Neonatal fungal sepsis in very low birth weight infants

When a baby weighing under 1500 grams arrives into the world weeks earlier than expected, every hour in the NICU is a calculated compromise between supportive intervention and unavoidable harm. Among the many threats these fragile infants face, fungal bloodstream infection stands out as one of the most insidious, often emerging quietly after days or weeks of intensive care. Incidence figures reported by the Australian and New Zealand Neonatal Network (ANZNN) sit within global ranges, with invasive Candida infection complicating the course of roughly 1–3% of very low birth weight (VLBW) admissions in major tertiary centres such as the Royal Women's Hospital in Melbourne and Westmead Hospital in Western Sydney. Although numbers appear small on a population level, attributable mortality remains stubbornly high, particularly when treatment is delayed beyond the first 48 hours of clinical deterioration.

The clinical challenge is that fungal sepsis in premature infants often mimics bacterial sepsis, necrotising enterocolitis, or simply a non-specific downturn, leading to empirical antibacterial therapy that may worsen the underlying problem by promoting fungal overgrowth. Strong awareness of risk factors remains the most reliable path to earlier diagnosis and timely initiation of antifungal therapy. As perinatal specialists gathered through the INS 2018 conference preview and successor meetings have repeatedly emphasised, prevention-oriented neonatal care is far more achievable when teams can anticipate which infants are most vulnerable.

Understanding neonatal fungal sepsis

Invasive fungal infection in the neonatal period is dominated by organisms of the Candida genus, with Candida albicans and Candida parapsilosis together responsible for the great majority of bloodstream isolates in NICU patients worldwide. Other species, including C. glabrata, C. krusei, and the emerging C. auris, appear less frequently but carry worrying resistance profiles and remain a subject of active surveillance through Australia's Therapeutic Goods Administration (TGA) and global antimicrobial resistance programmes.

Fungal sepsis differs from bacterial sepsis in several clinically important ways. The onset is typically later in the NICU stay, often after the first week of life rather than in the first 72 hours, and frequently follows a prolonged course of broad-spectrum antibiotics, central venous catheter use, or total parenteral nutrition. Haematogenous spread from a gastrointestinal or cutaneous source can lead to end-organ complications such as endophthalmitis, renal candidiasis, fungal meningitis, and peritoneal abscesses, all of which alter the prognosis significantly. Once established, infection also disrupts local microbiome resilience, complicating decisions about future empirical therapy.

Key risk factors in VLBW infants

Risk factors for invasive fungal sepsis are best understood as cumulative rather than independent. A VLBW infant with a birth weight below 1000 grams who requires prolonged parenteral nutrition through a central line, who has received multiple courses of broad-spectrum antibiotics, and who has been exposed to antenatal steroids is far more vulnerable than an infant with any single risk factor in isolation. Identifying these overlapping contributors is the cornerstone of any effective prevention programme.

Other commonly cited contributors include mucosal disruption from necrotising enterocolitis, histamine-2 receptor antagonist use where still prescribed, and abdominal surgery. Australian neonatal units have largely moved away from routine H2 blockers in preterm infants following updated local guidance, although proton pump inhibitor use continues in specific subgroups where reflux is clinically severe.

Clinical signs and diagnostic pathways

Recognising fungal sepsis requires a high index of suspicion. Apnoea, bradycardia, temperature instability, abdominal distension, and worsening respiratory status are common but non-specific. Thrombocytopenia, especially when disproportionately severe compared with overall clinical condition, is a recognised clue. Some infants develop a characteristic erythematous papular or maculopapular rash, although this finding is not consistent enough to rely on.

Blood culture remains the diagnostic cornerstone, though its sensitivity in low-colony-count fungaemia is imperfect. Adjunctive diagnostics, including beta-D-glucan assay and PCR-based Candida detection, are used selectively in larger Australian tertiary units and can shorten the time to a presumptive diagnosis. Lumbar puncture, abdominal ultrasound, and ophthalmological examination are recommended once a candidal bloodstream infection is confirmed, because of the propensity for end-organ seeding. Echocardiography is advisable when persistent fungaemia raises concern for endocarditis, particularly in infants with prolonged central venous access.

Empirical antifungal therapy should be considered in any VLBW infant with a deteriorating clinical picture who has been colonised with Candida on previous surface cultures and who has multiple cumulative risk factors, even before culture confirmation returns. The threshold for starting treatment is lower in units that have previously documented higher baseline incidence.

Prevention strategies and antifungal prophylaxis

Targeted prevention in everyday practice

• Limit the duration of broad-spectrum antibiotic courses through daily review of empirical therapy.
• Use peripheral access where possible and remove central lines as soon as feeding tolerance allows.
• Encourage early enteral feeding, ideally with maternal breast milk, to support healthy gut colonisation.
• Consider antifungal prophylaxis in infants with birth weight below 1000 grams and multiple additional risk factors.

Antifungal agents commonly used in Australia

• Fluconazole (Diflucan) — oral and IV formulations listed on the Pharmaceutical Benefits Scheme for specific indications.
• Amphotericin B (Fungizone) — reserved for invasive infection under hospital-only supply arrangements.
• Micafungin — listed through specialised authority mechanisms in selected Australian NICUs.
• Nystatin (Nilstat) — oral suspension for mucocutaneous candidiasis, not appropriate for fungaemia.

Unit-level bundles that combine antifungal prophylaxis with strict catheter-care protocols have been associated with substantial reductions in invasive candidiasis rates in published quality-improvement studies. Australian neonatal networks contribute data to the ANZNN registry, supporting local benchmarking and continuous review of prophylaxis policies. The role of prophylactic fluconazole in VLBW infants remains debated internationally, and local practice varies between tertiary units in Sydney, Melbourne, Brisbane, and Perth based on historical incidence and patient case-mix.

Outcomes and long-term implications

In survivors, neonatal fungal sepsis is associated with prolonged hospitalisation, additional neuro-imaging procedures, and higher rates of major neonatal morbidities including severe intraventricular haemorrhage, bronchopulmonary dysplasia, and retinopathy of prematurity. Long-term neurodevelopmental follow-up data suggest that infants who experienced invasive candidiasis are more likely to encounter motor and cognitive delays in early childhood compared with VLBW peers without fungal sepsis, although the contribution of confounding factors such as extreme prematurity itself remains substantial.

Parents often wish to understand why their baby became so unwell after a period of apparent stability. Sensitive, culturally aware conversations around the boundaries of care are essential, and exchanges through forums such as the FAOPS 2020 review of perinatal postmortem perspectives can support clinicians navigating these difficult discussions with families from diverse backgrounds.

Looking ahead, continued surveillance of antifungal resistance, refinement of prophylaxis policies, and stronger integration between midwifery and neonatal medical teams are central priorities. Perspectives shared on midwifery in Oceania and across the wider Asian region remind us that the smallest patients depend on collaborative, well-resourced perinatal systems wherever they are born.

If you are a clinician, midwife, or researcher interested in contributing to the FAOPS 2020 archive or joining the upcoming scientific dialogue on perinatal infections, we invite you to register your interest and share your work through our submission portal.