Endocrine Abstracts

Background: Hashimoto’s thyroiditis (HT) is the most common autoimmune disease worldwide. Prevalence of celiac disease (CD) is higher in HT patients and HT is the most prevalent coexisting disease in CD patients. In the last years, interest about gluten free diet (GFD) has increased for its supposed extraintestinal anti-inflammatory effect, so many patients with HT initiate GFD on their own.

Objectives: Aim of this meta-analysis is to examine all quantitative literature available data about the effect of a GFD period on TgAb, TPOAb, TSH, FT4 and FT3 levels in patients with HT and no histological confirm of CD.

Methods: The study was conducted according to MOOSE. The search was performed on databases Medline and Scopus. No language restriction was used. The last search was performed on November 16^th, 2022. Quality assessment was performed. Data collection included patients with HT and no gastrointestinal disease or presenting any gastrointestinal disease with negative histology for CD. Meta-analyses were performed using random-effect model. Statistical analyses were performed using StataSE 17.

Results: The online search retrieved 371 articles and 4 studies with total 87 patients were included for quantitative analysis. The risk of bias was generally low. We observed an overall reduction trend of antibody levels with effect size of -0.39 (CI: -0.81, 0.02; P = 0.06; I²=46.98%) for TgAb and of -0.68 (CI: -1.30, -0.05; P = 0.03; I²=74.73%) for TPOAb, after 6 months of GFD. These heterogeneities were solved with sub-analyses between patients with only HT and patients with HT and non-celiac gluten sensitivity (NCGS) (HT-NCGS) (TgAb P = 0.02) (TPOAb P = 0.01). TSH levels showed a significant reduction with effect size -0.35 (CI: -0.64, -0.05; P = 0.02; I²=0%) and FT4 levels showed a significant elevation with effect size 0.35 (CI: 0.06, 0.64; P = 0.02; I²=0%), after the same period of GFD (6 months). FT3 levels, instead, did not display a significative variation compared to the pre-diet levels (overall effect size 0.05; CI: -0.38, 0.48; P = 0.82; I²=51.24%). However, this heterogeneity was solved with a sub-analysis between HT patients and HT-NCGS patients (FT3 P = 0.04). The performed sub-analyses evidenced statistically significant changes for HT-NCGS patients in TgAb (Overall effect size -0.75; CI: -1.16, -0.34; P < 0.001; I²=0.00%) and TPOAb (Overall effect size -1.15; CI: -1.58, -0.72; P < 0.001; I²=0.00%) levels.

Conclusions: This is the first meta-analysis investigating the effect of GFD on HT. Our results seem to indicate a positive effect of the gluten-deprivation on thyroid function and its inflammation, in particular in patients with HT and NCGS. However, current evidences are not yet sufficient to recommend this dietary approach to all patients with diagnosis of HT.