Mitigating the risks: addressing smoking cessation in thyroid eye disease

Abstract

Objective

This study aimed to assess the effectiveness of verbal counselling and varenicline in achieving smoking cessation among patients with thyroid eye disease (TED) and to identify predictive factors associated with successful smoking cessation.

Methods and analysis

A cross-sectional analysis of data from the Iran TED (IrTED) Registry was conducted. Patients’ demographics and TED severity/activity were recorded. As a routine at the TED clinic in Rassoul Akram Hospital, since 2020, all smoking patients have undergone smoking cessation verbal counselling in the first and almost every subsequent visit session. The effectiveness of verbal counselling was assessed through a telephone survey. Retrospectively, the association between successful smoking cessation and disease severity/activity was evaluated. Prospectively, patients who failed or refused to quit smoking after counselling were offered varenicline. Acceptance, compliance and adherence rates were determined.

Results

While 379/685 patients of the IrTED 2013–2023 database (55.32%) reported no tobacco exposure, the remaining were active (n=117, 17.08%), passive (n=134, 19.56%) and former smokers (n=55, 8.02%). In 2020–2023, 73 active tobacco-smoking patients with TED were enrolled, all of whom received verbal cessation counselling at their first and subsequent visits. On the follow-up phone call, 51/73 were contacted, and 25/51 (49.01%) reported successful smoking cessation after verbal counselling. The remaining 26/51 smokers were offered varenicline (on the phone call); however, only 7/26 (26.92%) attended the prescription session. Of these patients, 3/7 (42.85%) completed the first month of treatment, and just 1/3 (33.33%) finished the 3-month course, successfully quitting smoking until the final follow-up (18 months). One patient (14.28%) reported constipation, and another one reported sleep paralysis (14.28%) as an adverse event.

Conclusion

Verbal counselling demonstrates effectiveness in smoking cessation for patients with TED, with nearly half of smokers successfully quitting the habit. Among those who failed to quit after verbal counselling, varenicline showed limited effectiveness, indicating poor compliance and motivation deficit.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• While smoking is known to exacerbate thyroid eye disease (TED) progression and hinder treatment, data on the efficacy of verbal counselling and varenicline for smoking cessation within the TED population remains limited.

WHAT THIS STUDY ADDS

• Verbal counselling effectively promotes smoking cessation in patients with TED, with almost half successfully quitting. For those who continued smoking after counselling, varenicline had limited effectiveness, suggesting poor compliance and motivation issues.

HOW MIGHT THIS STUDY AFFECT RESEARCH, PRACTICE OR POLICY

• Consistent, evidence-based verbal counselling is likely sufficient and highly effective for patients with TED willing to quit smoking, with varenicline offering only minimal additional benefits.

Introduction

Thyroid eye disease (TED) is an autoimmune-related ocular condition that can result in functional disabilities, such as restricted eye movement or compressive optic neuropathy, and cosmetic concerns due to lid abnormalities and exophthalmos.^{1 2} TED is most commonly observed in patients with Graves’ hyperthyroidism, with a prevalence ranging from 25% to 40% in this population.^{3 4} Although TED typically occurs in a hyperthyroid state, it has also been reported in hypothyroid and euthyroid states.5,7 The gender ratio for TED prevalence is approximately 2:1, indicating that the disease occurs about twice as frequently in women. However, this gender disparity tends to decrease with age, while disease severity increases.^{8 9}

Cigarette smoking is widely recognised as the most significant modifiable risk factor for the onset and progression of TED.^{10 11} The incidence of TED is notably higher among patients with Graves’ disease who smoke compared with non-smokers.¹² A dose-dependent relationship between smoking and TED severity has been demonstrated.¹³ Moreover, smokers tend to exhibit weaker and delayed treatment responses.¹⁴

Smoking cessation is beneficial in managing TED across multiple levels of prevention. At the primary prevention level, it aids in eliminating the key risk factor for TED onset; at the secondary level, it helps treat patients with mild or no complications; and at the tertiary level, it reduces disease-related disability.¹⁵ Furthermore, smoking cessation has been found to effectively reverse the epigenetic changes caused by smoking, resulting in reduced incidence and severity of TED.^{16 17} A prospective study on patients with Graves’ hyperthyroidism found that those who quit smoking at least a year before developing hyperthyroidism (ie, former smokers) had no increased risk of proptosis or diplopia compared with never-smokers.¹⁶ These findings reinforce the importance of smoking cessation in mitigating the risk of TED onset and progression.

