甲状腺功能亢进症与卒中发病相关性:一项基于前瞻性队列研究的Meta分析

Correlation between hyperthyroidism and risk of stroke:a meta-analysis based on prospective cohort studies

目的探讨甲状腺功能亢进症(HT)与卒中发病的相关性。方法中文以"甲状腺功能亢进症""卒中""脑梗死",英文以"thyroid dysfunction""thyroid disorder""thyroid disease""hyperthyroidism""thyrotoxicosis""stroke""cerebrovascular disease""cerebrovascular accident"为检索词。计算机检索Pubmed、EMbase、Google Scholar、中国期刊全文数据库(CNKI)、中国生物医学文献数据库(CBM)、万方数字化期刊全文数据库,并手工检索相关杂志。由2位研究者按照纳入和排除标准筛选文献,并进行资料提取,当意见不统一时,交由第3位研究者决定。采用Newcastle-Ottawa Scale(NOS)文献质量评价量表对符合纳入标准的文献进行质量评价,并采用Stata 12.0软件对纳入文献的结局指标进行Meta分析。结果共纳入11项前瞻性队列研究,NOS量表评分提示均为质量较高文献。Meta分析结果表明:HT患者发生卒中的风险是甲状腺功能正常(ET)人群的1.30倍(P<0.05)。亚组分析结果表明,亚洲地区的HT患者发生卒中的风险是亚洲ET人群的1.39倍(P<0.05);甲亢类型属于临床HT的患者发生卒中的风险是ET人群的1.32倍(P<0.05);不伴有心房颤动的HT患者发生卒中的风险是ET人群的1.33倍(P<0.05)。对校正了大多数混杂因素的研究进行的亚组分析表明,HT患者发生卒中的风险是ET人群的1.35倍(P<0.05)。Begg和Egger检验均表明无发表偏倚可能,剪补法校正后RR为1.26。敏感性分析提示本次研究结果稳定。结论 HT可能为卒中发病的一个危险因素,鉴于本研究的局限性,需要进一步开展相关研究进行验证。

Objective To investigate the correlation between hyperthyroidism (HT) and stroke.Methods The search terms included “hyperthyroidism”“stroke” and “cerebral infarction” in both Chinese and English, and“ thyroid dysfunction”“ thyroid disorder”“ thyroid disease”“ thyrotoxicosis”“cerebrovascular disease” and “cerebrovascular accident” in English. PubMed, Embase, Google Scholar,CNKI, CBM and Wanfang were searched, and the related magazines were also hand-searched. These studies were selected and related data were extracted by two authors according to the inclusion and exclusion criteria, and a third author would make a decision when the former two authors couldn't reach an agreement.The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of included studies, and the Stata 12.0 software was used for meta-analysis. Results Eleven prospective cohort studies were included. The NOS scores indicated that all the studies were high-quality studies. The meta-analysis showed that the risk of stroke in the HT group is 1.30 times as high as that in the euthyroidism (ET) group (P<0.05). Subgroup analyses showed that among Asian populations, the risk of stroke in the HT group is 1.39 times as high as that in the ET group (P<0.05); the risk of stoke in the clinical HT group is 1.32 times as high as that in the ET group (P<0.05 ); and the risk of stroke in the HT group without atrial fibrillation is 1.33 times as high as that in the ET group (P<0.05). In addition, the subgroup analysis of studies which had adjusted most of confounding factors showed that the risk of stroke in the HT group was 1.35 times as high as that in the ET group (P<0.05).Begg and Egger tests both showed that there were no published biases, and the RR value was 1.26 after conducting a trim and fill analysis. Sensitivity analysis indicated that the results were stable. Conclusion HT might be a risk factor of stroke, but due to the limitations of this meta-analysis, further relevant research is warranted for validation.