Candida auris Infection, a Rapidly Emerging Threat in the Neonatal Intensive Care Units: A Systematic Review

(1) Background: In recent years, a global epidemiological shift in candidemia has been observed, marked by the emergence of resistant non-albicans Candida species. Candida auris, in particular, has become a significant global concern, causing infections in both pediatric and adult populations within healthcare settings. Despite its widespread impact, there is a limited understanding of the clinical course and transmission dynamics of neonatal systemic Candida auris infections, hindering effective prevention and management. This study focused on the epidemiologic data, the clinical presentation, risk factors, and outcome of C. auris infection in neonatal population. (2) Methods: A systematic review of the literature using PubMed and Scopus databases until December 2023 was conducted. (3) Results: A total of 24 relevant studies were identified, encompassing 476 documented cases of Candida auris infection in neonates. Prematurity emerged as a primary risk factor, alongside total parenteral nutrition, central line insertion, mechanical ventilation, and prior broad-spectrum antibiotic use. The mortality rate reached approximately 42%, with therapeutic details sparingly reported in 12% of cases. Treatment strategies varied, with amphotericin B predominantly used as monotherapy, while combination antifungal agents were used in 44% of cases. Notably, 97.4% of cases exhibited fluconazole resistance, and 67.1% showed resistance to amphotericin B. Limited data were available on resistance to other antifungal agents. (4) Conclusions: Despite the rarity of neonatal Candida auris infections, their global occurrence necessitates comprehensive preparedness in patient care. A deeper understanding of Candida auris pathogenesis is crucial for developing effective strategies to control and prevent neonatal infections caused by this pathogen.


Introduction
Candida (C.) auris is a fungal pathogen, considered to be an emerging global health threat due to its rapid spread across the world and the emergence of strains resistant to multiple antifungal agents [1,2].It is associated with hospital-acquired and endemic nosocomial infections [2,3].There are two ways in which C. auris can affect the human body.It can either reside on some areas of the body, such as the skin, rectum, or mouth, infection in pediatric patients, with high mortality and hospital costs.Candida albicans and Candida parapsilosis are the most frequently isolated types of fungi in neonatal systemic candidiasis [21,22]; however, over the last few years, there has been a rise in cases of systematic C. auris infection reported in neonatal patients in NICUs worldwide.These outbreaks have a poor prognosis, despite antifungal therapy, as C. auris infection is often lethal in neonates [23][24][25].The incidence of invasive candidiasis cases in NICUs has risen during the last few years due to advances in healthcare.Many factors are involved in this rise, including invasive procedures, such as catheterizations of central veins and arteries, gastrointestinal tract invasive procedures, and the use of broad-spectrum antimicrobial therapies in the context of neonatal sepsis [23].These infections also complicate the clinical course of term neonates with underlying health problems that are hospitalized in the NICU [26].There are various international guidelines on the management, therapy, and prevention of invasive candidiasis in neonatal patients, most of which are based on what is known about neonatal C. albicans and C. parapsilosis infections [27][28][29].However, immune responses, clinical course, and the outcome of invasive candidiasis, in combination with antifungal prescription guidelines vary among Candida species [19,30,31].Furthermore, there is limited information about the clinical course of systemic C. auris infection and its spread, which undermines the sufficient prevention and management of fungemia in neonatal patients [32,33].Hence, there is a clinical need to understand the pathogenesis of C. auris infection during the neonatal period as well as to characterize the interactions between neonates and C. auris.This is the reason we conducted a systematic review in order to collect and update our knowledge and practices concerning the risk factors, clinical presentation, and outcomes of Candida auris infection in neonates.

Materials and Methods
We followed the methodology of a systematic review to identify, evaluate, and interpret the available research studies that addressed our research objective.A protocol was prepared, following the Preferred Reported Items for Systematic Reviews and Meta-analysis (PRISMA, presented as a Supplementary Material) guidelines [34], which is registered in the PROSPERO database (CRD42023417031) and used in this systematic review.

•
Randomized controlled trials (RCTs), observational studies, case reports and case series; • Population: neonates; • Study design: studies focused on Candida auris infection, defined as positive blood, urine, CSF, or other sterile anatomical site culture in neonates.

