Comparison of Different Corticosteroid Regimens for Refractory

Background: Refractory Methods: In this retrospective cohort study, 216 children with RMPP hospitalized between January 2023 and June 2025 were categorized by corticosteroid regimen: Group A (low-dose methylprednisolone, 1-2 mg/kg/day intravenously for 3-5 days, n=73), Group B (high-dose methylprednisolone pulse therapy, 10-30 mg/kg/day for 3 days, n=70), and Group C (step-down regimen: pulse therapy for 3 days followed by an oral prednisolone taper, n=73). The primary outcome was time to fever resolution after corticosteroid initiation. Multivariable logistic and Cox regression, inverse probability of treatment weighting (IPTW), and E-value analyses were performed. Results: Baseline characteristics were largely comparable, although white blood cell counts differed (P=0.013). Defervescence was faster in Groups B (2.40 ± 1.00 days) and C (2.00 ± 0.78 days) than in Group A (4.25 ± 1.71 days; both P<0.001); the Group C-versus-Group B difference (adjusted P=0.047) occurred largely before the oral taper began. Group C had the shortest hospital stay (8.26 ± 1.74 days) and the highest treatment effectiveness (89.04% vs 82.86% vs 64.38%; P=0.001), but the adjusted Group C-versus-Group B comparison was not significant (OR 2.17; 95% CI 0.77-6.12). Both high-dose regimens were independently associated with treatment effectiveness versus Group A (Group C: OR 5.26, 95% CI 2.08-13.32; E-value 2.16) and with faster defervescence; IPTW analyses yielded consistent results. Thirty-day readmission did not differ significantly (exact P=0.130). Conclusion: High-dose corticosteroid regimens were associated with faster defervescence and higher treatment effectiveness than low-dose therapy in children with RMPP. Whether an oral taper after pulse therapy provides additional benefit remains uncertain and warrants prospective validation.