Although limited research exists regarding smoking cessation interventions specifically for patients with TED, the available studies have yielded mixed results. On the one hand, a study conducted by Palomero et al found that oral counselling led to a cessation rate of 42.4% and a reduction in smoking for an additional 30.3% of patients.¹⁸ In contrast, another study reported that oral and behavioural interventions were not effective in smoking cessation among patients with TED.¹⁹ These conflicting findings underscore the need for further research on smoking cessation in TED patients.

Several pharmacological interventions are available for smoking cessation, including nicotine replacement therapies (eg, patches, chewing gums and tablets) as well as medications such as varenicline, bupropion and nortriptyline. Notably, varenicline and bupropion have received Food and Drug Administration approval for this indication. Among these options, varenicline has demonstrated efficacy in reducing the number of smoked cigarettes, craving and dependence,20,23 leading to its recommendation as the first-line treatment for smoking cessation.²¹

The primary objective of this study was to evaluate the effectiveness of verbal counselling and varenicline in facilitating smoking cessation among patients diagnosed with TED. Additionally, the study sought to identify factors that could predict successful smoking cessation among the studied patient group.

Materials and methods

Using data from the Iran TED (IrTED) Registry (https://orc.iums.ac.ir), this study examined patients who presented at Rassoul Akram Hospital’s oculoplastic clinic in Tehran, Iran, between March 2020 and March 2023.²⁴ The study complied with HIPAA and the Declaration of Helsinki.

As part of the standard care at the TED clinic in Rassoul Akram Hospital, since 2020, all smoking patients have received verbal counselling in the first and almost every subsequent visit session. The first author shared two evidence-based key points with the patients and answered their relevant questions:

• Smoking increases the incidence of proptosis by approximately twofold and triples the incidence of diplopia in patients with TED.¹⁶

• Response to treatment is delayed and considerably poorer in patients with TED who smoke.¹⁴

These two messages for the counselling were chosen partially based on a previous study on determinants of quality of life in Iranian patients with TED.²

The study commenced by extracting the smoking status of all registered patients (2013–2023) at their initial visit (non-smoker, former smoker, passive smoker or current smoker) from the registry.²⁵ Patients who smoked more than five cigarettes per week were classified as smokers. Active smokers at the first visit were selected for a follow-up telephone call to ascertain whether they had stopped or continued smoking after receiving the smoking cessation counselling (2020–2023) in the first visit. This self-reported information classified the smokers into two groups: persistent smoking and successful quitting. The inclusion criteria for the retrospective phase of the study comprised patients from both groups who had attended at least one subsequent clinical visit at least 3 months after their initial visit. During this phase of the study, the two groups were compared based on their baseline values and changes (from the first verbal counselling visit to the second clinical visit) in clinical indices to identify predictive factors associated with successful smoking cessation following verbal counselling. In the prospective phase of the study, patients who stated (on the follow-up phone call) that they refused or failed to quit smoking after verbal counselling were offered varenicline treatment. If they accepted, a clinical visit would be scheduled to provide them with the prescription and instructions for drug administration. A 3-month regimen of varenicline (0.5 mg per day for 3 days then 0.5 mg BD for 4 days then 1 mg BD for the rest of the regimen) was prescribed. Compliance and side effects of varenicline were recorded. Contraindications of starting varenicline included age under 18, pregnancy or breastfeeding.

TED Activity was defined as a CAS (clinical activity score) of or more equal to three. The severity of TED was scored following the EUGOGO (European Group on Graves' Orbitopathy) and NOSPECS system: (N) no signs and symptoms, (O) only signs, (S) soft tissue involvement, (P) proptosis, (E) extraocular muscle involvement, (C) corneal involvement and (S) sight loss.²⁶

The registration administrator (SAk) managed data collection, data cleaning and registry procedures to ensure adherence to quality standards throughout the study. Logistic regression analysis was applied to estimate the probability of adverse outcomes, such as active TED or sight-threatening TED. Sex variable was controlled as a potential confounding variable. For comparisons of quantitative (non-normally distributed) variables between groups, the Mann-Whitney U test was used. Categorical variables were analysed with dual tests of β Pearson χ² asymptotic significance (as a measure of association for nominal categorical variables) and γ Gamma approximate significance (as a measure of association for ordinal categorical variables) to examine the strength and direction of associations. All statistical analyses were performed using IBM SPSS Statistics V.25.

The current observational study adheres to the RECORD (Reporting of Studies Conducted Using Observational Routinely Collected Data) checklist to ensure comprehensive data reporting. Additionally, patients and the public were not involved in any aspect of the design, conduct, reporting or dissemination plans of the research.

