Afectarea sistemului nervos central în sepsisul cu debut precoce cu Candida parapsilosis la un nou-născut prematur: prezentare de caz și revizuire a literaturii

Introduction

Invasive candidiasis remains an important cause of morbidity and mortality among premature neonates, particularly in those requiring prolonged neonatal intensive care and exposure to invasive procedures. Prematurity, very low birth weight, central venous catheterization, parenteral nutrition, prolonged antimicrobial therapy, mechanical ventilation and extended hospitalization are consistently recognized as major predisposing factors for neonatal Candida infection^(1,2).

Among non-albicans Candida species, Candida parapsilosis has emerged as an important neonatal pathogen. A neonatal-specific systematic review and meta-analysis estimated that C. parapsilosis accounts for approximately one-third of neonatal Candida infections⁽¹⁾. More recent neonatal cohorts have confirmed its substantial contribution to candidemia, with reported proportions ranging from approximately 25% to 46% of invasive Candida infections in some neonatal populations^(2-5).

The epidemiology of C. parapsilosis is particularly relevant in the neonatal intensive care unit (NICU). The organism can colonize human skin, persist on environmental surfaces, adhere to intravascular devices and form biofilms. Molecular epidemiological studies have documented both persistent and sporadic clones during NICU outbreaks, while several investigations have implicated healthcare workers’ hands in horizontal transmission^(6-12). These characteristics also contribute to the strong association between C. parapsilosis candidemia and central venous catheter use^(2,13).

Although bloodstream infection represents its most common invasive manifestation, hematogenous dissemination may involve the central nervous system (CNS). Species-specific reports of neonatal C. parapsilosis CNS infection remain limited, but meningitis, meningeal inflammation, cerebral abscesses and subdural empyema have been documented in premature infants^(14,15). Broader studies of neonatal candidiasis further demonstrate that CNS involvement may occur even when conventional cerebrospinal fluid (CSF) parameters or microbiological findings are inconclusive^(16,17).

The diagnosis of CNS candidiasis can therefore be challenging. A negative CSF culture does not necessarily exclude CNS involvement; in a 10-year review of neonatal candidal meningitis, Candida was isolated from the CSF in only 74% of afflicted neonates⁽¹⁷⁾. Consequently, the assessment should integrate CSF microbiology and biochemistry with neurological examination and neuroimaging. Current candidiasis guidelines emphasize CNS evaluation in neonates with invasive Candida infection⁽¹⁸⁾.

We report the case of a preterm infant with C. parapsilosis candidemia associated with markedly abnormal CSF parameters and cranial ultrasonographic findings suggestive of ventriculitis and meningeal inflammation, despite a negative CSF fungal culture. This case illustrates the diagnostic uncertainty that may arise when microbiological, biochemical and neuroimaging findings do not fully converge and highlights the importance of a multimodal approach to suspected neonatal CNS candidiasis.

Literature review

Candida parapsilosis in preterm neonates

Prematurity is a recurrent characteristic of neonatal C. parapsilosis infection. Historical and contemporary reports consistently describe the organism among hospitalized premature and very-low-birth-weight infants^(13,19-21).

In the seven-year neonatal series of C. parapsilosis candidemia, 31 neonates were affected; almost all were premature and were diagnosed at a mean age of 19 days. The presence of central venous catheters and parenteral nutrition in almost all cases reflects the close relationship between invasive intensive care interventions and infection risk⁽¹³⁾.

More recent data confirm that C. parapsilosis remains clinically relevant. A recent study of healthcare-associated Candida infections found C. parapsilosis to be the most frequent species, accounting for 46.2% of neonatal bloodstream infections⁽³⁾. In another neonatal invasive candidiasis cohort, C. parapsilosis represented 25.5% of cases⁽⁴⁾, while a large retrospective study of neonatal invasive fungal disease reported C. parapsilosis in 30.5% of isolates⁽⁵⁾.

The predominance of C. parapsilosis varies among NICUs and geographical regions. Nonetheless, its consistent emergence as a predominant Candida species across diverse neonatal populations positions it as a major pathogen in premature infants, rather than a merely occasional opportunist.

