transcribed spacer (ITS) region and/or the D1/D2 domains of ribosomal (r)DNA or by the matrix-assisted laser desorption ionization time-of-flight mass spectroscopy (MALDI TOF-MS) [16][17][18][19].
Fungemia cases due to L. elongisporus are rare and are often related to immunosuppression or intravenous access devices [18][19][20][21]. Here, we describe the first case of L. elongisporus fungemia in an extremely low-birth-weight preterm neonate. The identity of the bloodstream isolate was confirmed by multiplex-PCR (mPCR) assay and PCR-sequencing of rDNA.

Case Report
An extremely low-birth-weight (ELBW, 710 g), preterm (23 weeks of gestational age), female neonate was born following spontaneous vaginal delivery. Antenatal complications included preterm labor, chorioamnionitis, and the administration of intrapartum antibiotics. The neonate was admitted to the neonatal intensive care unit (NICU) with ELBW, hyaline membrane disease and was managed medically with high-frequency oscillatory ventilation. She was cannulated with an umbilical arterial catheter and an umbilical venous catheter using a vacuum-assisted closure technique and was started on total parenteral nutrition together with empiric antibiotic treatment with ampicillin and amikacin planned for the next seven days. The septic screen performed on the day of admission showed no signs of sepsis. On Day 3, her abdominal ultrasound findings were normal, however, a cranial ultrasound indicated a tiny left choroid plexus cyst (1.8 mm). On Day 7, her follow-up ultrasound investigations revealed intraventricular bleeding with dilated lateral ventricles as well as the third ventricle. On Day 9, a complete septic workup was conducted again due to development of fever, including a blood culture. Antibiotic treatment was started with teicoplanin and amikacin empirically. However, with progressive deterioration, the antibiotic treatment was changed to tazocin and liposomal amphotericin B was also added to the regimen to cover for fungal infections. Despite the inclusion of an antifungal cover, the neonate expired on Day 9 due to sepsis. Subsequently, the blood sample retrieved on Day 9 yielded yeast growth.
The in vitro antifungal drug susceptibility testing (AST) of the L. elongisporus isolate Kw754/21 together with four other previous strains (Kw2486/06, Kw553/08, Kw3074/14 and Kw553/18) from Kuwait and the reference strain (CBS2605) was performed by the colorimetric microdilution-based MICRONAUT-AM AST panel for yeasts (Merlin Diagnostica GmbH, Bornheim, Germany). The minimum inhibitory concentration (MIC) values were read after 24 h of incubation at 37 • C and the data were interpreted by following the manufacturer's instructions and as described previously [24]. Although there are no clinical breakpoints for L. elongisporus, our isolate (Kw754/21) appeared susceptible to all nine antifungal drugs including four triazoles (fluconazole, voriconazole, itraconazole and posaconazole), three echinocandins (anidulafungin, micafungin and caspofungin) a polyene (amphotericin B) and a nucleotide analog (5-fluorocytosine) as the MIC values were considerably below the plasma levels normally achieved for these drugs (Table 1) [25].

Comments
Uncommon Candida species and other rare yeasts are increasingly being recognized as opportunistic pathogens. Rare yeasts species are generally resistant to one or more antifungal drugs [2,26]. These yeast species with reduced susceptibility to antifungal drugs exploit antifungal selection pressure during prophylactic/empiric treatment [24,26,27]. The increasing prevalence of these rare yeast species is attributable to many reasons, which include prolonged survival of critically ill ICU patients, extended use of broad-spectrum antibiotics, and the increasing use of intravascular catheters [27]. In addition, delays in identifying and treating uncommon yeast infections also contribute towards higher fatality rates associated with rare yeast infections [27][28][29]. As a result of the severe underlying conditions of the infected individuals, Candida infections are associated with a high crude mortality rate of 46 to 75% [30,31].
The actual incidence of invasive infections caused by L. elongisporus remains underreported as this species is frequently misidentified in clinical mycology laboratories as C. parapsilosis by phenotypic methods, such as the automated yeast identification systems like Vitek 2 or API 20C [18,19]. In the present study, we describe the first and fatal case of fungemia caused by L. elongisporus in a premature, extremely low-birth-weight neonate in Kuwait. The neonate deteriorated clinically rapidly after the onset of infection, likely due to prematurity and died due to signs and symptoms of sepsis despite initiation of treatment with liposomal amphotericin B soon after recognition of the signs of infection. A review of the literature has identified nine previous episodes of invasive infections due to L. elongisporus and all these cases occurred in adult (age range of 22 to 79 years) patients. The clinical details of these cases together with our case and outcome are presented in Table 2. The patients had several underlying conditions, indicating the debilitating nature of their illnesses before the onset of fungemia. Five of ten patients died, despite treatment with one or more antifungal drugs, which is comparable to the crude mortality rate of Candida infections ( Table 2). Three of five patients who succumbed to their infection were treated with echinocandin. Members of the C. parapsilosis complex, to which L. elongisporus is closely related, often show reduced in vitro susceptibility to echinocandins [17]. Current Infectious Diseases Society of America (IDSA) guidelines recommend using echinocandins to treat candidaemia caused by C. parapsilosis as echinocandin use does not negatively affects the outcome [34]. This is also reflected among fungemia cases due to L. elongisporus, and all five patients who survived the infection had also received echinocandins as the main drug or as part of their antifungal treatment regimen. This is also consistent with the AST data as all five clinical L. elongisporus isolates from Kuwait exhibited very low MICs for all three echinocandins and appeared susceptible to these drugs [18,19].
L. elongisporus was identified for the first time in 2008 as the causative agent of bloodstream infections [16]. In subsequent reports from other parts of the world, its identification in fungemia patients was reported (Table 2). Although C. parapsilosis isolates are common causes of fungemia in neonates, surprisingly, no previous invasive L. elongisporus cases in neonates have been described. This might be due to the fact that this yeast is still an uncommon pathogen in many geographical regions and/or it is misidentified as C. parapsilosis as it shares many phenotypic and physiological characteristics with members of the C. parapsilosis complex [16,17] and DNA sequencing or other molecular facilities are not readily available in many clinical mycology laboratories.

Conclusions
The first case of fungemia due to L. elongisporus in a premature, extremely low-birthweight neonate is presented. Major risk factors for invasive fungal infections in our neonate included hyaline membrane disease, extremely low-birth-weight, and prior treatment with broad-spectrum antibiotics. A bloodstream yeast isolate was cultured, which was initially misidentified as C. parapsilosis by VITEK 2. The correct identification of the isolate as L. elongisporus was obtained by a multiplex PCR and PCR-sequencing of rDNA. Despite the initiation of treatment with liposomal amphotericin B, the neonate died on the same day when the fungemia began. Funding: No specific grant was received for this study from any funding agency in the public, commercial or not-for-profit organization.

Institutional Review Board Statement:
The study was approved by the Ethics Committee, Health Sciences Center, Kuwait University, Kuwait (approval no. VDR/EC/3724 dated 31 January 2021).

Informed Consent Statement:
The need for informed consent was waived by the Ethics Committee, Health Sciences Center, Kuwait University, Kuwait as the clinical samples were obtained as part of routine patient care and the results are reported on deidentified samples without revealing patient identity.
Data Availability Statement: All relevant data are available in the manuscript.