“Balanced” Orbital Decompression for Severe Graves' Orbitopathy: Technique with Treatment Algorithm

Abstract

BACKGROUND: Graves' orbitopathy can produce proptosis, compressive optic neuropathy, and extraocular motility abnormalities; symptoms result from an increase in orbital volume due to expansion of intraorbital fat, with or without extraocular muscle involvement.

STUDY DESIGN: We conducted a chart review of patients who underwent orbital decompression to treat Graves' orbitopathy. Twenty-two orbits (13 patients) underwent orbital bone decompression, of which 17 orbits (9 patients) underwent the combined medial and lateral “balanced” decompression.

RESULTS: All patients had significant improvement with an average decrease in proptosis of 5.9 mm in the balanced decompression group. Restoration of normal optic nerve function was achieved in all patients with compressive optic neuropathy. Diplopia was noted in 4 patients (30.7%) preoperatively. Two patients had new postoperative diplopia (15.35%).

CONCLUSION: In our experience, “balanced” decompression results in a reduction of proptosis and improved optic nerve function and has a low incidence of complications.

Graves' orbitopathy is the most common extrathyroidal manifestation of Graves' disease. 1 The syndrome may produce eyelid retraction, proptosis, chemosis, diplopia, exposure keratopathy, and optic neuropathy. 2 Although the etiology is presumed to be autoimmune, the pathogenesis is not entirely understood. 3–5 Graves' orbitopathy is characterized by a variable increase in the volume of extraocular muscles and orbital connective tissue and fat (Table 1). Most patients with Graves' disease have mild ocular involvement, which benefits from supportive medical therapy and resolves spontaneously over 6 months to 1 year. In more severe forms of orbital disease, treatment planning is more complex. Corticosteroids with or without orbital radiotherapy are the mainstays of therapy in the 3% to 5% of patients who have the severe active phase of the disease. Surgery is reserved for cases of compressive optic neuropathy or severe proptosis if medical treatment fails or is contraindicated. For Graves' orbitopathy in the stable inactive phase, orbital decompression is the treatment of choice for persistent severe proptosis 6 (Fig 1).

MATERIALS AND METHODS

The charts of 410 patients who underwent treatment for Graves' orbitopathy over a 3-year period (1998 to 2000) at New York Presbyterian Hospital were reviewed; 85 orbital decompressions were performed in 45 patients. Bony decompression was performed in 22 orbits, and fat decompression was performed in 63 orbits. Indications for surgery, preoperative and postoperative Hertel measurements, optic nerve function, orthoptic measurements, and complications were analyzed (Table 2). Patients underwent either (1) combined endoscopic medial endonasal and lateral internal “balanced” decompression (Fig 2), (2) fat and 1 orbital wall (medial or lateral) compression, or (3) fat decompression only. The choice of surgery depended on the clinical findings and orbital pathology. Indications for balanced decompression procedure included compressive optic neuropathy and severe proptosis with enlarged extraocular muscle volume.

Table 1.

Assessment of severity of Graves' orbitopathy

Severity	Proptosis (mm)∗	Diplopia†	Optic neuropathy
Mild	4–5	Intermittent	Subclinical
Moderate	6–8	Inconstant	Decreased visual acuity 8/10 to 5/10
Severe	>8	Constant	Decreased visual acuity, <5/10

∗

Proptosis: change from baseline by expophthalmometry or CT/MRI measurement.

†

Diplopia: intermittent, present when fatigued; inconstant, present in secondary gaze positions; constant, present in primary gaze.

Modified from Bartalena L, Pinchera A, Marcocci C. Management of Graves' ophthalmopathy: reality and perspectives. Endocr Rev 2000;21:168–199.

All surgery was performed via a joint otolaryngology and ophthalmology team approach by the senior authors.

