Effects of probiotics and paraprobiotics on subjective and objective sleep metrics: a systematic review and meta-analysis

Abstract

Inadequate sleep (i.e., duration and/or quality) is becoming increasingly recognized as a global public health issue. Interaction via the gut-brain axis suggests that modification of the gut microbial environment via supplementation with live microorganisms (probiotics) or nonviable microorganisms/microbial cell fractions (paraprobiotics) may improve sleep health. This systematic review and meta-analysis aimed to clarify the effect of consuming probiotics/paraprobiotics on subjective and objective sleep metrics. Online databases were searched from 1980 to October 2019 for studies involving adults who consumed probiotics or paraprobiotics in controlled trials, during which, changes in subjective and/or objective sleep parameters were examined. A total of 14 studies (20 trials) were included in meta-analysis. Random effects meta-analyses indicated that probiotics/paraprobiotics supplementation significantly reduced Pittsburgh Sleep Quality Index (PSQI) score (i.e., improved sleep quality) relative to baseline (−0.78-points, 95% confidence interval: 0.395–1.166; p < 0.001). No significant effect was found for changes on other subjective sleep scales, nor objective parameters of sleep (efficiency/latency) measured using polysomnography or actigraphy. Subgroup analysis for PSQI data suggested that the magnitude of the effect was greater (although not statistically) in healthy participants than those with a medical condition, when treatment contained a single (rather than multiple) strain of probiotic bacteria, and when the duration of treatment was ≥8 weeks. Probiotics/paraprobiotics supplementation may have some efficacy in improving perceived sleep health, measured using the PSQI. While current evidence does not support a benefit of consuming probiotics/paraprobiotics when measured by other subjective sleep scales, nor objective measures of sleep; more studies using well-controlled, within-subject experimental designs are needed.