Results

The IrTED database (2013–2023) included 685 patients (66.56% women), with a mean age of 37.79 years (SD: 14.06, range 2–78). While 379 patients (55.32%) reported no tobacco exposure, the remaining were classified as active (n=117, 17.08%), passive (n=134, 19.56%) or former smokers (n=55, 8.02%). Significantly, active smoking rates differed between genders, with women at 7.24% and men at 36.68% (p<0.01). Figure 1 illustrates the severity distribution of patients in the data pool based on smoking status at the initial visit. Although univariate analysis suggests active smoking may be linked to increased activity and severity in TED in the first visit, after adjusting for the confounding variable of gender, smoking status was not found to be associated with disease activity and severity (table 1).

Figure 1. Distribution of severity profile of patients by smoking status at the first visit.

Table 1. Correlation between smoking status and TED severity and activity at initial visit.

	OR (95% CI) p value
	Active TED (vs inactive TED)		Sight-threatening TED (vs non-sight threatening)		Sight-threatening or moderate–severe TED (vs mild TED)
Comparison	Univariate analysis	Adjusted for sex	Univariate analysis	Adjusted for sex	Univariate analysis	Adjusted for sex
Current smoker versus passive smoker	OR: 3.66 (2.04 to 6.59) p<0.01	OR: 1.83 (0.88 to 3.80) p: 0.10	OR: 21.06 (2.75 to 161.51) p<0.01	OR: 5.66 (0.63 to 50.97) p: 0.12	OR: 2.25 (1.33 to 3.82) p<0.01	OR: 1.29 (0.67 to 2.49) p: 0.44
Current smoker versus non-smoker	OR: 2.47 (1.61 to 3.80) p<0.01	OR: 1.37 (0.85 to 2.21) p: 0.19	OR: 3.27 (1.64 to 6.55) p<0.01	OR: 1.67 (0.80 to 3.49) p: 0.17	OR: 1.56 (0.99 to 2.44) p: 0.05	OR: 1.28 (0.77 to 2.15) p: 0.34
Passive smoker versus non-smoker	OR: 1.48 (0.89 to 2.47) p: 0.13	OR: 1.07 (0.63 to 1.85) p: 0.78	OR: 6.42 (0.85 to 48.33) p: 0.07	OR: 3.88 (0.50 to 30.01) p: 0.19	OR: 1.44 (0.98 to 2.14) p: 0.06	OR: 1.15 (0.77 to 1.73) p: 0.50

EUGOGO, European Group on Graves' Orbitopathy; TED, thyroid eye disease.

After 2020, by incorporating verbal counselling as a routine portion of care at the TED clinic in Rassoul Akram Hospital, a substantial number of patients who reported smoking during their initial visit disclosed successful smoking cessation during a follow-up phone call (figure 2). Out of the 51 smoker patients contacted, 25 (49.01%) reported successful smoking cessation following verbal counselling. Based on the inclusion criteria, the case selection process yielded two groups: 12 patients who continued smoking (persistent smokers) and 16 patients who successfully quit (quitters). These groups were then compared based on their baseline characteristics (table 2) and the changes observed during follow-up (table 3). The average time between visits was 7.9 months (SD=5.2) for individuals who did not quit smoking and 10.9 months (SD=6.9) for those who quit (p=0.23). Persistent smokers in the study had a higher average age of 50.25 years compared with those who quit smoking, with a mean age of 42.50 years (p=0.07). In this study, no significant differences in clinical manifestations were observed between persistent smokers and successful quitters at the baseline (table 2). An analysis comparing previsit and postvisit indices indicated that patients who experienced worsening TED severity during their disease progression were more likely to quit smoking. The mean severity score change was −1.66 (SD=4.43) for persistent smokers and +4.06 (SD=7.04) for quitters (p=0.02). Even when changes in disease severity were categorised qualitatively (categorically) as worsening, stable or improving, the distinction between persistent smokers and those who successfully quit smoking remained statistically significant (table 3), both in statistical tests to measure association in nominal and ordinal categorical variables. This indicates that patients who endured TED severity aggravation throughout their disease progression showed a higher likelihood of smoking cessation. No statistically significant differences were found between the two groups regarding changes in CAS scores or shifts in disease activity classification (table 3).

Figure 2. Flow diagram outlining the steps involved in patient selection and varenicline treatment course. IrTED, Iran Thyroid Eye Disease.

Table 2. Comparison of clinical characteristics between successful quitters and persistent smokers.

