The increases in soft tissue volume characteristics of thyroid eye disease (TED) occur within a rigid bony cavity. Volume expansion within a fixed space leads generally, in a mechanical manner, to elevated pressure within the space. In the orbit, there are compensatory mechanisms limiting this volume–pressure relationship. Some of the pressure can be relieved by expanding the potential volume of the space anteriorly through the orbital outlet, manifesting as proptosis. Further pressure can be reduced by compensatory changes in blood flow and tissue density. However, there are limits to system capacity, and the long-term resistance of the orbital septum can overcome compensatory mechanisms, leading to a chronic elevation in intraorbital pressure for some patients affected by TED.1
Bone remodeling provides yet another mechanism for orbital pressure reduction. Previous radiologic studies have suggested that bony volume expansion occur in TED, particularly on the medial side of the orbit.2–4 However, the impact of these forces on the global bony orbital volume (BOV) is not well understood. In this study, the authors aim to assess bony changes in the orbit for patients affected by TED relative to normal controls.
This study adhered to the tenets of the Declaration of Helsinki and was performed in accordance with the Health Insurance Portability and Accountability Act. In this case–control study, eligible patients with a diagnosis of TED seen by a single specialist (D.B.R.) over a 2-year period were reviewed for study entry. Institutional review board approval for this study was provided by the University of California, Los Angeles. Participants included in the study were adults over 18 years of age with clinical evidence of TED and available high-resolution orbital CT studies. Exclusion criteria included the history of decompression surgery, in addition to medical or other ophthalmic conditions that could alter orbital anatomy.
Imaging studies for the control group were obtained from an institutional migraine clinic population and had been requested as part of the patient’s routine care. All control patient scans were reported as normal by a radiologist. The patients were matched for age and sex. To match the groups for race, only Caucasians were included in the study, as this population predominates in the region of the institution.
The primary outcome measure was BOV determined from measurements using imaging studies. Secondary outcome measures included the relationships between BOV, muscle volume (MV), fat volume (FV), Hertel measurements, duration of the disease, and the occurrence of dysthyroid optic neuropathy (DON) in patients with TED. All patients were clinically evaluated including calculation of the clinical activity score,5 Gorman diplopia score,6 Hertel exophthalmometry, and lagophthalmos. Length of disease from the onset of ophthalmic symptoms to the date of the clinic visit was also recorded.
The diagnosis of DON was made on clinical grounds and was defined as a combination of a decrease in visual acuity, the presence of a relative afferent pupillary defect, decrease in color vision, and Humphrey visual field defects. Other causes of optic neuropathy were excluded.
Only CT scans with slices of <1 mm thickness were used in this study. The scans were analyzed using the previously validated7 Mimics (Materialize, Leuven, Belgium 2017) analysis software tool. Three-dimensional reconstructions of the orbits were created, allowing the calculation of the BOV, MV, and FV, as described by Regensburg et al.7 (Fig. 1). The aforementioned technique involves measurement of the FV and MV only upto the bony aditus of the orbit. The authors modified this technique by measuring all the FV and MV posterior to the orbital septum. All reconstructions were performed by one of the authors (S.U.).
Statistical analysis was carried out using SPSS version 22.0 (SPSS, Inc., Chicago, IL). The groups were stratified by sex to reduce the influence of this variable as a confounding factor. The nonnormally distributed age of the groups was compared using the Mann–Whitney U test. The BOV of the 2 groups was compared using a 2-tailed t test. The generalized estimating equation8 was used to assess the relationship between BOV, MV, FV, Hertel measurements, duration of TED, age, and DON. This equation was used so that OU could be included for analysis without bias from the correlation between OU in the same patient. Statistical significance was defined as p < 0.05.
Intraobserver and Interobserver Variability.
For intraobserver variability, Shoaib Ugradar (SU) calculated the BOV, MV, and FV of 10 randomly selected patients on 3 consecutive days. Intraobserver variability was expressed using the coefficient of variation, calculated as the SD divided by 100 × the mean of the measurements. The percentage was calculated by multiplying this figure by 100.
The calculations of 2 observers were analyzed for interobserver variability. SU calculated the BOV, FV, and MV of 10 orbits twice. Thereafter, the second observer performed the same calculations independently on the same orbits (Fig. 2).
A total of 100 participants were included in the study, contributing 200 orbits. This sample comprised of 50 patients with TED (37 female, 13 male) and 50 controls (37 female and 12 male).
There were no significant differences between the control and patient groups in baseline characteristics. Demographic data are summarized in Table 1. The mean (SD) duration of time between the initial onset of symptoms of TED and the date of the scan used for analysis was 892.5 days (907.41) for males and 737.5 days (1,033.52) for females. Eleven patients (22%) had received a short course of oral steroids prior to imaging. Of the patients with TED, 14 demonstrated DON (15 orbits, 1 case was bilateral). At the time of the scan, the average clinical activity score (SD) for males was 4.1 (2.29) and 4.35 (2.63) for females.
