- Email: [email protected]
Orbital Decompression for Disfiguring Exophthalmos in Thyroid Orbitopathy
Orbital Decompression for Disfiguring Exophthalmos in Thyroid Orbitopathy Christopher]. Lyons, FRCS, Jack Rootman, MD, FRCSC Background: Progressive exophthalmos from dysthyroid orbitopathy may result in marked disfigurement. Commonly affecting middle-aged women, it can be a significant social and psychologic handicap. Cosmesis is increasingly recognized as an indication for orbital decompression. Many argue, however, that decompression carries risks of operative complications which are only warranted where vision is threatened. Methods: Between 1984 and 1990, the authors performed orbital decompression for cosmetic indications on 34 patients (65 orbits) with thyroid orbitopathy. The charts of these patients were reviewed to assess the results of surgery and the nature and incidence of complications. Results: The mean retroplacement achieved was 4 mm (range, -1.0 to 10 mm). In 29 (85%) of the 34 patients, there was a difference of 1 mm or less in the proptosis of the two eyes postoperatively. Diplopia arose de novo in five (18%) previously asymptomatic patients. Postdecompression strabismus was managed successfully with adjustable surgery (mean, 1.3 operations per patient). With the exception of transient infraorbital nerve hypoesthesia, there were no surgical complications. Conclusion: Decompression surgery is effective in reducing exophthalmos in dysthyroid orbitopathy. In this series of operations, complications were rare and treatable. Decompression is often the first of a series of operations that may be necessary to correct the cosmetic sequelae of this condition. It should be considered by experienced surgeons in carefully selected and counseled patients who have disfiguring thyroid orbitopathy. Ophthalmology 1994;101:223-230
The indications for orbital decompression in thyroid orbitopathy have evolved over the last 10 years. Optic nerve compression and corneal exposure remain indications for urgent surgical decompression, particularly if treatment with steroids or irradiation has proven ineffective. Enthusiasm for surgical decompression purely for cosmetic improvement has been tempered by the recorded complications of this procedure. In addition to a slight risk of Originally received: November 9, 1992. Revision accepted: August 23, 1993. From the Department of Ophthalmology, University of British Columbia, Vancouver, British Columbia, Canada. Supported by The Frost Trust, London, England (Dr. Lyons). Presented at the American Academy of Ophthalmology Annual Meeting, Dallas, November 1992. Reprint requests to Jack Rootman, Department of Ophthalmology, 2550 Willow St, Vancouver, British Columbia, V5Z 3N9 Canada.
blindness,I,2 earlier reports 3,4 have stressed the high incidence of postoperative diplopia due to increased extraocular muscle restriction. In this context, surgeons believed that cosmesis was rarely, if ever, an indication for decompression. 5 Until recently, it was reserved for patients with "grotesque orbital and peri-orbital distortion who suffered altered body image and self-concept to the point of reclusiveness."6 Dysthyroid orbitopathy predominantly affects females at a time in life when physical changes greatly undermine patient confidence. The psychologic burden of the progressive disfigurement resulting from this disease is well recognized. 6,7 We encountered many patients who clearly expressed great concern over their deteriorating appearance. Our initially tentative incursion into surgical decompression as a means of reducing proptosis and redressing disfigurement led us to believe that patients were greatly improved psychologically and socially by this pro-
223
Ophthalmology
Volume 101, Number 2, February 1994
Table 1. Surgical Approach Used for Decompressions Approach Lynch medial canthal incision Subciliary incision Lateral canthal + "swinging lid flap"
No.
42 17
-CD ...J
"'C C
o Unilateral. decompression
5
-a:E 4 -- 3 etS
6 (3 wall decompressions)
6
. .e-
rnE
c>. CD I... CD-
cedure. This, in conjunction with the lower complication rates associated with the medial canthal or inferior fornix approach to antro-ethmoidal decompression,8 encouraged us to lower our threshold for performing decompression surgery in disfiguring exophthalmos. We have reviewed the results of surgical decompression of the orbit where the indication for surgery was cosmetic disfigurement from thyroid orbitopathy.
The charts and photographs of all patients with thyroid orbitopathy who underwent orbital decompression for disfiguring exophthalmos between 1984 and 1990 inclusive were retrieved and analyzed. Thirty-four patients were identified from a total of 90 who underwent decompression over this period. There were 24 females and 10 males. Their average age was 39.6 years (range, 19-71 years). Three patients underwent unilateral surgery, and bilateral decompressions were performed on all other patients. A total of 65 orbits were decompressed. No patient had previously undergone strabismus surgery. All the patients in this series complained primarily of their appearance. Their main concern was not only the staring expression related to the widening of their palpebral fissure, but also the protuberance of their eyes. Other complaints included tearing, intermittent foreign body sensation, lid swelling (especially on waking) and, in three patients, spontaneous prolapse of the globe. Seven (21 %) of the 34 patients complained of diplopia in the primary or reading position preoperatively. All were deemed to be in a quiescent phase of their orbitopathy
Table 2. Symmetry of Postoperative Results Difference (mm)·
No. of Patients
0
17
2 3
12
2 3
• Difference in Hertel exophthalmometry between the two eyes.
224
(JetS
c£ CD..e 1...0. CD O
1
12
0
17
~x
ow
Pre
Patients and Methods
1
3: CD _E CDO .QE CD- 2
operative
Post
Figure 1. Symmetry of exophthalmometer measurements.
by virtue of nonprogression of soft-tissue signs, proptosis, and motility abnormalities. All patients had been followed preoperatively for a minimum of 6 months. Each preoperative and postoperative examination included visual acuity, testing for color vision and pupillary reactions, measurement of lid fissure height and scleral show, exophthalmometry, range of ocular movements, and prism cover test measurements in the cardinal positions of gaze and in the reading position. We also assessed soft tissue signs with a clinical quantitative method for pre-septal and pre-tarsal lid swelling, as well as chemosis and injection. All patients' orbits were decompressed into the maxillary antrum and ethmoidal air cells. Three patients underwent bilateral three-wall decompressions with additional decompression into the temporal fossa. The approach used for the decompressions is summarized in Table 1. The mean postoperative follow-up was 19.3 months. The overall average preoperative Hertel exophthalmometer reading was 24.06 mm. For males, the average was 25.7 mm (range, 20-30 mm), whereas for females it was 23.4 mm (range, 18-29 mm). The three patients undergoing bilateral three-wall decompression had a mean preoperative exophthalmometry of 25.7 mm. Postoperatively, the overall average exophthalmometry was 19.88 mm. The mean retroplacement was 4 mm in males (range, 1-6 mm) and 3.96 mm in females (range, -1-10 mm). The mean retroplacement achieved with a two-wall decompression was 3.7 mm, whereas a threewall decompression resulted in a mean of 7.3 mm retroplacement.