Exclusion Criteria
• Surveys that included children and adults in the same group with neonates and did not offer information that solely involved the neonatal population;

•
Review articles, systematic reviews and meta-analyses, and conference proceedings were excluded;

•
Studies not published in the English language.

Research Objective
To obtain information on the clinical presentation, risk factors, and outcomes of Candida auris infection in neonates.
Risk factors/characteristics of the affected population; 3.
Outcomes of the infection including mortality.

Search Strategy-Data Source
This systematic review was performed from May 2023 to December 2023.We used the PubMed and Scopus databases, with the research end date set at 31 December 2023.
The used combination of key words: "candida auris", "neonate", "newborn", "premature", and "preterm" with logical operators Boolean (AND, OR).Moreover, in order to cover the full expansion of the available literature and reduce the risk of losing studies, we manually searched and reviewed the references of each chosen study, individually, as well as the references of previous systematic reviews on the topic.

Resolving Conflicts
The data extraction and quality assessment were conducted independently by two researchers (P.K.T., A.E.P.), with conflicts being resolved through discussion and consensus between the researchers or, if needed, with the involvement of a third researcher (R.S.).

Data Synthesis and Presentation
We recorded the data on a table according to the day of life that the infection occurred, existing comorbidities, risk factors, prematurity, administration of total parenteral nutrition, placement of central catheters, mechanical ventilation and previous use of broad-spectrum antibiotics, number of participants, number of events, and other relevant criteria for the classification of the study population (subpopulations of neonates include preterm infants, very low birth weight infants, those undergoing cardiac or other surgical procedures, infants with congenital malformations, and so on) research study design, date of publication, regardless whether there was chronological ambiguity or not, number of participants, and other relative criteria in order to collect and, if possible, analyze the results of the study.We described potential gaps in the data and offered suggestions for future investigation.

Results
A total of 125 studies were retrieved from the research of the electronic literature databases; 42 were duplicate entries and were removed.The removal of duplicate studies was achieved with the help of a management tool, which is used to remove duplicate studies, installed in the digital literature reference program (EndNote X8).After meticulously studying the titles and abstracts of the remaining studies, 19 of them were excluded either due to the fact that their subject did not serve the purpose of the study or because they met some of the exclusion criteria that was already evident from their title and abstract, while 12 reviews concerned the same field of inquiry but were not focused on the neonatal population.A careful study of the entire text of the remaining 64 studies revealed that only 24 of them met all the inclusion criteria and were therefore included in this systematic review.The flowchart is presented in Figure 1.
Of all the studies included in the present review, three concerned public health surveillance programs, four were laboratory-based surveillance series, five were case series, three were cross-sectional studies, and the remaining nine were cohort studies.The characteristics of each study, concerning cases of neonatal C. auris infection, are extensively presented in Table 1.Of all the studies included in the present review, three concerned public health surveillance programs, four were laboratory-based surveillance series, five were case series, three were cross-sectional studies, and the remaining nine were cohort studies.The characteristics of each study, concerning cases of neonatal C. auris infection, are extensively presented in Table 1.In total, these studies assessed data from 476 neonates.The countries from which the cases of neonatal C. auris infection were documented are Bangladesh, China, Colombia, India, South Africa, and Venezuela (Figure 2).In total, these studies assessed data from 476 neonates.The countries from which the cases of neonatal C. auris infection were documented are Bangladesh, China, Colombia, India, South Africa, and Venezuela (Figure 2).The demographic data of the study population are presented in Table 2. Unfortunately, among the retrieved studies, there were lots of missing data.For the majority of cases, there were no data concerning the age of gestation.Of the 31 neonates with a documented gestational age, 20 were preterm neonates (64.5%), with a median gestational The demographic data of the study population are presented in Table 2. Unfortunately, among the retrieved studies, there were lots of missing data.For the majority of cases, there were no data concerning the age of gestation.Of the 31 neonates with a documented gestational age, 20 were preterm neonates (64.5%), with a median gestational age of 28 weeks (IQR [26][27][28][29][30][31][32][33][34].The median patient age (among the 264 cases with reported data) during disease onset was 21 (range: 1-71) days.Apart from prematurity, many other predisposing factors were documented, such as the administration of total parenteral nutrition (TPN), the placement of central line catheters, mechanical ventilation, and previous administration of broad-spectrum antibiotics (with piperacillin tazobactam and meropenem combined with vancomycin being the most used).As far as the underlying conditions are concerned, information was provided only for 42 neonates (8.8% of cases); congenital heart disease (14/42, 33.3%), congenital gastrointestinal disorders (10/42 neonates, 23.8%), and necrotizing enterocolitis (NEC, 4/42, 9.5%) were the most common comorbidities.From the available literature data, 20 out of the aforementioned neonates (48%) had a history of a previous abdominal or cardiac surgery.No neonate had previously received antifungal prophylaxis.
Candida auris was isolated from blood cultures (98.1%), urine (16.5%), peritoneal fluid specimen (2.4%), and cerebrospinal fluid (2.1%) (Table 3).Information concerning the type of therapy was only provided for 57 cases.The choice of antifungal medication differed significantly between studies, with amphotericin B (AMB) being the most frequently administered agent as monotherapy (Table 4).The mean duration of antifungal therapy (excluding neonates that passed away) was 27.85 days (range: 11-42 days).Information concerning the resistance of the pathogen to antifungal medication was only given for 79 cases.Specifically, resistance to fluconazole was mentioned in 77 out of the 79 cases (97.4%) and to AMB in 53 cases (67.1%).Data concerning the outcome of patients were provided for 93 cases, out of which the infection was successfully treated in 58.1% of patients, whereas 41.9% of neonates passed away.Information regarding survival was not available for 383 cases.