	Still smoking (n=12)	Quit smoking (n=16)	P value
Age (years) Median (range)	52.00 (31.0–64.0)	41.00 (26.0–62.0)	0.07*
Sex (male/female)	9/3	12/4	1.00¶
Smoking dose (fags/day) Median (range)	15.00 (5.0–40.0)	11.00 (3.0–30.0)	0.58*
TED duration (months) Median (range)	12.00 (2.0–103.0)	5.50 (1.0–110.0)	0.63*
The interval between the two visits (months) Median (range)	7.00 (3.0–20.0)	9.00 (3.0–31.0)	0.24*
History of eye surgery	2/12	0/16	0.17¶
History of systemic steroid	5/12	9/16	0.70¶
Ocular pain†	2/12	4/16	0.67¶
Tearing	9/12	7/16	0.13¶
Diplopia**	8/12	6/16	0.25¶
Proptosis‡ (mm) Median (range)	4.00 (0.0–10.5)	4.25 (0.0–10.5)	0.45*
Conjunctival redness	7/12	10/16	1.00¶
Chemosis	2/12	1/16	0.56¶
Clinical activity score (CAS) Median (range)	3.00 (0.0–4.0)	2.50 (0.0–5.0)	0.69*
Active thyroid eye disease§	8/12	8/16	0.45¶
NOSPECS severity score Median (range)	17.50 (5.0–40.0)	11.50 (3.0–30.0)	0.59*

^*Mann-Whitney U test.

^†Spontaneous or by movement.

^‡Worst eye.

^§CAS score equal to or more than 3.

^¶Fisher’s exact test significance (two-sided).

^**Primary gaze, extreme gaze, or intermittent diplopia.

NOSPECS, (N) no signs and symptoms, (O) only signs, (S) soft tissue involvement, (P) proptosis, (E) extraocular muscle involvement, (C) corneal involvement and (S) sight loss; TED, thyroid eye disease.

Table 3. Comparison of clinical manifestation changes (worsening or improving, as opposed to remaining stable) in continuous smokers and successful quitters between the second visit and the first (verbal cessation counselling) visit.

	Persistent smokers		Successful quitters		P value
Change	Worsening	Improving	Worsening	Improving
Ocular pain	2/12	1/12	2/16	3/16	0.72*	0.46‡
Activity class†	0/12	4/12	1/16	4/16	0.63*	0.47‡
NOSPECS score	2/12	7/12	11/16	4/16	0.02*	<0.01‡

^*Pearson χ² asymptotic significance (as a measure of association for nominal categorical variables).

^†Change in classification from active to inactive or vice versa (CAS ≥3 considered active).

^‡Gamma approximate significance (as a measure of association for ordinal categorical variables).

CAS, clinical activity score; NOSPECS, (N) no signs and symptoms, (O) only signs, (S) soft tissue involvement, (P) proptosis, (E) extraocular muscle involvement, (C) corneal involvement and (S) sight loss.

Among 51 cigarette smokers contacted, 26 reported persistent smoking (figure 2). When offered varenicline, 12/26 smokers agreed to attempt the medication and scheduled a visit. However, only seven attended and started the medication, with three complying with the first month of treatment. Of these, one patient quit smoking after the first month but discontinued varenicline and restarted smoking 1 month later. Another patient failed to quit smoking after taking varenicline for 60 days and discontinued the drug. Only one patient completed the 3-month regimen, successfully quitting after the first month and sustaining cessation until the last follow-up (18 months). Two patients experienced side effects while taking varenicline: one reported an unpleasant sleep-related event (sleep paralysis), leading to medication discontinuation and another experienced mild constipation.

Discussion

Smoking has been unequivocally recognised as the most significant risk factor for the onset and progression of TED. Research has demonstrated that smoking significantly increases the odds of developing TED by 7.7 times (95% CI, 4.3 to 13.7) when compared with healthy individuals who do not smoke.²⁷ Smoking escalates the risk of ophthalmopathy progression post-radioiodine therapy and reduces the effectiveness of orbital radiation and glucocorticoid therapies.²⁸ Furthermore, it is associated with a higher likelihood of requiring surgical intervention for TED.²⁹ In a study on patients with Graves’ disease, smoking was linked to a 1.3-fold increase in TED incidence, and a 2.6-fold and 3.1-fold increase in proptosis and diplopia incidence, respectively.¹⁶ The study revealed that former smokers had a considerably lower risk of developing proptosis and diplopia compared with active smokers who had a similar lifetime cigarette consumption. This finding emphasises the benefits of smoking cessation in reducing the risk of these TED-related complications.¹⁶