An important feature of its epidemiology is the repeated occurrence of NICU outbreaks. Molecular investigations have demonstrated both persistent and sporadic clones circulating among neonates⁽⁶⁾. Microsatellite genotyping studies have similarly identified clusters of genetically related strains^(8,10).

Several studies have implicated healthcare workers’ hands in transmission. In some outbreaks, neonatal isolates were molecularly related to isolates obtained from staff^(11,12). These findings support the concept that C. parapsilosis has a particular capacity for horizontal transmission through direct contact.

C. parapsilosis is also well adapted to device-associated infection, given its capacity to adhere to artificial surfaces and form biofilms. This key virulence determinant is particularly relevant in premature infants requiring prolonged central venous access. As a result, bloodstream infection may become persistent, and antifungal efficacy may be compromised unless the device is removed.

Late-onset sepsis and risk factors

The available neonatal literature strongly supports C. parapsilosis as a predominantly late-onset pathogen.

In the seven-year case series, the mean age at diagnosis was 19 days⁽¹³⁾. An outbreak associated with healthcare-worker transmission occurred at a mean postnatal age of approximately 2 weeks⁽¹¹⁾, while another NICU outbreak reported a mean onset of approximately 35 days⁽²⁰⁾.

These data contrast with true congenital or early-onset fungal infection data and suggest that prolonged NICU exposure is central to C. parapsilosis pathogenesis.

One neonatal candidiasis study documented invasive Candida infections during the first week of life⁽²²⁾. However, the available data from that study do not establish that those early cases were specifically caused by C. parapsilosis. Therefore, the current neonatal evidence is considerably stronger for C. parapsilosis as a late-onset healthcare-associated pathogen.

Repeatedly reported risk factors include:

• extreme prematurity;
• very low or extremely low birth weight;
• central venous catheterization;
• parenteral nutrition;
• mechanical ventilation;
• prolonged hospitalization;
• broad-spectrum antimicrobial exposure;
• previous fungal colonization;
• exposure to multiple invasive procedures.

In the recent neonatal case-control study, 51 neonates had invasive candidiasis, with C. parapsilosis responsible for approximately one-quarter of infections⁽⁴⁾. Earlier comparative and observational studies similarly identified premature neonates receiving intensive supportive care as the predominant population affected^(13,23-25).

The clinical presentation is generally nonspecific. Premature infants may develop apnea, recurrent desaturation, increased respiratory support, feeding intolerance, abdominal distension, temperature instability, lethargy, hypotension, or other signs indistinguishable from bacterial late-onset sepsis.

Benjamin et al. compared neonatal C. albicans and C. parapsilosis fungemia with coagulase-negative staphylococcal bacteremia, highlighting the diagnostic overlap between fungal and bacterial bloodstream infection⁽²⁵⁾.

Persistent candidemia is particularly important, because it may indicate an infected central venous catheter, biofilm-associated disease, inadequate antifungal activity, or an unrecognized site of dissemination. The 2022 neonatal cohort evaluating therapeutic strategies showed the clinical importance of persistent bloodstream infection and catheter management⁽²⁾.

CNS involvement and meningitis

The central nervous system represents an important potential site of dissemination during neonatal candidiasis. However, the published C. parapsilosis-specific CNS literature is substantially smaller than the candidemia literature.

Carter et al. reported a particularly informative case involving a premature infant born at 26+4 weeks of gestation. Both blood and cerebrospinal fluid cultures were positive for C. parapsilosis, and the infection was complicated by a subdural empyema related to migration of a central venous catheter⁽¹⁵⁾. This case illustrates that CNS disease may extend beyond isolated meningitis and may involve additional intracranial infectious complications.

More recently, a series of six premature infants with Candida CNS infection included two infants infected with C. parapsilosis⁽¹⁴⁾. The study described cerebrospinal fluid abnormalities, brain abscesses, meningeal inflammatory changes and MRI findings, evaluating the prolonged antifungal therapy.

