C-reactive protein, white blood cell counts and sequential interpretation of PfHRP2/pLDH for antibiotic stewardship in children under 5 years of age

In sub-Saharan Africa, the management of non-malaria acute febrile illnesses remains challenging in peripheral health centers without laboratory facilities. This study retrospectively assessed diagnostic approaches constructed through the integration of biological data and clinical information from established database to assess their potential impact on antibiotics prescribing practices within an antimicrobial stewardship framework in malaria endemic areas. Data from 396 febrile children (axillary temperature ≥37.5 °C) under 5 years of age, collected between April and December 2016 were retrospectively analyzed. Diagnostic approaches integrating malaria RDT results (sequential interpretation of PfHRP2/pLDH and PfHRP2-only), C-reactive protein (CRP), white blood cell (WBC) counts, microbiological findings (blood, stool, and urine) and recorded antibiotic prescription were assessed. Association between malaria diagnostic results and clinical/biological data were assessed using logistical regression, adjusted for age, sex, axillary temperature, CRP value, WBC count and microbiological findings. Malaria sequential diagnostic results with malaria RDT-PfHRP2/pLDH were interpreted and reported as either (i) positive when the T2-pLDH line appear, regardless of the results of T1-HRP2 line, (ii) negative when both T1-HRP2 and T2-pLDH lines do not appear, or (iii) undetermined when the T1-HRP2 line only appears. Using malaria sequential diagnostic approach, logistic regression of malaria-negative or undetermined results showed a negative correlation with axillary temperature > 38.5 °C (aOR 0.37; 95% CI 0.24–0.58; p 14 × 103/µL (aOR 2.57; 95% CI 1.39–4.79, p = 0.003), compared with malaria-positive results. In children with malaria-negative or undetermined sequential results, negative malaria tests showed a positive correlation with CRP ≥10 mg/L (aOR 3.46; 95% CI 1.51–8.82; p = 0.005), but a negative correlation with WBC counts > 14 × 103/µL (aOR 0.38; 95% CI 0.15–0.93; p = 0.037), compared with malaria-undetermined results. The optimal diagnostic approach combining PfHRP2-only results with CRP values and WBC counts predicted the need for antibiotic prescriptions in 18.7% (74/396), potentially identifying 9/23 sepsis cases. When using sequential malaria diagnostic approach combined with CRP values and WBC counts, antibiotic need was predicted in 25.5% (101/396), potentially identifying 18/23 sepsis cases. Integrating sequential malaria diagnostics with CRP, WBC counts, and clinical information improves differentiation of febrile illnesses and supports more targeted antibiotic use in malaria-endemic settings. Not applicable.