Background: Sensory stimulation has been used in the early rehabilitation of comatose patients with traumatic brain injury, but the effect of sensory stimulation involving family members is unclear.

Objectives: To evaluate the effects of family-centered sensory and affective stimulation on comatose patients with traumatic brain injury and explore the factors that affect the outcomes.

Design: A systematic review with a meta-analysis.

Data sources: Electronic databases including PubMed, Web of Science, Google Scholar, Cochrane Library, CINAHL, China National Knowledge Infrastructure, and WanFang were searched from October 2019 to May 2020.

Review methods: Two reviewers independently assessed eligibility of potential studies and extracted data. Quality of included studies was assessed according to the evaluation criteria of Cochrane Evaluation Manual 5.1.0. Outcome measures of the meta-analysis were the Glasgow Coma Scale scores, the Western Neuro Sensory Stimulation Profile scores, awakening time, and satisfaction rate. To explore whether there was a difference in the effect between variants of the intervention, variables as subgroups were time to start intervention, type of intervention, duration of each intervention, daily frequency of intervention, days of intervention, and patient's area.

Result: Seventeen randomized controlled trials were included in the review and meta-analysis. Most studies were of medium quality. The improvement of the Glasgow Coma Scale score is significantly greater with the intervention implemented within 24 h compared to the intervention implemented 24 h later (mean difference 3.91, 95% confidence interval 3.44-4.38 vs. mean difference 1.90, 95% confidence interval 1.69-2.12, respectively). The results of subgroup analyses show that auditory stimulation combined with tactile stimulation and multi-sensory stimulation are associated with better outcomes than a single use of auditory stimulation. Studies from Asia report more positive outcomes than those from America (mean difference 1.94, 95% confidence interval 1.73-2.16 vs. mean difference 0.44, 95% confidence interval -0.87-1.75). And the improvement of the Glasgow Coma Scale score with the stimulation performed by family members is greater than that with the stimulation implemented by nurses (mean difference 2.17, 95% confidence interval 1.67-2.66). Besides, it is associated with better awakening time, awakening rate, and satisfaction rate compared to routine care.

Conclusion: Early family-centered sensory and affective stimulation is more effective than routine care and nurse-implemented sensory stimulation in improving the level of consciousness and cognition of comatose patients with traumatic brain injury, and multi-sensory stimulation is more effective than single stimulation. More studies with larger sample size and high quality in different countries are warranted.