Lyons et al . Decompression in Dysthyroid Orbitopathy
Table 3. Patients in Whom Diplopia in Primary Position and Downgaze Was Present Both Preoperatively and Postoperatively: Prism Cover Test Findings and Management Preoperative Status
Principal Duction Limitation
XT 8, R/ L 25
E, Ab
L/ R 14
E,Ab
XT 6, R/L 14
E
ET 30, L/ R 12
E,Ab
Incision, Walls Removed Lynch Eth, Max Lynch Eth, Max Lynch Eth, Max Lynch Eth, Max
Postoperative Status
Management
ET 2, R/ L 8
Spontaneous orthophoria after 1 yr
ET 6, L/ R 15
RIR recession (adjustable)
R/ L 10
Symptoms controlled with prisms
ET 30, L/R 14
Bilateral IR recession, bilateral MR recession
Remained esotropic with L hypertropia -+ IR + MR recession Exotropic 30, L hypertropia 12 -+ LSR recession, advance LMR Orthotropia in primary + downgaze ET 4
Ab
L/R 18
E
XT 30, R/ L 6
E, Ab
Lynch Eth, Max Lynch Eth, Max L lid Eth, Max
+ L MR recession
ET 12
R
L/ R 8
Spontaneous orthophoria after 1 year
XT 50, L/R 10
Bilateral LR recession
XT = exotropia; R/L = right hypertropia; E = preoperative limitation of elevation; Ab = preoperative limitation of abduction; Eth = decompression into ethmoidal air cells; Max = decompression into maxillary antrum; L/ R = left hypertropia; ET = esotropia; RIR = right inferior rectus; IR = inferior rectus; MR = medial rectus; LSR = left superior rectus; LMR = left medial rectus; LR = lateral rectus.
The symmetry of the postoperative exophthalmometry is described in Table 2. The relationship between preoperative and postoperative symmetry is illustrated graphically on Figure I. The preoperative lid fissure height ranged from 8 to 19 mm (mean, 13.4 mm). The mean lid fissure height after decompression (but before eyelid surgery) was 12.6 mm (range, 9-17 mm).
Diplopia Seven patients who complained of diplopia preoperatively remained symptomatic postoperatively, although two of these became asymptomatic within the first year postoperatively. Four of the seven patients underwent strabismus surgery. The measurements of these patients' strabismus are described in Table 3.
Table 4. Patients with No Diplopia in Primary Position and Downgaze Preoperatively, and Diplopia Postoperatively: Prism Cover Test Findings and Management
Preoperative Status
Principal Duction Limitation
Orthophoric
Ab
Phoric (XT 6, L/R 5)
E(R > L)
Esophoria 4 (esotropic)
E, Ab
Orthophoria
Ab
Orthophoria
Incision, Walls Removed Lynch Eth, Max Lynch Eth, Max (R only) L lid Eth, Max, Lat Lynch Eth, Max Lynch Eth, Max
Postoperative Status
Management
14 ET
Bilateral MR recession
R hypotropia
Controlled with slight head posture; lost to follow-up
40 ET
Bilateral MR recession
25 ET
Bilateral MR recession
12 ET
Esophoric at 18 mos
Ab = preoperative limitation of abduction; Eth = decompression into ethmoidal air cells; Max = decompression into maxillary antrum; ET = esotropia; MR = medial rectus; XT = exotropia; L/R = left hypertropia; E = preoperative limitation of elevation; Lat = lateral decompression.
225
Ophthalmology
Volume 101, Number 2, February 1994
Table 5. Additional Surgery Performed on Patients Undergoing Decompression for Disfiguring Thyroid Orbitopathy Correction of Upper Lid Retraction (mullerectomy)
No.
Before decompression Post-decompression Lower lid elevation Lateral tarsorrhaphy Blepharoplasty
5 18 3 3 7
Five patients who did not complain of diplopia in the primary position or in downgaze preoperatively became symptomatic in those positions postoperatively. Their details are described in Table 4. In this group, three patients underwent surgery for postoperative strabismus. One patient's symptoms were controlled with prisms which were gradually reduced; he became asymptomatic ~fter 18 months. One patient, who controlled a small nght hypotropia with a head posture, was lost to follow-up. Eighteen of the 34 patients were found to have full ocular movements before decompression. Of these patients, strabismus developed in only one after decompression via a Lynch approach (Table 4, patient 5). This patient became spontaneously phoric within 18 months. In three patients, limitation of upgaze improved after decompression surgery. Of the 13 patients with restricted ocular motility, strabismus developed in 7 preoperatively and postoperatively and in 4 de novo. The incision used for decompression is noted in Tables 3 and 4; 10 of the 12 patients with strabismus postoperatively were decompressed via a Lynch incision. Decompression surgery is recognized to be only one of a series of procedures that may be needed to rehabilitate patients with dysthyroid orbitopathy.9 Table 5 describes the eyelid surgery performed on patients in this series. Infraorbital nerve hypoesthesia frequently was reported by these patients in the early postoperative period. This symptom never persisted longer than 6 months after surgery. Other than the cases of strabismus described ab?ve, there were no complications related to decompressIOn. There was no deterioration in visual acuity in any of these patients.
