Intraocular and extraocular manifestations of thyroid dysfunction in Danish patients – A Nationwide Study

The population in all our studies are Danish residents who received treatment at a Danish hospital. Therefore, the following sections contain a brief description of the Danish healthcare system and Danish nationwide registries. In studies I and II, the data was obtained from Danish nationwide registries, while for study III, we collected data retrospectively from patient records from the two Danish hospitals, namely Rigshospitalet and Zealand University Hospital. Thus, the population frame is the same in all three studies, but the data sources and the inclusion criteria differ between the studies.

All exudative AMD cases and patients with TED are treated at a hospital; therefore, we choose to use the Danish nationwide registries as our material source for studies I and II. For study III, we evaluated the outcomes of single muscle recessions performed using pure TA. Danish nationwide registries do not include details such as the type of anaesthesia used for the strabismus surgery and the surgical outcomes; thus, we used patient records to collect data.

Study I was a longitudinal nationwide cohort study using Danish nationwide registries. We included all residents of Denmark aged 50–100 years between 2008 and 2018 and studied the interaction between the exposures thyroid dysfunction (either hypo- or hyperthyroidism) and cataract surgery on the outcome of exudative AMD. We defined thyroid dysfunction as two prescriptions for thyroid hormone supplementary medication or antithyroid medication within 5 years. We defined cataract surgery as a diagnosis code for cataract and a procedure code for cataract surgery. We defined exudative AMD as a diagnosis code for AMD followed by an anti-VEGF injection treatment code. We excluded all patients who had been or were prescribed lithium or amiodarone during the time period of the study. Furthermore, we excluded all thyroidectomies and thyroid cancer diagnoses. We included the following confounders: sex, age, education level, hypertension and chronic obstructive pulmonary disease (COPD). Because smoking is not registered in the Danish registries, we used COPD as a confounder and a proxy for smoking. See Table A1 in Appendix for the diagnosis codes, procedure codes and prescription definitions used in this study.

To determine the thyroid status of the included patients, we studied the prescriptions 5 years prior to the TED diagnosis. If a patient received L-thyroxine and antithyroid medication, then we considered the patient to be hyperthyroid. See Table A2 in Appendix for the diagnosis codes, procedure codes and prescription definitions used in this study.

Study III was a retrospective case series of all consecutive single muscle recessions conducted at two Danish centres between 2007 and 2018 by one surgeon. We divided the two groups undergoing surgery into a TED and a non-TED group. None of the muscles operated on had previously undergone surgery. The patients underwent surgery with TA if they themselves asked for TA as their anaesthetic option or if the surgeon deemed GA could pose a systemic risk to the patient. The indication for surgery and the methods used—including the surgical plan, surgical techniques and pre- and post-surgical evaluation—were standardised according to local clinical practice. All patients with TED included in this study were in the state of inactive disease, meaning that their angle of deviation had remained unchanged for 6 months prior to surgery.

The postoperative examinations occurred 6 weeks after surgery. If available, we included data from their latest visit after the surgery.

The TA used was eye drops (0.4% oxybuprocaine hydrochloride) and gel (2% lidocaine gel). The drops and gel were applied at the insertion site of the muscle. The application was made twice with 5 minutes in between.

The patients were asked immediately after the surgery to report how much pain they had experienced during the surgery by using the numerical visual analogue pain score scale (NVAS) scale. The patients rated their pain from 0 to 10, with 0 being no pain at all and 10 being the highest/maximum pain level (McCormack et al., 1988).

We established the definitions of usability and tolerability in accordance with the three parameters defining a surgical procedure performed with TA: how the patient experienced the surgery (NVAS score), the outcome of the surgery (motor success/sensory success/binocular single vision) and the course of the surgery (the complication rate). We established the motor success criteria according to how much PD a participant without strabismus could tolerate without developing diplopia.

