HbA1c is widely used to assess glycemic control, and the American Diabetes Association (ADA) has recommended its use for diagnosing diabetes and prediabetes, with an HbA1c of 5.7–6.4% indicating prediabetes and ≥6.5% indicating diabetes.¹ Its convenience — no fasting requirement, low day-to-day biological variability, and reflection of average glycemia over the preceding 8–12 weeks — has made it central to both diagnosis and monitoring of diabetes. However, HbA1c formation depends not only on the prevailing glucose concentration but also on erythrocyte lifespan and turnover. Conditions that shorten or prolong erythrocyte survival — hemoglobinopathies, iron deficiency, chronic kidney disease, and disorders of erythropoiesis — can produce spuriously high or low HbA1c values independent of true glycemic status. Thyroid hormone stimulates erythropoiesis, and hypothyroidism characteristically produces a hypoproliferative, normocytic anaemia through reduced erythropoietin drive and marrow activity. Even in the absence of overt anaemia, the altered red-cell kinetics of hypothyroidism could theoretically prolong effective erythrocyte exposure to glucose and spuriously raise HbA1c. Prior studies, notably by Kim et al., have shown that non-diabetic patients with overt hypothyroidism have significantly higher A1c and glycated albumin levels than euthyroid controls, with reduction in A1c following thyroid hormone replacement paralleling a rise in erythropoietin, reticulocyte count and mean corpuscular haemoglobin — evidence that thyroid hormone-driven erythropoiesis, rather than glycemia, underlies the change.² In India, although the effect of diabetes on thyroid dysfunction has been extensively studied, there is a relative paucity of prospective data correlating thyroid hormone replacement with HbA1c in newly detected non-diabetic hypothyroid patients. Given the increasing reliance on HbA1c as a stand-alone diagnostic tool, this represents a clinically important gap: non-diabetic hypothyroid patients could be misclassified as having prediabetes or diabetes on the basis of an HbA1c value confounded by their underlying thyroid disorder. This study was undertaken to prospectively evaluate this relationship. OBJECTIVES Primary: To detect HbA1c levels in non-diabetic patients with newly detected hypothyroidism. Secondary: To determine the reduction in HbA1c levels with thyroid hormone replacement.

Study design and setting: This was a prospective observational study conducted among patients attending Subbaiah Medical College in Shimoga over a 24-month period. Participants: During the study period, patients newly detected to be hypothyroid on an outpatient and inpatient basis were screened for HbA1c. Those with HbA1c levels >5.6% were considered for inclusion; a total of 40 such patients (a mix of overt and subclinical hypothyroidism) were enrolled after written informed consent. Inclusion criteria: Age >18 years with newly detected hypothyroidism (elevated thyroid-stimulating hormone with low free thyroxine, or biochemical evidence of subclinical hypothyroidism). Exclusion criteria: (1) Diabetes mellitus; (2) anaemia; (3) renal insufficiency; (4) liver dysfunction; (5) severe hypertriglyceridemia. Investigations: Free thyroxine (fT4), thyroid-stimulating hormone (TSH), fasting and post-prandial blood glucose, HbA1c, complete hemogram, liver function tests, renal function tests, lipid profile and a baseline electrocardiogram were performed in all patients. Intervention and follow-up: Patients were initiated on levothyroxine replacement (25–100 µg/day, individualised to the severity of hypothyroidism and patient profile) and reviewed with repeat TSH and HbA1c at 4 and 8 weeks. Statistical analysis: Continuous variables are presented as mean ± SD and categorical variables as number (%). Paired Student's t-test was used to compare continuous variables (TSH, HbA1c) across the three time points, and the paired-proportion test was used to compare categorical shifts in TSH and HbA1c categories. A p-value <0.05 was considered statistically significant. Analyses were performed using SAS 9.2, SPSS 15.0, Stata 10.1, MedCalc 9.0.1 and R

