# Acupuncture and related therapies for insomnia symptoms in hypertensive patients: protocol for a network meta-analysis of randomized controlled trials

## TL;DR

This study will compare the effectiveness of acupuncture and related therapies for treating insomnia in people with high blood pressure.

## Contribution

The study introduces a Bayesian network meta-analysis protocol to evaluate various acupuncture therapies for insomnia in hypertensive patients.

## Key findings

- The study will assess the effectiveness of acupuncture therapies on insomnia severity in hypertensive patients.
- It will compare acupuncture against sham treatments, usual care, and pharmacotherapy.
- The study will evaluate outcomes like sleep quality, blood pressure, and adverse events.

## Abstract

Insomnia and hypertension frequently co-occur and may exacerbate cardiovascular risk. While acupuncture and related therapies (ARTs) are widely used for sleep symptoms, their comparative effectiveness across modalities remains unclear. This protocol outlines a Bayesian network meta-analysis (NMA) to compare the effectiveness of ARTs for insomnia in adults with hypertension and inform clinical decision-making.

This protocol follows preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) and is registered in PROSPERO. We will search PubMed, Embase, Cochrane Library (CENTRAL), Web of Science, and four Chinese databases (CNKI, Wanfang, VIP, and SinoMed) from inception to January 2026, and we will also screen WHO ICTRP, ClinicalTrials.gov, and ChiCTR without language restrictions. Eligible studies are randomized controlled trials enrolling adults with hypertension and insomnia, comparing an ART with sham, usual care, sleep hygiene, pharmacotherapy, or another ART. The primary outcome is change in global insomnia severity at post-treatment or at the first follow-up; secondary outcomes include sleep parameters, blood pressure, quality of life, mood, and adverse events. Pairwise meta-analyses will be performed using RevMan. A Bayesian random-effects NMA will be implemented in R (v4.4.1), with network plots and league tables generated in Stata (v15.1). The assumptions of transitivity and coherence will be assessed using design-by-treatment and node-splitting approaches; model fit will be evaluated using the deviance information criterion (DIC). Risk of bias will be assessed using RoB 2, and the certainty of evidence will be rated using GRADE adapted for NMA, with prespecified subgroup and sensitivity analyses.

PROSPERO; identifier CRD420251173289.