Discussion Surgical decompression for thyroid orbitopathy into the temporal fossa was first described in 1911 by Dollinger. 10 The technique has since been modified, with each wall of the orbit being transgressed in turn to provide additional space for the enlarged orbital contents. Currently, decompression is most commonly performed into the maxillary antrum and ethmoidal air cells. 8 Additional decompression into the temporal fossa and the intracranial space or "four-wall" decompression II is reserved for pa226
tients with severe proptosis. Recently, an alternative mode of reduction of proptosis in thyroid orbitopathy was reported,12 involving removal of orbital fat via an anterior approach. This technique has not yet been widely evaluated. Antro-ethmoidal decompression has been the operation of choice in thyroid orbitopathy for the last 40 years. This was performed via a Caldwell Luc approach (Ogura technique) until the early 1980s and continues to be performed this way in many centers. McCord8 has documented a trend away from this approach in favor of inferior fornix incisions, which has been accompanied by a reduction in postoperative motility problems. Our criteria for cosmetic decompression are generally based on exophthalmometry measurements of 23 mm or more in men and 20 mm or more in women. However, the patients' ages together with present and past appearance (from photographs) are also important. In addition, occupations with a great deal of public exposure as well as cultural expectations must be considered in the assessment of these patients' concerns. As a standard part of the preoperative interview, the patient is asked to bring photographs of themselves taken before the development of the orbitopathy to define the problem and set realistic goals for the decompression procedure. Surgery is only offered to patients who express significant discomfort about their appearance, often manifested by avoidance of new social contacts or the constant wearing of dark glasses. A detailed discussion with the patient precedes consent. The recorded complications of decompression surgery are presented, and the possibility of introducing or aggravating diplopia is fully explained. The patient also is warned that further surgery to correct lid malposition or strabismus may be necessary to achieve a satisfactory result. It is important for the disease to be quiescent and for the surgeon to choose the least potentially damaging procedure that will give good results. Factors such as facial contour, race, depth of orbit, previous physical appearance, and patient expectation should be considered when selecting the operation or set of operations that will achieve the desired outcome for each patient. The amount of decompression performed is tailored to the patient's exophthalmos and, in unilateral decompression, to the degree of contralateral proptosis. The aim is to achieve appropriate retroplacement of the globes. Previous authors also have stressed the importance of postoperative symmetry of the eyes in achieving satisfactory cosmesis.1 3 We previously reported l4 that the number of walls decompressed correlates well with the amount of decompression achieved. On average, a single wall decompression results in a 2.5-mm retroplacement, a floor and medial wall in 4 mm, and a three-wall decompression in 7 mm. These decompressive effects often improve after 6 months of follow-up. 14 These results are similar to those reported by other surgeons. 8 Single-wall decompression surgery and the resultant herniation of orbital tissue into a single sinus cavity may give rise to asymmetric displacement of the orbital contents; we prefer to "balance" the decompression by removing two walls at least. IS The surgical techniques used
Lyons et al . Decompression in Dysthyroid Orbitopathy
o Figure 2. This 34-year-old woman had a 4-year history of thyroid orbitopathy whose onset coincided with hyperthyroidism. She was stable for 8 months. Results of ocular examination showed orthophoria with exophthalmos of 27 mm in the left eye and 26 mm in the right eye and interpalpebral fissure of 18 mm in the right eye and 20 mm in the left (A and B). She underwent bilateral three-wall decompression via a swinging eyelid approach. Three months after decompression, she underwent bilateral mullerectomy. The final result, obtained 8 months after decompression, showed exophthalmometry of 19 mm bilaterally, orthophoria, and an interpalpebral fissure of 11 mm in the right eye and 12 mm in the left (e and D).
in this series of patients were described previously elsewhere. 1 The subciliary, swinging eyelid or medial canthal approaches were used to allow a controlled and directly observed removal of the medial and inferior walls of the orbit. Using these, release of the orbital contents into the
antro-ethmoidal complex was achieved with minimally visible skin incisions. The lower lid or inferior fornix approach to decompression allows internal blepharoplasty to be performed at the time of decompression which may further improve the cosmetic result.
227
Ophthalmology
Volume 101, Number 2, February 1994
Figure 3. This 29-year-old man had exophthalmos secondary to hyperthyroidism that had been stable for 8 months. On physical examination, he was orthophoric with exophthalmometry of 23 mm in the right eye and 24 mm in the left (A). He underwent bilateral medial wall and floor decompression. Four months after surgery, he was orthophoric with exophthalmometry of 21 mm bilaterally (B).
To date, there has been no case-controlled prospective study to evaluate results with each type of surgery. Rather than advocate specific procedures, we would emphasize the need to become familiar with the approach chosen and to individualize care according to the above principles. As the surgeon's skill improves, the choice of decompression may change and also may be combined with bony advancement and augmentation. Patients undergoing decompression for compressive optic neuropathy or corneal exposure often have severe apical crowding. This group of patients has a high prevalence of preoperative restriction of ductions and diplopia. 14 Because this was a major indication for decompression in earlier series, severe motility problems frequently were present preoperatively and even more prevalent postoperatively. The perception that, in experienced hands, surgical decompression is a safe operation has been accompanied by a gradual change in its indication in thyroid orbitopathy; patients are undergoing this surgery earlier and have a lower prevalence of preoperative ocular motility problems. In DeSanto'sl7 series of 200 patients who underwent decompression by the Ogura technique between 1969 and
228
1976, the incidence for decompression was cosmesis or steroid intolerance in 16.5% of the patients and the incidence of diplopia was 53% preoperatively, rising to 80% postoperatively. More recently, Shorr et al 6 reported a series of decompressions performed using the same technique, in which 60% of the patients were undergoing surgery for cosmetic indications. Diplopia was present in 32% of patients preoperatively, increasing to 50% postoperatively. They found that patients with restricted ocular motility preoperatively were more likely to have postoperative diplopia. Their findings are confirmed by this series: postoperative diplopia developed in only 1 of 18 patients with full ductions preoperatively, whereas 12 of the 16 patients with preoperative restrictions of gaze had diplopia postoperatively. In addition to a change in patient selection, changes in decompression techniques have contributed to the reduction in operative complications. Whereas Hurwitz and Birt 18 found a worsening of motility in 16 of 19 orbits decompressed via the trans-antral approach, McCord8 reported a 5.6% incidence of worsened muscle balance after orbital decompression via the trans-lid approach.
Lyons et al . Decompression in Dysthyroid Orbitopathy
Figure 4. This 33-year-old man had stable proptosis for 2 years and exophthalmometry of 26 mm in the right eye and 27 mm in the left (A). He underwent bilateral two-wall decompression. Six months after decompression, he was orthophoric with exophthalmometry of 21 mm bilaterally (B).