When studying the association between thyroid dysfunction and cataract surgery on exudative AMD (study I), our population consisted of 2 087 305 individuals. Regarding the thyroid status of our population, 1.6% (33 992) were hyperthyroid, 2.8% (59 318) were hypothyroid and the rest were euthyroid (95.5%; 1 994 065). In total, 1.3% (26 998) developed exudative AMD. Overall, 72.6% (1 516 379) of the individuals were followed up until the end date of the study. The rest either died or experienced a competing event (thyroidectomy, thyroid cancer or a prescription of lithium or amiodarone).

In the group that developed exudative AMD, the median age was 79.4 years, and a majority of the patients were women (63.6%; 17 184). At the time of inclusion, the thyroid dysfunction division was the following: 2.6% (702) were hyperthyroid, 4.8% (1289) were hypothyroid and the remaining patients were euthyroid (92.6%; 25 007).

Our primary model (interaction effects model) showed that both hypothyroidism (hazard ratio [HR] 1.17; 95% confidence interval [CI] 1.10–1.25; p < 0.001) and hyperthyroidism (hazard ratio [HR] 1.23; 95% confidence interval [CI] 1.13–1.34; p < 0.001) were associated with development of exudative AMD, but only in patients who had not undergone cataract surgery (p_interaction = 0.003). There was no association between thyroid disorder in patients who had undergone cataract surgery. Additionally, we evaluated a secondary model (main effects model) without interaction or stratification. This model showed an association between exudative AMD and hypothyroidism, hyperthyroidism and prior cataract surgery.

For study II, we identified 4106 incident TED cases in Denmark between 2000 and 2018. On average, the annual population sample comprised 4.3 million individuals, for an overall total of 8.22 × 10⁷ person years. The mean overall age at the time of the TED diagnosis was 51.3 years (standard deviation [SD] = 14.5 years). The majority of the incident cases were women (81.4%; 3344), with a female-to-male ratio of 4:1. We did not observe a change in the sex distribution throughout the study (p = 0.11). At the time of the TED diagnosis, most of the patients were hyperthyroid (73.5%), with a hyperthyroid-to-hypothyroid-to-euthyroid ratio of 6:1:1.

The overall average incidence of TED was 5.0 per 100 000 person years: 8.0 per 100 000 person years for women and 1.9 per 100 000 person years for men. The incidence rate changed throughout the study period for both women and men, following the same pattern for each sex. We saw a decrease for 2005–2010, followed by an increase until 2014 when the rate reached a stable level until the end of the study period. The lowest overall incidence was 3.0 per 100 000 person-years, while the highest overall incidence was 6.4 per 100 000 person-years (Figure 1).

The 4-year cumulative incidence for ophthalmological sequelae of TED was the following: 10% for a diagnosis of strabismus, 8% for strabismus surgery and 5% for orbital decompression surgery. Compared with women, men had a higher cumulative incidence of strabismus and strabismus surgery, but not of decompression surgery.

In study III, we included 111 patients who underwent single muscle recessions with pure TA (42 patients in the TED group and 69 patients in the non-TED group). We initially included 115 patients but excluded four patients without TED due to perioperative conversion to other types of anaesthesia. There were several significant differences between the groups: a higher percentage of women, a larger preoperative vertical deviation, a larger number of muscles recessed and a greater change in vertical deviation after surgery in the TED group compared with the non-TED group. The majority of the surgeries performed in the TED group were recessions of the inferior rectus muscle.

The mean NVAS score was 2.3 (SD = 1.3) in the TED group and 1.6 (SD = 1.1) in the non-TED group; the difference was not significant when adjusted for sex. Overall, the women in the TED group had a significantly higher pain level than the women in the non-TED group. When analysing pain scores according to muscles, patients with TED undergoing inferior rectus recession had a significantly higher pain score then patients without TED.

Postoperatively, motor success, sensory success and binocular vision were achieved in more than 90% of patients in both groups. There was no significant difference between the groups. Complications occurred in 5% of patients in the TED group and 12% of patients in the non-TED group. These complications consisted of erosions of the cornea and conjunctival rupture and the need for resuturing. The OCR was triggered in two patients in the non-TED group.

