Clinical characteristics of invasive candidiasis in infants born before 30 weeks of gestation: a nested case series from a multicenter cohort study in the Netherlands and Belgium
Rimke R. de Kroon, Aranka J. van Wesemael, Mirjam M. van Weissenbruch, Tim de Meij, Hendrik J. Niemarkt
TL;DR
This study identifies clinical features of invasive candidiasis in extremely preterm infants and highlights the need for early diagnosis and treatment to improve outcomes.
Contribution
The study reveals that vaginal delivery and specific clinical features may indicate vertical transmission of candidiasis, offering new prevention opportunities.
Findings
Infants with invasive candidiasis were born at an average of 25.7 weeks with low birth weight and often delivered vaginally.
Delayed antifungal treatment was common, and C-reactive protein and platelet count at diagnosis were linked to fatal outcomes.
Most cases were caused by Candida albicans, and affected infants required significant medical interventions like ventilation and transfusions.
Abstract
Invasive candidiasis poses a serious risk to preterm infants, due to its rapidly progressive and severe clinical course, resulting in considerable mortality and long-term morbidity. Therefore, the aim of the current study was to assess the clinical characteristics of invasive candidiasis in very preterm infants to increase awareness among clinicians. A multicenter cohort study database was screened for infants, born in one of ten Neonatal Intensive Care Units (NICUs) in the Netherlands and Belgium between October 2014 and May 2025, with blood- and/or cerebrospinal fluid culture-proven invasive candidiasis (gestational age < 30 weeks) in the first 29 days of life. Clinical data were retrospectively collected. Out of 2.824 infants, 24 were diagnosed with invasive candidiasis (0.8%), most frequently caused by Candida albicans (83%). Affected infants demonstrated distinct clinical features:…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
- —https://doi.org/10.13039/501100008359Maag Lever Darm Stichting
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAntifungal resistance and susceptibility · Neonatal and Maternal Infections · Parasitic Diseases Research and Treatment
Introduction
Invasive candidiasis, defined as a Candida infection of the blood and/or cerebrospinal fluid (CSF), is a threat to preterm infants, particularly to those with a gestational age (GA) below 32 weeks [1]. Although the incidence is low (~ 4–8% of all sepsis episodes) [2], the clinical course is severe and rapidly progressive, with potential for Candida dissemination to various organs [1]. Invasive candidiasis is associated with mortality rates of up to 60% with long-term effects, including neurodevelopmental impairment, in survivors [1].
Preterm infants are particularly susceptible to develop invasive candidiasis due to their immature immune system and impaired barrier function [3]. Candida colonization can occur as early as birth [4]. The use of medical devices (e.g. central venous catheters and invasive ventilation) compromises skin and mucosal barriers, enabling translocation of Candida from colonized sites into the bloodstream [5]. Additionally, disruption of the gut microbiota by broad-spectrum antibiotics may promote intestinal overgrowth of Candida [6], potentially increasing the risk for disease through translocation.
Targeted treatment is often delayed, as empirical therapy for suspected sepsis typically does not include antifungal agents. The turn-around time for fungal cultures is long [7], which may delay treatment initiation [8]. Some clinicians opt for prophylactic antifungal treatment. However, its use varies, due to the generally high number needed-to-treat, depending on the baseline incidence, and potential emergence of resistant strains [9, 10]. While Candida albicans is most frequently isolated, there appears to be a global shift toward non-albicans Candida species, including the emerging Candida auris, further complicating routine prophylaxis and treatment [11].
The aim of the current study was to comprehensively assess clinical characteristics of invasive candidiasis in infants born before 30 weeks of gestation to increase awareness among clinicians.