In our series, diplopia was present preoperatively in the primary or reading positions in 7 (21 %) of the 34 patients. Of these patients, two became asymptomatic within 1 year of the decompression surgery. One patient's diplopia, due to a IO-prism diopter hypotropia, was managed with prisms only. Three patients underwent one operation each which was sufficient to treat their symptoms in the primary and reading positions. One patient with a severe preoperative myopathy needed three strabismus operations to be eliminate diplopia in the primary position and downgaze. Of the 27 patients who did not complain of diplopia preoperatively, diplopia developed in 5 (18%). This persisted beyond 1 year in four (14%) patients. Eighteen percent of the patients who did not complain of diplopia preoperatively did so after decompression. Overall, 24% of our patients required postdecompression strabismus surgery. This incidence compares favorably with earlier series in which decompression was carried out via the maxillary antrum or in the presence of greater muscle involvement by the dysthyroid process. Where diplopia did occur de novo after decompression, it resolved spontaneously or was cured by one strabismus surgery (one patient was lost to follow-up). We agree with the previous authors 6 that
orbital decompression per se does not complicate the management of strabismus. The improvement reported by three patients who became asymptomatic after 1 year postoperatively, and the fourth whose diplopia was controlled with prisms only, may be due to symmetric involvement of the inferior recti, reducing the angle of the vertical strabismus. It is also important to note that these patients had a reduction in the size of their deviation after their decompression, a phenomenon that has been reported previously in a small proportion of patients. 5,8, 19 A-pattern esotropia and torsional diplopia are complications of decompression surgery that have been reported previously.20,21 This is due to increased mechanical advantage of the superior oblique muscle after retroplacement of the globe, or decreased inferior oblique action after inferonasal orbital decompression. Neither of these was found in this group, although torsion was not specifically measured. It is likely that these patterns are more prevalent in patients who have undergone large inferior rectus recession before orbital decompression. None of our patients had previous strabismus surgery (because this was an exclusion criterion for this study). In some cases, the cosmetic defect of thyroid orbitopathy can be remedied by correcting lid retraction only
229
Ophthalmology
Volume 101, Number 2, February 1994
without resorting to a potentially risky decompression. This has not always been our experience; five patients in this series had undergone levator recessions for cosmetic purposes before decompression became necessary to correct their disfigurement. Our overall results showed that total lid fissure height was unaffected by decompressive surgery. An improvement of lower lid retraction and worsening of upper lid retraction after decompression has been reported previousli,g but was noted in only a few of our patients; this was no more likely in patients who had had decompression via a lower lid incision than in those whose decompression was performed via a Lynch incision. Twenty-nine (85%) of the 34 patients in our series had 1 mm or less difference in the proptosis of their two eyes after orbital decompression. The relation between preoperative and postoperative symmetry of the proptosis of the two eyes is illustrated graphically in Figure 1. The cosmetic improvement achieved with orbital decompression IS illustrated in Figures 2, 3 and 4. Patients commonly report that their disfigurement from thyroid orbitopathy is a significant social and psychologic handicap. We have found surgical decompression to be reasonably safe and effective in reducing cosmetically disfiguring exophthalmos and improving cosmesis. In our view, cosmesis is a valid indication for this form of surgery in carefully selected and informed patients. We emphasize that this article represents an evolving view point that is governed by increasing experience and comfort with offering orbital decompression to patients who have major cosmetic concerns. In some respects, the risk associated with orbital decompression has been over-emphasized, and this procedure may have been withheld in circumstances in which patients could have been helped. Although we would not encourage a cavalier attitude toward what is a potentially sight-threatening procedure, a wellcontrolled approach by an experienced surgeon should be considered when dealing with the disfigurement of thyroid orbitopathy.
References 1. Ogura JH, Lucente FE. Surgical results of orbital decompression for malignant exophthalmos. Laryngoscope 1974;84:637-44. 2. Long JC, Ellis PP. Total unilateral visual loss following orbital surgery. Am J Ophthalmol 1971 ;71 :218-20. 3. Young JDH. Ocular complications of transantral decompression for thyrotrophic exophthalmos. Proc R Soc Med 1971 ;64:929-31.
230
4. DeSanto LW, Gorman CA. Selection of patients and choice of operation for orbital decompression in Graves ophthalmopathy. Laryngoscope 1973;83:945-59. 5. Linberg JV, Anderson RL. Transorbital decompression: indications and results. Arch Ophthalmol 1981 ;99: 113-19. 6. Shorr N, Neuhaus RW, Baylis HI. Ocular motility problems after orbital decompression for dysthyroid ophthalmopathy. Ophthalmology 1982;89:323-8. 7. Apers RC, Oosterhuis JA, Sedee E, Bierlaagh J. Results of transantral decompression in hyperthyroid exophthalmos. Doc Ophthalmol 1977;44:207-14. 8. McCord CD Jr. Current trends in orbital decompression. Ophthalmology 1985;92:21-33. 9. Shorr N, SeiffSR. The four stages of surgical rehabilitation of the patient with dysthyroid ophthalmopathy. Ophthalmology 1986;93:476-83. 10. Dollinger J. Die Druckentlastung der Augenhohle durch Entfernung der au/3eren Orbitalwand bei hochgradigen Exophthalmus (Morbus Basedowii) und konsekutiver Hornhauterkrankung. Dtsch Med W ochenschr 1911;37: 1888-90. 11. Maroon JC, Kennerdell JS. Radical orbital decompression for severe dysthyroid exophthalmos. J Neurosurg 1982;56: 260-6. 12. Olivari N. Transpalpebral decompression of endocrine ophthalmopathy (Graves' disease) by removal of intraorbital fat: experience with 147 operations over 5 years. Plast Reconstruct Surg 1991;87:627-43. 13. Mourits MP, Koornneef L, Wiersinga WM, et al. Orbital decompression for Graves' ophthalmopathy by inferomedial, by inferomedial plus lateral, and by coronal approach. Ophthalmology 1990;97:636-41. 14. Satorre J, Rootman J. Des compresion orbitaria en la orbitopatia distiroidea. Resultados preliminares. Arch Soc Esp Oftalmol 1991 ;60:721-8. 15. Leone CR Jr, Piest KL, Newman RJ. Medial and lateral wall decompression for thyroid ophthalmopathy. Am J Ophthalmol 1989; 108: 160-6. 16. Rootman J. Diseases of the Orbit: A Multidisciplinary Approach. Philadelphia: JB Lippincott, 1988. 17. DeSanto LW. The total rehabilitation of Graves' ophthalmopathy. Laryngoscope 1980;90: 1652-78. 18. Hurwitz JJ, Birt D. An individualized approach to orbital decompression in Graves' orbitopathy. Arch Ophthalmol 1985; 103:660-5. 19. Streeten DHP, Anderson GH Jr, Reed GF, Woo P. Prevalence, natural history and surgical treatment of exophthalmos. Clin Endocrinol 1987;27: 125-33. 20. Fells P. Orbital decompression for severe dysthyroid eye disease. Br J Ophthalmol 1987;71:107-11. 21. Garrity JA, Saggau DD, Gorman CA, et al. Torsional diplopia after transantral orbital decompression and extraocular muscle surgery associated with Graves' orbitopathy. Am J Ophthalmol 1992;113:363-73.