These observations confirm that C. parapsilosis can produce severe CNS disease in preterm infants, although the overall number of species-specific reports remains limited.

A broader neonatal study evaluating deep-tissue Candida invasion in 130 neonates examined CNS, ocular, cardiac, pulmonary, hepatic and renal dissemination⁽¹⁶⁾. Another large neonatal invasive fungal infection cohort identified end-organ involvement and included C. parapsilosis among the major fungal species⁽⁵⁾.

Since the cohorts combined different Candida species, species-specific rates of CNS involvement, particularly for C. parapsilosis, are not always available.

CSF and neuroimaging findings

Cerebrospinal fluid evaluation has an important role when CNS dissemination is suspected. Published C. parapsilosis CNS cases demonstrate that a positive CSF culture can confirm meningitis⁽¹⁵⁾. In the premature-infant CNS series, cerebrospinal fluid abnormalities and neuroimaging findings contributed to diagnosis and follow-up, and only a subset of the reported CNS infections was caused by C. parapsilosis⁽¹⁴⁾.

These studies also illustrate an important diagnostic principle: CNS Candida infection should be evaluated using the combination of microbiology, CSF cytology and biochemistry, neurological findings and imaging rather than relying on a single parameter.

However, the neonatal C. parapsilosis literature does not establish that an isolated increase in CSF protein or alteration in CSF glucose is diagnostic of fungal meningitis.

This distinction is particularly important in premature infants. Abnormal CSF biochemical parameters may occur in complex clinical situations and should be interpreted together with the CSF cell count, culture results and the imaging findings.

In proven or clinically suspected CNS disease, the reported abnormalities include meningeal inflammation, cerebral abscesses and other intracranial infectious lesions^(14,15).

Treatment and antifungal resistance

The therapeutic management of neonatal C. parapsilosis infection is complicated by device-associated infection, persistent candidemia and emerging antifungal resistance.

Neonatal studies describe amphotericin formulations, fluconazole and, in selected refractory cases, voriconazole.

Source control is particularly relevant, because C. parapsilosis is strongly associated with central venous catheters and biofilm formation. The 2022 neonatal candidemia cohort specifically evaluated therapeutic strategies, catheter management and outcomes, demonstrating the clinical importance of adequate intravascular source management⁽²⁾.

Altuncu et al. reported two preterm infants with persistent Candida parapsilosis fungemia despite high-dose liposomal amphotericin B therapy. After switching treatment to voriconazole, blood cultures became negative, and both infants were successfully treated without reported serious adverse effects⁽²⁶⁾.

Voriconazole was also evaluated in six premature infants with Candida CNS infection, including two with C. parapsilosis⁽¹⁴⁾. Although these reports are clinically informative, they involve small numbers of patients and should not be interpreted as evidence that voriconazole represents the preferred treatment for all neonatal C. parapsilosis infections.

Antifungal resistance represents a growing concern. Earlier reports had already documented the emergence of fluconazole resistance in C. parapsilosis strains responsible for NICU infections⁽²⁷⁾.

More recently, a cluster of C. parapsilosis displaying an unusual fluconazole-trailing phenotype was described among extremely premature and very-low-birth-weight (VLBW) infants⁽²⁸⁾.

This table represents a proposed clinical framework synthesized from published neonatal Candida CNS studies and current candidiasis guidelines rather than
a formally validated diagnostic classification^{(14,15,17,18)}.

Together, these findings demonstrate that antifungal susceptibility cannot be assumed solely from species identification. Susceptibility testing is especially important during persistent candidemia, outbreaks, or prior azole exposure.

Case presentation

Perinatal history and initial NICU course

A preterm male infant was born at 32 weeks of gestation, with a birth weight of 1700 g, by emergency caesarean section, after maternal preeclampsia. The infant was admitted to the neonatal intensive care unit, where noninvasive respiratory support was initiated. Central venous access was established for parenteral nutrition. Empirical antimicrobial therapy with ampicillin and gentamicin was administered, caffeine was given to prevent apnea of prematurity, and minimal enteral nutrition was initiated with trophic feeds.

