Vitamin B12 and thyroid autoimmunity: Comparative insights from Hashimoto's and Graves' disease

Introduction

Globally, thyroid pathologies constitute a major public health challenge, with prevalence rates reaching ~5% across populations. Predominant manifestations include autoimmune thyroid disease (AITD), particularly Hashimoto's thyroiditis (HT) and Graves' disease (GD). These conditions develop through immune-mediated inflammatory processes involving lymphocyte invasion and elevated concentrations of thyroid-specific antibodies (1). The thyroid is exceptionally susceptible to autoimmune dysfunction (2). Within populations maintaining an adequate iodine consumption, the aetiology of thyroid malfunction predominantly involves autoimmune pathways rather than dietary insufficiencies (3).

Building on this autoimmune framework, HT is a chronic autoimmune condition typically causing hypothyroidism. It is characterised by the destruction of thyroid follicles, leading to low T3/T4 and elevated thyroid-stimulating hormone (TSH) levels, and is often accompanied by anti-thyroid peroxidase (anti-TPO) antibodies (4). Conversely, GD causes hyperthyroidism with thyrotoxicosis, goitre and orbitopathy driven by stimulating thyrotropin receptor antibodies (1). Both conditions are markedly more common among women and can progress through hyper-, eu-, and hypothyroid phases as thyroid damage evolves (5).

Concurrent with these autoimmune thyroid disorders, vitamin B12 deficiency represents a clinically significant nutritional concern, as this water-soluble vitamin serves as a vital cofactor for DNA biosynthesis and cellular metabolism. Due to a low dietary intake or malabsorption, B12 deficiency is relatively common and may produce non-specific, yet severe haematological and neurological effects (6). The clinical evaluation of B12 status relies on total serum B12 as the primary biomarker, supplemented by holotranscobalamin for enhanced diagnostic precision (7,8).

Moreover, the association between B12 deficiency and autoimmune conditions warrants particular attention. While affecting only 3 to 4% of the general population, vitamin B12 deficiency demonstrates a significantly higher prevalence among patients with autoimmune diseases, including autoimmune thyroiditis (AIT) (9). Given these patterns, the B12-thyroid association is particularly relevant in the Iraqi population, among whom both AITDs and nutritional deficiencies, such as vitamin B12 deficiency, are prevalent health concerns (10,11). This association is strengthened by the frequent coexistence of AITDs with other autoimmune disorders, including type 1 diabetes, vitiligo and particularly pernicious anaemia, which directly impairs B12 absorption through autoimmune gastritis (1). Furthermore, recent research investigating the roles of B12, folate, vitamin D and anaemia in HT has suggested that the metabolic status can influence thyroid autoimmunity, underlining the importance of nutritional factors in thyroid immunity (12).

Considering such convergences, B12 deficiency symptoms mimic thyroid dysfunction, requiring diligent evaluation. Thus, the early identification of B12 deficiency is essential in patients with thyroid disorders. Symptoms such as fatigue, cognitive impairment and neuropathy often overlap with thyroid disorders, potentially delaying diagnosis and leading to suboptimal treatment outcomes (13). A direct comparison of the vitamin B12 status between patients with HT and GD could therefore reveal disease-specific patterns of deficiency, inform targeted screening strategies, and lead to an improvement in patient management protocols in clinical practice.

While serum vitamin B12 levels have been examined in the context of thyroid dysfunction, prior research reveals contradictory evidence (14). Direct comparisons of the vitamin B12 status between patients with HT and GD, particularly in Middle Eastern populations, such as in Iraq, remain limited, despite their clinical importance. Therefore, in order to strengthen the evidence-based research in this understudied area, the present case-control study was designed to assess and compare vitamin B12 levels across patients with HT, GD and healthy controls, while investigating correlations with thyroid function parameters [TSH and free thyroxine (FT4)] and autoimmune markers [thyroid peroxidase antibodies (anti-TPO) and thyrotropin receptor antibodies (TRAbs)].

Subjects and methods

Research design and sample size determination

Designed with a case-control method, the present study was conducted between December, 2024 and March, 2025 at the Thyroid Centre, Smart Health Tower, Sulaymaniyah, Iraq. Patient recruitment involved the random selection of cases among individuals with confirmed HT or GD. All diagnoses were established prior to study inclusion by a team of specialised physicians following standard clinical protocols. Healthy controls were recruited from among hospital staff and individuals presenting for routine check-ups, after ensuring they had no history of thyroid or autoimmune diseases.

In order to establish the required sample size, a power analysis was performed using G*Power statistical software for Windows (Universität Düsseldorf, Düsseldorf, Germany), setting included α at 0.05, power at 0.95 and effect size at 0.4 (14,15). The minimal sample size required was determined as 102 participants, which was increased to 120 to compensate for potential procedural errors. The three experimental arms consisted of the HT, GD and control groups, preserving a balanced 1:1:1 proportion across the enrolment process. Each group was composed of an equal number of 40 participants.

