for the main comparison.
| Acupuncture + baseline treatment versus baseline treatment alone | ||||||
|
Patient or population: adults with stroke Settings: inpatients Intervention: Acupuncture + baseline treatment Comparison: baseline treatment | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Risk with baseline treatment | Risk with Acupuncture + baseline treatment | |||||
| Improvement of dependency at the end of treatment assessed with Barthel Index | The mean improvement of dependency at the end of treatment was 0 | The mean improvement of dependency at the end of treatment in the intervention group was 9.19 undefined more (4.34 more to 14.05 more) | ‐ | 616 (9 RCTs) | ⊕⊝⊝⊝ VERY LOW1,2 | Substantial heterogeneity in results. Most studies were at high or unclear risk of bias. All of the studies were carried out in China |
| Improvement of global neurological deficit at the end of treatment assessed with Modified Edinburgh and Scandinavian Stroke Scale | The mean improvement of global neurological deficit at the end of treatment was 0 | The mean improvement of global neurological deficit at the end of treatment in the intervention group was 2.39 undefined fewer (3.34 fewer to 1.43 fewer) | ‐ | 240 (4 RCTs) | ⊕⊕⊝⊝ LOW1 | Most studies were at high or unclear risk of bias. All of the studies were carried out in China |
| Improvement of global neurological deficit at the end of treatment | Study population | OR 3.89 (1.78 to 8.49) | 543 (7 RCTs) | ⊕⊕⊝⊝ LOW1 | Most studies were at high or unclear risk of bias. All of the studies were carried out in China | |
| 674 per 1000 | 890 per 1000 (787 to 946) | |||||
| Moderate | ||||||
| 733 per 1000 | 914 per 1000 (830 to 959) | |||||
| Improvement of motor function at the end of treatment ‐ upper and lower extremities motor function (FMA) assessed with Fugl‐Meyer Assessment | The mean improvement of motor function at the end of treatment ‐ upper and lower extremities motor function was 0 | The mean improvement of motor function at the end of treatment ‐ upper and lower extremities motor function in the intervention group was 6.16 undefined more (4.2 more to 8.11 more) | ‐ | 245 (4 RCTs) | ⊕⊕⊝⊝ LOW1 | Most studies were at high or unclear risk of bias. All of the studies were carried out in China |
| Improvement of motor function at the end of treatment ‐ general motor function assessed with Motor assessment scale | The mean improvement of motor function at the end of treatment ‐ general motor function was 0 | The mean improvement of motor function at the end of treatment ‐ general motor function in the intervention group was 4.53 undefined more (2.99 more to 6.07 more) | ‐ | 60 (1 RCT) | ⊕⊕⊝⊝ LOW1 | |
| Improvment of general motor function at the end of follow up assessed with Fugl‐Meyer Assessment follow‐up: mean 3 months | The mean improvement of general motor function at the end of follow‐up was 0 | The mean improvement of general motor function at the end of follow‐up in the intervention group was 7.59 more (0.98 more to 14.2 more) | ‐ | (1 RCT) | ⊕⊕⊕⊝ MODERATE3 | |
| Improvement of motor function at the end of treatment assessed with Fugl‐Meyer Assessment | Study population | OR 2.41 (0.98 to 5.96) | 125 (2 RCTs) | ⊕⊕⊝⊝ LOW1 | ||
| 710 per 1000 | 855 per 1000 (705 to 936) | |||||
| Moderate | ||||||
| 720 per 1000 | 861 per 1000 (716 to 939) | |||||
| Improvement of cognitive function at the end of treatment assessed with Mini‐mental state examination | The mean improvement of cognitive function at the end of treatment was 0 | The mean improvement of cognitive function at the end of treatment in the intervention group was 2.54 undefined more (0.03 more to 5.05 more) | ‐ | 278 (5 RCTs) | ⊕⊝⊝⊝ VERY LOW1,2 | Substantial heterogeneity in results. Most studies were at high or unclear risk of bias. All of the studies were carried out in China |
