Hashimoto’s thyroiditis is the most common organ-specific autoimmune disease and the leading cause of thyroid dysfunction in iodine-sufficient regions, affecting nearly 10 percent of the population and predominantly women. For decades, clinicians have relied on two circulating autoantibodies—directed against thyroglobulin and thyroid peroxidase—as diagnostic hallmarks of the condition. Yet a substantial minority of patients carry the characteristic ultrasonographic signature of the disease, with diffuse hypoechogenicity and a heterogeneous thyroid echotexture, while testing negative for these antibodies in their blood. A new retrospective study from Bulgaria now suggests that this serological divide may matter more than expected, because antibody-negative and antibody-positive patients with the same diagnosis appear to carry different cardiometabolic risk profiles, even when their thyroid hormone levels are entirely normal.

The research, conducted at the Department of Endocrinology of St. George University Hospital in Plovdiv, enrolled 163 euthyroid patients with confirmed Hashimoto’s thyroiditis who were not receiving levothyroxine or any other medication known to influence thyroid status, carbohydrate or lipid metabolism, or body weight. Seventy-one patients were antibody-positive, defined by thyroid peroxidase antibody concentrations above 9 IU/L or thyroglobulin antibody concentrations above 4 IU/L, or both. The remaining 92 patients had both antibodies below these thresholds but an established diagnosis based on previous clinical evaluation and characteristic ultrasonographic findings. Thyroid ultrasonography was performed in all participants as part of the diagnostic workup, and all patients had provided written informed consent for the research use of their routinely collected clinical data.

The immunopathogenesis of antibody-negative Hashimoto’s thyroiditis is thought to rely predominantly on cell-mediated mechanisms rather than the humoral, B-cell-driven response that produces the classic autoantibodies. Proposed contributors include natural killer cell activity, regulatory T-cell dysfunction, local intrathyroidal antibody production that never reaches detectable circulating levels, and altered immune regulation influenced by vitamin D deficiency and epigenetic mechanisms. Environmental triggers such as infections, excessive iodine intake, or radiation exposure can release thyroid antigens, which dendritic cells and other antigen-presenting cells process and present to naive T cells in lymph nodes, fueling pro-inflammatory cytokines including interferon-gamma, interleukin 17, and tumor necrosis factor-alpha. However, the clinical and metabolic implications of these antibody-negative mechanisms had remained incompletely understood.

Thyroid hormones are well known to govern energy homeostasis, lipid and glucose metabolism, and blood pressure, but whether thyroid autoantibodies themselves track with metabolic disturbances in euthyroid patients has been controversial. Some earlier studies reported associations between thyroid autoantibodies and disturbances in glucose and lipid metabolism, whereas others found no significant relationship between thyroid peroxidase antibody status and lipid profiles. Evidence directly comparing metabolic and cardiometabolic profiles between antibody-positive and antibody-negative euthyroid patients with Hashimoto’s thyroiditis had been limited, which is precisely the gap the Bulgarian team set out to close.

The results revealed a striking pattern. In unadjusted analyses, the antibody-negative group exhibited a generally less favorable metabolic profile. Compared with antibody-positive patients, they had a significantly higher body mass index, with a median of 31.22 kg/m2 versus 24.85 kg/m2, along with higher fasting insulin, higher HOMA-IR values indicating greater insulin resistance, higher triglyceride concentrations, higher values on the atherogenic index of plasma and the TyG index, and higher uric acid concentrations. Their HDL cholesterol levels were significantly lower. No significant differences emerged between the groups in age, thyroid-stimulating hormone, free triiodothyronine, free thyroxine, total cholesterol, or the liver enzymes ASAT and ALAT, and the sex distribution was statistically similar, with women comprising the large majority in both arms.

The most technically important step came when the researchers repeated the between-group comparisons using analysis of covariance adjusted for age, sex, and BMI, because body mass index was the only baseline characteristic that differed between the groups. After adjustment, antibody-positive patients had 22 percent lower triglyceride concentrations than antibody-negative patients, a geometric mean ratio of 0.78 with a 95 percent confidence interval of 0.65 to 0.92. They also had lower atherogenic index of plasma values, an adjusted mean difference of minus 0.15, and lower TyG index values, an adjusted mean difference of minus 0.37. Differences in fasting insulin, HOMA-IR, and uric acid were attenuated and no longer statistically significant, while the modest advantage in HDL cholesterol persisted, with a geometric mean ratio of 1.10.