Discussion
Very little information is available for neonatal C. auris infection, while the potential severity of the disease requires timely recognition and appropriate management.To our knowledge, this is the first systematic review of the literature focused on the epidemiologic data, the clinical presentation, risk factors and outcome of C. auris infection in the neonatal population.
Invasive fungal infections (IFIs) are the main cause of morbidity and mortality in hospitalized neonates in the NICUs, especially preterm neonates and those with a very low birth weight (VLBW, <1500 g) [55].These neonates are immunocompromised, exposed to broad-spectrum antibiotics, have an immature epithelial barrier and are subjected to invasive procedures, and consequently, they are at an increased risk for opportunistic fungal infections.The majority of IFIs are caused by Candida species [56].Invasive Candidiasis (IC) is the third most common cause of late-onset sepsis in VLBW neonates and an important cause of morbidity and mortality in the NICU [56].The frequency of IC in NICUs ranges between 0.5 and 20% and varies between centers and patient population and is inversely proportional to gestational age and birth weight, with a higher frequency documented in neonates with an extremely low birth weight (ELBW) [26,56,57].However, cases of neonatal IFI caused by other fungal species are increasingly reported in the international literature.In recent years, we have witnessed a global change in candidemia epidemiology with the appearance of resistant, non-albicans species of Candida, with a particular emphasis on the increase in cases of C. auris infections.Candida auris is spreading globally and causing hospital infections in pediatric and adult populations, particularly in ICUs [58][59][60] and is included in the 2019 Emergency Threats Report of the Centers for Disease Control and Prevention (CDC) of the USA [12].As is evident from the history of the discovery of this fungus, C. auris can cause severe ear infections.However, its main threat lies on its isolation from the bloodstream.Wound and soft tissue infections, osteomyelitis, central nervous system infections, and generalized infections in immunosuppressed patients have been described in the international literature [60].According to our systematic review, we located 24 publications [23][24][25]32,[35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54] from 2013 until today, in which 476 cases of C. auris infections are reported.Unfortunately, there was a lot of information missing among the retrieved studies.For the majority of the recorded cases, data regarding the age of gestation were not provided.However, it is noteworthy that, among the overall number of infants with reported gestational age, 64.5% were premature, suggesting that prematurity, as in other cases of IFI, is one of the main contributing factors to C. auris infection.Apart from prematurity, other contributing factors noted were the administration of total parenteral nutrition, the insertion of umbilical or other central lines, mechanical ventilation and prior administration of broad-spectrum antibiotics, which were also recognized in adult and pediatric patients with C. auris infection [60][61][62].From the total number of patients with available data concerning underlying health problems, 33.3% had congenital heart disease, 23.8% had congenital gastrointestinal malformations, and 9.5% had necrotizing enterocolitis.These findings suggest that invasive infections caused by C. auris represent an emerging threat, not only for premature neonates but also for term neonates hospitalized in the NICU due to comorbidities, such as complex congenital heart disease or other congenital malformations requiring surgical correction [23,25].Fungemia is a condition with high mortality, often as high as 50%, while the mortality rate for C. auris is reported to be between 40 and 60% [4,6].The mortality rate we documented, in the total number of neonates with C. auris infection (93 cases), for which the studies provided information concerning the outcome of the infection, was quite high (approximately 42%), a finding consistent with the high mortality rate reported in studies focused on pediatric and adult patients [6,60].
C. auris has comparable characteristics with common candida species, concerning its force of infection and its ability to adhere to the host's cell and form a biomembrane.However, in contrast to C. albicans, where the source of infection is usually the patient's own intestinal microbiota, C. auris infections are mainly caused by the dispersion of the pathogen via the hands, from a colonized patient to another or from the colonized inanimate environment.C. auris causes systematic infections such as septicemia, soft tissue infections, and surgical site infections in seriously affected individuals of all ages, from premature infants to the elderly patients.It is also isolated from respiratory and urine samples, which it often colonizes [63].A distinctive characteristic of C. auris is that it colonizes the patient's skin and the hospital environment for extended periods of time and spreads from one patient to another within healthcare facilities, a characteristic not previously observed with other Candida species.These characteristics remind us of those seen in multi-resistant bacteria such as Acinetobacter or Klebsiella [63].C. auris adheres to the skin due to a unique agent it possesses which does not