Participant selection criteria

The inclusion criteria comprised of adults aged 25-60 years with a confirmed diagnosis of HT or GD for the case groups and the absence of autoimmune or major health conditions for the control group. The exclusion criteria included recent vitamin B12 supplementation over the past 6 months, the presence of non-thyroidal autoimmune diseases, a history of thyroidectomy, gastrointestinal surgery affecting nutrient absorption, a vegetarian diet and pregnancy or lactation. In order to minimise the potential for confounding variables, the ages and sex of the participants were balanced across the case and control groups (16).

Data collection and laboratory assessments

In order to capture a comprehensive patient profile for research, a structured questionnaire was designed to obtain demographic data, such as age and sex, medical backgrounds including surgical history, thyroid treatment use, a family history of autoimmune conditions and disease duration, alongside lifestyle factors, such as the use of supplements, dietary habits, smoking status and alcohol consumption, and any relevant health symptoms. The body mass index (BMI), measured in kg/m2, was determined by dividing the recorded weight by the height squared. Based on WHO standards, participants were classified as underweight, healthy weight, overweight, or obese according to their BMI thresholds of <18.5, 18.5 to 24.9, 25.0 to 29.9, and ≥30.0 in kg/m2, respectively (17).

All study participants provided a blood sample without prior fasting. For patients with thyroid disorders, sampling coincided with routine clinical visits for thyroid monitoring, whereas control participants provided samples within the same time frame. To minimise temporal variability, samples were collected at comparable times for all participants. A volume of 3 ml venous blood was drawn into VACUETTE® serum separator tubes (Greiner Bio-One) containing a clot activator and separation gel. The sample was allowed to clot at room temperature for ~20 min, and then centrifuged at 2,500 x g for 15 min at 20-25˚C using a Hettich RotoFix 32A centrifuge. The separated serum was immediately analysed for the quantitative determination of vitamin B12, TSH, FT4, anti-TPO and TRAb concentrations. Laboratory analyses were conducted using the cobas® pro e 801 analysers (Roche Diagnostics GmbH), which employ electrochemiluminescence technology for immunoassay detection with daily internal quality controls and standard calibration. Reference ranges were established according to manufacturer specifications as follows: Serum vitamin B12: Reference range, 197-771 pg/ml; categorised as deficiency: <197 pg/ml; borderline, 197-300 pg/ml; and normal, 300-771 pg/ml; TSH: Reference range, 0.4-4.2 µIU/ml; FT4: Reference range, 12-22 pmol/l; anti-TPO: Reference range, <35 IU/ml; and TRAbs: Reference range, <1.75 IU/l (18,19). While the guidelines exhibit some inconsistency in terms of the exact threshold values for vitamin B12 deficiency and no single universal cut-off is recognised, this scheme is concordant with commonly adopted international thresholds, which typically define deficiency at <200 pg/ml and a borderline range of 200-300 pg/ml (20,21). Anti-TPO was measured specifically in the HT cases, and TRAbs in the GD cases, while both antibodies were evaluated in the control group.

Ethical considerations

The present study was executed in strict adherence to the ethical principles of the Declaration of Helsinki, safeguarding the rights and dignity of the participants. Formal approval was granted by the Ethics Board of the College of Health and Medical Technology at Sulaimani Polytechnic University (reference no. 30/245, dated December 1, 2024). After receiving a thorough explanation of the purpose of the study, processes, possible risks and advantages, participants provided written informed consent. Additionally, they were made aware of their freedom to discontinue participation at any point during the study without consequence. All data were collected and stored in accordance with institutional privacy and confidentiality guidelines (22).

Statistical analysis

The evaluation of data in the present study was carried out using IBM SPSS statistical software, version 26.0 (IBM Corp.), to execute a range of statistical computations necessary for the analysis. The normality of continuous variables was tested using the Shapiro-Wilk test to ascertain their distribution patterns. It was observed that the majority of continuous variables, including TSH, FT4, anti-TPO, TRAbs and vitamin B12 concentrations, did not conform to a normal distribution, leading to the decision to report the results as the median and quartile range (QR) for a precise depiction of central tendencies and variability. For comparisons between two distinct groups, the Mann-Whitney U test was used for non-parametric data, and the independent samples t-test was applied for parametric data to detect significant differences, while the Kruskal-Wallis test was applied for analyses involving three or more groups to assess variations across categories. When the Kruskal-Wallis test indicated statistical significance, post hoc pairwise comparisons were performed using Dunn's test with Bonferroni correction to adjust for multiple testing. Categorical data, encompassing variables such as sex, smoking status, residential location, familial history, vitamin B12 status, and BMI classifications-were summarised as frequencies and corresponding percentages, with differences being evaluated using either the Chi-squared test or Fisher's exact test, selected based on the expected frequencies in contingency tables. To explore the potential correlations between serum vitamin B12 levels and thyroid-related markers, such as TSH, FT4, anti-TPO, and TRAbs, Spearman's rank correlations were reported with 95% confidence intervals (CIs) to indicate the precision of association estimates. Independent associations between vitamin B12 status and thyroid autoimmunity were initially evaluated using univariate logistic regression and subsequently assessed with multivariate logistic regression adjusted for BMI, residence, and smoking. All statistical tests were conducted as two-tailed to ensure a thorough evaluation. A value of P<0.05 was considered to indicate a statistically significant difference.

