Is thyroid autoimmunity associated with gestational diabetes mellitus? Paquita Montaner a, ⁎ , Lluisa Juan b , Raquel Campos c , Llorenç Gil c , Rosa Corcoy d a Servei de Medicina Interna, Hospital Sant Joan de Déu de Martorell, 08760 Martorell, Spain b Servei de Bioquímica Clínica, Hospital Sant Joan de Déu de Martorell, 08760 Martorell, Spain c Servei de Ginecologia i Obstetrícia, Hospital Sant Joan de Déu de Martorell, 08760 Martorell, Spain d Servei d’Endocrinologia i Nutrició, Hospital de la Santa Creu i Sant Pau, 08025, Barcelona, Spain Received 1 June 2007; accepted 12 November 2007 Abstract Because inflammatory markers have been associated with incident diabetes mellitus, we aimed to evaluate if thyroid peroxidase antibodies (TPOAbs) in early pregnancy are predictive of gestational diabetes mellitus (GDM). Six hundred nineteen pregnant women without former diabetes mellitus were evaluated for TPOAb positivity after booking. A universal GDM screening protocol and the Third Workshop–Conference criteria were used for GDM detection. In addition to bivariate analysis, multivariate logistic regression models were constructed with GDM as the dependent variable and TPOAb positivity as one of the potential predictive ones. The rate of TPOAb positivity was 10%; and that of GDM, 6.9% (6.8% in women without and 8.1% in women with TPOAb positivity, not significant). Thyroid peroxidase antibodies did not enter the multivariate logistic regression model to predict GDM that identified the following independent predictive variables: maternal age (odds ratio [OR] 1.5, 95% confidence interval [CI] 0.57-4.0 for the second tertile; OR 2.84, 95% CI 1.16-6.96 for the third tertile), prior GDM (OR 9.38, 95% CI 3.34-26.39), and diabetes mellitus in first-degree relatives (OR 3.22, 95% CI 1.65-6.27). In conclusion, we have not identified TPOAb positivity in early pregnancy as a predictor of GDM. © 2008 Elsevier Inc. All rights reserved. 1. Introduction Inflammatory markers have been related to incident diabetes mellitus in middle-aged and elderly subjects [1,2], Pima Indians [3], and pregnant women [4,5]. The underlying mechanism appears to be increased insulin resistance more than decreased insulin secretion [6]. A high prevalence of positivity for thyroid peroxidase antibodies (TPOAbs) has been reported in both women with gestational diabetes mellitus (GDM) [7] and those with polycystic ovary syndrome [8]. The potential link between thyroid auto- immunity and these conditions could be insulin resistance because serum levels of inflammatory cytokines are increased in patients with thyroid autoimmunity [9]. Similarly, patients with subclinical hypothyroidism of autoimmune origin display increased C-reactive protein, which is not modified with thyroid function status [10]. This feature is parallel to the increase in C-reactive protein described for prediabetic subjects with islet cell autoimmu- nity in some [11] but not all articles [12]. We hypothesized that positivity for TPOAbs is associated with GDM, presumably through increased insulin resistance. 2. Material and methods 2.1. Sample size calculation The sample size was calculated using Sample Power 1.0 with the following assumptions: (1) rate of TPOAb positivity = 25% in women with GDM [7] and 10% in women without GDM [13], (2) a 10% prevalence of GDM [14], (3) a bilateral a error of .05 and a b error of .80. The estimated sample size was 600. 2.2. Inclusion criteria • First analytical assessment at a gestational age less than 20 weeks. 2.3. Exclusion criteria • Prepregnancy diabetes mellitus. • Inadequate assessment of TPOAb positivity or glucose tolerance during pregnancy. Available online at www.sciencedirect.com Metabolism Clinical and Experimental 57 (2008) 522 – 525 www.elsevier.com/locate/metabol ⁎ Corresponding author. Tel.: +34 93 774 20 20; fax: +34 93 775 3706. E-mail address: 12657fmb@comb.es (P. Montaner). 0026-0495–/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.metabol.2007.11.015