Table 2.
Summary of pooled estimate of efficacy outcomes stratified by study design, heterogeneity among studies, and P-interaction between study designs.
| Outcome | Study design (number of studies) | Pooled effect estimates | Heterogeneity between studies | P-interaction between study types (RCT and OBS) |
|---|---|---|---|---|
| ASIA motor score improvement | RCT (3 studies) | Difference in means: 1.38; 95% CI: -11.9, 14.7; | I2 = 53.6%; P-heterogeneity = 0.12 |
P = 0.43 |
| OBS (2 studies) | Difference in means: 9.62; 95% CI: -5.92, 25.1; | I2 = 90.6%; P-heterogeneity<0.01 |
||
| Neurological improvementa | RCT (1 study) | RR: 1.50; 95% CI: 0.12, 18.4; |
NA (1 study) | P = 0.62 |
| OBS (2 studies) | RR: 0.78; 95% CI: 0.54, 1.13; |
I2 = 0%; P-heterogeneity = 0.34 |
Abbreviations: OBS: Observational study; RCT = Randomized controlled trial.
a
Neurological improvement was measured by one letter grade improvement from baseline in ASIA impairment scale or Frankel impairment scale.