The indications for orbital decompression in thyroid orbitopathy have evolved over the last 10 years. Optic nerve compression and corneal exposure remain indications for urgent surgical decompression, particularly if treatment with steroids or irradiation has proven ineffective. Enthusiasm for surgical decompression purely for cosmetic improvement has been tempered by the recorded complications of this procedure. In addition to a slight risk of Originally received: November 9, 1992. Revision accepted: August 23, 1993. From the Department of Ophthalmology, University of British Columbia, Vancouver, British Columbia, Canada. Supported by The Frost Trust, London, England (Dr. Lyons). Presented at the American Academy of Ophthalmology Annual Meeting, Dallas, November 1992. Reprint requests to Jack Rootman, Department of Ophthalmology, 2550 Willow St, Vancouver, British Columbia, V5Z 3N9 Canada.
blindness,I,2 earlier reports 3,4 have stressed the high incidence of postoperative diplopia due to increased extraocular muscle restriction. In this context, surgeons believed that cosmesis was rarely, if ever, an indication for decompression. 5 Until recently, it was reserved for patients with "grotesque orbital and peri-orbital distortion who suffered altered body image and self-concept to the point of reclusiveness."6 Dysthyroid orbitopathy predominantly affects females at a time in life when physical changes greatly undermine patient confidence. The psychologic burden of the progressive disfigurement resulting from this disease is well recognized. 6,7 We encountered many patients who clearly expressed great concern over their deteriorating appearance. Our initially tentative incursion into surgical decompression as a means of reducing proptosis and redressing disfigurement led us to believe that patients were greatly improved psychologically and socially by this pro-
223
Ophthalmology
Volume 101, Number 2, February 1994
Table 1. Surgical Approach Used for Decompressions Approach Lynch medial canthal incision Subciliary incision Lateral canthal + "swinging lid flap"
No.
42 17
-CD ...J
"'C C
o Unilateral. decompression
5
-a:E 4 -- 3 etS
6 (3 wall decompressions)
6
. .e-
rnE
c>. CD I... CD-
cedure. This, in conjunction with the lower complication rates associated with the medial canthal or inferior fornix approach to antro-ethmoidal decompression,8 encouraged us to lower our threshold for performing decompression surgery in disfiguring exophthalmos. We have reviewed the results of surgical decompression of the orbit where the indication for surgery was cosmetic disfigurement from thyroid orbitopathy.
The charts and photographs of all patients with thyroid orbitopathy who underwent orbital decompression for disfiguring exophthalmos between 1984 and 1990 inclusive were retrieved and analyzed. Thirty-four patients were identified from a total of 90 who underwent decompression over this period. There were 24 females and 10 males. Their average age was 39.6 years (range, 19-71 years). Three patients underwent unilateral surgery, and bilateral decompressions were performed on all other patients. A total of 65 orbits were decompressed. No patient had previously undergone strabismus surgery. All the patients in this series complained primarily of their appearance. Their main concern was not only the staring expression related to the widening of their palpebral fissure, but also the protuberance of their eyes. Other complaints included tearing, intermittent foreign body sensation, lid swelling (especially on waking) and, in three patients, spontaneous prolapse of the globe. Seven (21 %) of the 34 patients complained of diplopia in the primary or reading position preoperatively. All were deemed to be in a quiescent phase of their orbitopathy
Table 2. Symmetry of Postoperative Results Difference (mm)·
No. of Patients
0
17
2 3
12
2 3
• Difference in Hertel exophthalmometry between the two eyes.
224
(JetS
c£ CD..e 1...0. CD O
1
12
0
17
~x
ow
Pre
Patients and Methods
1
3: CD _E CDO .QE CD- 2
operative
Post
Figure 1. Symmetry of exophthalmometer measurements.
by virtue of nonprogression of soft-tissue signs, proptosis, and motility abnormalities. All patients had been followed preoperatively for a minimum of 6 months. Each preoperative and postoperative examination included visual acuity, testing for color vision and pupillary reactions, measurement of lid fissure height and scleral show, exophthalmometry, range of ocular movements, and prism cover test measurements in the cardinal positions of gaze and in the reading position. We also assessed soft tissue signs with a clinical quantitative method for pre-septal and pre-tarsal lid swelling, as well as chemosis and injection. All patients' orbits were decompressed into the maxillary antrum and ethmoidal air cells. Three patients underwent bilateral three-wall decompressions with additional decompression into the temporal fossa. The approach used for the decompressions is summarized in Table 1. The mean postoperative follow-up was 19.3 months. The overall average preoperative Hertel exophthalmometer reading was 24.06 mm. For males, the average was 25.7 mm (range, 20-30 mm), whereas for females it was 23.4 mm (range, 18-29 mm). The three patients undergoing bilateral three-wall decompression had a mean preoperative exophthalmometry of 25.7 mm. Postoperatively, the overall average exophthalmometry was 19.88 mm. The mean retroplacement was 4 mm in males (range, 1-6 mm) and 3.96 mm in females (range, -1-10 mm). The mean retroplacement achieved with a two-wall decompression was 3.7 mm, whereas a threewall decompression resulted in a mean of 7.3 mm retroplacement.
