Early loss of platelet response and rehospitalization following high-dose dexamethasone in patients with newly diagnosed ITP

Abstract Background: High-dose dexamethasone (HD-DXM) and prednisone are both standard first-line treatments for newly diagnosed immune thrombocytopenia (ITP). Current international guidelines recommend considering additional HD-DXM cycles every 2–4 weeks based on treatment response. Although HD-DXM is associated with a rapid initial platelet response, data regarding early post-treatment platelet decline and risk of rehospitalization remain limited. Methods: A single-center retrospective cohort study of consecutive adult patients hospitalized in a tertiary academic hospital, with newly diagnosed ITP between January 2012 and December 2024, who were treated with steroids as first-line therapy. Patients were stratified by initial corticosteroid regimen into three groups: Group A received dexamethasone 40 mg daily for 4 days, with additional courses given as needed based on platelet counts during ambulatory follow-up; Group B received oral prednisone with tapering down; and Group C received dexamethasone 40 mg for 4 days followed by oral prednisone with tapering down. The primary outcome was rehospitalization within 30 days due to low platelet counts. Secondary outcomes included 90-day rehospitalization, overall response (OR), defined as platelet count > 30×10⁹/L, and complete response (CR), defined as platelet count > 100×10⁹/L, Both responses were assessed on day 14 among patients who had not been rehospitalized prior to this time point. Steroid-free status at 6 weeks was also evaluated. Group comparisons were performed using Chi-square test for categorical variables, and ANOVA test for continuous variables, followed by pairwise comparisons when overall significance was reached. Cox proportional hazards models were used to assess the association between treatment group and time to rehospitalization. A p-value < 0.05 was considered statistically significant. Results: A total of 226 patients were included: 81 in Group A, 113 in Group B, and 32 in Group C. Median age in Group A was 60 years (IQR 39–73); 59% were female. Baseline characteristics and comorbidities were comparable, except for higher ischemic heart disease in Group C (15.6%) vs. Group A (1.2%) and Group B (7.1%). Platelet counts at admission were lower in Group C (median 5 × 10⁹/L, IQR 2.8–14.5) compared to Group A (13 × 10⁹/L, IQR 6–24) and Group B (14 × 10⁹/L, IQR 6–22.8; p = 0.006). Group C patients were also more likely to receive intravenous immunoglobulin during hospitalization (53% vs. 25% and 27% in Groups A and B; p = 0.008). Median hospital stay was longer in Group C (6 days, range 3–17), compared to 4 days in Groups A and B (ranges 2–25 and 2–24; p = 0.01). Rehospitalization within 30 days occurred in 37.0% (95% CI 27.3–47.9) of Group A, 11.5% (95% CI 6.8–18.7) of Group B, and 3.2% (95% CI 0.6–15.7) of Group C (p < 0.001). In a Cox regression model, Group A had a higher risk of 30-day readmission vs. Group B (HR = 1.35, 95% CI 0.70–2.63), although not statistically significant (p = 0.363). Median time to rehospitalization in Group A was 13.5 days (range 7–27), and median platelet count at readmission was 6 × 10⁹/L (range 2–25). Most rehospitalizations occurred within 30 days. Ninety-day readmission rates were 43.2% in Group A, 15% in Group B, and 9.3% in Group C (p < 0.001). Among the 30 patients in Group A rehospitalized within 30 days, 17 (57%) were readmitted before day 15, and 5 (17%) between days 15–30, despite a second dexamethasone cycle in the ambulatory setting. At day 14, OR among patients not rehospitalized was 75.6% (34/45; 95% CI 62.5–88.6) in Group A and 91.8% (89/97; 95% CI 86.2–97.3) in Group B (p = 0.008). In Group C, 86.2% (25/29; 95% CI 72.9–99.6) achieved an overall response. CR at day 14 was 48.9% (95% CI 34.3–63.5) in Group A, 56.7% (95% CI 46.8–66.6) in Group B, and 44.8% (95% CI 26.7–62.9) in Group C. These differences were not statistically significant. Conclusions: Initial treatment with high-dose dexamethasone in newly diagnosed ITP was associated with a high rate of early platelet decline and significantly increased rehospitalization. In contrast, regimens including oral prednisone at discharge were linked with more stable platelet counts and lower early readmissions. These findings challenge the assumption that a single HD-DXM cycle followed by reassessment at day 14 is sufficient as first-line therapy, highlighting the need for more durable early approaches.