All 44 subjects completed the treatment protocol. Their entry characteristics at week 0 are summarised by treatment allocation in Table 1. The treatment groups were generally well matched, except for a small difference in HDL cholesterol
and a non-significant difference in body mass index. Three subjects reported constipation and increased flatus, one on placebo and two on Lactobacillus. Based on a count of returned capsules at study closure, all subjects except one onsumed more than 75% of the amount dispensed. One subject complaining of increased flatus on Lactobacillus halved intake over the final two weeks of study. Table 2 summarises the key lipid and lipoprotein outcomes. Total and LDL cholesterol levels did not change significantly over time, (based on examination of confidence intervals), but there
was modest downward trend on both arms of treatment. Mean LDL cholesterol was 5.2% lower on placebo and 7.0% lower on Lactobacillus, but with no significant difference between treatments. HDL cholesterol trended slightly lower on
both arms, but with no significant difference over time or between treatments. There were no significant changes in triglycerides or differences between treatments. The percentage change in LDL cholesterol was evaluated in those below and above the week 0 median LDL cholesterol of 3.95 mmol/L. In those with lower LDL cholesterol, LDL cholesterol showed
a 0.1% (95% confidence interval 24 to þ18) fall on placebo and a 3.3% (16 to þ11) fall on Lactobacillus. In those with higher LDL cholesterol, LDL cholesterol showed a 10.7% (27 to 1.2) fall on placebo and a 10.4% (31 to þ1.2) fall on
Lactobacillus. The changes in liver enzymes, glucose, creatinine and CRP between week 0 and week 8 are
summarised in Table 3. There were no significant changes over time or between treatment arms. Body weights did not change significantly over the study duration.
Discussion
Although LDL cholesterol trended downwards on both treatment arms over the 10 weeks of treatment, this placebo-controlled evaluation demonstrated no significant effect of Lactobacillus fermentum in the dose selected on total or LDL
cholesterol, HDL cholesterol or serum triglycerides. We cannot exclude an effect from a still higher dose, but this approach would most likely be limited by potential adverse effects. There were already 3 out of 46 subjects reporting bowel
side effects at the present dose. Table 1 Baseline characteristics by treatment group Placebo (n ¼ 21) Lactobacillus