Lyons et al . Decompression in Dysthyroid Orbitopathy
Table 3. Patients in Whom Diplopia in Primary Position and Downgaze Was Present Both Preoperatively and Postoperatively: Prism Cover Test Findings and Management Preoperative Status
Principal Duction Limitation
XT 8, R/ L 25
E, Ab
L/ R 14
E,Ab
XT 6, R/L 14
E
ET 30, L/ R 12
E,Ab
Incision, Walls Removed Lynch Eth, Max Lynch Eth, Max Lynch Eth, Max Lynch Eth, Max
Postoperative Status
Management
ET 2, R/ L 8
Spontaneous orthophoria after 1 yr
ET 6, L/ R 15
RIR recession (adjustable)
R/ L 10
Symptoms controlled with prisms
ET 30, L/R 14
Bilateral IR recession, bilateral MR recession
Remained esotropic with L hypertropia -+ IR + MR recession Exotropic 30, L hypertropia 12 -+ LSR recession, advance LMR Orthotropia in primary + downgaze ET 4
Ab
L/R 18
E
XT 30, R/ L 6
E, Ab
Lynch Eth, Max Lynch Eth, Max L lid Eth, Max
+ L MR recession
ET 12
R
L/ R 8
Spontaneous orthophoria after 1 year
XT 50, L/R 10
Bilateral LR recession
XT = exotropia; R/L = right hypertropia; E = preoperative limitation of elevation; Ab = preoperative limitation of abduction; Eth = decompression into ethmoidal air cells; Max = decompression into maxillary antrum; L/ R = left hypertropia; ET = esotropia; RIR = right inferior rectus; IR = inferior rectus; MR = medial rectus; LSR = left superior rectus; LMR = left medial rectus; LR = lateral rectus.
The symmetry of the postoperative exophthalmometry is described in Table 2. The relationship between preoperative and postoperative symmetry is illustrated graphically on Figure I. The preoperative lid fissure height ranged from 8 to 19 mm (mean, 13.4 mm). The mean lid fissure height after decompression (but before eyelid surgery) was 12.6 mm (range, 9-17 mm).
Diplopia Seven patients who complained of diplopia preoperatively remained symptomatic postoperatively, although two of these became asymptomatic within the first year postoperatively. Four of the seven patients underwent strabismus surgery. The measurements of these patients' strabismus are described in Table 3.
Table 4. Patients with No Diplopia in Primary Position and Downgaze Preoperatively, and Diplopia Postoperatively: Prism Cover Test Findings and Management
Preoperative Status
Principal Duction Limitation
Orthophoric
Ab
Phoric (XT 6, L/R 5)
E(R > L)
Esophoria 4 (esotropic)
E, Ab
Orthophoria
Ab
Orthophoria
Incision, Walls Removed Lynch Eth, Max Lynch Eth, Max (R only) L lid Eth, Max, Lat Lynch Eth, Max Lynch Eth, Max
Postoperative Status
Management
14 ET
Bilateral MR recession
R hypotropia
Controlled with slight head posture; lost to follow-up
40 ET
Bilateral MR recession
25 ET
Bilateral MR recession
12 ET
Esophoric at 18 mos
Ab = preoperative limitation of abduction; Eth = decompression into ethmoidal air cells; Max = decompression into maxillary antrum; ET = esotropia; MR = medial rectus; XT = exotropia; L/R = left hypertropia; E = preoperative limitation of elevation; Lat = lateral decompression.
225
Ophthalmology
Volume 101, Number 2, February 1994
Table 5. Additional Surgery Performed on Patients Undergoing Decompression for Disfiguring Thyroid Orbitopathy Correction of Upper Lid Retraction (mullerectomy)
No.
Before decompression Post-decompression Lower lid elevation Lateral tarsorrhaphy Blepharoplasty
5 18 3 3 7
Five patients who did not complain of diplopia in the primary position or in downgaze preoperatively became symptomatic in those positions postoperatively. Their details are described in Table 4. In this group, three patients underwent surgery for postoperative strabismus. One patient's symptoms were controlled with prisms which were gradually reduced; he became asymptomatic ~fter 18 months. One patient, who controlled a small nght hypotropia with a head posture, was lost to follow-up. Eighteen of the 34 patients were found to have full ocular movements before decompression. Of these patients, strabismus developed in only one after decompression via a Lynch approach (Table 4, patient 5). This patient became spontaneously phoric within 18 months. In three patients, limitation of upgaze improved after decompression surgery. Of the 13 patients with restricted ocular motility, strabismus developed in 7 preoperatively and postoperatively and in 4 de novo. The incision used for decompression is noted in Tables 3 and 4; 10 of the 12 patients with strabismus postoperatively were decompressed via a Lynch incision. Decompression surgery is recognized to be only one of a series of procedures that may be needed to rehabilitate patients with dysthyroid orbitopathy.9 Table 5 describes the eyelid surgery performed on patients in this series. Infraorbital nerve hypoesthesia frequently was reported by these patients in the early postoperative period. This symptom never persisted longer than 6 months after surgery. Other than the cases of strabismus described ab?ve, there were no complications related to decompressIOn. There was no deterioration in visual acuity in any of these patients.
