Table 3.
Meta-analysis results of outcomes reported by studies comparing MMF vs. AZA a.
| Outcome | Study Design (N) | Time in months | Relative Risk b | Statistics c | |
|---|---|---|---|---|---|
| (95% CI) | p | I2 | |||
| Total Infections | RCT [49,51,52,53] (919) | 12 | 1.17 (1.03, 1.33) | 0.01 | 0 |
| CMV | RCT [49,51,52,53] (919) | 12 | 0.94 (0.82, 1.03) | 0.17 | 41 |
| Abdominal pain | RCT [49,50,53] (873) | 12 | 1.40 (1.06, 1.83) | 0.02 | 14 |
| Diarrhea | RCT [49,50,53] (873) | 12 | 1.49 (1.17, 1.90) | 0.001 | 10 |
| Nausea | RCT [49,50,53] (873) | 12 | 0.98 (0.69, 1.39) | 0.91 | 41 |
| Vomiting | RCT [49,50,53] (873) | 12 | 1.54 (1.10, 2.15) | 0.01 | 0 |
| Malignancies | RCT [49,53] (652) | 12 | 1.52 (0.81, 2.82) | 0.19 | 0 |
| Withdraw | RCT [49,50,53] (873) | 12 | 1.21 (0.77, 1.92) | 0.40 | 66* |
a Results reaching statistical significance are in bold font. b Relatives risk values of <1 favor treatment with MMF. c p: p-value for relative risk estimation;I2: test for heterogeneity. * RCT Sadek 2002 [53] is largely responsible for the heterogeneity among RCTs that reported withdraw. Sensitivity analysis showed a relative risk of 1.93 (1.06 to 3.52), and dramatically reduced heterogeneity (p =0.95, I2 = 0%) when this trial was removed from the analysis.