The children all began the trial with a fasting blood cholesterol concentration greater than 4.8 mmol/L. The experiment was double-blind so 300mg of garlic extract was administered to the experimental group and an identical placebo was given to the control group by a patient-nurse coordinator and supervising physician who were unaware of the group distinctions (McCrindle et al., 1998).
The results were measured by the changes in fasting lipid profiles (when compared to the baseline) observed in participant blood samples. The garlic extract did not cause blood samples to exhibit a significant decrease in fasting cholesterol levels (+0.6% [95% confidence interval, -5.8% to +6.9%]) or LDL-C (-0.5% [95% confidence interval, -8.7% to +7.6%]). There were also no significant changes in concentrations of HDL-C or triglycerides. This led the researchers involved in the study to conclude that garlic extract therapy has no significant benefit to the cardiovascular health of pediatric patients with familial hyperlipidemia (McCrindle et al., 1998).
This study possesses many strengths that support the validity of its data. For instance, the experiment was designed to preserve a strong correlation between the independent and dependent variables (garlic extract supplementation and improving blood lipid profiles, respectively). This design was created by establishing baseline characteristics for the children allowed to participate in the study such as family history of hypercholesterolemia and premature atherosclerotic heart disease, and a small range of initial fasting total cholesterol concentrations.
Some weaknesses were also present in the study. For example, the sample size was very small and consisted of children who were all pediatric patients at the same hospital in Ontario, Canada. Therefore, the lack of variation could have limited the generalizability of the data.
The results of this study clearly demonstrate the ineffectiveness of garlic supplementation as means of lowering blood cholesterol. The confounding biological variables which could have been produced by other studies were controlled for in this McCrindle et al. (1998) study because the children had a predisposition of hypercholesterolemia. Since familial hypercholesterolemia could manifest itself in patients by either producing malfunctioning lipoprotein receptors or non-functioning receptors, the effects of garlic (or lack thereof) were more accurately measured. Therefore, this study strongly supports the insignificance of garlic extract treatment on hypercholesterolemia.
On a similar note, the study by Gardner et al., (2001) evaluated the effect of two doses of a garlic preparation (powdered) on plasma lipids in fifty-one adults (both male and female) with moderate forms of hypercholesterolemia. The experiment had a null hypothesis that fasting plasma lipid profiles would be unaffected by the doses of garlic administered to participants. The study lasted twelve weeks, was double-blind and randomly divided adults into three groups: one placebo-controlled, an experimental half-dose (500 mg garlic powder/day), and an experimental full-dose (1,000 mg garlic powder/day). Participants began with fasting plasma LDL-C of 157.418.7, mean S.D., and were assessed during 9 clinical follow-up visits which measured weight, blood pressure, plasma total cholesterol and triacylglycerol from fasting blood samples, and the number of pills not consumed (for compliance).
The results of the study showed average changes in LDL-C of 0.0 4.3, +1.44.8, and -10.16.8 mg/dl from the placebo, half-dose, and full-dose participants, respectively (Gardner et al., 2001). There was no significant decrease in LDL-C nor any significant differences in HDL-C or triacylglycerol levels (P = 0.5). No difference was observed between the two dosages either. Therefore, the authors concluded that the garlic preparation used in the experiment had no significant effect on the blood lipid profiles of moderately hypercholesterolemic adults.
The double-blinding technique used in the study was very effective because participants were unable to distinguish between the experimental garlic supplements and the placebo tablets. The technique strengthened the reliability of the data by most likely contributing to the extremely-low dropout rate of only two participants (one half-dose, one full-dose) and fairly high compliance rate (92%). The study was also strengthened by a plethora of controls to safeguard against as many confounding variables as possible because the data can be deemed more valid.
The study also had some weaknesses. For example, the low sample size of the experiment produced insignificant statistics because it made differences more difficult to detect. In addition, the diversity among participants was limited as a side-effect of having a small sample size. The lack of diversity makes the data somewhat questionable due to the homogeneous nature of the data.
The results of this study strongly oppose the use of garlic preparations as an independent method of significantly lowering plasma cholesterol levels in moderately hypercholesterolemic adults. Because so many confounding variables were controlled for (e.g., No pregnant women or smokers; no people suffering from heart disease, diabetes, or have used blood pressure medication within the past month; specified weight, diet and activity levels maintained [using food records, physical activity questionnaires]; presence of placebo), it is extremely difficult to refute the data collected.
Many peer-review studies have been published and refuted by other researchers in hopes of arriving at a common conclusion.