TABLE 2.
Meta-analysis of intervention trials reporting the effect of apple compared with control on serum lipids and body composition1
| Net change (95% CI) (I2) | ||||
|---|---|---|---|---|
| Studies (subjects), n | Analysis 1 | Analysis 2 | Analysis 3 | |
| TC (mg/dL) | 5 (324) | −4.10 (−13.02, 4.82) (46.0%) | −5.14 (−16.31, 6.03) (64.7%) | −4.50 (−14.45, 5.45) (54.8%) |
| LDL (mg/dL) | 4 (289) | −4.75 (−10.40, 0.90) (19.0%) | −7.09 (−15.54, 1.35) (61.4%) | −5.29 (−11.77, 1.18) (34.4%) |
| HDL (mg/dL) | 4 (289) | −0.79 (−2.58, 0.99) (0.0%) | −0.97 (−2.76, 0.82) (0.0%) | −0.75 (−2.55, 1.05) (0.0%) |
| TG (mg/dL) | 5 (324) | 8.91 (−10.32, 28.14) (59.1%) | 9.66 (−9.05, 28.36) (54.0%) | 8.60 (−14.93, 32.13) (72.4%) |
| Body Weight (kg) | 5 (393) | 0.14 (−0.45, 0.73) (0.0%) | 0.14 (−0.45, 0.73) (0.0%) | 0.14 (−0.45, 0.73) (0.0%) |
| BMI (kg/m2) | 3 (229) | −0.39 (−0.59, −0.20) (0.0%) | ||
Meta-analyses were conducted using the random-effects model. Analysis 1: for one 3-arm parallel trial Chai 2012, the main analyses included 3-mo intake time-point because it was the closest to the final time points in the other studies. Analysis 2: for one 3-arm parallel trial Chai 2012, sensitivity analysis was conducted using the 6-mo time point. Analysis 3: For one 3-arm parallel trial Chai 2012, sensitivity analysis was conducted using the 12-mo time point. I2 is an indicator of between-comparison heterogeneity. I2 >50% was deemed as having significant heterogeneity. HDL, high-density lipoprotein cholesterol; LDL, low-density lipoprotein cholesterol; TC, total cholesterol; TG, triglycerides.