The mean (SD) lagophthalmos was 1.72 (0.63) for males and 1.45 (0.62) for females. The mean (SD) Gorman diplopia score was 1.21 (1.07) for males and 1.10 (0.96) for females.
Soft Tissue Volume.
The mean (SD) MV in males with TED was 13,416.12 mm3 (3,451.92) while that of females was 11,010.93 mm3 (2,553.97). The mean (SD) FV in males with TED was 9,981.46 mm3 (2,459.14) and 5,574.16 (2,209.85) in females with TED. The difference between MV and FV in males and females with TED was significant (p < 0.01 and p < 0.01, respectively). The difference between MV and FV in patients with TED (matched for age and sex) was significant (p < 0.01). When patients who had received steroids were matched for age and sex against patients who did not receive steroids, no significant differences in the FV and MV were found (p = 0.77).
Bony Orbital Volume.
The mean (SD) BOV for males in the TED group was 28.62 ml (3.47), while that of the control group was 24.59 ml (2.19). This difference was significant (p < 0.01). The mean (SD) BOV for females with TED was 24.33 ml (2.39), while that of the female control group was 20.97 ml (1.84). This was again significant (p < 0.01). Patients who had received steroids prior to the scan used in this study were matched for age and sex against TED patients who did not have steroid therapy. No significant difference was found in the BOV between these groups (p = 0.86).
There were no significant relationships between BOV and: MV (p = 0.15), duration of disease (p = 0.705), Hertel measurements (p = 0.212), age (p = 0.9), and DON (p = 0.12, Table 2). There was a significant relationship between the BOV and FV (p < 0.05; Table 3).
Intraobserver and Interobserver Variability.
For 10 randomly selected orbits, SU found mean differences in the calculation (twice for each orbit) expressed in percentages of 0.83% for the BOV, 0.81% for FV, and 0.97% for the MV. Intraobserver variability of the 2 calculations on the 10 different orbits was 1.1% for the BOV, 0.89% for FV, and 1.34% for the MV.
The authors found the BOV to be significantly greater in patients with TED. While this was related to the FV, it was not related to MV, Hertel exophthalmometry, the occurrence of DON, or duration of the disease. This study suggests that TED is associated with general bony remodeling of the orbit.
Previous research has demonstrated bony orbital changes in response to localized mechanical pressure.4 Tan et al.3 provided a potential link between increased soft tissue volume and localized bony changes by showing that the angle of the inferomedial orbital strut was greater, while the angle of the medial wall was smaller, in patients with TED. Chan et al.9 reasoned that this represents a compensatory mechanism after their study found that bony orbital angles were wider in patients who did not have DON.
Of note, in both of these studies and ours, an association between BOV and proptosis was not demonstrated. The authors may expect that expanding proptosis is a precursor to bony remodeling or alternatively, reduction in proptosis response may be related to increased orbital pressure and thus orbital remodeling. It is of course possible that these situations overlap, and it may be challenging to peel their differential effects on bone remodeling apart without some data on premorbid eye position and direct orbital pressure measurements. These are both subjects of ongoing research.
This study found a significant increase in FV and MV in patients with TED compared with reference values for controls, suggested by Wiersinga et al.10 However, in their study, the MV was much smaller than that in the present study. This is due to their use of the bony orbital aditus (frontozygomatic suture to the anterior lacrimal crest) as a cut-off for measuring the FV and MV. This meant that for eyes that were proptotic, a large amount of FV and MV was not measured (reflected by a FV + MV/OV ratio < 1). They also excluded the oblique muscles from their measurements.
In the present study, the authors found a significant relationship between the FV and BOV. No relationship was found between the MV and BOV, despite the MV being significantly larger than the FV in males and females. Therefore, a purely mechanical process induced by the expansion of soft tissue volume does not explain the overall increase in BOV in patients with TED.
The authors would postulate that the mechanism for this process could be multifactorial. Under normal conditions, bone tissue is dynamic and continuously undergoes remodeling throughout life. This process is regulated by both systemic and local mediators. Systemic factors that are known to be involved in TED such as thyroid hormone, insulin like growth factor -1, and Interleukin 1 have also been shown to play a key role in bone metabolism.11 These factors could clearly be important in the process of bone remodeling demonstrated in this study.
