Fifty-one studies (n=1,666) were included for review. A variety of study designs were included, of which six were double-blind placebo-controlled parallel-design randomised controlled trials (RCTs; n=150).
The methodological quality of the included trials was not formally assessed, however the author stated that many of the included trials lacked randomisation and blinding, had small sample sizes and short treatment durations.
A higher baseline serum cholesterol level was associated with a greater decrease in total serum cholesterol (correlation coefficient = -0.585, p=0.000). Higher baseline vitamin C levels appeared to be associated with a greater decrease in total serum cholesterol (correlation coefficient 0.500, p=0.003), although this is unclear as the axes of the graph appear to have been mislabelled. There was no association between change in total serum cholesterol and change in plasma vitamin C levels.
When studies were compared according to the baseline total serum cholesterol levels, the group with the most severe baseline cholesterol demonstrated the greatest percentage reduction in total serum cholesterol levels (WMD 14.30 ±8.3) compared with the normal cholesterol (WMD 0.91 ±6.8) and borderline high cholesterol groups (WMD 3.90 ±5.78). The most severe cholesterol group also had the lowest baseline vitamin C levels (weighted baseline plasma vitamin C: 0.55 ±0.19 mg/dL versus 1.24 ±0.17 mg/dL in the normal cholesterol group).