Methods
This case series is embedded in a prospective multicenter cohort study, for which daily fecal samples are collected in the first month of life (GA < 30 weeks from October 2014 until March 2021, < 28 weeks from March 2021 onwards) in 10 Neonatal Intensive Care Units (NICUs) in the Netherlands and Belgium. Exclusion criteria are chromosomal abnormalities and/or congenital gastrointestinal (GI) diseases. The study was approved by the local medical ethical review boards (protocol 2014.386, amendment A2016.363) and written informed consent was obtained from parents or legal guardians. The study database between October 2014 and May 2025 was screened for infants with invasive candidiasis, defined as a positive blood and/or CSF culture for Candida resulting in systemic infection. For all screened infants, key baseline characteristics (GA, BW, mode of delivery) and sepsis and/or GI disease were obtained (Figure S1), facilitating direct comparison between affected infants and the rest of the cohort. For all affected infants, detailed clinical characteristics were retrospectively collected. Statistical analyses are described in Appendix A. Routine prophylaxis with fluconazole was administered in 1/10 study sites (Table S1). Treatment delay was defined as antifungal treatment initiation ≥ 1 days after withdrawal of blood culture for suspected infection. End-organ disseminated disease was defined as invasive candidiasis spread to distant organs. Culture-proven sepsis was defined as a blood and/or CSF culture-proven infection within or after the first 72 h of life (early-onset (EOS) and late-onset sepsis (LOS), respectively). GI disease included necrotizing enterocolitis (NEC) modified Bell’s stage ≥ 2 A and/or focal intestinal perforation (FIP).
Results
Out of 2.824 screened infants, 24 (0.8%) were diagnosed with invasive candidiasis (Figure S1, Table S2) with affected infants evenly distributed over the study period. Baseline characteristics of the affected infants are displayed in Table 1. Notably, the mean GA was 25 weeks + 5 days (SD ± 9 days), BW was 827 g (SD ± 198 g) and 89% was born through vaginal delivery. The three key baseline parameters were significantly different in the affected infants compared to the rest of the cohort (Table S3). The effects of adhering to two or more key baseline parameters (GA < 27 weeks, BW < 1000 g, and vaginal delivery) on the likelihood to develop invasive candidiasis was assessed, and demonstrated an increased risk (OR = 6.57, CI = 2.09–20.66,* p* < 0.001) (Table S4). Invasive candidiasis incidence decreased with increasing GA, from 3.7% at 24–25 weeks to 0.4% at 28–29 weeks. Table 1. General patient characteristics and clinical characteristics of very preterm infants (gestational age < 30 weeks) with invasive candidiasisInfants diagnosed with invasive candidiasis (n = 24)General patient characteristics****DemographicsGestational age, mean (weeks + days) ± SD (days)25 + 5 ± 9Birth weight, mean (grams) ± SD (grams)827 ± 198Mode of delivery, vaginal delivery, n [%]21 [88]Biological sex, female, n [%]9 [38]Singleton, yes, n [%]16 [66]Apgar score at 5 min < 7, n [%]8 [33]Medication practicesReceived surfactant within 72 h after birth, n [%]14 [58]Ratio cumulative IV antibiotics in first month of life, days, median (IQR)0.74 (0.33)Exposure to IV antibiotics prior to onset of disease, n [%]23 [96]Exposure to probiotics prior to onset of disease, n [%]7 [29]Feeding practicesReached FEF in first month of life^1^, n [%]16 [66]Time to FEF, days, median (IQR)13 (11)Received exclusively human milk at first day of FEF^2^, n [%]24 [100]Mortality29-day mortality, n [%]6 [25]Mortality during NICU