Obstructive sleep apnea (OSA) is defined by frequent episodes of reduced or complete cessation of airflow during sleep and is linked to negative health outcomes. Understanding the genetic factors influencing expression of OSA may lead to new treatment strategies. Electronic health records (EHRs) can be leveraged to both validate previously reported OSA-associated genomic variation and detect novel relationships between these variants and comorbidities. We identified candidate single nucleotide polymorphisms (SNPs) via systematic literature review of existing research. Using datasets available at Geisinger (n = 39,407) and Vanderbilt University Medical Center (n = 24,084), we evaluated associations between 40 previously implicated SNPs and OSA diagnosis, defined using clinical codes. We also evaluated associations between these SNPs and OSA severity measures obtained from sleep reports at Geisinger (n = 6571). Finally, we used a phenome-wide association study approach to help reveal pleiotropic genetic effects between OSA candidate SNPs and other clinical codes and laboratory values available in the EHR. Most previously reported OSA candidate SNPs showed minimal to no evidence for associations with OSA diagnosis or severity in the EHR-derived datasets. Three SNPs in LEPR, MMP-9, and GABBR1 validated for an association with OSA diagnosis in European Americans; the SNP in GABBR1 was associated following meta-analysis of results from both clinical populations. The GABBR1 and LEPR SNPs, and one additional SNP, were associated with OSA severity measures in European Americans from Geisinger. Three additional candidate OSA SNPs were not associated with OSA-related traits but instead with hyperlipidemia and autoimmune diseases of the thyroid. To our knowledge, this is one of the largest candidate gene studies and one of the first phenome-wide association studies of OSA genomic variation. Results validate genetic associates with OSA in the LEPR, MMP-9 and GABBR1 genes, but suggest that the majority of previously identified genetic associations with OSA may be false positives. Phenome-wide analyses provide evidence of mediated pleiotropy. Future well-powered genome-wide association analyses of OSA risk and severity across populations with diverse ancestral backgrounds are needed. The comprehensive nature of the analyses represents a platform for informing future work focused on understanding how genetic data can be useful to informing treatment of OSA and related comorbidities.

中文翻译：

---

使用电子健康记录表征阻塞性睡眠呼吸暂停的遗传和表型异质性。

阻塞性睡眠呼吸暂停 (OSA) 的定义是睡眠期间气流减少或完全停止的频繁发作，并且与负面健康结果有关。了解影响 OSA 表达的遗传因素可能会导致新的治疗策略。可以利用电子健康记录 (EHR) 来验证先前报告的 OSA 相关基因组变异并检测这些变异与合并症之间的新关系。我们通过对现有研究的系统文献回顾确定了候选单核苷酸多态性 (SNP)。使用 Geisinger (n = 39,407) 和范德比尔特大学医学中心 (n = 24,084) 提供的数据集，我们评估了 40 个先前涉及的 SNP 与 OSA 诊断之间的关联，使用临床代码定义。我们还评估了这些 SNP 与从 Geisinger (n = 6571) 的睡眠报告中获得的 OSA 严重程度测量值之间的关联。最后，我们使用了一种现象范围的关联研究方法来帮助揭示 OSA 候选 SNP 与 EHR 中可用的其他临床代码和实验室值之间的多效性遗传效应。大多数先前报告的 OSA 候选 SNP 在 EHR 衍生数据集中显示出与 OSA 诊断或严重程度相关的证据很少甚至没有。LEPR、MMP-9 和 GABBR1 中的三个 SNP 经验证与欧裔美国人的 OSA 诊断相关；在对两个临床人群的结果进行荟萃分析后，GABBR1 中的 SNP 相关联。GABBR1 和 LEPR SNP 以及一个额外的 SNP 与来自 Geisinger 的欧裔美国人的 OSA 严重程度测量相关。另外三个候选 OSA SNP 与 OSA 相关性状无关，但与高脂血症和甲状腺自身免疫性疾病相关。据我们所知，这是最大的候选基因研究之一，也是 OSA 基因组变异的第一个现象级关联研究之一。结果证实了 LEPR、MMP-9 和 GABBR1 基因与 OSA 的遗传关联，但表明大多数先前确定的与 OSA 的遗传关联可能是假阳性。全现象分析提供了介导多效性的证据。未来需要对具有不同祖先背景的人群进行 OSA 风险和严重程度的强有力的全基因组关联分析。