Results

The baseline demographic and clinical characteristics of the 120 study participants are presented in Table I. The median age of all the participants was 42.5 years, with no significant difference observed between the patient group (median, 42.0 years) and the healthy group (median, 43.0 years) (P=0.640). Similarly, there was no significant difference in sex distribution between the patients (65.8% males and 67.1% females) and the healthy controls (34.2% males and 32.9% females) (P=0.890). The mean BMI was comparable between the patient group (29.50±5.4) and the healthy group (28.83±4.58), with no statistically significant difference (P=0.515). Significant differences were found in terms of the family history of autoimmune thyroiditis (AIT), smoking status and residence of the participants. A significantly higher proportion of patients (78.9%) reported a family history of AIT compared to the healthy individuals (21.1%) (P=0.007). As regards smoking status, a significantly greater percentage of patients (93.3%) were smokers compared to the healthy group (6.7%) (P=0.019). Furthermore, a significantly higher proportion of patients resided in rural areas (93.2%) compared to the healthy participants (6.8%), while a larger percentage of healthy individuals resided in urban areas (48.7%) compared to the patients (51.3%) (P<0.001) (Table I).

Table I Foundational demographic and clinical attributes of the study participants.

Table I

Foundational demographic and clinical attributes of the study participants.

Variables	Total	Patients	Healthy	P-value
Age (years), median (QR)	42.5 (35.0-47.0)	42.0 (34.5-47.0)	43.0 (35.5-46.5)	0.640
Sex, n (%)
Male	38 (100.0)	25 (65.8)	13 (34.2)	0.890
Female	82 (100.0)	55 (67.1)	27 (32.9)
BMI (mean ± SD)	29.3±5.1	29.50±5.4	28.83±4.58	0.515
Family history of AIT, n (%)
Yes	57 (100.0)	45 (78.9)	12 (21.1)	0.007
No	63 (100.0)	35 (55.6)	28 (44.4)
Smoking status, n (%)
Yes	15 (100.0)	14 (93.3)	1 (6.7)	0.019
No	105 (100.0)	66 (62.9)	39 (37.1)
Place of residence, n (%)
Urban	76 (100.0)	39 (51.3)	37 (48.7)	<0.001
Rural	44 (100.0)	41 (93.2)	3 (6.8)

[i] QR, quartile range; SD, standard deviation; AIT, autoimmune thyroiditis; BMI, body mass index.

Among the 120 study participants, vitamin B12 deficiency was identified in 13 individuals (10.8%), with the highest prevalence observed in the healthy control group (n=6, 46.2%), followed by the participants with GD (n=5, 38.5%) and HT (n=2, 15.4%). Borderline vitamin B12 levels were documented in 38 participants (31.7%), with the majority concentrated in the HT group (n=18, 47.4%), while the GD and healthy control groups each contributed 10 participants (26.3% each). A normal vitamin B12 status was found in 69 participants (57.5%), distributed relatively evenly across the three groups: GD (n=25, 36.2%), healthy controls (n=24, 34.8%) and HT (n=20, 29.0%). Statistical analysis demonstrated no significant association between the vitamin B12 status categories and study group classification (P=0.215). In alignment with the categorical distribution of the vitamin B12 status, the analysis of serum vitamin B12 concentrations revealed no statistically significant differences among the study groups (P=0.556). The median vitamin B12 level for the total study population was 318.0 pg/ml (QR, 255-427.5). When assessed by group, the HT group demonstrated a median value of 301.5 pg/ml (QR, 250.5-370.5), the GD group had a median of 321.0 pg/ml (QR, 269.0-445.0), and the healthy control group exhibited a median of 347.5 pg/ml (QR, 247.0-438.0) (Table II).

Table II Comparison of vitamin B12 status, thyroid function parameters and thyroid autoantibodies among participants with Hashimoto's thyroiditis, Graves' disease and healthy controls.

Table II

Comparison of vitamin B12 status, thyroid function parameters and thyroid autoantibodies among participants with Hashimoto's thyroiditis, Graves' disease and healthy controls.