Discussion Surgical decompression for thyroid orbitopathy into the temporal fossa was first described in 1911 by Dollinger. 10 The technique has since been modified, with each wall of the orbit being transgressed in turn to provide additional space for the enlarged orbital contents. Currently, decompression is most commonly performed into the maxillary antrum and ethmoidal air cells. 8 Additional decompression into the temporal fossa and the intracranial space or "four-wall" decompression II is reserved for pa226
tients with severe proptosis. Recently, an alternative mode of reduction of proptosis in thyroid orbitopathy was reported,12 involving removal of orbital fat via an anterior approach. This technique has not yet been widely evaluated. Antro-ethmoidal decompression has been the operation of choice in thyroid orbitopathy for the last 40 years. This was performed via a Caldwell Luc approach (Ogura technique) until the early 1980s and continues to be performed this way in many centers. McCord8 has documented a trend away from this approach in favor of inferior fornix incisions, which has been accompanied by a reduction in postoperative motility problems. Our criteria for cosmetic decompression are generally based on exophthalmometry measurements of 23 mm or more in men and 20 mm or more in women. However, the patients' ages together with present and past appearance (from photographs) are also important. In addition, occupations with a great deal of public exposure as well as cultural expectations must be considered in the assessment of these patients' concerns. As a standard part of the preoperative interview, the patient is asked to bring photographs of themselves taken before the development of the orbitopathy to define the problem and set realistic goals for the decompression procedure. Surgery is only offered to patients who express significant discomfort about their appearance, often manifested by avoidance of new social contacts or the constant wearing of dark glasses. A detailed discussion with the patient precedes consent. The recorded complications of decompression surgery are presented, and the possibility of introducing or aggravating diplopia is fully explained. The patient also is warned that further surgery to correct lid malposition or strabismus may be necessary to achieve a satisfactory result. It is important for the disease to be quiescent and for the surgeon to choose the least potentially damaging procedure that will give good results. Factors such as facial contour, race, depth of orbit, previous physical appearance, and patient expectation should be considered when selecting the operation or set of operations that will achieve the desired outcome for each patient. The amount of decompression performed is tailored to the patient's exophthalmos and, in unilateral decompression, to the degree of contralateral proptosis. The aim is to achieve appropriate retroplacement of the globes. Previous authors also have stressed the importance of postoperative symmetry of the eyes in achieving satisfactory cosmesis.1 3 We previously reported l4 that the number of walls decompressed correlates well with the amount of decompression achieved. On average, a single wall decompression results in a 2.5-mm retroplacement, a floor and medial wall in 4 mm, and a three-wall decompression in 7 mm. These decompressive effects often improve after 6 months of follow-up. 14 These results are similar to those reported by other surgeons. 8 Single-wall decompression surgery and the resultant herniation of orbital tissue into a single sinus cavity may give rise to asymmetric displacement of the orbital contents; we prefer to "balance" the decompression by removing two walls at least. IS The surgical techniques used
Lyons et al . Decompression in Dysthyroid Orbitopathy
o Figure 2. This 34-year-old woman had a 4-year history of thyroid orbitopathy whose onset coincided with hyperthyroidism. She was stable for 8 months. Results of ocular examination showed orthophoria with exophthalmos of 27 mm in the left eye and 26 mm in the right eye and interpalpebral fissure of 18 mm in the right eye and 20 mm in the left (A and B). She underwent bilateral three-wall decompression via a swinging eyelid approach. Three months after decompression, she underwent bilateral mullerectomy. The final result, obtained 8 months after decompression, showed exophthalmometry of 19 mm bilaterally, orthophoria, and an interpalpebral fissure of 11 mm in the right eye and 12 mm in the left (e and D).
in this series of patients were described previously elsewhere. 1 The subciliary, swinging eyelid or medial canthal approaches were used to allow a controlled and directly observed removal of the medial and inferior walls of the orbit. Using these, release of the orbital contents into the
antro-ethmoidal complex was achieved with minimally visible skin incisions. The lower lid or inferior fornix approach to decompression allows internal blepharoplasty to be performed at the time of decompression which may further improve the cosmetic result.
227
Ophthalmology
Volume 101, Number 2, February 1994
Figure 3. This 29-year-old man had exophthalmos secondary to hyperthyroidism that had been stable for 8 months. On physical examination, he was orthophoric with exophthalmometry of 23 mm in the right eye and 24 mm in the left (A). He underwent bilateral medial wall and floor decompression. Four months after surgery, he was orthophoric with exophthalmometry of 21 mm bilaterally (B).
To date, there has been no case-controlled prospective study to evaluate results with each type of surgery. Rather than advocate specific procedures, we would emphasize the need to become familiar with the approach chosen and to individualize care according to the above principles. As the surgeon's skill improves, the choice of decompression may change and also may be combined with bony advancement and augmentation. Patients undergoing decompression for compressive optic neuropathy or corneal exposure often have severe apical crowding. This group of patients has a high prevalence of preoperative restriction of ductions and diplopia. 14 Because this was a major indication for decompression in earlier series, severe motility problems frequently were present preoperatively and even more prevalent postoperatively. The perception that, in experienced hands, surgical decompression is a safe operation has been accompanied by a gradual change in its indication in thyroid orbitopathy; patients are undergoing this surgery earlier and have a lower prevalence of preoperative ocular motility problems. In DeSanto'sl7 series of 200 patients who underwent decompression by the Ogura technique between 1969 and
228
1976, the incidence for decompression was cosmesis or steroid intolerance in 16.5% of the patients and the incidence of diplopia was 53% preoperatively, rising to 80% postoperatively. More recently, Shorr et al 6 reported a series of decompressions performed using the same technique, in which 60% of the patients were undergoing surgery for cosmetic indications. Diplopia was present in 32% of patients preoperatively, increasing to 50% postoperatively. They found that patients with restricted ocular motility preoperatively were more likely to have postoperative diplopia. Their findings are confirmed by this series: postoperative diplopia developed in only 1 of 18 patients with full ductions preoperatively, whereas 12 of the 16 patients with preoperative restrictions of gaze had diplopia postoperatively. In addition to a change in patient selection, changes in decompression techniques have contributed to the reduction in operative complications. Whereas Hurwitz and Birt 18 found a worsening of motility in 16 of 19 orbits decompressed via the trans-antral approach, McCord8 reported a 5.6% incidence of worsened muscle balance after orbital decompression via the trans-lid approach.
Lyons et al . Decompression in Dysthyroid Orbitopathy
Figure 4. This 33-year-old man had stable proptosis for 2 years and exophthalmometry of 26 mm in the right eye and 27 mm in the left (A). He underwent bilateral two-wall decompression. Six months after decompression, he was orthophoric with exophthalmometry of 21 mm bilaterally (B).
