Table 4. Dose–Response Subgroup Analyses of Association Between High-Density Lipoprotein Cholesterol and Risk of Total Stroke and Ischemic Stroke, Systematic Review and Meta-Analysis of Prospective Cohort Studies.
| Characteristics | Total Stroke |
Ischemic Stroke |
||||||
|---|---|---|---|---|---|---|---|---|
| No. of Studies | Relative Risk (95% CI) | I 2 | P Valuea | No. of Studies | Relative Risk (95% CI) | I 2 | P Valuea | |
| All studies | 18 | 0.82 (0.76–0.89) | 42.9 | .03 | 22 | 0.75 (0.69–0.82) | 50.1 | .004 |
| Sex | ||||||||
| Men/women | 7 | 0.82 (0.70–0.96) | 41.8 | .11 | 8 | 0.78 (0.68–0.89) | 51.0 | .046 |
| Men | 8 | 0.79 (0.64–0.99) | 31.9 | .17 | 9 | 0.76 (0.65–0.90) | 39.6 | .10 |
| Women | 6 | 0.74 (0.52–1.04) | 60.0 | .03 | 8 | 0.71 (0.56–0.88) | 49.7 | .05 |
| Region | ||||||||
| Asian | 10 | 0.81 (0.70–0.95) | 28.6 | .18 | 10 | 0.76 (0.68–0.85) | 8.0 | .37 |
| Non-Asian | 8 | 0.77 (0.63–0.93) | 59.2 | .02 | 12 | 0.74 (0.66–0.84) | 65.8 | .001 |
| Follow-up period | ||||||||
| <10 years | 7 | 0.92 (0.81–1.05) | 44.4 | .10 | 9 | 0.78 (0.66–0.91) | 57.7 | .03 |
| ≥10 years | 11 | 0.77 (0.70–0.85) | 31.3 | .15 | 13 | 0.74 (0.66–0.82) | 47.2 | .03 |
| Publication year | ||||||||
| ≤2010 | 4 | 0.53 (0.39–0.72) | 4.6 | .37 | 8 | 0.72 (0.63–0.82) | 51.1 | .05 |
| >2010 | 14 | 0.85 (0.78–0.92) | 27.0 | .17 | 14 | 0.77 (0.68–0.87) | 51.5 | .01 |
| Sample size | ||||||||
| <10,000 | 8 | 0.67 (0.55–0.82) | 30.4 | .19 | 8 | 0.79 (0.60–1.05) | 59.0 | .004 |
| ≥10,000 | 10 | 0.85 (0.78–0.93) | 41.2 | .08 | 14 | 0.75 (0.68–0.81) | 47.9 | .02 |
| Alcohol drinking | ||||||||
| No | 6 | 0.88 (0.79–0.98) | 6.5 | .38 | 8 | 0.74 (0.66–0.83) | 59.1 | .02 |
| Yes | 12 | 0.76 (0.68–0.85) | 48.0 | .03 | 14 | 0.76 (0.66–0.88) | 47.8 | .02 |
| Education | ||||||||
| No | 13 | 0.80 (0.71–0.91) | 31.8 | .12 | 15 | 0.74 (0.68–0.81) | 34.6 | .09 |
| Yes | 5 | 0.74 (0.52–1.05) | 71.3 | .02 | 7 | 0.76 (0.62–0.95) | 70.4 | .002 |
| Body mass index | ||||||||
| No | 6 | 0.89 (0.78–1.02) | 29.2 | .22 | 6 | 0.80 (0.70–0.91) | 59.4 | .03 |
| Yes | 12 | 0.78 (0.71–0.87) | 46.3 | .04 | 16 | 0.72 (0.64–0.81) | 48.6 | .02 |
| Systolic blood pressure | ||||||||
| No | 6 | 0.82 (0.71–0.94) | 45.3 | .10 | 6 | 0.74 (0.65–0.84) | 67.6 | .009 |
| Yes | 12 | 0.82 (0.74–0.90) | 46.7 | .04 | 16 | 0.76 (0.67–0.86) | 42.3 | .04 |
| Physical activity | ||||||||
| No | 10 | 0.87 (0.74–1.01) | 23.6 | .23 | 12 | 0.76 (0.68–0.84) | 52.9 | .02 |
| Yes | 8 | 0.73 (0.61–0.88) | 53.7 | .04 | 10 | 0.74 (0.63–0.88) | 51.8 | .03 |
| Lipid lowering medication use | ||||||||
| No | 9 | 0.69 (0.53–0.91) | 59.2 | .01 | 14 | 0.72 (0.65–0.80) | 61.6 | .001 |
| Yes | 9 | 0.85 (0.77–0.93) | 3.6 | .41 | 8 | 0.83 (0.73–0.93) | 0 | .62 |
| Other lipid profiles parameters | ||||||||
| No | 10 | 0.84 (0.69–1.03) | 55.2 | .02 | 12 | 0.75 (0.67–0.84) | 61.0 | .003 |
| Yes | 8 | 0.77 (0.69–0.86) | 0 | .47 | 10 | 0.75 (0.64–0.87) | 34.9 | .13 |
a
Based on the DerSimonian and Laird random-effects model.