Background

Studies exploring the associations between perfluoroalkyl acids (PFAAs) and type 2 diabetes mellitus (T2DM) are rather limited and have reported conflicting results. All studies to date, including prospective ones, have relied on a single blood sample to study this association. Similarly, studies investigating how T2DM status may influence the longitudinal changes in PFAA concentrations have not been previously performed. As PFAA concentrations in humans have changed considerably over the last two decades, and as individuals diagnosed with T2DM usually undergo lifestyle changes that could influence these concentrations, a single blood sample may not necessarily reflect the life-time exposure to PFAA concentrations. Hence, repeated measurements from the same individuals will extend our understanding of how PFAAs are associated with T2DM. The present study, therefore, aimed to explore associations between pre- and post-diagnostic PFAA blood profiles and T2DM and assess factors associated with longitudinal changes in PFAAs in T2DM cases and controls.

Methods

Questionnaire data and blood samples from women participating in the Norwegian Women and Cancer study were used to conduct a nested case-control study among 46 T2DM cases matched to 85 non-diabetic controls. PFAAs were measured in blood samples collected prior to (2001/02) and after (2005/6) T2DM diagnosis. We investigated the association between PFAAs and incident and prevalent T2DM using conditional logistic regression. We assessed the longitudinal changes in PFAA concentrations within and between matched cases and controls using t-tests and linear regression models.

Results

We observed no significant associations between pre-diagnostic PFAA concentrations and T2DM incidence. Similar results were observed for the post-diagnostic PFAA concentrations and T2DM prevalence. Decrease over time in PFAA concentrations were observed for PFOA and ∑PFOS concentrations, whereas increase over time were observed for PFNA, PFDA and PFUnDA concentrations. Longitudinal trends in PFAA concentrations among T2DM cases were similar to the changes observed in controls.

Conclusions

The study did not find evidence of association between PFAAs and incident or prevalent T2DM. The longitudinal changes in PFAAs concentrations were not influenced by T2DM status.

中文翻译：

---

2型糖尿病病例和对照中全氟烷基酸的诊断前和诊断后血液概况。

背景

探索全氟烷基酸（PFAAs）与2型糖尿病（T2DM）之间关联的研究非常有限，并且报告了相互矛盾的结果。迄今为止，包括前瞻性研究在内的所有研究均依靠单一血样来研究这种关联。同样，以前尚未进行过研究T2DM状态如何影响PFAA浓度的纵向变化的研究。在过去的二十年中，由于人类的PFAA浓度已经发生了巨大变化，而且诊断为T2DM的个体通常会经历可能影响这些浓度的生活方式变化，因此一次血液采样不一定能反映出其终生暴露于PFAA浓度的情况。因此，来自同一个人的重复测量将扩展我们对PFAA如何与T2DM相关的理解。

方法

参加挪威妇女与癌症研究的妇女的问卷调查数据和血液样本被用于对46例T2DM病例和85例非糖尿病对照进行巢式病例对照研究。在T2DM诊断之前（2001/02）和之后（2005/6）采集的血液样本中测量PFAA。我们使用条件对数回归研究了PFAA与事件和流行的T2DM之间的关联。我们使用t检验和线性回归模型评估了匹配病例与对照之间及其之间PFAA浓度的纵向变化。

结果

我们观察到诊断前PFAA浓度与T2DM发病率之间无显着关联。对于诊断后的PFAA浓度和T2DM患病率，观察到相似的结果。PFAA和∑PFOS浓度的PFAA浓度随时间下降，而PFNA，PFDA和PFUnDA的浓度则随时间增加。T2DM病例中PFAA浓度的纵向趋势与对照组中观察到的变化相似。

结论

该研究没有发现PFAA与事件或流行的T2DM之间存在关联的证据。PFAAs浓度的纵向变化不受T2DM状态的影响。