Aims/hypothesis

Children and adults born preterm have an increased risk of type 1 diabetes. However, there is limited information on risk patterns across the full range of gestational ages, especially after extremely preterm birth (23–27 weeks of gestation). We investigated the risk of type 1 diabetes in childhood and young adulthood across the full range of length of gestation at birth.

Methods

Data were obtained from national registers in Finland, Norway and Sweden. In each country, information on study participants and gestational age was collected from the Medical Birth Registers, information on type 1 diabetes diagnoses was collected from the National Patient Registers, and information on education, emigration and death was collected from the respective national register sources. Individual-level data were linked using unique personal identity codes. The study population included all individuals born alive between 1987 and 2016 to mothers whose country of birth was the respective Nordic country. Individuals were followed until diagnosis of type 1 diabetes, death, emigration or end of follow-up (31 December 2016 in Finland, 31 December 2017 in Norway and Sweden). Gestational age was categorised as extremely preterm (23–27 completed weeks), very preterm (28–31 weeks), moderately preterm (32–33 weeks), late preterm (34–36 weeks), early term (37–38 weeks), full term (39–41 weeks; reference) and post term (42–45 weeks). HRs and 95% CIs from country-specific covariate-adjusted Cox regression models were combined in a meta-analysis using a common-effect inverse-variance model.

Results

Among 5,501,276 individuals, 0.2% were born extremely preterm, 0.5% very preterm, 0.7% moderately preterm, 4.2% late preterm, 17.7% early term, 69.9% full term, and 6.7% post term. A type 1 diabetes diagnosis was recorded in 12,326 (0.8%), 6364 (0.5%) and 16,856 (0.7%) individuals at a median age of 8.2, 13.0 and 10.5 years in Finland, Norway and Sweden, respectively. Individuals born late preterm or early term had an increased risk of type 1 diabetes compared with their full-term-born peers (pooled, multiple confounder-adjusted HR 1.12, 95% CI 1.07, 1.18; and 1.15, 95% CI 1.11, 1.18, respectively). However, those born extremely preterm or very preterm had a decreased risk of type 1 diabetes (adjusted HR 0.63, 95% CI 0.45, 0.88; and 0.78, 95% CI 0.67, 0.92, respectively). These associations were similar across all three countries.

Conclusions/interpretation

Individuals born late preterm and early term have an increased risk of type 1 diabetes while individuals born extremely preterm or very preterm have a decreased risk of type 1 diabetes compared with those born full term.

Graphical Abstract

中文翻译：

---

出生胎龄与儿童期和青年期 1 型糖尿病：芬兰、挪威和瑞典的一项全国登记研究

目标/假设

早产儿童和成人患 1 型糖尿病的风险增加。然而，关于整个孕龄范围内的风险模式的信息有限，特别是在极度早产（妊娠 23-27 周）之后。我们调查了儿童期和青年期在出生时整个妊娠期中患 1 型糖尿病的风险。

方法

数据来自芬兰、挪威和瑞典的国家登记册。在每个国家，有关研究参与者和胎龄的信息是从医学出生登记处收集的，有关 1 型糖尿病诊断的信息是从国家患者登记处收集的，有关教育、移民和死亡的信息是从各自的国家登记处收集的。个人层面的数据使用独特的个人身份代码进行链接。研究人群包括 1987 年至 2016 年间出生的所有活生生，母亲的出生国是相应的北欧国家。对个体进行随访，直至诊断出1型糖尿病、死亡、移民或随访结束（芬兰2016年12月31日，挪威和瑞典2017年12月31日）。胎龄分为极早产（23-27 周）、极早产（28-31 周）、中度早产（32-33 周）、晚期早产（34-36 周）、早产（37-38 周） ，全学期（39-41 周；参考）和学期后（42-45 周）。使用共效应逆方差模型将国家特定协变量调整 Cox 回归模型的 HR 和 95% CI 合并到荟萃分析中。

结果

在 5,501,276 人中，0.2% 为极度早产，0.5% 为严重早产，0.7% 为中度早产，4.2% 为晚期早产，17.7% 为早产，69.9% 为足月，6.7% 为足月后出生。在芬兰、挪威和瑞典，有 12,326 名（0.8%）、6364 名（0.5%）和 16,856 名（0.7%）人被诊断为 1 型糖尿病，中位年龄分别为 8.2 岁、13.0 岁和 10.5 岁。与足月出生的同龄人相比，晚产或早产的个体患 1 型糖尿病的风险增加（汇总、多重混杂因素调整后 HR 1.12，95% CI 1.07、1.18；1.15、95% CI 1.11、1.18） ， 分别）。然而，极早产或极早产的婴儿患 1 型糖尿病的风险较低（调整后 HR 分别为 0.63、95% CI 0.45、0.88；调整后 HR 分别为 0.78、95% CI 0.67、0.92）。这三个国家的这些关联都是相似的。

结论/解释

与足月出生的人相比，晚产和早产的人患 1 型糖尿病的风险增加，而极早产或极早产的人患 1 型糖尿病的风险降低。

图形概要