Table 4.
GRADE evidence profile of the effect of corticosteroid injection on function
| Quality assessment | Summary of findings | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Comparison | No. of trials | Limitations | Inconsistency | Indirectness | Imprecision | Publication bias | Participants | Effect size (95% CI)a |
GRADE | |
| Corticosteroid injection | Comparator | |||||||||
| Corticosteroid injection vs physical therapy | ||||||||||
| Short term | 2 [25, 52] | Very serious limitationsb | Serious inconsistencyc | No serious indirectness | No serious imprecision | Undetected | 49 | 50 | −0.69 (−1.31, − 0.07) | Low |
| Medium term | 2 [25, 52] | Very serious limitationsb | Serious inconsistencyc | No serious indirectness | Serious imprecisiond | Undetected | 49 | 50 | −0.55 (− 1.14, 0.03) | Very low |
| Corticosteroid injection vs foot orthoses | ||||||||||
| Short term | 2 [61, 64] | Very serious limitationsb | Serious inconsistencyc | No serious indirectness | Serious imprecisiond | Undetected | 70 | 73 | −0.78 (−1.81, 0.25) | Very low |
| Corticosteroid injection vs extracorporeal shockwave therapy | ||||||||||
| Short term | 2 [41, 58] | Very serious limitationsb | No serious inconsistency | No serious indirectness | Serious imprecisiond | Undetected | 90 | 86 | 0.11 (−0.18, 0.41) | Very low |
| Medium term | 2 [41, 58] | Very serious limitationsb | No serious inconsistency | No serious indirectness | Serious imprecisiond | Undetected | 90 | 86 | 0.21 (−0.08, 0.51) | Very low |
| Corticosteroid injection vs platelet-rich plasma injection | ||||||||||
| Short term | 3 [21, 36, 58] | Very serious limitationsb | No serious inconsistency | No serious indirectness | No serious imprecision | Undetected | 94 | 93 | −0.18 (−0.47, 0.10) | Low |
| Medium term | 3 [21, 36, 58] | Very serious limitationsb | No serious inconsistency | No serious indirectness | Serious imprecisiond | Undetected | 94 | 93 | 0.10(−0.18, 0.39) | Very low |
| Longer term | 3 [21, 36, 58] | Very serious limitationsb | No serious inconsistency | No serious indirectness | No serious imprecision | Undetected | 94 | 93 | 0.21 (−0.08, 0.49) | Low |
| Corticosteroid injection vs botulinum toxin-A injection | ||||||||||
| Short term | 2 [27, 28] | Very serious limitationsb | Serious inconsistencyc | No serious indirectness | Serious imprecisiond | Undetected | 45 | 45 | 0.76 (−0.24, 1.76) | Very low |
Abbreviations: CI Confidence interval, GRADE Grading Recommendations Assessment, Development and Evaluation
a Negative values indicate that the effect size (SMD) favours corticosteroid injection
b All participants for this outcome were from trials rated at high risk of bias
c Rated down 1 level for consistency as there was significant heterogeneity (i.e. I2 greater than 40%)
d Rated down 1 level as the upper and lower boundaries of the confidence intervals represent different conclusions