Procedure

Endoscopic medial endonasal decompression. The nose and the middle meatus were topically vasoconstricted with two 1/2–inch-wide cottonoid strips saturated with 0.05% oxymetazolin solution. The middle turbinate was endoscopically medialized and an uncinectomy was performed. A complete endoscopic ethmoidectomy was then performed with suction Blakesley forceps. The anatomic limits of the dissection were the frontal sinus ostium anteriorly, sphenoid sinus posteriorly, fovea ethmoidalis superiorly, lamina papyracea laterally, and middle turbinate medially. A frontal sinusotomy was done to ensure patency of the ostium. The sphenoid sinus was then entered using a sphenoid punch to create a 4 × 4-mm opening. Nasal antral windows were fashioned by cannulating the natural ostium of the maxillary sinus and removing the entire medial wall of the sinus with straight suction Blakesley and backbiting forceps. The window's borders formed by were the inferior turbinate, the posterior wall of the maxillary sinus, and the orbital floor. An orbital decompression was performed by removing the entire medial wall of the orbit, extending to the floor of the orbit as far lateral as the bony canal of the infraorbital nerve, which was left intact. The bone at the lamina papyracea was fractured with a Cottle elevator. The fragments were then removed with the straight suction Blakesley forceps and backbiting forceps, keeping the septum orbitale intact. The posterior and anterior limits of this bone were the sphenoid sinus and the frontal sinus ostium, respectively. In the superior and inferior dimensions, the boundaries were the fovea ethmoidalis and hard bony junction of the medial and inferior orbital walls. The medial aspect of the floor of the orbit was then fractured from the junction of the medial and inferior orbital walls with the Cottle elevator. Bone excision continued laterally to the infraorbital nerve bony canal and from the sphenoid sinus to the agger nasi cell anteriorly. Bone fragments were removed as described above. Orbital contents herniated into the ethmoid and maxillary sinuses, thereby decreasing proptosis and pressure within the orbit.

In cases requiring periorbital incision, a 4–0 suture was placed beneath the tendon of the medial rectus muscle to facilitate anterolateral retraction. This maneuver tented the septum orbitale adjacent to the medial rectus muscle and prevented injury of the muscle by the endoscopic surgeon (Fig 3). The septum orbitale was then incised above and below the muscle with a fine curved scalpel, from posterior to anterior to avoid obstruction of the operative field by the herniated fat.

Lateral internal decompression. After completion of the intranasal endoscopic surgery, the patient was re-prepped with povidone-iodine (Betadine) solution. A sterile contact lens was placed in the eye. A lateral canthotomy as well as superior and inferior cantholysis was performed with a No. 15 blade and Stevens scissors. The subcutaneous tissue was undermined with Bovie cautery. With a malleable retractor and 2 vein retractors under the superior and inferior skin flaps, a periosteal incision was made with the Bovie instrument. A Freer elevator was used to reflect the periosteum nasally, and bipolar cautery was used to achieve hemostasis. With the malleable retractor in place, the lateral orbital rim was thinned with a cutting burr. The bone resection was carried out posteriorly and inferiorly to the level of the infraorbital canal and superiorly to the level of the lacrimal fossa. Laterally, the bone was resected to the lateral orbital periosteum underlying the temporalis muscle. The temporalis muscle was not exposed to prevent it from filling the orbital space, thereby decreasing the orbital volume. The wound was copiously irrigated to remove any residual bony fragments. The periosteum was closed with a 4–0 monofilament (Biosyn; US Surgical, Norwalk, CT) absorbable suture. The lateral canthal angle was recreated with 4–0 monofilament absorbable suture, which was then attached to the reapproximated periosteum. The deep tissues were reapproximated with 4–0 Biosyn and 5–0 chromic, and a running 6–0 nylon was used for skin closure.

Fig. 1.

Management algorithm for Graves' orbitopathy. EOM, Extraocular muscle hypertrophy; +, mild; ++, moderate; +++, severe.

Table 2.

Indications for decompression

Indication	Surgical approach
Fat only, moderate EOM, moderate Proptosis	Fat
Asymmetric proptosis, moderate EOM, severe proptosis	Lateral or medial approach with fat
Optic neuropathy, severe EOM, severe proptosis	Combined medial and lateral approach

EOM, Extraocular muscle hypertrophy.

