Profile of Invasive Fungal Infections in a Paediatric Haemato-Oncology Unit: A Five-Year Retrospective Cohort Study From a Tertiary Care Centre in South India.

BACKGROUND

Invasive fungal infections (IFIs) are a leading cause of morbidity and mortality in paediatric haemato-oncology (PHO) patients. Empirical antifungal therapy in high-risk children must be informed by local epidemiological, species-level and antifungal-susceptibility data, yet contemporary Indian paediatric cohort data remain limited.

OBJECTIVE

To describe the clinical, microbiological, therapeutic and outcome profile of probable and proven IFI in a tertiary care South Indian PHO unit over five years, and to benchmark the findings against contemporary international literature.

METHODS

Retrospective observational cohort study of children aged ≤18 years admitted with probable or proven IFI (EORTC/MSGERC 2020 criteria) between January 2019 and December 2024. Baseline demographics, underlying disease, neutropenia profile, clinical presentation, mycological workup (cultures with CLSI M27 susceptibility testing, serum galactomannan (GM), beta-D-glucan and bronchoalveolar lavage), imaging, antifungal prophylaxis and therapy, central venous access device (CVAD) management and infection-episode outcomes were extracted and analysed descriptively.

RESULTS

Among 580 febrile PHO admissions, 33 episodes of probable/proven IFI occurred in 23 patients (overall incidence 5.8%). Median age was 8.5 years (range 1.5-17); 12/23 (52.2%) were male. Haemato-lymphoid malignancy was the underlying diagnosis in 17/23 (73.9%); two children had inborn errors of immunity (autoimmune lymphoproliferative syndrome, n = 1; hyper-IgM syndrome, n = 1). Three patients (13.0%) were haploidentical haematopoietic stem cell transplantation recipients. Neutropenia (ANC < 1500 cells/mm³) was present in 19/23 (82.6%), and profound neutropenia (ANC < 100 cells/mm³) in 7/23 (30%) at IFI onset. Ten patients (43.5%) had a breakthrough infection on prophylaxis. Candida spp. and Aspergillus spp. were the two dominant pathogen groups (each 43.5%; n = 10), followed by Mucorales (8.7%; n = 2) and Geotrichum (Magnusiomyces) capitatum (4.3%; n = 1). All 20 Candida episodes presented as fungemia; Candida tropicalis was the single most common isolate (12/20; 60.0%), and 20/20 (100%) were fluconazole-susceptible on CLSI M27 testing. Fluconazole was the first-line therapeutic agent in candidemia, informed by universal in vitro susceptibility and CVAD was removed in all 10 Candida-infected patients (100%). Serum GM was positive in 8/23 (34.8%). Radiological fungal pneumonia was documented in 10/23 (43.5%), with air-crescent, ground-glass, nodular and consolidative patterns in 11%, 10%, 7% and 15% of the cohort, respectively; sinusitis was identified in 2/23 (8.6%). Crude all-cause mortality was 6/23 (26.1%): 0 in the Candida group, 3/10 (30%) in aspergillosis, 2/2 (100%) in mucormycosis and 1/1 (100%) in Geotrichum infection.

CONCLUSION

In this contemporary South Indian PHO cohort, IFI incidence was 5.8% with a Candida-Aspergillus co-dominance, a marked C. tropicalis preponderance, universal fluconazole susceptibility permitting fluconazole-based first-line candidemia therapy, no candidaemia-attributable deaths alongside 100% CVAD removal, and high mortality in the small numbers of patients with mould infection. These data support locally informed empirical antifungal selection, aggressive CVAD-based source control and structured deployment of imaging and biomarker-driven diagnosis of invasive mould disease in Indian PHO.