admittance, n [%]10 [42]CLINICAL CHARACTERISTICS****DiagnosisPostnatal age at disease onset (days)^3^, median (IQR)11 (14)Onset of invasive candidiasis- Early onset (< 72 h), n [%]2 [8]- Late onset (≥ 72 h), n [%]22 [92]Exposure to invasive medical device (e.g. central venous or arterial catheters, invasive ventilation, umbilical lines, and/or peripheral IVs) in 48 h prior to disease onset, n[%]23 [96]Exposure to solely a peripheral IV prior to onset of disease, n [%]5 [20]Candida colonization prior to disease onset^4^:- Fungal pathogen cultured from urine, n [%]3 [13]- Fungal skin infection that required treatment with miconazole, n [%]9 [38]- Enteral nystatine administration, n [%]2 [9]Previous comorbidities:- Clinical sepsis, n [%]3 [13]- Bacterial culture-proven sepsis, n [%]6 [25]- Gastrointestinal disease (e.g. NEC or FIP), n [%]6 [25]- Absence of comorbidities, n [%]13 [54]MicrobiologyCandida cultured from:- Only blood culture, n [%]20 [83]- Only CSF culture, n [%]2 [8.3]- Both blood and CSF cultures, n [%]2 [8.3]Cultured pathogen:- Candida albicans, n [%]20 [83]- Candida tropicalis, n [%]1 [4.2]- Candida parapsilosis, n [%]1 [4.2]- Candida glabrata, n [%]1 [4.2]- Candida guillermondii, n [%]1 [4.2]Resistant strain isolated^5^, n [%]2 [10]Co-infection with bacterial pathogens^6^, n [%]3 [17]Antifungal TreatmentReceived fluconazole IV prior to disease onset^7^:- Routine prophylactic fluconazole, n [%]2 [8]- Clinical indication for fluconazole before diagnostic work-up, n [%]5 [21]- No exposure to fluconazole iv prior to diagnostic work-up, n [%]17 [71]Delay in initiation of IV antifungal treatment^8^, n[%]14 [58]Time to start IV antifungal treatment (days), in case treatment had not yet been initiated prior to diagnostic workup:- On the same day, n [%]3 [13]- 1 day after work-up, n [%]6 [25]- 2 days after work-up, n [%]4 [16]- 3 days after work-up, n [%]3 [13]- Mortality prior to start antifungal treatment, n [%]^9^1 [4.2]Antifungal treatment:- Monotherapy, n [%]14 [58]- Dual therapy, n [%]6 [25]- Triple therapy, n [%]2 [8.3]- Received no treatment due to mortality, n [%]1 [4.2]Type of IV antifungal treatment^10^:- Fluconazole, n [%]22 [96]- Liposomal amphotericin B, n [%]7 [30]- Micafungin, n [%]2 [8.7]- Caspofungin, n [%]1 [4.3]Disease severityLaboratory findings^11^:- C-reactive protein, mg/L, median (IQR; min, max)28 (68; 0, 221)- Platelets, 10^9^/L, median (IQR; min, max)85 (177; 21, 560)- Leukocytes, 10^9^/L, median (IQR; min, max)24 (17; 5, 55)Requirement for invasive ventilation, n [%]14 [58]Requirement for hemodynamic support:- Inotropics (e.g. hydrocortisone, dopamine, dobutamine), n [%]8 [36]- IV fluid resuscitation, n [%]10 [42]- Inotropics and/or IV resuscitation, n [%]10 [42]Requirement for transfusion:- Red blood cell transfusion, n [%]17 [71]- Fresh frozen plasma transfusion, n [%]4 [17]- Platelet transfusion, n [%]8 [33]End-organ disseminated candidiasis^12^:- Liver, n [%]2 [8.3]- Kidneys, n [%]2 [8.3]- Eyes, n [%]1 [4.2]- Joints, n [%]1 [4.2]- Central nervous system, n [%]6 [25]Absence of concurrent comorbidities, n [%]16 [66]^1^8 infants did not reach FEF (mortality prior to FEF (n = 6), reached FEF after first month of life (n = 2), respectively). ^2^Exclusively human milk may consist solely of mother’s own milk, donor milk, or a combination. ^3^The day on which the diagnostic work-up was performed, marked by at least a positive blood and/or CSF culture for Candida, was considered the day of onset. ^4^Missing data: nystatine administration (n = 1), urine cultures (n = 1). ^5^C. guillermondii was isolated with resistance to anidulafungin and micafungin (n = 1). C. albicans was isolated with