RESULTS

Twenty-two orbits in 13 patients underwent bony orbital decompression. Eleven patients were women and 2 were men (mean age, 50.3 years; age range, 26 to 70 years). Indications for decompression were symmetric proptosis in 4 patients, asymmetric proptosis in 4 patients, and optic neuropathy in 5 patients. Hertel measurements of the orbits revealed asymmetric proptosis in 4 patients and symmetric proptosis secondary to severe muscle hypertrophy in 9 patients.

Nine patients (17 orbits) underwent balanced decompression; 4 patients (5 orbits) underwent fat removal and orbital wall decompression on either the medial or lateral wall only (3 lateral and 1 medial-bilateral orbits). The reduction of proptosis achieved with fat and lateral or medial internal decompression was an average of 3.5 mm (range, 3 to 4 mm). With combined endoscopic medial endonasal and lateral internal decompression, the average reduction was 5.9 mm (range, 3 to 10 mm). In all 5 patients, preoperative optic neuropathy was reversed.

Fig. 2.

(A) “Balanced” medial and lateral decompression. (B) Orbit showing extent of medial and lateral bone removal (in dark gray; arrow marks the infraorbital nerve).

Four patients (30.7%) had preoperative diplopia. Two patients (30.7%) had persistent diplopia (2 of 13, 15.35%), 2 patients had a worsening of diplopia (2 of 13, 15.35%), and 2 patients without preoperative diplopia had new-onset postoperative diplopia (2 of 13, 15.35%) (Table 3). Eight patients underwent lid or extraocular muscle surgery after orbital decompression, to correct diplopia.

DISCUSSION

Graves' orbitopathy can be divided into an acute (active) phase and a chronic (stable) phase. Patients in the active phase can have severe ocular involvement characterized by compressive optic neuropathy. The treatment options include corticosteroids, orbital external beam irradiation, or surgical decompression. In our treatment algorithm, surgical decompression in the acute phase is reserved for patients for whom medical treatment fails or cases in which there is a medical contraindication to these modalities. Less commonly advocated therapies include somatostatin analogs and intravenous immunoglobulins; however, neither of these has proven efficacy. Patients enter the chronic (stable) phase after an average of 1 to 3 years. Mild orbital involvement requires only supportive measures. More severe chronic orbital involvement may require orbital decompression, followed by extraocular muscle or eyelid surgery.

Fig 3.

(A) Stitch in medial rectus tendon for anterolateral retraction to protect the muscle. (B) Stitch tenting the septum orbitale adjacent to the medial rectus muscle; septum orbitale is then incised above and below the muscle (arrows).

Table 3.

Incidence of preoperative and postoperative diplopia, muscle, and lid surgery

	Incidence, n/13
Preoperative diplopia	4 (30.7%)
Postoperative persistent diplopia	2 (15.35%)
Postoperative worsening of diplopia	2 (15.35%)
Postoperative new-onset diplopia	2 (15.35%)
Postoperative muscle surgery	6 (46.05%)
Lid surgery	2 (15.35%)

At our institution the surgical management is tailored to the anatomic cause of proptosis (fat compartment versus extraocular muscle hypertrophy), the severity of Graves' orbitopathy, and/or the presence of asymmetry between eyes. 7,8 High-resolution computed tomography or magnetic resonance imaging was performed in all of our patients to define the orbital pathology (Fig 4). In our series, 63 orbits in 32 patients underwent decompression of fat only. Patients with asymmetric proptosis of >5 mm underwent orbital wall and fat removal. Patients with severe proptosis or optic neuropathy underwent combined endoscopic medial endonasal and lateral internal decompression to achieve a balanced decompression (Fig 5).