Clinical deterioration and diagnosis of candidemia

Subtle clinical deterioration was first noted on the third day of life, with reduced reactivity and a diminished sucking reflex. Cranial ultrasonography excluded intraventricular hemorrhage. Inflammatory markers were only mildly elevated, while the leukocyte count decreased slightly. The blood culture obtained on the first day of life remained negative. Given the unfavorable clinical evolution, antibiotic therapy was escalated.

By the fifth day of life, the infant’s condition had regressed to overt sepsis, presenting with mottled, ashen-gray skin, apnea, accompanied by a further increase in inflammatory markers. Blood cultures subsequently yielded Candida parapsilosis, confirming candidemia and establishing neonatal invasive candidiasis.

Cerebrospinal fluid evaluation

Given the risk of hematogenous dissemination associated with neonatal candidemia, lumbar puncture was performed. Approximately 3 mL of clear, xanthochromic cerebrospinal fluid was obtained under increased pressure.

CSF microbiological analysis:

• bacterial culture – negative;
• fungal culture – negative for Candida.

The abnormal CSF profile raised concern for possible CNS involvement; however, microbiological confirmation of fungal meningitis was not obtained. The case was therefore interpreted as C. parapsilosis candidemia with abnormal CSF findings and suspected CNS involvement, but without microbiologically proven meningitis.

Cranial ultrasonography and neurological assessment

Serial cranial ultrasonography demonstrated increased echogenicity and irregularity of the ventricular ependyma, raising suspicion of ventriculitis. The cortical sulci also appeared widened and echogenic, a nonspecific finding that may reflect meningeal or subarachnoid inflammatory changes.

Apart from these abnormalities, there was no ventricular dilatation, focal parenchymal lesion, or other sonographic evidence of intracranial infectious complications. Brain MRI was not performed.

Because CNS candidiasis may occur despite a negative CSF culture, we interpreted the microbiological and biochemical CSF findings together with the neurological examination and serial neuroimaging findings.

Antifungal treatment and microbiological course

Antifungal susceptibility testing showed that the C. parapsilosis isolate was susceptible to fluconazole, and this drug was initiated.

Despite five days of therapy, the infant remained clinically unwell, the inflammatory markers increased, and blood cultures continued to yield C. parapsilosis. Given the persistent candidemia and the lack of clinical improvement, antifungal therapy was escalated to liposomal amphotericin B. Liposomal amphotericin B was administered for 28 days, with treatment continued for three weeks after the first documented negative blood culture.

Outcome and follow-up

The infant showed progressive clinical and laboratory improvement, being discharged at 38 weeks of corrected gestational age. Neurological examination at discharge was within normal limits.

Reference values based on Srinivasan et al.⁽²⁹⁾ and Zimmermann and Curtis⁽³⁰⁾.

A pediatric neurology consultation was recommended to establish an individualized neurological and neurodevelopmental follow-up plan, considering the potential long-term impact of prematurity, the invasive candidiasis and the abnormal CSF and cranial ultrasonographic findings.

Discussion

The presented case illustrates several clinically relevant aspects of neonatal Candida parapsilosis infection. The infant was born prematurely at 32 weeks of gestation and was exposed early to central venous catheterization, parenteral nutrition and antimicrobial therapy, all of which are recognized risk factors for invasive candidiasis^(1,2,13). Although C. parapsilosis is more commonly described as a healthcare-associated pathogen occurring later during NICU hospitalization, invasive infection may also occur within the first postnatal week. In the presented case, the onset of subtle clinical signs on the third day of life, followed by progression to overt sepsis on the fifth day, supports an early-onset presentation of the infection. This evolution also emphasizes that the initial manifestations of invasive fungal infection may be nonspecific and difficult to distinguish from bacterial sepsis^(24,25).