resistance to fluconazole and voriconazole (n = 1). Missing data: 4 infants. ^6^Co-infection with *Escherichia coli *(n = 1), Staphylococcus aureus (n = 1), and Coagulase-negative Staphylococci (n = 2). ^7^Among the two infants who received fluconazole prophylaxis before developing invasive candidiasis, one had a blood culture positive for C. albicans resistant to both fluconazole and voriconazole, while the other had a blood culture positive for non-resistant C. albicans. ^8^Treatment delay was defined as initiation of antifungal medication on day after work-up or at a later timepoint. ^9^The infant (born with GA 26–27 weeks) who passed away without initiation of antifungal therapy died on the day the diagnostic work-up was conducted in the third week of life. ^10^Dosage per antifungal medication according to the local protocols: fluconazole 25 mg/kg/day on day 1 followed by 12 mg/kg/day from day 2, liposomal amphotericin B 3–5 mg/kg/day, micafungin 10 mg/kg/day, and caspofungin 2 mg/kg/day. ^11^For all three parameters, the most aberrant measurement within 48 h following work-up was collected. Missing data: CRP (n = 1), platelets (n = 1), leukocytes (n = 2). ^12^Dissemination to the liver, kidneys, and heart was confirmed by ultrasound. Ocular involvement was assessed through an ophthalmologic examination. Joint dissemination was established by a positive Candida culture of the synovial fluid. Central nervous system dissemination was confirmed by a positive culture of the CSF and/or by cerebral ultrasound findings. Abbreviations: CRP C-reactive protein, CSF cerebrospinal fluid, FEF full enteral feeding, FIP focal intestinal perforation, IV intravenous, IQR interquartile range, NEC necrotizing enterocolitis; NICU Neonatal Intensive Care Unit, SD standard deviation
Diagnosis
Detailed clinical characteristics were obtained from all infants with invasive candidiasis (n = 24, Table 1). The median postnatal age was 11 days (IQR 14) at time of diagnostic work-up. Candida colonization was present in 46% prior to clinical onset. 46% was diagnosed with clinical or culture-proven sepsis (38%) and/or GI disease (25%) prior to clinical onset. The 29-day incidence of GI disease was higher in affected infants (46%) compared to those with non-Candida sepsis (16%) or no sepsis (9%, p < 0.001).
Microbiology
C. albicans was the predominant isolated Candida species (83%) (Table 1). From 2 infants, a resistant strain was isolated (Candida guillermondii resistant to anidulafungin and micafungin and C. albicans resistant to fluconazole and voriconazole, respectively).
Antifungal treatment
In 7 infants (29%), fluconazole had already been administered prior to empirical antibiotics, either as part of routine prophylaxis or based on clinical indicators, such as skin lesions or severe NEC (Table 1). Initiation of antifungal treatment was delayed in 58% of infants. In 3 infants (13%), initiation of adequate treatment took up to 3 days.
Disease severity
Invasive ventilation and hemodynamic support was required in 58% and 42% of infants, respectively (Table 1). Platelets were low (median 85 × 10^9^/L, IQR 177) and 33% received a platelet transfusion. Moreover, 71% of infants received a red blood cell transfusion. C-reactive protein (CRP) was higher in the group with a fatal outcome (median 68 mg/L (IQR 112) in deceased infants vs. 7 mg/L (IQR 74) in survivors, p = 0.040), while platelets were lower (median 53 × 10^9^/L (IQR 55) in deceased infants vs. 124 × 10^9^/L (IQR 245) in survivors, p = 0.010). In 33%, disseminated candidiasis was diagnosed. Central nervous system involvement was most common (25%). Notably, 42% died during NICU admittance.