There are several advantages to combined endoscopic medial endonasal and lateral internal decompression. “Balanced” decompression leads to maximal decrease in proptosis and reversal of optic neuropathy when present (average decompression of 5.9 mm in our series). Furthermore, balanced decompression may reduce the incidence of postoperative strabismus and minimizes the need for rehabilitative eye muscle surgery because it does not lead to an inferomedial displacement of the globe (Fig 6). In contrast, solely endoscopic medial and inferior endonasal decompression or transantral decompression has high rates of postoperative diplopia, approaching 66% in some series. 9 It has been noted in the literature that this traditional endoscopic approach produces an asymmetric decompression, with a shift of the globe medially and inferiorly, which may contribute to the higher incidence of new or worsened diplopia. 10 The average reduction of proptosis has been reported as 4.7 mm in the transantral technique group. 9 In studies evaluating only medial endoscopic techniques, the average decrease in proptosis ranged from 3.79 to 4.4 mm. 10–13 Kennedy et al 12 in 1990 reported an average reduction of 4.7 mm; however, in 2 of the 5 orbits studied, the endoscopic approach was coupled with Caldwell-Luc antrostomies. The reduction in proptosis is improved with the addition of lateral orbital approaches 14 (Table 4).

Fig. 4.

Computed tomography scan (axial and coronal) showing fat and muscle hypertrophy.

Fig. 5.

(A) Preoperative visual field defect at 40 mg/d prednisone. (B) Postoperative visual field with improvement after decompression at 0 mg prednisone.

Shepard et al 15 in 1998 published their data on “balanced” orbital decompression in 11 patients. The mean reduction of proptosis in their series was 4.6 mm. Two patients had new-onset diplopia (18.2%), and 2 patients required lid or muscle surgery even though 7 patients had persistent or new-onset diplopia. 15 In contrast, our series had a 15.35% incidence of new-onset diplopia, and 8 patients (6 muscle surgeries and 2 lid surgeries) required surgery to correct diplopia or eyelid abnormalities.

Both combined endoscopic medial endonasal and lateral internal decompression and transantral decompression have the advantage of no visible external scar; and each offers maximal access to the orbital apex. 13

A variation of the endoscopic medial and lateral internal decompression or transantral decompression is the transconjunctival or transcaruncular approach. 16 This variation enables removal of the medial and inferior walls with a low incidence of new-onset postoperative diplopia (range, 6% to 41%). However; decompression of the posterior ethmoids and orbital apex is more difficult. A 3-wall orbital decompression can be achieved by combining a transantral approach with a lateral translid approach or a coronal approach. 17 This approach has a lower incidence of diplopia than transantral approach alone. The disadvantages of this approach include less effective decompression of the orbital apex and possible injury to the infraorbital nerve. Kalmann et al 17 achieved an average reduction of proptosis by 4.3 mm using this technique.

Fig. 6.

A composite picture showing preoperative Hertel measurements of right eye (24 mm) and left eye (25 mm) (A), postoperative Hertel measurements of right eye (18 mm) and left eye (18 mm) (B) after lid and muscle surgery (C).

Table 4.

Surgical approach and reduction of proptosis

Surgical approach	Mean reduction in proptosis (mm)
Combined medial and lateral∗	5.9
Combined medial and lateral†	4.6
Transantral‡	4.7
Three wall decompression§	4.3

∗

Our series.

†

Shepard KG, Levin PS, Terris DJ. Balanced orbital decompression for Graves' ophthalmopathy. Laryngoscope 1998;108(11 Pt 1):1648–1653.

‡

Garrity JA, Fatourechi V, Bergstralh EJ, et al. Results of transantral orbital decompression in 428 patients with severe Graves' ophthalmopathy. Am J Ophthalmol 1993;116:533–547.

Kalmann R, Mourits MP, van der Pol JP, et al. Coronal approach for rehabilitative orbital decompression in Graves' ophthalmopathy. Br J Ophthalmol 1997;81:41–45.

Fig. 7.

Prolapsed orbital contents with patent maxillary and sphenoid ostium.

