Table 4.
Summary of findings.
| Outcomes | No. of participants (RCTs) | Anticipated absolute effects (95% CI) | Relative effect (95% CI) | I 2 value | Quality of evidence (GRADE) | Comments | |
|---|---|---|---|---|---|---|---|
| Risk with control group | Risk with CHM group | ||||||
| Total analysis | |||||||
| OFS | 704 (5) | — | SMD 2.31 lower (1.48–3.14 lower) | — | 94% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| SS | 2,289 (22) | — | SMD 1.82 higher (1.26–2.38 higher) | — | 97% | ⊕○○○ Very low |
Risk of bias (−1) Publication bias (−1) Inconsistency (−2) |
| LVA-BF | 1,778 (17) | — | MD 5.70 higher (4.18–7.22 higher) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−2) Strong association (+1) |
| RVA-BF | 1,778 (17) | — | MD 4.83 higher (3.37–6.29 higher) | — | 97% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−2) Publication bias (−1) Strong association (+1) |
| BA-BF | 1,888 (18) | — | MD 5.58 higher (4.24–6.92 higher) | — | 96% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−2) Publication bias (−1) Strong association (+1) |
| TER | 3,582 (33) | 295 per 1,000 | 450 per 1,000 (419–499) | RR 1.55 (1.42–1.69) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| ET level | 342 (3) | — | MD 14.57 lower (6.81–22.32 lower) | — | 96% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−2) |
| CGRP level | 426 (4) | — | MD 6.24 higher (4.37–8.11 higher) | — | 96% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−2) |
| Fib level (vs. AT) | 348 (4) | — | MD 0.31 lower (0.12–0.50 lower) | — | 97% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−2) |
| TC level (vs. AT) | 348 (4) | — | MD 0.56 lower (0.31–0.82 lower) | — | 71% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
|
| |||||||
| Subgroup analysis according to the comparison types | |||||||
| CHM plus AD vs. AD | |||||||
| OFS (vs. betahistine) | 98 (1) | — | MD 7.80 lower (6.02–9.58 lower) | — | N/A | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
| SS | 886 (7) | — | SMD 2.45 higher (1.32–3.58 higher) | — | 98% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| SS (vs. flunarizine) | 264 (3) | — | SMD 2.16 higher (0.44–3.87 higher) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| SS (vs. betahistine) | 363 (2) | — | SMD 1.29 higher (0.34 lower–2.91 higher) | — | 98% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| SS (vs. flunarizine and betahistine) | 143 (1) | — | MD 6.98 higher (6.48–7.48 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| SS (vs. diphenidol) | 116 (1) | — | MD 2.67 higher (2.41–2.93 higher) | — | N/A | ⊕⊕⊕○ Moderate |
Imprecision (−1) |
| LVA-BF | 791 (7) | — | MD 5.39 higher (3.33–7.45 higher) | — | 98% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| LVA-BF (vs. flunarizine) | 226 (2) | — | MD 3.96 higher (1.91–6.01 higher) | — | 94% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| LVA-BF (vs. betahistine) |
172 (2) | — | MD 8.73 higher (5.49–11.97 higher) | — | 94% | ⊕○○○ Very low |
Risk of bias (−1) Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| LVA-BF (vs. flunarizine and betahistine) | 143 (1) | — | MD 4.59 higher (3.28–5.90 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| LVA-BF (vs. diphenidol) | 116 (1) | — | MD 5.51 higher (4.39–6.63 higher) | — | N/A | ⊕⊕⊕○ Moderate |
Imprecision (−1) |
| LVA-BF (vs. nimodipine) | 134 (1) | — | MD 2.40 higher (1.90–2.90 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| RVA-BF | 791 (7) | — | MD 5.28 higher (3.38–7.18 higher) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| RVA-BF (vs. flunarizine) | 226 (2) | — | MD 4.80 higher (4.23–5.38 higher) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| RVA-BF (vs. betahistine) | 172 (2) | — | MD 7.77 higher (7.17–8.37 higher) | — | 25% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| RVA-BF (vs. flunarizine and betahistine) | 143 (1) | — | MD 5.04 higher (3.85–6.23 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| RVA-BF (vs. diphenidol) | 116 (1) | — | MD 4.69 higher (3.77–5.61 higher) | — | N/A | ⊕⊕⊕○ Moderate |
Imprecision (−1) |
| RVA-BF (vs. nimodipine) | 134 (1) | — | MD 1.82 higher (1.35–2.29 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| BA-BF | 791 (7) | — | MD 5.28 higher (3.97–6.59 higher) | — | 92% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| BA-BF (vs. flunarizine) | 226 (2) | — | MD 4.85 higher (4.04–5.65 higher) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| BA–BF (vs. betahistine) | 172 (2) | — | MD 5.70 higher (5.15–6.24 higher) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| BA–BF (vs. flunarizine and betahistine) | 143 (1) | — | MD 6.92 higher (5.74–8.10 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| BA-BF (vs. diphenidol) | 116 (1) | — | MD 6.23 higher (4.42–8.04 higher) | — | N/A | ⊕⊕⊕○ Moderate |