Other pathways regulated through the osteoprotegerin and receptor activator of NF-kappa B pathways12 may also underscore the physiology demonstrated in these findings. Local influence on the osteoprotegerin/receptor activator of NF-kappa B system may take the form of inflammation or mechanical stress. Mechanical stress found in conditions, such as increased intracranial pressure13 and benign juxta-osseous masses,14 have been associated with remodeling of bone in other areas of the body. Such forces may have an impact on the bony orbit in TED, due to the expansion of orbital soft tissue volume within a rigid bony structure, leading to chronically increased intraorbital pressure.1,15 Inflammation is also a key factor in TED pathophysiology16 and similar nonspecific inflammation has been associated with bony remodeling in other conditions such as rheumatoid arthritis17 and periodontal disease.18 It is not clear which of these mechanisms predominate or how much overlap there may be between them; however, there are a number of known pathways that could be affected by TED physiology explaining the phenomenon the authors have noted here.
Limitations include the potential impact of steroids on bony remodeling. In this study, 11 (22%) patients had received steroids prior to the scan used in this study. The authors did not find a significant difference in BOV, FV, or MV, between these patients and other TED patients who had not received steroids. A further potential limitation is the use of a semiautomated segmentation technique for the calculation of the volumes in the orbit. However, the authors calculated the interclass and intraclass correlation and found that the process was accurate and repeatable.
This study suggests that there is an increase in the BOV of patients with TED. This is likely to have a multifactorial etiology that could include metabolic disturbance (hyperthyroidism), inflammation, and local mechanical pressure.
1. Otto AJ, Koornneef L, Mourits MP, Deen-van Leeuwen L. Retrobulbar pressures measured during surgical decompression of the orbit. Br J Ophthalmol 1996;80:1042–5.
2. Weis E, Heran MK, Jhamb A, et al. Quantitative computed tomographic predictors of compressive optic neuropathy in patients with thyroid orbitopathy: a volumetric analysis. Ophthalmology 2012;119:2174–8.
3. Tan NYQ, Leong YY, Lang SS, et al. Radiologic parameters of orbital bone remodeling in thyroid eye disease. Invest Ophthalmol Vis Sci 2017;58:2527–33.
4. Bokman CL, Ugradar S, Rootman DB. Measurement of medial wall bowing and clinical associations in thyroid eye disease. Ophthal Plast Reconstr Surg 2018;34:557–9.
5. Mourits MP, Koornneef L, Wiersinga WM, et al. Clinical criteria for the assessment of disease activity in Graves’ ophthalmopathy: a novel approach. Br J Ophthalmol 1989;73:639–44.
6. Bahn RS, Gorman CA. Choice of therapy and criteria for assessing treatment outcome in thyroid-associated ophthalmopathy. Endocrinol Metab Clin North Am 1987;16:391–407.
7. Regensburg NI, Kok PH, Zonneveld FW, et al. A new and validated CT-based method for the calculation of orbital soft tissue volumes. Invest Ophthalmol Vis Sci 2008;49:1758–62.
8. Glynn RJ, Rosner B. Comparison of alternative regression models for paired binary data. Stat Med 1994;13:1023–36.
9. Chan LL, Tan HE, Fook-Chong S, et al. Graves ophthalmopathy: the bony orbit in optic neuropathy, its apical angular capacity, and impact on prediction of risk. AJNR Am J Neuroradiol 2009;30:597–602.
10. Wiersinga WM, Regensburg NI, Mourits MP. Differential involvement of orbital fat and extraocular muscles in Graves’ ophthalmopathy. Eur Thyroid J 2013;2:14–21.
11. Hadjidakis DJ, Androulakis II. Bone remodeling. Ann N Y Acad Sci 2006;1092:385–96.
12. Kohli SS, Kohli VS. Role of RANKL-RANK/osteoprotegerin molecular complex in bone remodeling and its immunopathologic implications. Indian J Endocrinol Metab 2011;15:175–81.
13. Du Boulay GH, El Gammal T. The classification, clinical value and mechanism of sella turcica changes in raised intracranial pressure. Br J Radiol 1966;39:422–42.
14. Som PM, Lawson W, Lidov MW. Simulated aggressive skull base erosion in response to benign sinonasal disease. Radiology 1991;180:755–9.
15. Riemann CD, Foster JA, Kosmorsky GS. Direct orbital manometry in patients with thyroid-associated orbitopathy. Ophthalmology 1999;106:1296–302.
16. Naik VM, Naik MN, Goldberg RA, et al. Immunopathogenesis of thyroid eye disease: emerging paradigms. Surv Ophthalmol 2010;55:215–26.
17. Baum R, Gravallese EM. Impact of inflammation on the osteoblast in rheumatic diseases. Curr Osteoporos Rep 2014;12:9–16.
© 2019 by The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc., All rights reserved.
18. Gosset M, Rochefort GY. Inflammation and bone remodeling pathologies. Int J Orthop 2014;1:124–9.