Division of General Internal Medicine, Department of Medicine, St. Luke's International Hospital, Tokyo, Japan.
Department of Epidemiology, Graduate School of Public Health, St. Luke's International University, Tokyo, Japan.
Eur J Clin Microbiol Infect Dis. 2019 Aug;38(8):1539-1545. doi: 10.1007/s10096-019-03584-1. Epub 2019 May 20.
Little is known about the association between glycemic status and herpes zoster. The aim of this study was to evaluate whether glycemic status, including both high and low hemoglobin A1c(HbA1c), is associated with subsequent herpes zoster. We conducted a retrospective longitudinal study in a large teaching hospital in Tokyo, Japan, from 2005 to 2016. We included all participants who underwent voluntary health check-ups at the hospital. Our primary outcome was the incidence of herpes zoster in groups of individuals stratified by HbA1c levels, which were compared using the generalized estimating equation (GEE), adjusting for participants' demographic characteristics, social history, body mass index, and comorbidities. A total of 81,466 participants were included in this study. The mean age (standard deviation) was 46.5 (12.1), and 39,643 (48.7%) participants were male. Among them, 1751 (2.1%) were diagnosed with diabetes prior to their first visits. After a median follow-up of 1784 [interquartile range (IQR), 749-3150] days, 673 (0.8%) participants developed herpes zoster. The incidence of herpes zoster was 1.45 per 1000 person-years. Compared with the reference group (HbA1c of 5.0-6.4%), the lowest HbA1c group (HbA1c of < 5.0%) had a significantly higher adjusted odds ratio (OR) (OR 1.63; 95% confidence interval (CI), 1.07-2.48) of developing herpes zoster. The group with an HbA1c of ≥ 9.5% had a higher but nonsignificant OR than the reference group (OR 2.15; 95% CI, 0.67-6.94). Our longitudinal study demonstrated that individuals in the lowest (< 5.0%) HbA1c group had a significantly higher risk of developing herpes zoster than the reference group (HbA1c of 5.0-6.4%) after adjusting for covariates.
关于血糖状态与带状疱疹之间的关系知之甚少。本研究旨在评估血糖状态（包括高和低血红蛋白 A1c[HbA1c]）是否与随后发生带状疱疹有关。我们在日本东京的一家大型教学医院进行了一项回顾性纵向研究，时间为 2005 年至 2016 年。我们纳入了在医院接受自愿健康检查的所有参与者。我们的主要结局是根据 HbA1c 水平分层的个体中带状疱疹的发生率，使用广义估计方程（GEE）进行比较，同时调整了参与者的人口统计学特征、社会史、体重指数和合并症。这项研究共纳入了 81466 名参与者。平均年龄（标准差）为 46.5（12.1）岁，39643 名（48.7%）参与者为男性。其中，1751 名（2.1%）在首次就诊前被诊断患有糖尿病。在中位数为 1784[四分位距（IQR），749-3150]天的随访后，673 名（0.8%）参与者发生带状疱疹。带状疱疹的发生率为每 1000 人年 1.45 例。与参考组（HbA1c 为 5.0-6.4%）相比，HbA1c 最低组（HbA1c<5.0%）发生带状疱疹的调整后比值比（OR）显著更高（OR 1.63；95%置信区间（CI），1.07-2.48）。HbA1c≥9.5%的组与参考组相比，OR 虽较高但无统计学意义（OR 2.15；95%CI，0.67-6.94）。我们的纵向研究表明，在校正了协变量后，HbA1c 最低（<5.0%）组的个体发生带状疱疹的风险明显高于参考组（HbA1c 为 5.0-6.4%）。
