Table 5.
Subgroup analysis in the propensity-score matched corticosteroid group.
| Initial dose of methylprednisolone (N = 106) | Administration period (N = 118) | Timing of corticosteroids initiation (N = 118) | ||||
|---|---|---|---|---|---|---|
| Pulse/semi-pulse | Standard dose | ≤ 10 days | > 11 days | On day 1 | on day 2 or 3 | |
| (N = 30) | (N = 76) | (N = 55) | (N = 63) | (N = 70) | (N = 48) | |
| Time to PCR negativity of the swab solution | ||||||
| Median [IQR]—days | 22 [15, 24] | 15 [9, 24] | 15 [8, 24] | 21 [14, 26] | 15 [9, 24] | 20 [15, 27] |
| Hazard ratio (95%CI) | 0.854 (0.537–1.358) | 1.437 (0.968–2.132) | 0.753 (0.504–1.124) | |||
| p-value | 0.505 | 0.072 | 0.165 | |||
| Duration of fever | ||||||
| Median [IQR]—days | 8 [4, 16] | 10 [5, 19] | 6 [4, 15] | 11 [6, 28] | 8 [4, 15] | 10 [4, 19] |
| Hazard ratio (95%CI) | 1.037 (0.639–0.1.683) | 1.337 (0.886–2.019) | 0.927 (0.610–1.407) | |||
| p-value | 0.882 | 0.167 | 0.720 | |||
| Time to improvement in radiological findings | ||||||
| Median [IQR]—days | 11 [4, 23] | 8 [6, 16] | 7 [5, 14] | 9 [6, 18] | 7 [5, 16] | 8.5 [6, 18] |
| Hazard ratio (95%CI) | 0.880 (0.554–1.399) | 1.124 (0.760–1.662) | 0.855 (0.576–1.271) | |||
| p-value | 0.589 | 0.557 | 0.439 | |||
| Invasive mechanical ventilation—No. (%) | 17 (56.7) | 17 (22.4) | 9 (16.4) | 31 (49.2) | 18 (25.7) | 22 (45.8) |
| Duration of invasive mechanical ventilation | ||||||
| Median [IQR]—days | 8 [6, 10] | 15 [10, 26] | 10 [8, 16] | 11 [8, 18] | 10 [7, 13] | 11.5 [9, 23] |
| Hazard ratio (95%CI) | 2.831 (1.347–5.950) | 1.247 (0.588–2.646) | 0.713 (0.374–1.358) | |||
| p-value | 0.006 | 0.565 | 0.304 | |||
| Mortality—No. (%) | ||||||
| On day 14 | 2 (6.7) | 5 (6.6) | 5 (9.1) | 2 (3.2) | 3 (4.3) | 4 (8.3) |
| On day 28 | 4 (13.3) | 6 (7.9) | 6 (10.9) | 6 (9.5) | 6 (8.6) | 6 (12.5) |
| During the entire observation period | 4 (13.3) | 8 (10.5) | 7 (12.7) | 7 (11.1) | 6 (8.6) | 8 (16.7) |
Categorical variables were presented as numbers (percentages). Continuous variables related to time were presented as median [interquartile ranges]. Cox proportional hazards model was used to calculate the hazard ratio and its 95% confidence interval for the treatment effect between groups. A p value of < 0.05 was considered statistically significant. IQR, interquartile ranges; CI, confidence interval; PCR, polymerase chain reaction.