Variables	Total	Hashimoto's thyroiditis	Graves' disease	Healthy controls	HT vs. GDa	HT vs. healthy a	GD vs healthya	P-value
Vitamin B12 status
Deficient	13 (100.0)	2 (15.4)	5 (38.5)	6 (46.2)	-	-	-	0.215
Borderline	38 (100.0)	18 (47.4)	10 (26.3)	10 (26.3)
Normal	69 (100.0)	20 (29.0)	25 (36.2)	24 (34.8)
Vitamin B12, median (QR)	318.0 (255-427.5)	301.5 (250.5-370.5)	321.0 (269.0-445.0)	347.5 (247.0-438.0)	-	-	-	0.556
					-	-	-
TSH level, median (QR)	1.875 (0.925-3.990)	5.045 (2.020-8.505)	0.626 (0.005-2.935)	1.565 (1.330-2.075)	<0.001	<0.001	<0.001	<0.001
FT4, median (QR)	15.85 (13.45-18.50)	16.25 (12.47-18.75)	16.0 (12.5-32.20)	15.50 (14.35-17.55)	-	-	-	0.876
Anti-TPO, median (QR)	26.55 (11.90-280.50)	280.50 (146.0-511.5)	-	11.90 (9.72-15.55)	-	-	-	<0.001
TRAb antibody, median (QR)	1.37 (0.87-3.55)	-	3.55 (2.11-9.28)	0.88 (0.80-1.05)	-	-	-	<0.001

[i] ^aPairwise comparisons were conducted only for variables with significant overall group differences. QR, quartile range; TSH, thyroid-stimulating hormone; FT4, free thyroxine; anti-TPO, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody; HT, Hashimoto's thyroiditis; GD, Graves' disease.

Thyroid function parameters demonstrated distinct patterns across the study groups. The TSH levels varied significantly among the participants, with the HT group exhibiting the highest median concentration at 5.045 µIU/ml (QR, 2.020-8.505), while the GD group exhibited the lowest at 0.626 µIU/ml (QR, 0.005-2.935) and the healthy controls had intermediate values at 1.565 µIU/ml (QR, 1.330-2.075). Statistical analysis confirmed a significant difference in the TSH concentrations between the three groups (P<0.001) (Table II and Fig. 1A). Conversely, the FT4 levels remained comparable across all groups, with values of 16.25 pmol/l (QR, 12.47-18.75), 16.0 pmol/l (QR, 12.5-32.20) and 15.50 pmol/l (QR, 14.35-17.55) for HT, GD and healthy controls, respectively (P=0.876) (Table II).

Figure 1 Dot plots demonstrating the serum (A) TSH, (B) anti-TPO and (C) TRAb concentrations for Hashimoto's thyroiditis, Graves' disease and the control groups. Each dot represents a participant. Median values are displayed as follows: Solid line, healthy controls; dashed line, Hashimoto's thyroiditis; dotted line, Graves' disease. Logarithmic scales are applied to the y-axes. TSH, thyroid-stimulating hormone; anti-TPO, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody.

Thyroid autoantibody measurements revealed characteristic patterns specific to each disease group. The anti-TPO antibody concentrations were substantially higher in the HT group, with a median of 280.50 IU/ml (QR, 146.0-511.5). This contrasted sharply with the healthy controls, who had a median of 11.90 IU/ml (QR, 9.72-15.55). The overall median anti-TPO antibody level across all participants was 26.55 IU/ml (QR,11.90-280.50), with statistical analysis confirming a highly significant difference between groups (P<0.001). TRAb concentrations exhibited a similar pattern of group-specific elevation, with participants in the GD group exhibiting markedly increased levels at 3.55 IU/l (QR, 2.11-9.28) compared to healthy controls at 0.88 IU/l (QR, 0.80-1.05). The overall median TRAb concentration was 1.37 IU/l (QR, 0.87-3.55), with statistical analysis confirming significant differences between the groups (P<0.001) (Table II, and Fig. 1B and C).

These distinct biochemical and immunological profiles are clearly illustrated in Fig. 1. TSH values cluster at elevated levels for HT and are suppressed in GD. The anti-TPO concentrations exhibited a pronounced increase in the HT group compared to the controls, and TRAb values were distinctly elevated in the GD group, but remained low in the controls. This graphical representation reinforces the specific diagnostic and pathophysiological features distinguishing each group (Table II and Fig. 1).

The analysis of the vitamin B12 status concerning demographic and clinical characteristics revealed that obesity was the predominant BMI category among the participants with vitamin B12 deficiency (61.5%), followed by the normal BMI (23.1%) and overweight (15.4%) categories. A similar pattern was observed in the borderline group, where obesity also constituted the largest subgroup (44.7%), with the overweight (39.5%) and normal BMI (15.8%) categories also represented. Among the individuals with normal vitamin B12 levels, the overweight category was the most prevalent (47.8%), while the obesity (31.9%) and normal BMI (20.3%) categories were also noted (Table III).

Table III Distribution of demographic and clinical characteristics by vitamin B12 status.

Table III

Distribution of demographic and clinical characteristics by vitamin B12 status.