Overall, the thesis has contributed new knowledge within the field of ophthalmological manifestations of patients with thyroid disorders and TED. We have demonstrated a possible association between hyper- and hypothyroidism and exudative AMD in patients without previous cataract surgery. The association we found was modest, meaning that thyroid disorder is only a contributing factor to the development of exudative AMD. However, exudative AMD is one of the most common vision-threatening diseases and, therefore, the public health impact can still be significant. Furthermore, the treatment and outcomes of exudative AMD can be costly for society. Additionally, our study was the first to investigate exudative AMD specifically and to also include cataract surgery in the study model. This approach has provided more detailed information on the association between thyroid disorder and AMD.

Through this thesis we have provided real-life estimates on the nationwide incidence of TED and the cumulative incidence of strabismus, strabismus surgery and decompression surgery in patients with TED. We have also provided insight into the fluctuation of the incidence of this disease over time and the characteristics of the patients in terms of sex, age and thyroid status at time of diagnosis.

We have also reported the first case series that compares the outcome of single muscle recessions performed with pure TA in patients with and without TED. We found that TA is a suitable anaesthetic method to use in single muscle recessions in patients with TED. This new information is valuable as it informs strabismus surgeons that TA is an option for single muscle recession small angle surgery in patients with TED.

We must emphasise that the findings from this thesis can only be generalised to an iodine-sufficient, predominantly Caucasian population with free access to healthcare.

Our finding that there is an association between exudative AMD and thyroid disorder is novel and contributes to the body of knowledge on the aetiology of exudative AMD. Given that previous studies only investigated the association between any form of AMD, and not specifically the exudative form, and did not include cataract surgery as a confounder, it is imperative that the study is repeated in other populations to see if our results can be confirmed. From a clinical perspective, our results show that an ophthalmologist should be aware that patients with hyper- or hypothyroidism have an increased risk of developing exudative AMD if they have not undergone cataract surgery and ensure that the patients are informed about the symptoms of exudative AMD so that they can receive timely treatment if they develop the condition. Hopefully, the results from our study can add to the body of knowledge on factors that contribute to the development of exudative AMD. Better understanding of this disorder can facilitate future preventive strategies and better treatments so that fewer patients develop debilitating symptoms such as blindness.

Our study on the incidence of TED and its ophthalmological sequalae provides valuable information for both clinicians and patients. More specifically, clinician have an estimate of how many new cases of TED can be expected per year and a description of the group of patients that develop the condition. Patients with thyroid disorder can now be informed on the nationwide incidence of TED. Moreover, once a patient develops TED, they can now receive information on their risk of developing strabismus or having to undergo strabismus surgery or decompression surgery. This information can facilitate the patient's understanding of their condition and give them a better picture of the potential sequelae of their disease.

Our findings can be used to compare the Danish patients with TED to other populations; this comparison could potentially provide more insight into the nature of the disease. Furthermore, our results can be used for future evaluations of new treatments and implementations in the field of TED treatment in Denmark by studying the changes in the incidence of TED and the cumulative incidence of ophthalmological sequelae over time and extending beyond 2018.

Until now, the topic of using TA for single muscle recessions in patients with TED has remained controversial, with some surgeons not willing to perform surgery on this group of patients when they are awake, fearing pain and discomfort. Hopefully, our results can contribute to encourage surgeons to use TA when performing single muscle recessions in patients with TED. Undergoing their surgery in TA, patients will spend less time in the hospital and avoid the potential side effects of GA.

The methods and outcomes from this thesis will be used for future projects in the field of ophthalmological manifestations of thyroid disorders. Indeed, we have already started working on several projects.

We will continue to use the Danish nationwide registries to study the risk factors for the development of TED. Previous studies have proposed that patients who are treated with RAI have a higher risk of developing TED compared with those who underwent thyroidectomy or were treated with solely medication. However, the findings on this topic have been conflicting and the number of patients included in these studies has been relatively small (Marcocci et al., 1992; Stein et al., 2015; Tallstedt et al., 1992). Using the Danish nationwide registries, we will report on the incidence of TED in patients with Graves' disease separated by the treatment received. This study is being performed in collaboration with Stanford University in the United States.