Baseline characteristics The mean age of the 40 patients was 40.25 ± 15.89 years (range: 17–50+ years), with the largest proportion (30%) aged 21-30 years. The cohort was overwhelmingly female (38/40, 95%). The commonest presenting context was evaluation for surgical/physical fitness (30%), followed by neck swelling (15%), febrile illness (12.5%), and menstrual irregularity (10%); the remainder presented with non-specific complaints such as fatigability, generalised weakness, and weight gain. Table 1. Baseline demographic profile (n = 40) Age (years) 17–20 2 (5.0) 21–30 12 (30.0) 31–40 11 (27.5) 41–50 11 (27.5) >50 4 (10.0) Sex Female 38 (95.0) Male 2 (5.0) Mean age ± SD: 40.25 ± 15.89 years Thyroid and glycemic profile at baseline The mean baseline free T3 was 255.85 ± 70.18 pg/dl, and free T4 was 1.17 ± 0.45 ng/dl, consistent with a spectrum from subclinical to overt hypothyroidism. Fasting and post-prandial glucose were normal in all patients by design (mean FBS 89.88 ± 9.82 mg/dl; mean PPBS 108.25 ± 13.94 mg/dl), confirming the non-diabetic status of the cohort. Change in TSH with levothyroxine therapy. Mean TSH fell progressively and significantly from 33.36 ± 30.77 mIU/L at baseline to 21.44 ± 21.74 mIU/L at 4 weeks and 10.33 ± 8.72 mIU/L at 8 weeks (p<0.001 for both comparisons versus baseline, paired t-test). The proportion of patients with TSH <10 mIU/L rose from 20% at baseline to 62.5% at 8 weeks, a 42.5% shift (p<0.001, paired proportion test). Table 2. TSH (mIU/L) at baseline, 4 weeks and 8 weeks Baseline 33.36 ± 30.77 — — 4 weeks 21.44 ± 21.74 11.92 <0.001** 8 weeks 10.33 ± 8.72 23.03 <0.001** **p ≤ 0.01 (Student's t-test, paired) Change in HbA1c with levothyroxine therapy. Mean HbA1c fell in parallel with TSH normalisation: from 6.02 ± 0.31% at baseline to 5.79 ± 0.25% at 4 weeks and 5.56 ± 0.27% at 8 weeks (p<0.001 at both time points, paired t-test). The proportion of patients with HbA1c in the prediabetic/elevated range (≥6%) fell from 52.5% (21/40) at baseline to 20% (8/40) at 4 weeks and 10% (4/40) at 8 weeks — a 42.5% reduction (p<0.001, paired proportion test). No patient in the cohort had a diabetic-range HbA1c (>9%) at any time point. Table 3. HbA1c (%) at baseline, 4 weeks and 8 weeks Baseline 6.02 ± 0.31 — — 4 weeks 5.79 ± 0.25 0.23 <0.001** 8 weeks 5.56 ± 0.27 0.46 <0.001** **p ≤ 0.01 (Student's t-test, paired) Table 4. Distribution of patients by HbA1c category over follow-up <6% 19 (47.5) 32 (80.0) 36 (90.0) 6–9% 21 (52.5) 8 (20.0) 4 (10.0) >9% 0 (0) 0 (0) 0 (0) Reduction of 42.5% in the ≥6% category from baseline to 8 weeks was statistically significant (paired proportion test). Hemoglobin As anaemia was an exclusion criterion, haemoglobin values remained within the normal range throughout the study: 77.5% of patients had haemoglobin between 10–12 g/dl, 15% had haemoglobin>12 g/dl, and only 7.5% had haemoglobin <10 g/dl (all above the exclusion threshold used). This confirms that the observed reduction in HbA1c could not be attributed to correction of anaemia or a change in erythrocyte turnover secondary to iron repletion, and instead reflects the direct effect of correcting the hypothyroid state.