Imprecision (−1) |
| BA-BF (vs. nimodipine) | 134 (1) | — | MD 2.74 higher (2.19–3.29 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| TER | 1,529 (13) | 311 per 1,000 | 461 per 1,000 (420–538) | RR 1.53 (1.35–1.73) | 21% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| TER (vs. flunarizine) | 610 (6) | 407 per 1,000 | 590 per 1,000 (472–773) | RR 1.48 (1.16–1.90) | 50% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TER (vs. betahistine) | 526 (4) | 206 per 1,000 | 322 per 1,000 (262–459) | RR 1.68 (1.27–2.23) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| TER (vs. flunarizine and betahistine) | 143 (1) | 286 per 1,000 | 562 per 1,000 (369–858) | RR 1.97 (1.29–3.00) | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| TER (vs. diphenidol) | 116 (1) | 259 per 1,000 | 362 per 1,000 (207–632) | RR 1.40 (0.80–2.44) | N/A | ⊕⊕○○ Low |
Imprecision (−2) |
| TER (vs. nimodipine) | 134 (1) | 328 per 1,000 | 433 per 1,000 (279–669) | RR 1.32 (0.85–2.04) | N/A | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
| ET level (vs. diphenidol) | 116 (1) | — | MD 11.14 lower (9.49–12.79 lower) | — | N/A | ⊕⊕⊕○ Moderate |
Imprecision (−1) |
| CGRP level | 200 (2) | — | MD 8.89 higher (0.76 lower–18.54 higher) | — | 98% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−2) |
| CGRP level (vs. flunarizine) | 84 (1) | — | MD 13.89 higher (11.48–16.30 higher) | — | N/A | ⊕⊕○○ Low |
Imprecision (−2) |
| CGRP level (vs. diphenidol) | 116 (1) | — | MD 4.04 higher (3.68–4.40 higher) | — | N/A | ⊕⊕⊕○ Moderate |
Imprecision (−1) |
|
| |||||||
| CHM plus MT vs. MT | |||||||
| OFS | 246 (2) | — | SMD 3.17 lower (6.48 lower–0.15 higher) | — | 98% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−2) |
| SS | 726 (7) | — | SMD 1.33 higher (0.12–2.54 higher) | — | 98% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| LVA-BF | 358 (3) | — | MD 6.24 higher (1.36–11.12 higher) | — | 98% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| RVA-BF | 358 (3) | — | MD 5.62 higher (1.03–10.21 higher) | — | 98% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| BA-BF | 358 (3) | — | MD 4.62 higher (0.32–8.91 higher) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TER | 614 (6) | 235 per 1,000 | 406 per 1,000 (320–508) | RR 1.71 (1.36–2.16) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| ET level | 226 (2) | — | MD 16.48 lower (33.31 lower–0.34 higher) | — | 98% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−2) |
| CGRP level | 226 (2) | — | MD 4.63 higher (2.25–7.00 higher) | — | 93% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
|
| |||||||
| CHM plus AT vs. AT | |||||||
| OFS | 70 (1) | — | MD 1.91 lower (1.37–2.45 lower) | — | N/A | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
| SS | 677 (8) | — | SMD 1.72 higher (1.33–2.11 higher) | — | 79% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| LVA-BF | 629 (7) | — | MD 5.81 higher (2.92–8.70 higher) | — | 95% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| RVA-BF | 629 (7) | — | MD 4.03 higher (1.05–7.01 higher) | — | 96% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| BA-BF | 739 (8) | — | MD 6.43 higher (2.97–9.89 higher) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TER | 1,149 (13) | 307 per 1,000 | 471 per 1,000 (405–546) | RR 1.54 (1.32–1.78) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
|
| |||||||
| CHM plus MT plus AT vs. MT plus AT | |||||||
| OFS | 290 (1) | — | MD 7.06 lower Risk of bias (−1) Imprecision (−1; 6.27–7.85 lower) |
— | N/A | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| TER | 290 (1) | 290 per 1,000 | 407 per 1,000 (296–563) | RR 1.40 (1.02–1.94) | N/A | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
|
| |||||||
| Subgroup analysis according to the CHM prescription names | |||||||
| BBTT plus active controls vs. active controls | |||||||
| OFS | 410 (2) | — | SMD 3.44 lower (0.69–6.20 lower) | — | 98% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| SS | 220 (2) | — | MD 5.15 higher (4.81–5.50 higher) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| LVA-BF | 184 (2) | — | MD 4.44 higher (3.18–5.69 higher) | — | 71% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| RVA-BF | 184 (2) | — | MD 3.85 higher (2.29–5.41 higher) | — | 84% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| BA-BF | 184 (2) | — | MD 3.48 higher (0.04–6.92 higher) | — | 95% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| TER | 1,168 (9) | 329 per 1,000 | 486 per 1,000 (424–559) | RR 1.48 (1.29–1.70) |
33% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| ET level | 100 (1) | — | MD 25.13 lower (21.29–28.97 lower) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| CGRP level | 100 (1) | — | MD 5.89 higher (4.78–7.00 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
|
| |||||||