| Improvment of cognitive function at the end of follow‐up
assessed with Mini‐mental state examination follow‐up: 1 month |
The mean improvement of cognitive function at the end of follow up was 0 | The mean improvement of cognitive function at the end of follow‐up in the intervention group was 3.47 undefined more (2.43 more to 4.51 more) | ‐ | 71 (1 RCT) | ⊕⊕⊝⊝ LOW1 | |
| Improvement of cognitive function at the end of treatment assessed with: Mini‐mental state examination | Study population | OR 3.82 (1.89 to 7.72) | 166 (3 RCTs) | ⊕⊕⊝⊝ LOW1 | ||
| 512 per 1000 | 800 per 1000 (665 to 890) | |||||
| Moderate | ||||||
| 533 per 1000 | 814 per 1000 (684 to 898) | |||||
| Improvement of depression at the end of treatment assessed with Hamilton Depression Scale | The mean improvement of depression at the end of treatment was 0 | The mean improvement of depression at the end of treatment in the intervention group was 2.58 undefined fewer (3.28 fewer to 1.87 fewer) | ‐ | 552 (6 RCTs) | ⊕⊝⊝⊝ VERY LOW1,2 | Substantial heterogeneity in results. Most studies were at high or unclear risk of bias. All of the studies were carried out in China |
| Improvement of depression at the end of treatment assessed with Hamilton Depression Scale | Study population | OR 2.03 (1.10 to 3.72) | 342 (4 RCTs) | ⊕⊕⊝⊝ LOW1 | Most studies were at high or unclear risk of bias. All of the studies were carried out in China | |
| 784 per 1000 | 880 per 1000 (799 to 931) | |||||
| Moderate | ||||||
| 807 per 1000 | 894 per 1000 (821 to 939) | |||||
| Improvement of swallowing function at the end of treatment | The mean improvement of swallowing function at the end of treatment was 0 | The mean improvement of swallowing function at the end of treatment in the intervention group was 1.11 undefined fewer (2.08 fewer to 0.14 fewer) | ‐ | 200 (2 RCTs) | ⊕⊝⊝⊝ VERY LOW1,2 | |
| Improvement of pain at the end of treatment assessed with Visual Analogue Scale | The mean improvement of pain at the end of treatment was 0 | The mean improvement of pain at the end of treatment in the intervention group was 2.88 undefined fewer (3.68 fewer to 2.09 fewer) | ‐ | 118 (2 RCTs) | ⊕⊕⊝⊝ LOW1 | |
| Improvement of sleep quality at the end of treatment assessed with Rhone Planck Sleepiness Scale | The mean improvement of sleep quality at the end of treatment was 0 | The mean improvement of sleep quality at the end of treatment in the intervention group was 1.09 undefined fewer (2.37 fewer to 0.19 more) | ‐ | 60 (1 RCT) | ⊕⊕⊕⊝ MODERATE3 | |
| Improvement of spasticity at the end of treatment assessed with Modified Ashworth Spasticity Rating Scale | The mean improvement of spasticity at the end of treatment was 0 | The mean improvement of spasticity at the end of treatment in the intervention group was 0.4 undefined fewer (0.64 fewer to 0.16 fewer) | ‐ | 60 (1 RCT) | ⊕⊕⊝⊝ LOW1 | |
| Improvement of quality of life at the end of treatment assessed with MOS SF‐36 | The mean improvement of quality of life was 0 | The mean improvement of quality of life in the intervention group was 2.73 undefined more (0.54 fewer to 6 more) | ‐ | 71 (1 RCT) | ⊕⊕⊝⊝ LOW1 | |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; OR: Odds ratio; | ||||||
| GRADE Working Group grades of evidence High quality: We are very confident that the true effect lies close to that of the estimate of the effect Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect | ||||||
1 Downgraded by two levels for very serious risk of bias (none of the trials used adequate allocation concealment, nor blinding of participants or researchers. Also, most of them were at risk of attrition bias). 2 Downgraded by one level for serious inconsistency (due to substantial heterogeneity, I2 = 57%). 3 Downgraded by one level for this trial did not use blinding of participants or researchers.