Categorical analyses against established risk cutoffs reinforced this picture. Obesity was substantially more frequent among antibody-negative patients, affecting 54.5 percent of that group compared with 27.5 percent of antibody-positive patients, a difference that was highly significant with a Cramer’s V of 0.316. High-risk values on the atherogenic index of plasma were also more frequent in antibody-negative individuals, at 24.7 percent versus 10.4 percent. By contrast, categorical differences in HOMA-IR, fasting glucose within the reference range, HDL cholesterol, triglycerides, and total cholesterol did not reach statistical significance, suggesting that the association with antibody status is more evident for selected lipid-related cardiometabolic measures than for metabolic risk as a whole.

Correlation analyses of antibody titers provided complementary, although modest, evidence. Higher thyroid peroxidase antibody concentrations were weakly associated with lower fasting insulin, lower HOMA-IR, and weak inverse correlations with the atherogenic index of plasma, triglycerides, and the TyG index, while showing no association with fasting glucose, total cholesterol, HDL cholesterol, uric acid, or body mass index. Thyroglobulin antibody concentrations were not significantly associated with any of the examined glucose, lipid, or cardiometabolic parameters. The authors caution that the magnitude of these correlations was small and that they should not be interpreted as evidence of a direct metabolic effect or a protective role of thyroid autoantibodies themselves.

The attenuation of the insulin resistance differences after adjustment for body mass index is particularly relevant given the well-established relationship between adiposity, chronic low-grade inflammation, and impaired insulin signaling. Adipose tissue is increasingly recognized as an active immunometabolic organ, and obesity-related inflammation contributes to metabolic dysfunction, while thyroid dysfunction, adiposity, insulin resistance, and other components of the metabolic syndrome are closely interconnected. The persistence of the triglyceride, HDL cholesterol, atherogenic index, and TyG differences after adjustment, however, indicates that the observed variation in these selected cardiometabolic parameters cannot be attributed solely to differences in adiposity. The findings align with previous reports of inverse associations between thyroid autoimmunity and selected metabolic abnormalities, including a national survey linking thyroglobulin antibody positivity to lower risks of glucose and lipid abnormalities and a Mendelian randomization study identifying an inverse association between genetic susceptibility to thyroid peroxidase antibody positivity and metabolic syndrome, although other population-based studies have reported neutral or less favorable metabolic associations with thyroid autoimmunity.

From a clinical perspective, the study does not support using thyroid autoantibody status as an independent marker of overall metabolic risk. Metabolic risk assessment in patients with Hashimoto’s thyroiditis should continue to be based primarily on established metabolic risk factors, while antibody status may represent an additional characteristic deserving further study. The retrospective, cross-sectional, single-center design precludes causal inference and limits generalizability, and residual confounding related to adiposity and other unmeasured metabolic factors cannot be excluded, since body mass index does not fully reflect body composition or visceral adiposity and information on waist circumference, dietary habits, physical activity, smoking status, and disease duration was not consistently available. Prospective longitudinal studies incorporating more detailed measures of body composition, lifestyle factors, and disease duration are needed to clarify whether the antibody-related cardiometabolic differences observed here hold up over time and what they mean for the individualized follow-up of the millions of people living with this common autoimmune condition.

Subject of Research: Thyroid autoantibody status and cardiometabolic characteristics in euthyroid Hashimoto's thyroiditis

Article Title: Thyroid autoantibody status and cardiometabolic characteristics in euthyroid Hashimoto’s thyroiditis

Article References: Nacheva, S., Nyagolova, P., Yaneva, A., Ivanovska, M., Troev, D., Mihaylova, A., & Orbetzova, M. (2026). Thyroid autoantibody status and cardiometabolic characteristics in euthyroid Hashimoto’s thyroiditis. Pharmacia, 73, Article e213001. https://doi.org/10.3897/pharmacia.73.e213001

Image Credits: AI Generated

DOI: 10.3897/pharmacia.73.e213001

Keywords: Hashimoto's thyroiditis, thyroid autoantibodies, insulin resistance, triglycerides, HDL cholesterol, cardiometabolic risk, TyG index, atherogenic index of plasma, Thyroid, autoantibody, status, cardiometabolic