	Vitamin B12 status
Parameters, n (%)	Deficient (n=13)	Borderline (n=38)	Normal (n=69)	P-value
BMI group
Underweight	0 (0.0)	0 (0.0)	0 (0.0)	0.168
Healthy weight	3 (23.1)	6 (15.8)	14 (20.3)
Overweight	2 (15.4)	15 (39.5)	33 (47.8)
Obese	8 (61.5)	17 (44.7)	22 (31.9)
Sex
Male	2 (15.4)	11 (28.9)	25 (36.2)	0.326
Female	11 (84.6)	27 (71.1)	44 (63.8)
Age group, years				0.533
25-30	0 (0.0)	4 (10.5)	9 (13.0)
31-35	2 (15.4)	5 (13.2)	12 (17.4)
36-40	1 (7.7)	8 (21.1)	10 (14.5)
41-45	8 (61.5)	8 (21.1)	13 (18.8)
46-50	2 (15.4)	8 (21.1)	16 (23.2)
51-55	0 (0.0)	4 (10.5)	6 (8.7)
56-60	0 (0.0)	1 (2.6)	3 (4.3)
Disease durationa				0.649
<1 year	2 (28.6)	7 (25.0)	15 (33.3)
From 1 to 3 years	3 (42.9)	6 (21.4)	10 (22.2)
From 3 to 5 years	2 (28.6)	7 (25.0)	8 (17.8)
>5 years	0 (0.0)	8 (28.6)	12 (26.7)

[i] ^aThis was calculated only for the patients (Hashimoto's thyroiditis and Graves' disease groups). BMI, body mass index.

As regards sex distribution, females constituted the majority across all vitamin B12 classifications, notably among those exhibiting deficiency (84.6%) and borderline levels (71.1%). As for age, vitamin B12 deficiency was notably clustered within the 41-45-year cohort (61.5%). By contrast, the participants with a borderline status demonstrated a more uniform age distribution, primarily spanning the 36-50-year range. Those with normal vitamin B12 levels exhibited a wider age spread, with the 46-50-year category containing the largest proportion (23.2%) (Table III).

The assessment of disease duration within the patients with AIT indicated that vitamin B12 deficiency was most prevalent among individuals diagnosed for 1 to 3 years (42.9%); the remaining deficient cases were equally divided between disease durations of <1 year and 3 to 5 years (28.6% each). Participants with borderline vitamin B12 levels displayed a varied distribution of disease duration: 28.6% had a duration >5 years, 25.0% each were in the <1 year and 3 to 5 year categories, and 21.4% had a duration of 1 to 3 years. Among the individuals with normal vitamin B12 levels, the largest subgroup (33.3%) had a disease duration of <1 year, followed by those with a duration of >5 years (26.7%) (Table III).

Spearman's correlation analysis was performed to assess the correlation between serum vitamin B12 concentrations and various thyroid-related parameters, including TSH, FT4, TRAbs and anti-TPO antibodies. The choice of Spearman's correlation analysis was based on the non-normal distribution of the data. The analysis revealed that the correlation coefficient (rs) between vitamin B12 and TSH was 0.007, with a P-value of 0.942, indicating no observable correlation between these variables (95% CI, -0.158 to 0.183). Similarly, the correlation between vitamin B12 and FT4 was weakly positive (rs=0.075; P=0.418), suggesting a negligible association (95% CI, -0.098 to 0.234). For the thyroid autoantibodies, TRAbs displayed a slight negative correlation with vitamin B12 (rs=-0.034; P=0.763; 95% CI, -0.275 to 0.214), while anti-TPO antibodies exhibited a modest negative correlation (rs=-0.125, P=0.268; 95% CI, -0.339 to 0.088). However, none of these associations reached statistical significance, as all P-values exceeded the conventional threshold (Table IV and Fig. 2).

Figure 2 Spearman's rank correlation between the serum vitamin B12 level and thyroid-related parameters among study participants. (A-D) Scatter plots depicting the correlations between serum vitamin B12 levels and thyroid function or autoimmune markers: (A) TSH, (B) FT4, (C) TRAb, and (D) anti-TPO. Logarithmic scales are applied to the y-axes. TSH, thyroid-stimulating hormone; FT4, free thyroxine; TRAb, thyrotropin receptor antibody; Anti-TPO, anti-thyroid peroxidase antibodies.

Table IV Spearman's rank correlation analysis between serum vitamin b12 levels and thyroid-related parameters in study participants.

Table IV

Spearman's rank correlation analysis between serum vitamin b12 levels and thyroid-related parameters in study participants.

Variables	No. of participants	rs	P-value	95% CI lower	95% CI upper
Vitamin B12 level and TSH	120	.007	.942	-0.158	0.183
Vitamin B12 level and FT4	120	.075	.418	-0.098	0.234
Vitamin B12 level and TRAba	80	-.034	.763	-0.275	0.214
Vitamin B12 l level and anti-TPOb	80	-.125	.268	-0.339	0.088

[i] ^aThis was calculated only for the healthy control and the Graves' disease group;

[ii] ^bthis was calculated only for the healthy control and the Hashimoto's thyroiditis group. Significance of the correlation is established at the 0.01 level (two-tailed). TSH, thyroid-stimulating hormone; FT4, free thyroxine; anti-TPO, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody.