The most important diagnostic challenge was the possible CNS involvement. Lumbar puncture yielded clear, xanthochromic CSF under increased pressure, with mild pleocytosis, markedly elevated protein and reduced glucose. Although these findings suggested an inflammatory CNS process, fungal culture remained negative. Abnormal CSF biochemistry alone cannot establish the diagnosis of Candida meningitis, but a sterile CSF culture does not reliably exclude it. Fernandez et al. reported that Candida was isolated from CSF in only approximately 74% of neonates diagnosed with candidal meningitis, illustrating the limited sensitivity of CSF culture in this population⁽¹⁷⁾. Therefore, neonatal CNS candidiasis should be evaluated by integrating microbiological, cytochemical, neurological and imaging findings^(14,17,18).

Cranial ultrasonography further increased the suspicion of CNS involvement. Serial examinations revealed increased echogenicity and irregularity of the ventricular ependyma, suggestive of ventriculitis, together with widened and echogenic cortical sulci that could reflect meningeal or subarachnoid inflammation. These findings are not specific for Candida infection and should not be considered diagnostic in isolation. However, when associated with candidemia and markedly abnormal CSF findings, they become more clinically meaningful. Published neonatal Candida CNS cases have described meningeal inflammatory changes, cerebral abscesses and other intracranial complications^(14,15). Zhu et al. reported CNS Candida infection in premature infants with CSF abnormalities and neuroimaging evidence of meningeal inflammation and cerebral abscess formation; two of the six infants were infected with C. parapsilosis⁽¹⁴⁾. Carter et al. also described culture-proven C. parapsilosis meningitis complicated by subdural empyema in a premature neonate⁽¹⁵⁾.

The absence of ventricular dilatation, focal parenchymal lesions, or sonographic evidence of cerebral abscess formation argued against advanced focal CNS disease. However, because brain MRI was not performed, subtle parenchymal or meningeal abnormalities could not be definitively excluded.

Persistent candidemia despite documented fluconazole susceptibility also deserves consideration. In vitro susceptibility does not necessarily predict rapid clinical or microbiological response, particularly when device-associated infection or biofilm formation is present. Fluconazole resistance and trailing phenotypes have also been reported among neonatal C. parapsilosis isolates^(27,28). In the presented case, persistent positive blood cultures and the lack of clinical improvement justified the reassessment of antifungal therapy and escalation to liposomal amphotericin B.

The favorable neurological examination at discharge was reassuring; nevertheless, prematurity, invasive candidiasis and possible CNS inflammatory involvement support continued neurological and neurodevelopmental follow-up. Overall, this case highlights the difficulty of confirming or excluding neonatal CNS candidiasis when cerebrospinal fluid culture is negative despite biochemical and neuroimaging abnormalities, emphasizing the importance of a multimodal diagnostic approach.

Conclusions

Neonatal Candida parapsilosis candidemia may present with nonspecific clinical deterioration and can be complicated by suspected CNS involvement, particularly in premature infants exposed to central venous access and parenteral nutrition. In the presented case, markedly abnormal cerebrospinal fluid findings together with sonographic features suggestive of ventriculitis and meningeal inflammation raised concern for CNS dissemination despite a negative fungal CSF culture.

A negative CSF culture should therefore not be considered sufficient to exclude CNS candidiasis when other clinical and paraclinical findings remain suggestive. At the same time, abnormal CSF biochemistry and nonspecific neuroimaging findings should not be regarded as proof of fungal meningitis. The diagnosis should be based on the integrated assessment of bloodstream and cerebrospinal fluid microbiology, CSF cytochemistry, neurological findings and serial neuroimaging.

Persistent candidemia despite documented antifungal susceptibility should additionally prompt reassessment of the therapeutic strategy, investigation of intravascular and deep-tissue sources, and careful consideration of source control. Long-term neurological and neurodevelopmental surveillance remains appropriate in preterm infants recovering from invasive candidiasis, particularly when CNS involvement cannot be definitively excluded.

This work is permanently accessible online free of charge and published under the CC-BY.

Corresponding author: Miruna-Mihaela Rotar E-mail: miruna-mihaela.aldescu@ drd.umfcd.ro

Conflict of interests: none declared.

Financial support: none declared.