Discussion
We described a case series of infants born before 30 weeks of gestation with invasive candidiasis. These infants generally had low GA, extremely low BW, and were born vaginally. Adherence to two or more key baseline criteria (GA < 27 weeks, BW < 1000 g, and vaginal delivery) was strongly associated with the likelihood of invasive candidiasis (OR = 6.57, CI = 2.09–20.66, p < 0.001), indicating that these features contribute to disease vulnerability. Moreover, affected infants were frequently exposed to clinical and/or culture-proven sepsis (38%) and/or GI disease (25%) prior to onset, and the 29-day GI disease incidence was significantly higher in the affected infants. Extremely preterm infants have a more immature immune system and impaired barrier function [3], which, combined with previous comorbidities, potentially facilitates disease development. Vaginal delivery may result in Candida colonization, particularly in the presence of maternal urogenital colonization, allowing vertical transmission [12]. The observed interval between birth and onset of disease (median 11 days postnatal age) could reflect the time required for birth-acquired colonization to progress to invasive disease. However, the potential role of vertical transmission in invasive candidiasis should be interpreted cautiously as, for example, maternal colonization status was not assessed in this study.
Candida as a cause for sepsis is often considered late in the diagnostic evaluation. However, timely consideration is of importance, considering that early initiation of (empirical) antifungal treatment is essential to ensure favorable outcomes [8]. Our study highlights that treatment delay occurred in 58%, which may in part be explained by the long turnaround time for fungal blood cultures [7], as well as the guidelines on standard empirical antimicrobial therapy for suspected sepsis in the Netherlands and Belgium that solely includes antibacterial agents. Antifungal treatment is frequently deferred until invasive candidiasis is confirmed by blood and/or CSF-culture results, resulting in delayed initiation of therapy for invasive candidiasis relative to the onset of clinical symptoms.
Given the rarity of invasive candidiasis, routine screening or empirical treatment with antifungals in all preterm infants is not justified. Our findings suggest that in infants with distinct clinical features (extreme prematurity, extremely low BW, vaginal delivery, and sepsis and/or GI disease prior to clinical onset), clinicians should maintain a high index of suspicion, particularly in units with a higher incidence of Candida colonization and/or infection. These findings may support clinicians in considering antifungal therapy, when two or more baseline criteria are met, as part of the empirical treatment regimen for infants with suspected sepsis. In countries with a different pathogen distribution and/or resistance patterns, alternative antifungal treatment (e.g. micafungin, amphotericin B) should be considered as empirical therapy, as opposed to fluconazole, while awaiting culture results. Notably, some infants developed invasive candidiasis despite receiving antifungal prophylaxis, emphasizing that prophylaxis alone should not reassure clinicians. When invasive candidiasis is suspected or confirmed, dissemination assessment, including CSF analysis and imaging of liver, brain, kidneys, eyes, and musculoskeletal system, should be considered as one third of cases had disseminated disease. This advice is supported by the severe disease progression, frequently requiring invasive ventilation, hemodynamic support, and red blood cell and/or platelet transfusion [8].
Our study stands out due its multicenter approach and detailed clinical characterization of invasive candidiasis. We were able to assess key baseline characteristics as well as sepsis and/or GI disease in the total cohort. However, additional clinical data from the unaffected infants was unavailable, hampering the identification of additional risk factors. Nonetheless, we observed relevant trends warranting further investigation and identified parameters (CRP and platelets) associated with a fatal outcome. Notably, future (prospective) studies should include additional maternal data collection, including fungal colonization and pregnancy complications, such as chorioamnionitis. Extrapolation of our findings to other countries may be challenging, one of the reasons being that C. albicans was the most common pathogen, hampering direct comparison with countries where non-albicans species are a larger contributor [11]. Lastly, our study focused on blood- and CSF-culture proven invasive candidiasis, and therefore may underestimate the overall burden of Candida-related disease by excluding other clinically significant manifestations of candidiasis, such as urinary tract infections, peritonitis, and congenital cutaneous candiasis.
In conclusion, this study provides a detailed clinical profile of infants with invasive candidiasis, highlighting shared clinical characteristics, delayed antifungal treatment, and substantial disease burden. These findings may increase awareness among clinicians and support the development of strategies for earlier recognition and targeted diagnostics. Our observation that invasive candidiasis more frequently occurs following vaginal delivery suggests that colonization and subsequent invasive infection may result from vertical transmission. This finding potentially opens new avenues for understanding early pathogenesis and potentially tailoring preventive strategies.
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 88.8 KB)