To further explore the determinants of the vitamin B12 status among the study participants, a univariate logistic regression analysis was initially conducted (Table V). BMI exhibited an odds ratio (OR) of 0.946 (95% CI, 0.846-1.058; P=0.329), rural residence had an OR of 0.483 (95% CI, 0.125-1.859; P=0.290), and smoking status had an OR of 1.806 (95% CI, 0.218-14.989; P=0.584); none of these factors reached statistical significance. A subsequent multivariate logistic regression model including the same variables confirmed these finding. Similarly, none of the examined factors emerged as statistically significant predictors of the vitamin B12 status. Specifically, BMI [odds ratio (OR), 1.046; 95% CI, 0.933-1.174; P=0.441], the place of residence (urban vs. rural: OR, 0.507; 95% CI, 0.131-1.969; P=0.327) and smoking status (OR, 1.506; 95% CI, 0.172-13.157; P=0.711) did not demonstrate independent associations with vitamin B12 deficiency following adjustment for confounding factors.

Table V Univariate and multivariate logistic regression of factors associated with vitamin B12 status.

Table V

Univariate and multivariate logistic regression of factors associated with vitamin B12 status.

	Univariate analysis					Multivariate analysis
Variables	B	SE	Exp(B) (OR)	95% CI for OR	P-value	B	SE	Exp(B) (OR)	95% CI for OR	P-value
BMI	-0.056	0.057	0.946	0.846-1.058	0.329	-0.045	0.059	1.046	0.933-1.174	0.441
District (residence)
Urbana
Rural	-0.728	0.688	10.483	0.125-1.859	0.290	-0.679	0.692	10.507	0.131-1.969	0.327
Smoking status
Yesa
No	-0.591	1.080	11.806	0.218-14.989	0.584	-0.409	1.106	11.506	0.172-13.157	0.711

[i] ^aReference category for comparison in logistic regression model. Categorical variable coding: Urban/Yes=reference; Rural/No=comparison group. B, regression coefficient; SE, standard error; OR, odds ratio; CI, confidence interval.

Discussion

The analysis of the demographic data in the present study revealed patterns that are largely consistent with both regional and international literature. The pronounced female predominance in the HT and GD groups is a well-established epidemiological feature of AITD, consistent with the findings from the studies by Zoori and Mousa (23) in Nasiriya, Iraq, and by Aon et al (24) in Kuwait, which similarly identified a notably higher occurrence of AITD among females in their respective regions. The median age of the participants in the present study was 42.5 years, which was also comparable to the mean age of 42.48 years reported in the Indian study by Kaur et al (25) and that of 41.21 years in the Iraqi study by Abed et al (26) for patients with an autoimmune hypothyroid and GD, positioning the present study cohort within a typical age range for AITD diagnosis. The significantly higher rates of a positive family history of AITD in the patient groups herein reinforce the strong genetic component of these diseases. The observed associations with smoking and rural residence are novel findings that warrant further investigation, as they may point to specific environmental triggers or lifestyle factors relevant to our regional population.

The present study validated participant grouping with distinct thyroid function and autoantibody profiles across HT, GD and the controls, demonstrating significant differences in TSH, anti-TPO and TRAb levels (all P<0.001), while FT4 remained comparable (P=0.876). The hormonal and immunological patterns align with the synthesis in the study by Vargas-Uricoechea (27), who confirmed TSH and FT4 as the foundational laboratory tests in the initial assessment of thyroid dysfunction and noted the defining roles of TRAb and anti-TPO antibodies in differentiating autoimmune thyroid disease entities. The comparable FT4 levels across groups suggest that a number of participants were in subclinical stages or under effective therapy. This biochemical distinction confirms the precision of the disease classification and underpins the reliability of the analyses in the present study.

The principal finding of the present study was the lack of a statistically significant association between the serum B12 level and the presence of either HT or GD in the Iraqi cohort (P=0.215). The analysis demonstrated no significant variations in themedian B12 concentrations or the prevalence of B12 deficiency among the HT, GD and healthy control groups (P=0.556). Moreover, vitamin B12 levels did not significantly correlate with any of the measured thyroid function or autoimmune markers. This outcome contributes a crucial, albeit null, finding to a field characterised by markedly conflicting evidence and highlights the complexity of the relationship between vitamin B12 and autoimmune thyroid disorder.

The absence of significant group differences may be explained by several physiological and genetic factors. For example, pernicious anaemia, a manifestation of autoimmune gastritis, leads to vitamin B12 malabsorption. Notably, intrinsic factor antibodies and anti-parietal cell antibodies, which play a decisive role in B12 absorption (28), were not screened in the present study cohort. Their absence may partially explain the lack of B12 deficiency detected, although they are recognised mediators in the link between gastric and thyroid autoimmunity. Moreover, pernicious anaemia is observed only in a subset of patients with thyroid disorders. According to Vaqar and Shackelford (28), up to one-quarter of individuals with autoimmune gastritis develop pernicious anaemia, and these patients often present with coexisting autoimmune conditions, including thyroiditis. However, such an overlap may vary across populations, potentially reflecting a low prevalence of gastric autoantibody positivity in some regions or differing genetic backgrounds compared with populations in which thyroid-gastric associations are more common. A recent genome-wide association study identified susceptibility loci for pernicious anaemia, such as PTPN22, HLA-DQB1, IL2RA and AIRE, that overlap with loci implicated in autoimmune thyroid disease (29). In other words, only patients with thyroid disorders carrying these susceptibility variants are prone to B12 malabsorption, while those without such predispositions are likely to maintain normal B12 levels. Consequently, the combined effects of the incomplete expression of gastric autoimmunity and genetic heterogeneity may explain the null association observed in our study.

