Effect size of improvement rate, reduction in pain and increase in grip strength for corticosteroid injection
| Short term | Intermediate term | Long term | |
|---|---|---|---|
| 4–12 weeks | 26 weeks | 52 weeks | |
| Overall improvement RR (95% CI) RR>1 favours treatment | |||
| CSI vs no intervention or NSAIDs | |||
| Bisset | 2.94 (1.90 to 4.45)* | 0.55 (0.41 to 0.73)* | 0.75 (0.62 to 0.90)* |
| Coombes | 7.32 (2.83 to 18.94)* | 0.68 (0.50 to 0.92)* | 0.91 (0.77 to 1.06) |
| Hay | 1.60 (1.18 to 2.17)* | 0.77 (0.60to 0.98)* | 1.07 (0.88 to 1.30) |
| Smidt | 2.86 (1.96 to 4.16)* | – | 0.84 (0.68 to1.02) |
| Toker | 2.27 (1.04 to 4.97) * | – | – |
| Pooled | – | 0.66 (0.53 to 0.81) * | 0.87 (0.73 to 1.04) |
| Heterogeneity | >65% | p=0.21, I2=35% | p=0.07, I2=58% |
| Pain (negative value favours treatment) SMD (95% CI) | |||
| CSI vs no intervention or NSAIDs | |||
| Bisset | −1.43 (−1.83 to −1.04)* | 0.40 (0.04 to 0.76)* | 0.27 (−0.08 to 0.62) |
| Coombes | −2.14 (−2.68 to −1.60)* | 0.16 (−0.28 to 0.59) | 0.08 (−0.35 to 0.52) |
| Hay | −1.05 (−1.45 to−0.66)* | 0.42 (0.04 to 0.80)* | 1.35 (0.94 to 1.76)* |
| Smidt | −1.49 (1.89 to−1.08)* | 0.27 (−0.09 to 0.63) | 0.15 (−0.20 to 0.51) |
| Toker | −1.14 (−2.07 to −0.22)* | – | – |
| Pooled | −1.43 (−1.64 to −1.23)* | 0.32 (0.13 to 0.51)* | |
| Heterogeneity | p=0.032, I2=62% | p=0.79, I2=0% | >65% |
| CSI vs lidocaine injection | |||
| Lindenhovius | −0.25 (−0.74 to 0.24) | 0.27 (−0.30 to 0.84) | – |
| Price 1 | −1.06 (−1.63 to −0.49)* | 3.13 (2.31 to 3.95)* | – |
| Price 2 | −3.37 (−4.20 to −2.54)* | 1.55 (0.93 to 2.17)* | |
| Pooled | – | – | – |
| Heterogeneity | >65% | >65% | – |
| All above pooled | – | – | – |
| Heterogeneity | >65% | >65% | – |
| CSI, exercise and stretching vs exercise and stretching | |||
| Newcomer† | 0.16 (−0.49 to 0.81) | −0.37 (−1.04 to 0.30) | – |
| Maximum grip strength (positive value favours treatment) SMD (95% CI) | |||
| CSI vs no intervention or NSAIDs | |||
| Smidt | −1.42 (−1.82 to −1.03)* | −0.38 (−0.74 to −0.02)* | −0.36 (−0.72 to 0.002) |
| No pooling | – | – | |
| CSI vs lidocaine injection | |||
| Lindenhovius | −0.19 (−0.68 to 0.30) | 0.07 (−0.50 to 0.64) | |
| Price 1 | −0.06 (−0.59 to 0.48) | −0.98 (−1,58 to −0.38)* | |
| Price 2 | 2.31 (1.62 to 3.00)* | −0.86 (−1.44 to −0.29)* | – |
| Pooled | – | – | |
| Heterogeneity | >65% | >65% | |
| All above pooled | – | −0.48 (−0.73 to −0.24)* | – |
| Heterogeneity | >65% | p=0.04, I2=64% | |
| CSI, exercise and stretching vs exercise and stretching | |||
| Newcomer† | −0.17 (−0.61 to 0.27) | – | – |
*Statistically significant (p<0.05); Price 1: hydrocortisone versus lidocaine and change in pain-free grip strength and Price 2: triamcinolone versus lidocaine.
†The values for Newcomer are given as change in pain.
CSI, corticosteroid injection; NSAIDS, non-steroidal anti-inflammatory drugs; RR, relative risk; SMD, standard mean difference.