Table 1.
Characteristics of included trials
| Author (year) reference | Diagnosis of PCP | Baseline treatment for PCP | Oxygenation entry criteria | Corticoid (route)/initial daily dose/duration (days) | Interval (max.) * | Blinding patients/care givers/assessors | Concealed allocation/n centres | Sample size calculation/stopped early | Randomised individuals to I/C | Total deaths in I/C |
| Clement et al. (1989) [9] | BAL, sputum | 88% TMP-SMX, 12% Pentamidine | PaO2<51 mmHg (room air) | Methylprednisolon (IV)/240 mg/8 d | Unlimited | Yes/Yes/Unclear | Unclear/1 | Unclear/No | 19/22 | 9/9 at 56 days |
| Montaner et al. (1990) [8] | BAL | TMP-SMX, Pentamidine, Dapsone-TMP | 85–90% O2-Saturation † | Prednisone (oral)/60 mg/7 d with 14 d tapering | 48 h | Yes/Yes/Unclear | Yes/1 | Yes/Yes | 18/19 | 1/0 at 30 days 2/1 at 90 days |
| Bozzette et al. (1990) [16] | 75% BAL, 15% sputum + presumed | 80% TMP-SMX, 18% Pentamidine, 2% Dapsone-TMP | Hypoxemia ratio >75 ‡ | Prednisone (oral)/80 mg/21 d or as baseline treatment | 36 h | No/No/Unclear | Yes/6 | Yes/No | 123/128 | 13/28 at 31 days 20/33 at 84 days |
| Gagnon et al. (1990) [7] | BAL, biopsy, sputum | TMP-SMX | PaO2<75 mmHg (35% oxygen) | Methylprednisolon (IV)/160 mg/7–10 d | 72 h | Yes/Yes/Unclear | Unclear/1 | Yes/Yes | 12/11 | 3/9 at 28 days 5/9 at 120 days |
| Nielsen et al. (1992) [6] | BAL, biopsy | TMP-SMX | PaO2<67.5 mmHg (room air) | Methylprednisolon (IV)/2 mg/kg/10 d | 24 h | No/No/Unclear | Unclear/3 | Unclear/Yes | 30/29 | 2/9 at 34 days 4/9 at 90 days |
| Walmsley et al. (1995) [19] | BAL, biopsy, sputum | 82% TMP-SMX, 17% Pentamidine, 1% Dapsone-TMP | PaO2<70 mmHg (room air) § | Methylprednisolon (IV)/80 mg/10 d | 24 h | Yes/Yes/Unclear | Yes/3 | Yes/No | 40/38 | 4/6 at 35 days |
Abbreviations: PCP, Pneumocystis jiroveci pneumonia; n, number; I/C, intervention/control group; BAL, bronchoalveolar lavage; TMP, trimethoprim; SMX, sulfamethoxazole; IV, intravenous.
* Maximal interval between initiation of baseline treatment for Pneumocystis jiroveci pneumonia and initiation of corticosteroid.
† Or 5% decrease on exercise.
‡ PaO2 divided by fraction of inspired oxygen.
§ Or alveolar-arterial oxygen gradient >40 mmHg if arterial blood gases could not be assessed on room air.