The results of the present study are strongly corroborated by a notable subset of studies that also failed to find a clear association. Notably, the study by Al-Mousawi et al (9) performed in Duhok, Iraq, aligns with the findings of the present study, as it reported no significant difference in total vitamin B12 levels between patients with subclinical hypothyroidism and healthy controls. This regional alignment suggests that in some Iraqi populations, a strong link may not be present. Similarly, the large meta-analysis performed by Benites-Zapata et al (14), which included >28,000 participants, provides nuanced support for the findings of the present study. Whereas it did find lower B12 levels in patients with overt hypothyroidism, it notably found no significant difference in B12 levels for patients with hyperthyroidism, AITD or subclinical hypothyroidism when compared to healthy controls (14), which perfectly mirrors the results from our well-defined AITD cohorts. Further support comes from the study by Aon et al (24), who found no statistically significant variation in the occurrence of vitamin B12 deficiency among the hypothyroid groups compared to the control group. Additionally, the literature reviews performed by Collins and Pawlak (13), and Kacharava et al (30) also emphasise the highly variable and inconsistent findings across the field, highlighting that the link is far from established and validating the importance of null findings such as those of the present study. Conversely, the findings of the present study are in contrast to certain other scientific studies that have reported a significant association between thyroid dysfunction and B12 deficiency. For instance, both Chatterjee et al (31) and Kaur et al (25) reported a high prevalence of B12 deficiency (68 and 70%, respectively) in their hypothyroid cohorts within their studies conducted in India.

Furthermore, another notable result of the present study was the non-existence of a statistically significant correlation with the 95% CI values for these correlations all including zero, reinforcing the absence of any meaningful association between vitamin B12 levels and the main thyroid function parameters (TSH) or autoimmunity (anti-TPO and TRAb). Similarly, this result is supported by the findings reported in the studies by Bhuta et al (32), Sinha et al (33) and Chatterjee et al (31), who all reported no significant correlations between TSH levels and vitamin B12 in their hypothyroid cohorts (31-33). In addition to this, the observation in the present study of a non-significant association with thyroid autoantibodies is consistent with several key investigations, including the studies by Aon et al (24) and Kumari et al (34), who also failed to establish a significant associative link between B12 status and anti-TPO levels. Similarly, the regional Iraqi study by Al-Mousawi et al (9) failed to establish a significant link between B12 status and hypothyroid autoantibodies However, this body of evidence stands in stark contrast to other published research. For instance, Chatterjee et al (31) demonstrated a strong negative correlation between vitamin B12 levels and the level of anti-TPO, a finding echoed in the studies by Aktaş (35) and Kacharava et al (36), who also reported weak, yet significant negative correlations between B12 and anti-TPO levels. This clear dichotomy in the literature, in which the present study aligns firmly with one perspective, underlines the ongoing scientific debate. This suggests that the association between vitamin B12 and thyroid autoimmunity is not necessarily direct or universal, and may instead be influenced by other unmeasured confounding factors, such as co-occurring autoimmune conditions or specific population genetics, which vary between study cohorts.

A granular examination of the present study cohort revealed several descriptive trends that, while not reaching statistical significance, provide valuable context. The apparent preponderance of females in the B12-deficient group, which comprised 84.6% of diagnosed cases, is most plausibly interpreted as a reflection of the underlying demographics of autoimmune thyroid disorder, a pattern well-documented by studies, such as those by Zoori and Mousa (23), and Aon et al (24), rather than an independent sex-based risk for deficiency. Furthermore, the descriptive clustering of deficiency within the 41-45-year age bracket and its prevalence among patients with a 1-3-year disease duration suggests a potential temporal window of vulnerability post-diagnosis. However, the lack of statistical significance in the present study is in agreement with the previous study by Kumari et al (34), who also reported no significant correlation between B12 levels and either patient age or disease duration. This convergence suggests that while these patterns may be observable, they do not represent robust, independent predictors of B12 status in AITD populations.

Univariate and multivariate analyses revealed no independent predictive role for BMI, residential location, or smoking status in determining vitamin B12 deficiency. This finding contributes to a complex and inconsistent body of literature on the B12-obesity association. For example, while the study on adults by Sun et al (37) in the USA reported a significant inverse association, other research such as a large randomised trial in a European cohort by de Araghi et al (38) found no significant link. The results of the present study, however, align more closely with evidence from the Middle East, such as the study by Abu-Shanab et al (39) on Jordanian adults, in which similar demographic variables showed poor predictive value. Furthermore, although a national Iraqi survey identified rural-urban disparities in B12 level (10), the adjusted model used herein did not detect a significant residence effect. Collectively, these results suggest that BMI, residential location and smoking status are not robust predictors of vitamin B12 deficiency in patients with AIT.

