Chantzichristos Dimitrios, Persson Anders, Eliasson Björn, Miftaraj Mervete, Franzén Stefan, Svensson Ann-Marie, Johannsson Gudmundur
Department of Internal Medicine and Clinical Nutrition, Institute of Medicine at Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.
Department of Endocrinology-Diabetes-Metabolism, Sahlgrenska University Hospital, Gothenburg, Sweden.
Eur J Endocrinol. 2018 Jan;178(1):113-120. doi: 10.1530/EJE-17-0751. Epub 2017 Oct 24.
We determined the incidence and prevalence of Addison's disease (AD) among persons with or without type 1 diabetes mellitus (T1DM) in nationwide, matched cohort studies.
Persons with T1DM were identified from the Swedish National Diabetes Register and each was matched for age, sex, year and county to five controls randomly selected from the general population. Persons with AD were identified from the Swedish National Inpatient Register. Baseline demographics and seasonal onset variation of AD were presented by descriptive statistics. Prevalence and incidence were estimated by proportions and incidence rates, respectively. Times to AD were analyzed using the Cox proportional hazard model.
Between 1998 and 2013, 66 persons with T1DM were diagnosed with AD at a mean age (s.d.) of 36.4 (13.0) years among 36 514 persons with T1DM, while 32 were diagnosed with AD at a mean age of 42.7 (15.2) years among 182 570 controls. The difference in mean age at diagnosis of AD between the groups was 6.3 years ( value = 0.036). The incidence of AD for a person with or without T1DM was therefore 193 and 18 per million person-years, respectively. The adjusted relative risk increase of developing AD in T1DM was 10.8 (95% CI: 7.1-16.5). The highest incidence of AD was observed during February-March and September-October. The prevalence of AD in persons with or without T1DM in December 2012 was 3410 and 208 per million, respectively. The odds ratio for AD in persons with T1DM vs controls was 16.5 (95% CI: 11.1-24.5).
The risk to develop AD among persons with T1DM is more than 10 times higher than in persons without T1DM. Persons with T1DM develop AD at a younger age. The incidence of AD may have a seasonal pattern.
在全国范围的匹配队列研究中,我们确定了1型糖尿病(T1DM)患者和非T1DM患者中艾迪生病(AD)的发病率和患病率。
从瑞典国家糖尿病登记处识别出T1DM患者,并根据年龄、性别、年份和郡县,将每名患者与从普通人群中随机选取的5名对照进行匹配。从瑞典国家住院登记处识别出AD患者。通过描述性统计展示AD的基线人口统计学特征和季节性发病变化。患病率和发病率分别通过比例和发病率进行估算。使用Cox比例风险模型分析发生AD的时间。
1998年至2013年期间,在36514名T1DM患者中,有66人被诊断为AD,平均年龄(标准差)为36.4(13.0)岁;而在182570名对照中,有32人被诊断为AD,平均年龄为42.7(15.2)岁。两组AD诊断时的平均年龄差异为6.3岁(P值 = 0.036)。因此,患有或未患有T1DM的人AD发病率分别为每百万人年193例和18例。T1DM患者发生AD的校正相对风险增加为10.8(95%置信区间:7.1 - 16.5)。AD发病率最高出现在2月至3月以及9月至10月。2012年12月,患有或未患有T1DM的人AD患病率分别为每百万3410例和208例。T1DM患者与对照相比患AD的比值比为16.5(95%置信区间:11.1 - 24.5)。
T1DM患者发生AD的风险比非T1DM患者高10倍以上。T1DM患者患AD的年龄更小。AD发病率可能具有季节性模式。