Effect of Smoking on Thyroid-Associated Ophthalmopathy


Eye and orbital changes are found in about 50% of patients with Graves’ hyperthyroidism and in 1-2% of patients with Hashimoto’s thyroiditis. The various eye and orbital signs and symptoms are called Graves’ ophthalmopathy or orbitopathy (GO). Various antibodies are linked with the ophthalmopathy, of which those targeting the TSH receptor (TSH-R) and the IGF1- Receptor may be the best candidates to play an etiological role in the development of this complex and controversial eye and orbital disorder.

Antibodies targeting the calcium-binding protein calsequestrin (CSQ), which is expressed five times more in eye muscle than other skeletal muscle, are found in the majority of patients with recent-onset Graves’ hyperthyroidism and orbital inflammatory changes and therefore should also be considered candidates. In earlier studies, we identified serum antibodies against the so-called “64 kDa protein”, shown to be the flavoprotein (Fp) subunit of the mitochondrial enzyme succinate dehydrogenase, G2s the terminal region of the transcription factor of FOXP1 and the connective tissue protein type XIII collagen (COLLXIII). While the three eye muscle antibodies (CSQ, Fp and G2s) are all good markers of ophthalmopathy in patients with Graves’ disease and Hashimoto’s thyroiditis, they are probably secondary to damage of the eye muscle and OCT proteins as they are all intracellular and only exposed to the immune reactions after their release from damaged orbital cells.

On the other hand, COLL XIII is an intracellular transmembrane protein in the orbital fibroblast and therefore a candidate for a target antigen in the orbital connective tissue (OCT) component of ophthalmopathy. Overall, the prevalence of ophthalmopathy in patients with newly diagnosed Graves’ hyperthyroidism who smoke is approximately 60% compared to 20% in nonsmokers. The mechanism for this association is unclear, but possibilities include: an action of chemical factors in cigarette smoke, a non-specific effect of smoking on the immune system or a more specific link between nicotine addiction and the orbital reactions.

It seems likely, however, that the association is due to a genetically based, but poorly understood, neuroimmunological process that is enhanced in smokers.

Here, we have addressed a possible association between smoking and serum levels of antibodies against CSQ, G2s, Fp and collagen XIII in patients with Graves’ ophthalmopathy. We studied patients with Graves’ ophthalmopathy or isolated upper eyelid retraction and lag. Slightly fewer than half of the patients were smokers. All patients had chemosis, conjunctival injection and periorbital swelling and some had eye muscle involvement (double vision, reduced upward gaze as well). Patients with only upper eyelid retraction and/or lag only were included as a separate group which we have termed “upper eyelid changes disease”. All patients had moderately severe or severe eye disease which was active and were selected at random from representative patients seen at the Nepean Hospital Thyroid clinics during the period 2005 – 2015.

Serum antibody results were obtained from the patient files. Eye muscle and COL XIII antibody tests had been carried out in the context of their clinical management. The patients’ smoking status (yes, no, number of pack years) was obtained by telephone. Serum antibodies were measured by standard. Enzyme-Linked Immunosorbent Assay (ELISA) and positive tests were determined by reference to a standard range for normal subjects. Mean serum levels of the 4 orbital anti- bodies were compared for smokers and nonsmokers with ophthalmopathy using one-way ANOVA and students t tests statistical analyses and correlations between antibody levels and smoking severity, measured as pack years, were assessed using Pearson’s correlation coefficient.

Mean serum antibody levels for antibodies against CSQ, Fp, G2s, and COLL XIII, were compared for smokers and nonsmokers with ophthalmopathy. Mean serum antibody levels of all four antibodies were significantly greater in smokers than non-smokers with ophthalmopathy. Mean antibody levels were slightly, but not significantly, increased for 3 of the 4 antibodies, in smokers with isolated upper eyelid disease.

We correlated smoking severity, measured as pack years, with serum antibody levels in smokers with ophthalmopathy or upper eye lid sign only. There were no significant correlations between the amount of cigarette smoking and serum orbital antibody level for CSQ, Fp and G2s. However, the correlation for COLXIII antibodies was significant (P – 0.0289, Pearson’s correlation coefficient), for COLL antibodies only.

The pathogenesis of Graves’ ophthalmopathy is not well understood but most likely involves an antibody reaction against the TSH-R and the IGF1- receptor that results in the activation of T cells against antigens in the retro- orbital space that share antigenic epitopes with thyroid follicular cells. These immune processes lead to orbital inflammation, with lymphocyte infiltration of the orbital tissues including the eye muscles, and release of cytokines that stimulate orbital fibroblasts to multiply and produce glycosaminoglycans, which absorb water, leading to increase in orbital volume.

Smoking is a well-recognized risk factor for the development of ophthalmopathy in patients with Graves’ hyperthyroidism. In addition, patients who smoke appear to be more likely to experience worsening of their ophthalmopathy if treated with radioactive iodine. Although the mechanism for this is not known, possible factors include the action of chemical elements in cigarette smoke which includes many well-known poisons, or a nonspecific effect of smoking on the immune system tending to enhance autoimmune reactions. It seems most likely however that the effect is complex and involving both genetic and environmental factors that are linked to the neuropsychological factors that characterize those people who are addicted to nicotine.

Recent studies have demonstrated that autoimmunity against the IGF-1 receptor alone or in conjunction with the TSH receptor may be the best candidate for a specific etiological reaction that can explain the development of ophthalmopathy in patients with Graves’ disease. A similar approach could be used to examine the role of smoking on the targeting in the orbit of these two proteins.

Crystal Jakson BSc (Hons) Hooshang Lahooti BSc PhD
Bernard Champion MB BS, MMed, FRACP
Honorary Professor of Medicine, Macquarie University, Sydney, Consultant Endocrinologist, The Bays Hospital, Mornington Vic Australia



Help us to Help Others


  • Community publicity & advocacy
  • Educational webinars
  • Information resources on our website

Patient Support

  • Toll Free Help line and email
  • Website thyroid news
  • Thyrobulletin newsletter


  • Annual Thyroid Research grants
  • Thyroid research news
  • Thyroid research articles