exist in other Candida species called Surface Colonization Factor (Scf1) [64].This agent provides the fungus with the ability to effectively adhere to the skin and central line catheters, resulting in infection.This discovery, recently published in the scientific journal Science, is significant because it might potentially lead to a new approach in treating this fungus through the development of new agents that inhibit the Scf1 factor and the fungus's ability to adhere.Although reports of C. auris infections among neonates and children are rare, its appearance in different countries and continents worldwide represents a new reality that requires proper preparation across all aspects of patient's healthcare.A better understanding of the mechanisms of pathogenesis will lead to strategies that aim to control and prevent C. auris infections in neonates.Information about experimental models of diffuse C. auris infection in adult mice have been reported.In these models, adult mice were intravenously infected with C. auris, revealing that the tissues with the highest fungal burden and invasion during the infection were the kidneys and the heart [65,66], as opposed to the experimental model of neonatal candidiasis conducted by Flores-Maldonado et al. [19], who realized that the liver and the brain of animal neonates were the tissues with the highest fungal burden and invasion during infection, indicating that the spread of the fungus during infection is different between neonates and adults.Therefore, there is great need to understand the pathogenesis of C. auris infection during the neonatal period and the characteristics of the interaction between C. auris and neonates.
Furthermore, these studies indicate that the pathogenesis and the outcome of systematic neonatal candidiasis varies between different strains of Candida.This is probably because different strains of Candida exhibit different patterns of virulence [67], antifungal sensitivity [68], and species-dependent recognition of the immune system [69].Specifically, C. auris is a strain of fungus that causes different immune responses in comparison to other strains, such as C. albicans [70].These differences in immune response are probably due to the composition of the cell wall of C. auris, where mannans and mannoproteins are the main inducers of the immune response, while in C. albicans, the main inducers of the immune response are glucans [71].
According to our systematic review, which includes 476 neonates with C. auris infection, the most common type of invasive infection was bloodstream infection (98.1%, 263/268 patients with available data on the site of infection).In our study, besides primary sepsis, other sites of infection were urinary tract infection (16.5%), meningitis (2.4%), peritonitis (2.4%), and skin abscess (1.2%).These findings are in accordance with information derived from studies on adult patients as well as with a recent systematic analysis conducted by Ashkenazi-Hoffnung and Rosenberg Danziger [60] which provides data on C. auris infection with a focus on children.According to the experimental model of Flores-Maldonado et al. [19], the spleen, kidneys, and lungs were the tissues of the mouse with the lowest colonization, while the liver showed the highest fungal invasion in the early stages of the infection and was associated with the presence of cellular infiltration.In contrast, the brain was the most susceptible tissue to invasion from C. auris, and the invasion persisted until the advanced stages of the infection, findings that were not confirmed by our systematic review.This could be primarily due to the high percentage of missing data (50.4%) in our review regarding the isolation site of C. auris.Additionally, we must take into consideration that in the experimental model, data concerning C. auris dissemination were evaluated via fungal burden and histopathological tissues analysis.
The clinical presentation of C. auris infections, in most cases, is nonspecific, and the providing a distinction from other types of systemic infections is often a challenge [72].This fact is also reflected in our systematic review where, in the majority of neonates, C. auris infection had a similar clinical presentation with bacterial sepsis.As with other Candida infections, the diagnosis of invasive C. auris infection is made by isolating the yeast in blood cultures or other biological specimens from infected sites, clinically or radiographically ascertained.Identifying C. auris in cultures is often a prolonged procedure in comparison to more common types of Candida.The diagnosis of C. auris is extremely difficult as it is often confused with other rare fungi species (C.haemulonii ή C. duobushaemulonii, C. famata, C. lusitaniae, Rhodotorula glutinis ή Saccharomyces cerevisiae) using conventional fungal identification methods and demands the use of newer methods of identification, such as mass spectrometry MALDI-TOF MS, which are not widely available [60,73].In order to overcome this problem, new precise and rapid diagnostic methods have been recently developed.Furthermore, morphological distinction detectors can be used for the rapid assessment of resistance mechanisms to echinocandins and azoles.C. auris has been included in the database of the VITEK-2 automated system since 2018.Despite being less available for routine testing, the molecular determination of sequence using polymerase chain reaction (PCR) analyses offers a definite identification of C. auris.Several molecular analyses have been developed, including conventional PCR assays, real-time PCR, magnetic resonance T2 sensing, and loop-mediated isothermal amplification (LAMP) assays [74,75].As opposed to automated biochemical systems and MALDI-TOF MS, which depend on specimen cultures, DNA can be directly isolated from the patient's samples, without the need for culture.