In this prospective study of 40 non-diabetic patients with newly detected hypothyroidism, HbA1c was elevated at baseline (mean 6.02 ± 0.31%, with over half of the cohort in the ≥6% "prediabetic" range) despite normal fasting and post-prandial glucose levels. Following levothyroxine replacement, HbA1c fell significantly and progressively in parallel with normalisation of TSH, with a 42.5% reduction in the proportion of patients misclassified in the elevated HbA1c category by 8 weeks. These findings support the hypothesis that hypothyroidism itself, independent of glycemia, can spuriously elevate HbA1c. This is consistent with the findings of Kim et al., who reported significantly higher A1c and glycated albumin in non-diabetic patients with overt hypothyroidism compared with euthyroid controls, with A1c levels falling after thyroid hormone replacement; in their cohort, 20% of subjects were misclassified as having prediabetes by ADA HbA1c criteria despite normalisation of HbA1c once hypothyroidism was corrected.² The proposed mechanism relates to thyroid hormones' stimulatory effect on erythropoiesis: hypothyroidism produces a hypoproliferative erythropoietic state with reduced erythropoietin, reticulocyte count and mean corpuscular haemoglobin, effectively lengthening the exposure of circulating erythrocytes to ambient glucose and raising the glycated fraction independent of true glycemic control. Handisurya et al. similarly demonstrated that subclinical and overt hypothyroidism are characterised by attenuated basal insulin secretion and increased glucose-induced insulin secretion, both of which partially reverse with T4 replacement, reflecting a broader disturbance of glucose–insulin homeostasis in hypothyroidism beyond erythrocyte kinetics alone.³ Cinemre et al. showed that subclinical hypothyroidism is associated with lower serum iron and that L-T4 replacement alone can reverse iron deficiency, underscoring that thyroid status can influence hematologic and iron parameters that are themselves determinants of HbA1c.⁴ The relationship between thyroid status and glucose homeostasis is bidirectional and complex. While hyperthyroidism is well recognised to cause insulin resistance and impaired glucose tolerance through increased hepatic gluconeogenesis and glycogenolysis, hypothyroidism has more recently also been linked to decreased insulin sensitivity, particularly in adipose tissue and skeletal muscle, through reduced expression of the insulin-sensitive glucose transporter GLUT-4.⁵ This peripheral insulin resistance could theoretically raise glycemia and secondarily HbA1c; however, in the present cohort, fasting and post-prandial glucose remained essentially unchanged and within the normal range throughout, arguing against a glycemia-mediated mechanism and favouring an erythrocyte-kinetic explanation for the observed HbA1c elevation and its correction. Importantly, in this study, patients with anaemia, renal insufficiency, hepatic dysfunction and severe hypertriglyceridemia — all of which can independently alter HbA1c — were excluded a priori, and haemoglobin remained normal throughout follow-up. This strengthens the inference that the fall in HbA1c reflects a direct effect of correcting the hypothyroid state rather than a confounding hematologic change. It is also notable that the reduction in HbA1c tracked closely with the degree of biochemical improvement in thyroid status (TSH), consistent with a dose–response relationship between the severity of hypothyroidism and the magnitude of HbA1c distortion described in other cohorts, where the effect on serum lipids and other glycated proteins has similarly been shown to be more marked with higher baseline TSH.⁶ These observations have direct clinical relevance. The 2009 International Expert Committee report and subsequent ADA guidelines endorse HbA1c ≥6.5% as diagnostic of diabetes and 5.7–6.4% as indicative of prediabetes.¹ Applying these cut-offs uncritically in a patient with untreated hypothyroidism, as demonstrated here, risks over-diagnosis of prediabetes or diabetes, unnecessary patient anxiety, and inappropriate initiation of lifestyle or pharmacological intervention for a condition that resolves with thyroid hormone replacement alone. Given the high background prevalence of thyroid dysfunction in the Indian population and its frequent co-existence with — or misclassification as — dysglycemia, clinicians should assess thyroid status before relying on HbA1c for diagnosis, and should preferably reassess HbA1c only after a euthyroid state has been achieved, or corroborate the diagnosis with a fasting or oral glucose tolerance test in patients with active, untreated hypothyroidism. Limitations: This was a single-centre study with a modest sample size and a follow-up period limited to 8 weeks; longer-term data on the durability of HbA1c normalisation, and its relationship to different degrees of hypothyroidism (subclinical versus overt), would strengthen these findings. The cohort was predominantly female, reflecting the epidemiology of hypothyroidism at this centre, which may limit generalizability. Erythrocyte lifespan was not directly measured, and the precise mechanistic contribution of altered red-cell kinetics versus other glycation-modulating factors in hypothyroidism remains to be defined by dedicated red-cell survival studies

HbA1c is subject to spurious elevation in non-diabetic patients with newly detected hypothyroidism, and this elevation falls significantly with correction of the hypothyroid state following levothyroxine replacement. These findings support caution in using HbA1c as a stand-alone diagnostic criterion for diabetes or prediabetes in patients with untreated hypothyroidism, and highlight the need for further studies, including red-cell lifespan studies, to fully elucidate the underlying mechanism.

1. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33:S62–69. 2. Kim MK, Kwon HS, Baek KH, et al. Effects of thyroid hormone on A1C and glycated albumin levels in nondiabetic subjects with overt hypothyroidism. Diabetes Care. 2010;33:2546–48. 3. Handisurya A, Pacini G, Tura A, Gessl A, Kautzky-Willer A. Effects of T4 replacement therapy on glucose metabolism in subjects with subclinical (SH) and overt hypothyroidism (OH). Clin Endocrinol (Oxf). 2008;69:963–9. 4. Cinemre H, Bilir C, Gokosmanoglu F, Bahcebasi T. Hematologic effects of levothyroxine in iron-deficient subclinical hypothyroid patients: a randomised, double-blind, controlled study. J Clin Endocrinol Metab. 2009;94:151–6. 5. Al Sayed A, Al Ali N, Bo Abbas Y, Alfadhli E. Subclinical hypothyroidism is associated with early insulin resistance in Kuwaiti women. Endocr J. 2006;53(5):653–657. 6. Sunanda V, Sangeeta S, Prabhakar Rao B. Study of lipid profile in hypothyroidism. Int J Biol Med Res. 2012;3:1373–1376. 7. Maratou E, Hadjidakis DJ, Kollias A, et al. Studies of insulin resistance in patients with clinical and subclinical hypothyroidism. Eur J Endocrinol. 2009;160:785–790. 8. Larsen P, Davies T. Hypothyroidism and thyroiditis. In: Williams Textbook of Endocrinology, 10th ed. Larsen PR, Kronenberg HM, Melmed S, Polonsky KS, eds. Saunders Elsevier; 2002:423–455. 9. World Health Organisation. Use of Glycated Haemoglobin (HbA1c) in the Diagnosis of Diabetes Mellitus: Abbreviated Report of a WHO Consultation. 2011. 10. Balasubramaniam S, et al. Effect of Iron Deficiency on Glycation of Haemoglobin. J Clin Diagn Res. 2013;7(1):15–17. 11. Son JI, et al. Haemoglobin A1c may be an inadequate diagnostic tool for diabetes mellitus in anaemic subjects. Diabetes Metab J. 2013;37:343–348. 12. Christy AL, et al. Is anaemia the cause of elevated HbA1c in hypothyroidism? J Clin Diagn Res. 2013;7(11):2442–2444. 13. Hare MJL, Shaw JE, Zimmet PZ. Current controversies in the use of haemoglobin A1c. J Intern Med. 2012;271:227–236. 14. Dagogo-Jack S. Pitfalls in the use of HbA1c as a diagnostic test: the ethnic conundrum. Nat Rev Endocrinol. 2010;6:589–93. 15. Brenta G. Why can insulin resistance be a natural consequence of thyroid dysfunction? J Thyroid Res. 2011; Article ID 152850.