长新冠嗅觉功能障碍恢复中的无创神经调控技术：系统综述与贝叶斯荟萃分析

文献信息

中文摘要

目的： 长新冠是一种多系统疾病，其中嗅觉障碍是最常见的表现之一。无创神经调控已成为一种有前景的COVID后遗症治疗方法。本系统综述与贝叶斯荟萃分析旨在综合现有证据，评估神经调控对长新冠患者嗅觉改善的作用。
材料与方法： 根据系统综述和荟萃分析优先报告条目2020指南，对PubMed和/或MEDLINE、Embase和Cochrane Library从建库至2026年1月22日进行了系统检索。纳入标准为：研究对象为接受任何无创神经调控方式治疗嗅觉障碍的长新冠患者，并设有对照组。采用Cochrane随机对照试验偏倚风险2工具评估随机对照试验的偏倚风险，采用非随机干预研究偏倚风险工具评估非随机干预研究的偏倚风险。使用R语言brms包进行贝叶斯随机效应荟萃分析，效应量为风险比（RR）和95%可信区间（CrI）。
结果： 共纳入4项研究，使用的神经调控方式异质性较大。完全嗅觉恢复（正常嗅觉）的合并RR为1.59（95% CrI，0.41-6.28），获益的后验概率为73.9%（RR > 1）；然而，CrI包含无效值。对于有临床意义的嗅觉改善，合并RR为2.20（95% CrI，0.63-7.52），获益的后验概率为88.8%，尽管区间略微包含无效值。使用更宽先验的敏感性分析产生了显著不同的估计值，表明结果依赖于先验。各研究均未报告严重不良事件。
结论： 现有证据提示无创神经调控对长新冠嗅觉功能障碍可能具有有利的效应方向，但确定性极低。这些发现应被解读为产生假设。在提出任何临床建议之前，需要采用标准化嗅觉测试和预先设定刺激方案的大型试验。

英文摘要

OBJECTIVES: Long COVID is a multisystemic condition in which olfactory disorders are among the most prevalent manifestations. Noninvasive neuromodulation has emerged as a promising therapeutic approach for post-COVID sequelae. This systematic review and Bayesian meta-analysis aimed to synthesize the available evidence on neuromodulation for olfactory improvement in patients with Long COVID.
MATERIALS AND METHODS: A systematic search was conducted in PubMed and/or MEDLINE, Embase, and Cochrane Library from inception through January 22, 2026, in accordance with Preferred Reporting items for Systematic Reviews and Meta-Analyses 2020 guidelines. Studies were eligible if they included patients with Long COVID undergoing any noninvasive neuromodulation modality for olfactory disturbances and had a control group. Risk of bias was assessed using Cochrane risk of bias 2 tool for randomized controlled trials and the risk of bias in nonrandomized studies of interventions tool for nonrandomized intervention studies. A Bayesian random-effects meta-analysis was performed using the brms package in R, with risk ratios (RRs) and 95% credible intervals (CrIs) as effect measures.
RESULTS: A total of four studies were included, using heterogeneous neuromodulation modalities. The pooled RR for complete olfactory recovery (normosmia) was 1.59 (95% CrI, 0.41-6.28), with a 73.9% posterior probability of benefit (RR > 1); however, the CrI included the null value. For clinically meaningful olfactory improvement, the pooled RR was 2.20 (95% CrI, 0.63-7.52), with an 88.8% posterior probability of benefit, although the interval marginally included the null value. Sensitivity analyses using wider priors yielded substantially different estimates, indicating that the results were prior-dependent. No serious adverse events were reported across the studies.
CONCLUSIONS: The available evidence suggests a possible favorable direction of effect of noninvasive neuromodulation on olfactory dysfunction in Long COVID, but the certainty is very low. These findings should be interpreted as hypothesis-generating. Larger trials using standardized olfactory testing and prespecified stimulation protocols are warranted before any clinical recommendations can be made.