| BYT plus active controls vs. active controls | |||||||
| SS | 134 (2) | — | MD 2.04 higher (1.35–2.72 higher) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| LVA-BF | 134 (2) | — | MD 1.72 higher (0.57–2.87 higher) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| RVA-BF | 134 (2) | — | MD 1.80 lower (0.72–2.88 lower) | — | 0% | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| BA-BF | 134 (2) | — | MD 0.43 higher (0.68 lower–1.55 higher) | — | 0% | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
| TER | 134 (2) | 224 per 1,000 | 284 per 1,000 (157–511) | RR 1.27 (0.70–2.28) | 0% | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
|
| |||||||
| DXT plus active controls vs. active controls | |||||||
| OFS | 126 (1) | — | MD 5.68 lower (4.36–7.00 lower) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| SS | 715 (7) | — | SMD 1.67 higher (0.20–3.14 higher) | — | 98% | ⊕⊕⊕○ Moderate |
Inconsistency (−1) |
| LVA-BF | 579 (5) | — | MD 5.13 higher (3.87–6.40 higher) | — | 78% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| RVA-BF | 579 (5) | — | MD 5.12 higher (3.42–6.83 higher) | — | 90% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| BA-BF | 579 (5) | — | MD 5.14 higher (2.66–7.62 higher) | — | 92% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TER | 789 (8) | 265 per 1,000 | 431 per 1,000 (352–517) | RR 1.61 (1.33–1.95) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| ET level | 242 (2) | — | MD 9.71 lower (6.61–12.81 lower) | — | 76% | ⊕○○○ Very low |
Risk of bias (−1) Inconsistency (−1) Imprecision (−1) |
| CGRP level | 326 (3) | — | MD 6.41 higher (4.15–8.67 higher) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
|
| |||||||
| GGT plus active controls vs. active controls | |||||||
| OFS | 98 (1) | — | MD 7.80 lower (6.02–9.58 lower) | — | N/A | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
| SS | 489 (5) | — | SMD 1.92 higher (0.99–2.85 higher) | — | 94% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| LVA-BF | 579 (5) | — | MD 7.29 higher (3.51–11.07 higher) | — | 99% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| RVA-BF | 579 (5) | — | MD 6.18 higher (3.12–9.24 higher) | — | 99% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| BA-BF | 579 (5) | — | MD 5.19 higher (3.50–6.88 higher) | — | 96% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TER | 609 (6) | 299 per 1,000 | 485 per 1,000 (395–595) | RR 1.62 (1.32–1.99) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
|
| |||||||
| GJT plus active controls vs. active controls | |||||||
| SS | 200 (1) | — | MD 2.00 higher (1.75–2.25 higher) | — | N/A | ⊕⊕○○ Low |
Risk of bias (−1) Imprecision (−1) |
| TER | 200 (1) | 88 per 1,000 | 187 per 1,000 (87–425) | RR 2.19 (0.99–4.86) | N/A | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
|
| |||||||
| YCT plus active controls vs. active controls | |||||||
| OFS | 70 (1) | — | MD 1.91 lower (1.37–2.45 lower) | — | N/A | ⊕○○○ Very low |
Risk of bias (−1) Imprecision (−2) |
| SS | 531 (5) | — | SMD 1.79 higher (0.93–2.64 higher) | — | 94% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| LVA-BF | 302 (3) | — | MD 7.63 higher (4.69–10.57 higher) | — | 80% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| RVA-BF | 302 (3) | — | MD 7.34 higher (6.02–8.66 higher) | — | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| BA-BF | 412 (4) | — | MD 11.01 higher (4.46–17.56 higher) | — | 96% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TER | 682 (7) | 328 per 1,000 | 504 per 1,000 (420–604) | RR 1.54 (1.28–1.84) | 0% | ⊕⊕⊕○ Moderate |
Risk of bias (−1) |
| Fib level | 348 (4) | — | MD 0.31 lower (0.12–0.50 lower) | — | 97% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
| TC level | 348 (4) | — | MD 0.56 lower (0.31–0.82 lower) | — | 71% | ⊕⊕○○ Low |
Risk of bias (−1) Inconsistency (−1) |
If the evidence of more than 10 studies showed MD <4 for the change in the blood flow velocity in the vertebrobasilar artery or RR >2 for the total effective rate, it was considered that there was a strong association for a treatment effect. AD, anti-vertigo drugs; AT, acupuncture therapy; BA-BF, basal artery blood flow; BBTT, Banxia Baizhu Tianma Tang; BYT, Buzhong Yiqi Tang; CHM, Chinese herbal medicine; CI, confidence interval; CGRP, calcitonin gene-related peptide; DXT, Dingxuan Tang; ET, endothelin; Fib, fibrinogen; GGT, Gegen Tang; GJT, Gegen Jieji Tang; GRADE, the grading of recommendations assessment, development, and evaluation; LVA-BF, left vertebral artery blood flow; MD, mean difference; MT, manual therapy; OFS, overall functional score; RCT, randomised controlled trial; RR, risk ratio; RVA-BF, right vertebral artery blood flow; SMD, standardised mean difference; SS, simple score; TC, total cholesterol; TER, total effective rate; YCT, Yiqi Congming Tang.