Effects of Lactobacillus plantarum PS128 on Depressive Symptoms and Sleep Quality in Self-Reported Insomniacs: A Randomized, Double-Blind, Placebo-Controlled Pilot Trial

Abstract

Recent animal studies have supported that Lactobacillus plantarum PS128 (PS128) can reduce the severity of anxiety and depression. However, previous studies did not focus on the sleep quality and mood of humans. This study determines whether PS128 reduces the severity of anxiety and depressive symptoms, regulates autonomic nervous system function, and improves sleep quality. Forty participants between 20 and 40 years of age with self-reported insomnia were randomly assigned to two groups, a PS128 group and a placebo group, in a double-blind trial. Participants took two capsules of either PS128 or a placebo after dinner for 30 days. Study measures included subjective depressive symptoms, anxiety and sleep questionnaires, and miniature-polysomnography recordings at baseline and on the 15th and 30th days of taking capsules. Overall, all outcomes were comparable between the two groups at baseline and within the 30-day period, yet some differences were still found. Compared to the control group, the PS128 group showed significant decreases in Beck Depression Inventory-II scores, fatigue levels, brainwave activity, and awakenings during the deep sleep stage. Their improved depressive symptoms were related to changes in brain waves and sleep maintenance. These findings suggest that daily administration of PS128 may lead to a decrease in depressive symptoms, fatigue level, cortical excitation, and an improvement in sleep quality during the deep sleep stage. Daily consumption of PS128 as a dietary supplement may improve the depressive symptoms and sleep quality of insomniacs, although further investigation is warranted.

Keywords: Lactobacillus plantarum PS128; anxiety; depression; heart rate variability; insomnia.

Figure 1

Study schedule. Participants underwent miniature-polysomnography at baseline, mid-test, and end-test. PSQI, Pittsburgh Sleep Quality Index; ISI, Insomnia Severity Index; ESS, Epworth Sleepiness Scale; BDI-II, Beck Depression Inventory-II; BAI, Beck Anxiety Inventory; STAI, State-Trait Anxiety Inventory State; MEQ, Morningness-Eveningness Questionnaire; VAS, Visual Analogue Scale.

Figure 2

Participant flow: 202 potential participants contacted us. We evaluated participant eligibility using questionnaires, interviews, and sleep dairies. In the end, 40 participants were included.

Figure 3

Effect of PS128 and placebo treatment on sleep/mood-related scores after 4 weeks of treatment. Within-groups analysis: both groups showed significant decreases in PSQI, ISI, ESS, BDI-II, and BAI scores compared to baseline. There was a significant decrease between groups in BDI-II and VAS fatigue level before sleep. † p < 0.05, ‡ p < 0.01 vs. baseline using repeat measurement ANOVA. * p < 0.05 vs. the control group by generalized estimating equations. Values are presented as mean ± SEM. All: Con, n = 19, PS128, n = 21; Inso: Con, n = 9, PS128, n = 12; Mis: Con, n = 10, PS128, n = 9. Con, control; Inso, insomniac; Mis, misperception; BMI, body mass index; PSQI, Pittsburgh Sleep Quality Index; ISI, Insomnia Severity Index; ESS, Epworth Sleepiness Scale; BDI-II, Beck Depression Inventory-II; BAI, Beck Anxiety Inventory; VAS, Visual Analogue Scale; WASO, wake after sleep onset; SE, sleep efficiency.

Figure 4

Effects of PS128 on sleep EEG. On day 30, the PS128 group showed significant decreases in awakenings in N3 compared to the control group. † p < 0.05, ‡ p < 0.01 vs. baseline by repeat measurement ANOVA. * p < 0.05 vs. the control group by generalized estimating equations. Values are presented as mean ± SEM. All: Con, n = 19, PS128, n = 21; Inso: Con, n = 9, PS128, n = 12; Mis: Con, n = 10, PS128, n = 9. Con, control; Inso, insomniac; Mis, misperception; N3, non-rapid eye movement sleep stage 3.

Figure 5

Effects of PS128 on sleep stages. Changes in N1%, N2%, N3%, and REM% on day 0, day 15, and day 30. † p < 0.05 vs. baseline by repeated measurement ANOVA. * p < 0.05 vs. the control group by generalized estimating equations. Values are presented as mean ± SEM. All: Con, n = 19, PS128, n = 21; Inso: Con, n = 9, PS128, n = 12; Mis: Con, n = 10, PS128, n = 9. N1, non-rapid eye movement sleep (NREM) stage 1; N2, NREM stage 2; N3, NREM stage 3; REM, rapid eye movement sleep. Con, control; Inso, insomniac; Mis, misperception.

Figure 6

Two dimensional scatter plots displaying the relationship between the change in BDI-II scores and the change in objective parameters from baseline to 4 weeks of treatment. The figure shows those parameters that are correlated to the △BDI. Control, n = 10; PS128, n = 17. * p < 0.05, Spearman rank correlation analysis. BDI-II, Beck Depression Inventory-II; REM, rapid eye movement stage; N2, non-REM (NREM) stage 2; N3, NREM stage 3; △ in scores as the day 0 score minus the day 30 score.