One of the main reasons of global concern about C. auris spreading is its sensitivity profile, which limits the choice of therapy.Most C. auris strains are resistant to fluconazole, cross-resistant to other azoles, and have variable increased minimum inhibitory concentrations (MICs) for amphotericin B [76,77].Echinocandins have the lowest MIC for C. auris in comparison to other categories of systematic antifungal medication.However, resistance has been described even to this medication [78].Data over time suggest that resistance rates to echinocandins are increasing [79].Information derived from our systematic review, regarding resistance to antifungal medication, was only available for 79 cases.More specifically, 97.4% of the cases were resistant to fluconazole, and 67.1% of the cases were resistant to AMB.Data concerning the resistance to other categories of antifungal medication were limited.This finding raises concern over the management of neonatal C. auris infections as fluconazole and AMB are the most widely used antifungal medicine in NICUs.Moreover, fluconazole is, up to date, the first choice medication used as a prophylaxis against fungal infections in extremely low birth weight infants [80].The treatment of C. auris presents with important challenges as it is resistant to various categories of antifungal medication, demanding changes in therapy due to treatment failure.Based on the information given above, the concern on resistance to azoles and AMB has led the CDC and Public Health England to recommend echinocandins as the first-line therapy against C. auris infections [75,81].However, so far, no correlation has been established between in vitro sensitivity testing and clinical results.Observational data from hospital cases show high mortality rates between patients infected by C. auris, regardless of the choice of antifungal agent, both in adults [35,82] and children [32,44].Additionally, past exposure to fluconazole and echinocandins has consistently been related to a high risk of C. auris infection [38,83].In our review, data concerning the type of therapy were provided only for 12% of cases.The choice of antifungal medication differed significantly between studies, with AMB mostly used as monotherapy.The mean duration of antifungal treatment (excluding neonates that passed away) was 27.85 days (range 11-42).In 44% of cases, an invasive infection was treated with a combination of antifungal agents.Conclusions concerning differences in mortality rates according to antifungal treatment could not be drawn due to the small number of patients.In cases of C. auris pediatric infection, treatment recommendations are derived from the Infectious Diseases Society guidelines of 2016 [81] for the therapy of candidiasis, which suggest caspofungin or micafungin as both medications are approved by the FDA for use in children above 2 months of age.For neonates, amphotericin B deoxycholate is the first-line therapy choice.If sensitivity reports show increased (MIC) in all categories of antifungals, there is evidence suggesting that combination therapy with fluconazole and echinocandin, AMB, or azole effectively reduces MIC in vitro [84][85][86].
In settings with high rates of C. auris, some authors recommend antifungal prophylaxis with echinocandins in low birth weight premature neonates as an alternative to the typical prophylaxis with fluconazole [27].This suggestion comes as a result of the high resistance that C. auris momentarily presents to fluconazole [32,87].A small, comparative clinical study reported that the use of micafungin, in comparison to fluconazole, for prophylaxis against fungal infections in ELBW neonates was connected with lower rates of C. albicans infection [88].Furthermore, the safety and pharmacokinetics of micafungin have been evaluated in VLBW [89,90].As a result, within the context of an endemic-epidemic in a NICU, the prophylactic use of micafungin could be taken into consideration for high risk populations.However, as aforementioned, there is a concern that the echinocandins used could exert selective pressure, leaning in favor of the emergence of C. auris.
C. auris poses a serious challenge to the healthcare system due to its unique characteristics, including the extended transmission among patients, the persistence in hospital environments, misidentification via traditional laboratory methods, antifungal resistance, and the association with high mortality rates.The serious threat posed by C. auris has urged public healthcare services, throughout the world, to detect and report cases to health authorities.
Our study has certain limitations, and its findings should be interpreted with caution.Our main limitation was the high rate of missing data.The gaps and significant heterogeneity in recording basic information, such as the timing of disease manifestation, prenatal risk factors, demographic characteristics of neonates, mode of C. auris transmission, and diagnostic method, among the studies included in our review are evident.This fact served as a primary inhibitor to conducting a meta-analysis.