A 4-wall orbital decompression is reserved for the most severe cases of proptosis (>35 to 40 mm). Although this approach produces the greatest reduction in proptosis, severe complications have been reported in the literature, including intracranial hemorrhage, damage to frontal lobes, meningitis, and pulsatile proptosis. 18,19

Complications are not an infrequent occurrence, with a reported rate of 21% for the conventional transantral approach and 10% for coronal 3-wall decompression. These rates excluded diplopia. 20 In our study, there were no early (bleeding/orbital hematoma) or delayed (sinusitis) complications. The absence of delayed complication of sinusitis may be attributed to the fact that we perform large maxillary, frontal, and sphenoid sinusotomies, which are still patent after the orbital contents have prolapsed into the ethmoid bowl (Fig 7).

CONCLUSION

A balanced orbital decompression can be performed by the combined endoscopic medial endonasal and lateral internal decompression. This technique allows for maximal decompression, including the orbital apex, and has a low incidence of new-onset diplopia. Inferior support is maintained and the incidence of injury to the medial rectus muscle is minimized. The average reduction of proptosis in our series was 5.9 mm. This is the preferred approach of orbital decompression in patients with severe proptosis with optic neuropathy at our institution.

References

Burch

Wartofsky

Graves' ophthalmopathy: Current concepts regarding pathogenesis and management. Endocr Rev 1993;14:747–93.

Weisman

Osguthorpe

Orbital decompression in Graves' disease. Arch Otolaryngol Head Neck Surg 1994;120:831–4.

Bahn

Heufelder

Pathogenesis of Graves' ophthalmopathy. N Engl J Med 1993;329:1468–75.

Gorman

Pathogenesis of Graves' ophthalmopathy. Thyroid 1994;4:379–83.

Heufelder

Pathogenesis of Graves' ophthalmopathy: Recent controversies and progress. Eur J Endocrinol 1995;132:532–41.

Bartalena

Pinchera

Marcocci

Management of Graves' ophthalmopathy: Reality and perspectives. Endocr Rev 2000;21:168–99.

Kazim

Trokel

Acaroglu

. Reversal of dysthyroid optic neuropathy following orbital fat decompression. Br J Ophthalmol 2000;84:600–5.

Trokel

Kazim

Moore

Orbital fat removal. Decompression for Graves orbitopathy. Ophthalmology 1993; 100:674–82.

Garrity

Fatourechi

Bergstralh

. Results of transantral orbital decompression in 428 patients with severe Graves' ophthalmopathy. Am J Ophthalmol 1993;116:533–47.

10.

Lund

Larkin

Fells

, et al Orbital decompression for thyroid eye disease: A comparison of external and endoscopic techniques. J Laryngol Otol 1997;111:1051–5.

11.

Asaria

RHY

Koay

Elston

. Endoscopic orbital decompression for thyroid eye disease. Eye 1998;12(Pt 6):990–5.

12.

Kennedy

Goodstein

Miller

. Endoscopic transantral orbital decompression. Arch Otolaryngol Head Neck Surg 1990;116:275–82.

13.

Michel

Oberlander

Neugebauer

. Follow-up of transnasal orbital decompression in severe Graves' ophthalmopathy. Ophthalmology 2001;108:400–4.

14.

Metson

Dallow

Shore

Endoscopic orbital decompression. Laryngoscope 1994;104:950–7.

15.

Shepard

Levin

Terris

Balanced orbital decompression for Graves' ophthalmopathy. Laryngoscope 1998;108(11 Pt 1):1648–53.

16.

Shorr

Baylis

Goldberg

. Transcaruncular approach to the medial orbit and orbital apex. Ophthalmology 2000;107:1459–63.

17.

Kalmann

Mourits

van der Pol

. Coronal approach for rehabilitative orbital decompression in Graves' ophthalmopathy. Br J Ophthalmol 1997;81:41–5.

18.

Tallstedt

Surgical treatment of thyroid eye disease. Thyroid 1998;8:447–52.

19.

Kennerdell

Maroon

An orbital decompression for severe dysthyroid exophthalmos. Ophthalmology 1982;89:467–72.

20.

Hutchison

Kyle

Long-term visual outcome following orbital decompression for dysthyroid eye disease. Eye 1995;9(Pt 5):578–81.