In the present study, the observed relationship between smoking status and TED severity or activity, at the cross-section of the first visit to the oculoplastic clinic, did not persist after controlling for gender (table 1). In other words, when women and men are analysed separately, the smoking habit is not associated with increased severity or activity at first visit. This highlights that male patients tend to present more advanced and active stages of the disease. A possible explanation may be men’s reluctance to seek access to health services. Delayed healthcare seeking in men is supported by a growing body of evidence.^{30 31}

While TED severity is not significantly different among active smokers, passive smokers and non-smokers at the cross-section of the initial presentation, this finding does not challenge the recognised role of smoking in triggering TED onset and progression in longitudinal studies.³²

Notably, the active smoking rates among Iranian patients with TED in this study were found to be almost double the national average. The study reported rates of 7.24% for women and 36.68% for men, compared with the average national smoking rates of 4.1% for women and 19.9% for men.³³ This emphasises the importance of implementing targeted cessation counselling in oculoplastic centres to address the heightened impact of smoking on TED incidence within this population.

Verbal counselling in the current study included conveying predetermined evidence-based facts^{14 34} to the patients regarding how smoking deteriorates TED. A previous study³⁵ showed that educating patients about the risks of smoking significantly increased the success rate of quitting from 1/23 (4.3%) to 9/34 (26.5%), resulting in a 22.1% difference (95% CI: 1.7% to 39.1%). Similarly, another study reported a 42.4% smoking cessation rate.¹⁸ In the present study, cessation counselling was regularly provided to all patients during almost every session, which has contributed to a cessation success rate of 25/51 (49.01%). This indicates that integrating verbal counselling into TED care can significantly increase the likelihood that an active smoker will quit or perhaps reduce smoking. In agreement with previous findings, patients whose TED severity was progressively worsening demonstrated a greater likelihood of quitting smoking (table 3). Active and severe TED, stable employment and increased family support have been previously identified as significant predictors of smoking cessation success.¹⁸

One of the primary objectives of the present study was to determine whether varenicline could be beneficial in smoking patients with TED. This question was specifically addressed through the telephone surveillance phase of the study (figure 2). Remarkably, varenicline was only offered to active smokers who failed to quit smoking after receiving verbal counselling. Therefore, it is important to consider that the limited effectiveness of varenicline observed in this study should be carefully attributed to the specific subset of smokers who have failed a previous verbal counselling and are perhaps more heavily dependent on tobacco, or possess lower motivation for cessation.

This study shows that if a smoking patient with TED is open to quitting, verbal counselling is likely sufficient and varenicline does not seem to offer additional effectiveness. One might argue that among the three patients who complied with varenicline for at least 1 month, one managed to quit and sustain a long-term smoking abstinence. This may appear as a reasonable 1/3 efficacy, suggesting that varenicline is a beneficial second-line intervention, provided that this is confirmed in future larger studies. Nonetheless, we would like to draw attention to the difference between drug effectiveness (in real-world practice) and efficacy (in an optimal situation with all presumptions fulfilled; ie, willingness to take the drug, compliance with the correct dosage, withstanding minor side-effects, adherence to the complete treatment period, etc.). The results indicate that among persistent smoker patients who were offered varenicline, only 1/26 achieved a successful cessation, a clear clue of limited effectiveness.

Some limitations of the present study belong to the cross-sectional methodology, with no certainty about the time order between exposure and outcome. While data suggest that smokers with more severe TED are more likely to quit smoking, another, though less conceivable, interpretation would be that those who stop smoking will develop more severe grades of the disease.³⁶ Another potential source of bias is the subjective definition of smoking cessation. It is possible that patients with more severe grades of TED feel guilty or stigmatised for their poorer outcome and thus tend to hide their continuous smoking behaviour, that is, give a false report of smoking cessation. In the absence of an objective test to confirm smoking cessation, we cannot rule out this possibility. Despite potential false reports from some patients, the substantial percentage of smoking cessation self-reported through verbal counselling (49.01%) implies that this approach is highly effective.

In conclusion, this study demonstrates that regularly providing patients with TED with evidence-based, quantitative information on the adverse effects of smoking (particularly the increased risk of proptosis, diplopia and treatment failure) can lead to a cessation rate of up to 50%. Patients experiencing progressive TED deterioration are more likely to quit smoking successfully. For those who do not quit smoking after verbal counselling, varenicline has been shown to have minimal effectiveness.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.