Conclusions
In conclusion, this study portrays that even though reports of neonatal C. auris infections are relatively rare, they have emerged in NICUs and seem to be replacing IFI caused by other fungal species.Their global emergence in many countries and across various continents represents a new reality that demands the effective planning of patient's healthcare, including laboratory preparedness for adequate detection and high clinical awareness among healthcare providers for any unrecognized or rarely encountered species of Candida, in order to promptly apply infection control measures.Despite effective intervention and dramatic improvement in compliance with health protocols, the transmission of C. auris may persist even to a small extent.In this context, it is emphasized that prevention and early intervention are undoubtedly the preferred means to address this challenging pathogen.A future investigation focused on the risk stratification of patients as well as optimal choices of pre-symptomatic screening will offer important information in order to help healthcare facilities in preemptively identifying these outbreaks within the constraints of resources.The associations of C. auris with hospital outbreaks in the NICUs, invasive infections, high levels of resistance to antifungal medication, and high rates of mortality underline the importance of global collective efforts to raise awareness and limit its spread.Future research should address knowledge gaps concerning appropriate antifungal therapy.

Figure 1 .
Figure 1.Flowchart of the systematic review.

Figure 1 .
Figure 1.Flowchart of the systematic review.

Figure 2 .
Figure 2. Map of countries where cases of Candida auris infection were documented in neonatal population.

Figure 2 .
Figure 2. Map of countries where cases of Candida auris infection were documented in neonatal population.

Table 1 .
Studies concerning cases of neonatal C. auris infection.

Table 2 .
Demographic data of the study population.
Data are presented as absolute frequencies (percentages).

Table 3 .
The isolation site of C. auris.

Table 4 .
Treatment strategies and survival rate.
Data are presented as absolute frequencies (percentages).