In our series, diplopia was present preoperatively in the primary or reading positions in 7 (21 %) of the 34 patients. Of these patients, two became asymptomatic within 1 year of the decompression surgery. One patient's diplopia, due to a IO-prism diopter hypotropia, was managed with prisms only. Three patients underwent one operation each which was sufficient to treat their symptoms in the primary and reading positions. One patient with a severe preoperative myopathy needed three strabismus operations to be eliminate diplopia in the primary position and downgaze. Of the 27 patients who did not complain of diplopia preoperatively, diplopia developed in 5 (18%). This persisted beyond 1 year in four (14%) patients. Eighteen percent of the patients who did not complain of diplopia preoperatively did so after decompression. Overall, 24% of our patients required postdecompression strabismus surgery. This incidence compares favorably with earlier series in which decompression was carried out via the maxillary antrum or in the presence of greater muscle involvement by the dysthyroid process. Where diplopia did occur de novo after decompression, it resolved spontaneously or was cured by one strabismus surgery (one patient was lost to follow-up). We agree with the previous authors 6 that
orbital decompression per se does not complicate the management of strabismus. The improvement reported by three patients who became asymptomatic after 1 year postoperatively, and the fourth whose diplopia was controlled with prisms only, may be due to symmetric involvement of the inferior recti, reducing the angle of the vertical strabismus. It is also important to note that these patients had a reduction in the size of their deviation after their decompression, a phenomenon that has been reported previously in a small proportion of patients. 5,8, 19 A-pattern esotropia and torsional diplopia are complications of decompression surgery that have been reported previously.20,21 This is due to increased mechanical advantage of the superior oblique muscle after retroplacement of the globe, or decreased inferior oblique action after inferonasal orbital decompression. Neither of these was found in this group, although torsion was not specifically measured. It is likely that these patterns are more prevalent in patients who have undergone large inferior rectus recession before orbital decompression. None of our patients had previous strabismus surgery (because this was an exclusion criterion for this study). In some cases, the cosmetic defect of thyroid orbitopathy can be remedied by correcting lid retraction only
229
Ophthalmology
Volume 101, Number 2, February 1994
without resorting to a potentially risky decompression. This has not always been our experience; five patients in this series had undergone levator recessions for cosmetic purposes before decompression became necessary to correct their disfigurement. Our overall results showed that total lid fissure height was unaffected by decompressive surgery. An improvement of lower lid retraction and worsening of upper lid retraction after decompression has been reported previousli,g but was noted in only a few of our patients; this was no more likely in patients who had had decompression via a lower lid incision than in those whose decompression was performed via a Lynch incision. Twenty-nine (85%) of the 34 patients in our series had 1 mm or less difference in the proptosis of their two eyes after orbital decompression. The relation between preoperative and postoperative symmetry of the proptosis of the two eyes is illustrated graphically in Figure 1. The cosmetic improvement achieved with orbital decompression IS illustrated in Figures 2, 3 and 4. Patients commonly report that their disfigurement from thyroid orbitopathy is a significant social and psychologic handicap. We have found surgical decompression to be reasonably safe and effective in reducing cosmetically disfiguring exophthalmos and improving cosmesis. In our view, cosmesis is a valid indication for this form of surgery in carefully selected and informed patients. We emphasize that this article represents an evolving view point that is governed by increasing experience and comfort with offering orbital decompression to patients who have major cosmetic concerns. In some respects, the risk associated with orbital decompression has been over-emphasized, and this procedure may have been withheld in circumstances in which patients could have been helped. Although we would not encourage a cavalier attitude toward what is a potentially sight-threatening procedure, a wellcontrolled approach by an experienced surgeon should be considered when dealing with the disfigurement of thyroid orbitopathy.
References 1. Ogura JH, Lucente FE. Surgical results of orbital decompression for malignant exophthalmos. Laryngoscope 1974;84:637-44. 2. Long JC, Ellis PP. Total unilateral visual loss following orbital surgery. Am J Ophthalmol 1971 ;71 :218-20. 3. Young JDH. Ocular complications of transantral decompression for thyrotrophic exophthalmos. Proc R Soc Med 1971 ;64:929-31.
230
4. DeSanto LW, Gorman CA. Selection of patients and choice of operation for orbital decompression in Graves ophthalmopathy. Laryngoscope 1973;83:945-59. 5. Linberg JV, Anderson RL. Transorbital decompression: indications and results. Arch Ophthalmol 1981 ;99: 113-19. 6. Shorr N, Neuhaus RW, Baylis HI. Ocular motility problems after orbital decompression for dysthyroid ophthalmopathy. Ophthalmology 1982;89:323-8. 7. Apers RC, Oosterhuis JA, Sedee E, Bierlaagh J. Results of transantral decompression in hyperthyroid exophthalmos. Doc Ophthalmol 1977;44:207-14. 8. McCord CD Jr. Current trends in orbital decompression. Ophthalmology 1985;92:21-33. 9. Shorr N, SeiffSR. The four stages of surgical rehabilitation of the patient with dysthyroid ophthalmopathy. Ophthalmology 1986;93:476-83. 10. Dollinger J. Die Druckentlastung der Augenhohle durch Entfernung der au/3eren Orbitalwand bei hochgradigen Exophthalmus (Morbus Basedowii) und konsekutiver Hornhauterkrankung. Dtsch Med W ochenschr 1911;37: 1888-90. 11. Maroon JC, Kennerdell JS. Radical orbital decompression for severe dysthyroid exophthalmos. J Neurosurg 1982;56: 260-6. 12. Olivari N. Transpalpebral decompression of endocrine ophthalmopathy (Graves' disease) by removal of intraorbital fat: experience with 147 operations over 5 years. Plast Reconstruct Surg 1991;87:627-43. 13. Mourits MP, Koornneef L, Wiersinga WM, et al. Orbital decompression for Graves' ophthalmopathy by inferomedial, by inferomedial plus lateral, and by coronal approach. Ophthalmology 1990;97:636-41. 14. Satorre J, Rootman J. Des compresion orbitaria en la orbitopatia distiroidea. Resultados preliminares. Arch Soc Esp Oftalmol 1991 ;60:721-8. 15. Leone CR Jr, Piest KL, Newman RJ. Medial and lateral wall decompression for thyroid ophthalmopathy. Am J Ophthalmol 1989; 108: 160-6. 16. Rootman J. Diseases of the Orbit: A Multidisciplinary Approach. Philadelphia: JB Lippincott, 1988. 17. DeSanto LW. The total rehabilitation of Graves' ophthalmopathy. Laryngoscope 1980;90: 1652-78. 18. Hurwitz JJ, Birt D. An individualized approach to orbital decompression in Graves' orbitopathy. Arch Ophthalmol 1985; 103:660-5. 19. Streeten DHP, Anderson GH Jr, Reed GF, Woo P. Prevalence, natural history and surgical treatment of exophthalmos. Clin Endocrinol 1987;27: 125-33. 20. Fells P. Orbital decompression for severe dysthyroid eye disease. Br J Ophthalmol 1987;71:107-11. 21. Garrity JA, Saggau DD, Gorman CA, et al. Torsional diplopia after transantral orbital decompression and extraocular muscle surgery associated with Graves' orbitopathy. Am J Ophthalmol 1992;113:363-73.