The discordant findings within the literature and the contrast between the results of the present study and studies reporting a positive association may likely be attributed to marked methodological heterogeneity. Firstly, a number of studies aggregate all hypothyroid patients, regardless of aetiology, whereas the present study differentiated between HT and GD. Secondly, the lack of a standardised diagnostic threshold for vitamin B12 deficiency, with cut-offs varying across different studies, greatly affects reported prevalence rates. As demonstrated by Kumari et al (34), raising the cut-off value from <145 to <200 pg/ml increased the observed prevalence from 45.5 to 55%. Of note, beyond such methodological heterogeneity, deeper population-based factors play a critical role in generating divergent results across studies. Genetic predisposition, such as varying frequencies of HLA types and immunoregulatory gene variants, differs substantially among ethnicities and may influence both thyroid disease risk and comorbid gastrointestinal autoimmunity such as pernicious anaemia (29). Nutritional habits, particularly with respect to animal product intake, impact baseline B12 status; this was demonstrated in a study from northern India, in which 86% of patients with vitamin B12 deficiency were pure vegetarians (40). It was found that, among those with both B12 deficiency and hypothyroidism, 17 of 19 were strict vegetarians (40). Collectively, these considerations underline that results from one study population may not be directly translatable to another, thereby highlighting the necessity of region-specific research and interpretation.

The present study possesses several methodological strengths that enhance the validity of its findings. A key strength is the inclusion of two distinct, well-defined AITD cohorts (HT and GD) alongside a healthy control group. This design allows for a more granular analysis than studies that aggregate all hypothyroid patients or focus on only one AITD type. The groups were well-matched for age and sex, and we utilised robust statistical power analysis to ensure an adequate sample size. The carefully crafted inclusion and exclusion criteria minimised the potential for confounding variables to influence the results. Additionally, the present study addressed a key research gap, as there is insufficient evidence to establish a direct correlation between vitamin B12 levels and TRAb titres in GD. Finally, the use of a single, high-precision ECLIA platform for all hormonal and vitamin assays reduced inter-assay variability and increased the reliability and consistency of our measurements.

Notwithstanding the merits of the present case-control study, certain constraints need to be recognised. Initially, although age and sex were matched between cases and controls, the omission of matching for additional demographic and lifestyle variables may have affected the findings. Moreover, the significant overrepresentation of female participants, in line with the epidemiology of AITD, limits the ability to thoroughly examine sex-specific associations and impedes direct comparisons between males and females. Additionally, despite drawing participants from varied regions within the province, the single-centre framework may constrain the applicability of the findings to other populations with distinct healthcare environments or demographic characteristics. Finally, the focus of the present study precludes any determination of causality concerning B12 deficiency. Future longitudinal or multi-centre studies, which should include thorough matching for these demographic and dietary factors, as well as relevant biomarkers and factors, such as gastric autoantibodies, are essential for advancing comprehension of this intricate association.

The findings of the present study suggest that routine screening for vitamin B12 deficiency in all patients with autoimmune thyroid disease may not be justified in low-prevalence populations. Instead, testing should be considered for those with clinical symptoms, risk factors, such as vegetarian diets, or concomitant gastric autoimmunity. This targeted approach can improve resource allocation and reduce unnecessary interventions. When B12 deficiency is detected, timely supplementation remains critical to prevent neurological and haematological complications. Ultimately, the present study supports adopting individualised screening strategies in clinical practice rather than blanket testing for all thyroid autoimmunity cases.

In conclusion, the present study demonstrates no significant association between serum vitamin B12 levels and autoimmune thyroid diseases, including HT and GD, with similar findings for thyroid function markers and disease-specific autoantibodies. These null results contribute to the existing literature by highlighting the variability in B12-AITD associations across populations, potentially influenced by genetic, nutritional, or methodological factors. While routine B12 screening may not be warranted for all patients with AITDs in similar settings, targeted assessment is advisable for those with relevant risk factors. In order to better elucidate potential influences and long-term patterns, future multicentre longitudinal studies are recommended.

Acknowledgements

The authors extend their sincere gratitude to the medical team of Dr Abdul Wahid at the Thyroid Centre, Smart Health Tower, Sulaymaniyah, Iraq, for their collaboration during the data collection process.

Funding

Funding: No funding was received.

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

Both authors (BHA and KRK) contributed equally to the writing, revising and finalisation of the manuscript, as well as to data analysis and table creation. BHA and KRK confirm the authenticity of all the raw data. Both authors have read and approved the final manuscript.

Ethics approval and consent to participate

The present study received official approval from the Ethics Board of the College of Health and Medical Technology at Sulaimani Polytechnic University (reference no. 30/245, dated December 1, 2024). All participants gave written informed consent after receiving a comprehensive explanation of the study's objectives, procedures, potential risks, and benefits. They were also informed of their right to withdraw from the study at any time.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References