Table 5.
GRADE for quality of evidence profile.
| Included studies | Outcomes | No. of RCTs (participants, E/C) | Certainty assessment | Effect estimate (95% CI) | P-value | Quality of evidence | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Risk of bias | Inconsistency | Indirectness | Imprecision | Publication bias | ||||||
| Song et al. (2016) | Total effectiveness rate | 6 (272/268) | Seriousa | Not serious | Not serious | Not serious | Seriousd | RR 1.09 (1.03, 1.17) | 0.005 | Low |
| Pain score | 4 (179/160) | Seriousa | Not serious | Not serious | Not serious | Seriousd | SMD −0.17 (−0.39, 0.05) | 0.12 | Low | |
| Li et al. (2016) | WOMAC scale | 2 (157/165) | Seriousa | Not serious | Not serious | Seriousc | Seriousd | MD 17.63 (−23.15, 58.41) | 0.40 | Very low |
| WOMAC pain score | 2 (157/165) | Seriousa | Not serious | Not serious | Seriousc | Seriousd | MD 13.45 (−26.99, 53.89) | 0.51 | Very low | |
| SF-36 scale | 2 (171/177) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD 0.98 (−0.97, 2.93) | 0.32 | Low | |
| Wang et al. (2017) | Total effectiveness rate | 8 (292/288) | Seriousa | Not serious | Not serious | Not serious | Seriousd | RR 1.18 (1.10, 1.27) | <0.0001 | Low |
| Pain score | 3 (124/107) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD −1.35 (−1.67, −1.02) | <0.0001 | Low | |
| Lysholm score | 3 (85/80) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD 1.45 (0.82, 2.08) | <0.0001 | Low | |
| Ma et al. (2017) | Pain score | 4 (179/160) | Seriousa | Seriousb | Not serious | Not serious | Not serious | MD 0.64 (0.02, 1.27) | 0.04 | Low |
| WOMAC scale | 2 (171/177) | Not serious | Not serious | Not serious | Not serious | Not serious | MD 1.86 (1.50, 2.22) | <0.0001 | High | |
| SF-36 Scale | 2 (157/165) | Not serious | Not serious | Not serious | Not serious | Not serious | MD 0.94 (−0.04, 1.91) | 0.06 | High | |
| Zhang et al. (2017) | Total effectiveness rate | 10 (433/429) | Seriousa | Not serious | Not serious | Not serious | Seriousd | OR 3.26 (2.12, 5.02) | <0.0001 | Low |
| Fan et al. (2018) | Total effectiveness rate | 10 (439/437) | Seriousa | Not serious | Not serious | Not serious | Seriousd | RR 0.47 (0.33, 0.67) | None | Low |
| Pain score | 7 (298/281) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | SMD 1.53 (−2.11, −0.95) | None | Very Low | |
| WOMAC scale | 2 (134/134) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | SMD −0.85 (−1.11, −0.59) | None | Very low | |
| Lu et al. (2019) | Total effectiveness rate | 5 (184/179) | Seriousa | Not serious | Not serious | Not serious | Not serious | OR 3.68 (1.72, 7.87) | <0.05 | Moderate |
| WOMAC pain score | 2 (133/138) | Not serious | Not serious | Not serious | Not serious | Seriousd | MD −2.22 (−3.21, −1.24) | <0.05 | Moderate | |
| Lysholm score | 2 (70/63) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD −7.79 (4.21, 11.37) | <0.05 | Low | |
| Li et al. (2019) | Total effectiveness rate | 11 (505/507) | Seriousa | Not serious | Not serious | Not serious | Not serious | RR 1.21 (1.14, 1.28) | <0.0001 | Moderate |
| Pain score | 7 (301/266) | Seriousa | Seriousb | Not serious | Not serious | Not serious | MD −2.71 (−4.90, 0.52) | 0.02 | Low | |
| WOMAC scale | 4 (207/205) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | MD −6.79 (−12.35, −1.23) | 0.02 | Very Low | |
| WOMAC pain score | 3 (203/199) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | MD −1.34 (−2.50, −0.17) | 0.02 | Very Low | |
| Lysholm score | 1 (40/40) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD 12.13 (6.87, 17.39) | <0.01 | Low | |
| Lequesne index | 1 (30/30) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD −4.22 (−5.74, −2.70) | <0.01 | Low | |
| Deng et al. (2020) | Total effectiveness rate | 9 (478/449) | Seriousa | Not serious | Not serious | Not serious | Seriousd | OR 3.19 (2.07,4.90) | <0.0001 | Low |
| Pain score | 5 (205/206) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD −1.66 (−2.16, −1.16) | <0.0001 | Low | |
| WOMAC scale | 2 (104/104) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | MD −1.95 (−4.52,0.62) | 0.14 | Very Low | |
| Lequesne index | 1 (46/46) | Seriousa | Not serious | Not serious | Not serious | Seriousd | MD −5.29 (−5.97, −4.61) | <0.0001 | Low | |
| Zhang et al. (2021) | Total effectiveness rate | 12 (659/617) | Seriousa | Not serious | Not serious | Not serious | Seriousd | RR 1.13 (1.08, 1.18) | <0.0001 | Low |
| Pain score | 7 (273/273) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | SMD −1.41 (−2.07, −0.75) | <0.0001 | Very low | |
| WOMAC scale | 4 (187/186) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | SMD −1.23 (−2.39, −0.08) | 0.04 | Very low | |
| WOMAC pain score | 4 (182/180) | Seriousa | Seriousb | Not serious | Not serious | Seriousd | SMD −0.91 (−1.47, −0.34) | 0.002 | Very low | |
E, experimental group; C, control group.
a
The risk of bias is unclear in most of the studies.
b
The confidence interval overlap less, the heterogeneity test P was very small, and the I2 was larger (I2 threshold value: 50%).
c
The sample size is small, and the CI is wide.
d
Funnel plot was not symmetrical, or the number of included studies was small and all were positive results (sample size threshold value: 300).