Smoking is a known risk factor for development of TED. Furthermore, it has been suggested that high cholesterol levels are also a risk factor for the development of TED (Bartalena et al., 2020). However, it has not been studied if other diseases that are caused by smoking and high levels of cholesterol increase the risk of developing TED in patients with Graves' disease. We will investigate this topic in collaboration with Vanderbilt University in the United States.

I will also participate in a Mendelian randomisation study investigating the correlation between thyroid dysfunction and several ophthalmological conditions such as AMD and glaucoma. There are few studies that use Mendelian randomisation. This work will provide new insights on the genetic correlation between thyroid disorder and ophthalmological disease.

This thesis has opened the door to new ideas and collaborations on the topic of ophthalmological manifestations of thyroid disorder, which will provide more knowledge and, hopefully, a better understanding of this area. I look forward to pursuing these new projects that will add to the body of knowledge on the topic and hopefully lead to improved treatment in these patients.

Chair:

Associate Professor Marie Louise Roed Rasmussen, MD, PhD, Department of Ophthalmology, Rigshospitalet-Glostrup, University of Copenhagen, Denmark

Assessors:

Professor Anja Eckstein, MD, DMSci, Department of Ophthalmology, Essen University Hospital, University of Duisburg-Essen, Germany

Professor Henrik Vorum, MD, DMSci, Department of Ophthalmology, Aalborg University Hospital, University of Aalborg, Denmark

Principal supervisor:

Professor Steffen Heegard, MD, DMSci, Department of Ophthalmology, Rigshospitalet-Glostrup, University of Copenhagen, Denmark

Primary co-supervisor:

Professor Christina Ellervik, MD, PhD, DMSci, Department of Clinical Biochemistry, Zealand University Hospital, University of Copenhagen, Denmark

Co-supervisors:

Helena Buch Hesgaard, MD, PhD, Institute of Neuroscience and Physiology, Gothenburg University, Sweden

Professor Toke Bek, MD, DMSci, Department of Ophthalmology, Aarhus University Hospital, Aarhus University, Denmark

Associate Professor Peter Bjerre Toft, MD, DMSci, Department of Ophthalmology, Rigshospitalet-Glostrup, University of Copenhagen, Denmark

Associate Professor Birte Nygaard, MD, PhD, Department of Endocrinology, Herlev-Gentofte Hospital, University of Copenhagen, Denmark

Professor Henrik Enghusen Poulsen, MD, DMSci, Department of Endocrinology, Bispebjerg Fredriksberg Hospital, University of Copenhagen, Denmark

Study I:

Boulakh, L., Isaksen, J.L., Poulsen, H.E., Faber, J., Heegaard, S., Nygaard, B., Kanters, J.K., Toft, P.B., Udholm, P.M., Bek, T., Hesgaard, H.B. & Ellervik, C. Thyroid dysfunction, cataract surgery, and exudative age-related macular degeneration—a longitudinal nationwide registry-based cohort study. Submitted.

Study II:

Boulakh, L., Nygaard, B., Bek, T., Faber, J., Heegaard, S., Toft, P.B., Poulsen, H.E., Toft-Petersen, A.P., Hesgaard, H.B. & Ellervik, C. (2022) Nationwide incidence of thyroid eye disease and cumulative incidence of strabismus and surgical interventions in Denmark. JAMA Ophthalmology, 140(7), 667–673. https://doi.org/10.1001/jamaophthalmol.2022.1002

Study III:

Boulakh, L., Toft-Petersen, A.P., Severinsen, M., Toft, P.B., Ellervik, C., Buch Hesgaard, H. & Heegaard, S. (2022) Topical anaesthesia in strabismus surgery for Graves' orbitopathy: a comparative study of 111 patients. Acta Ophthalmologica, 100(4), 447–453. https://doi.org/10.1111/aos.15024

Throughout this thesis, the studies are referred to with Roman numerals.