Risk Factors and Clinical Management Strategies for Bloodstream Infections in Pediatric Hematological Malignancies.

PURPOSE: Chemotherapy remains the primary treatment for haematological malignancies (HMs). While recent therapeutic advances have improved patient survival, treatment-induced immunosuppression and prolonged neutropenia significantly elevate the risks of bloodstream infection (BSI), contributing to higher mortality. This study identifies risk factors for BSI in pediatric HM patients, aiming to establish evidence-based protocols for infection prevention and risk stratification, thereby informing targeted healthcare policies to reduce complications and improve treatment outcomes.

PATIENTS AND

METHODS

We retrospectively analyzed 682 pediatric HM patients presenting with fever at our institution, including 98 BSI-confirmed cases.Results were statistically compared across multiple aspects.

RESULTS

Pediatric HM patients with bloodstream infection showed significantly higher AML prevalence (47.96% vs 33.22%, P = 0.005) and Gram-negative bacteria predominance (66.33%), notably Escherichia coli (E. coli) (27.55%). Stenotrophomonas maltophilia infection rates were significantly higher in lymphoma patients than in other hematological malignancies (P < 0.05). Multivariable analysis identified four modifiable risk factors: peak body temperature (OR = 5.468, 95% CI 3.407-8.775, P < 0.001), duration of neutropenia (OR = 1.181, 95% CI 1.120-1.245, P < 0.001), febrile neutropenia (OR = 8.193, 95% CI 3.574-18.780, P < 0.001), and invasive procedures (OR = 4.265, 95% CI 1.920-9.474, P < 0.001).

CONCLUSION

To reduce BSI complications in pediatric HM patients, we recommend implementing risk-stratified temperature protocols and strict neutropenia control (≤7 days), emphasizing rigorous clinical criteria for invasive procedures (PICC). These findings provide an evidence base for healthcare policies aimed at reducing infection-related mortality through optimized treatment protocols and enhanced clinical care standards.