INDICATIONS

Duane's syndrome is a condition of aberrant innervation that results in co-contraction of the medial and lateral rectus muscles in the affected eye. This aberrant innervation produces several manifestations, any one of which may be severe enough to warrant surgery in a given patient. These manifestations include a face turn with strabismus in primary position, an upshoot or downshoot during adduction, retraction during adduction, and enophthalmos. None of these findings is an absolute indication for surgery; the need for surgery depends on the severity of the manifestations and the degree to which the patient feels functionally compromised by the face turn or disfigured by the enophthalmos or bizarre eye movements.

An unacceptable face turn is by far the most common indication for surgical treatment. The face turn is a secondary manifestation of strabismus in primary position and develops to permit fusion. If it is sufficiently large, the face turn may be a functional handicap in addition to being disfiguring. Patients who have Duane's syndrome with exotropia in primary position usually have a face turn away from the affected eye. More commonly, an esodeviation in primary position leads to a face turn toward the side of the affected eye. This face turn usually is more pronounced with distant fixation. The face turn and measured deviation with near fixation may be less, but this is not a contraindication for surgery to correct the distant fixation deviation; overcorrection at near is uncommon.

An occasional Duane's syndrome patient may have bilateral involvement or a vertical deviation that precludes fusion with any head posture. In these cases, the strabismus itself rather than the secondary head posture can be the main indication for surgical correction, which should be performed at an earlier age than when fusion is present.

When the affected eye is adducted, an upshoot, downshoot, or retraction can be sufficiently disturbing to the patient or parents to warrant surgical treatment. In severe cases, there is significant enophthalmos and pseudoptosis, even in primary position. A reduction of 50% or more of the width of the palpebral fissure during adduction compared with primary position has been suggested as an indication for surgical treatment of the retraction.¹

Increasing the field of single binocular vision is desirable but not a primary goal in the treatment of Duane's syndrome. Patients with Duane's syndrome rarely complain of diplopia in the restricted field of gaze. Therefore, procedures designed to improve the field of single binocular vision should not be undertaken in Duane's syndrome at the expense of compromising other treatment objectives or increasing the risk of secondary complications.

CONTRAINDICATIONS

Many patients with Duane's syndrome are orthophoric in primary position or have only an insignificant face turn.^2,3 In these cases, surgery is not indicated unless another manifestation of Duane's syndrome, such as an upshoot or downshoot, is causing hardship. The fact that a patient seeks evaluation does not necessarily indicate a desire for surgical correction. Often, it is not the manifestations of the condition themselves that have motivated the consultation, but rather a concern that they indicate some ominous underlying intracranial disease, especially for parents who have suddenly noticed their child's eye movement abnormality. If discussion with the patient or the child's parents reveals that fear of life-threatening disease rather than a manifestation of Duane's syndrome itself is the chief concern, then explanation and reassurance may be all that is needed.

Because fusion usually can be obtained by means of a face turn, most children with Duane's syndrome have relatively normal binocular function and stereopsis. Therefore, in contradistinction to congenital esotropia—where the goal is to restore ocular alignment at as early an age as possible—the goal in Duane's syndrome should be to avoid disrupting normal binocular development. Thus, young age is a relative contraindication for surgery. Early treatment may be warranted if the face turn is very large, but usually it is preferable to delay surgery until 4 to 5 years of age. The visual system is relatively mature at this age and less susceptible to damage from a temporary disruption of binocularity, as can occur postoperatively if there is an unfavorable response to surgery. Also, older children can cooperate more adequately for the detailed examination necessary to appreciate the many subtle variations of Duane's syndrome for which the surgical plan may need to be adjusted.

While Duane's syndrome patients rarely complain of diplopia preoperatively, they often do not adapt easily to a postsurgical reduction of their field of single binocular vision to the contralateral side. Procedures should be planned to avoid producing significant new incomitance (exotropia in right gaze for a typical esotropic left Duane's syndrome patient), particularly in adults, who are more likely than children to be distressed by changes in the pattern of incomitance.

PROCEDURES

Because the manifestations of Duane's syndrome show great variability, no single surgical plan can be recommended. Rather, the choice of procedure or combination of procedures must be tailored to address the specific manifestations that are symptomatic in each patient.

Ipsilateral Medial Rectus Muscle Recession for Esotropic Duane's Syndrome

Recession of the ipsilateral medial rectus muscle is the mainstay of surgical treatment for Duane's syndrome.^1,4 This improves the face turn and the esotropia by relieving restriction and weakening the antagonist of the lateral rectus muscle. There also can be a modest improvement in abduction, but sometimes at the expense of reduced adduction. Medial rectus recession alone also may improve enophthalmos and vertical overshoots in adduction, in part by limiting adduction of the eye. However, additional measures usually are needed to deal adequately with these problems if they are severe.

It seems that indications for medial rectus muscle recession in sixth nerve palsy also should apply to Duane's syndrome; however, there are some important differences. In sixth nerve palsy, the absence of abduction suggests that medial rectus muscle recession is ineffective and a transposition procedure to provide some abduction force is necessary for any long-term improvement. Although preoperative abduction ability is a good prognostic sign in Duane's syndrome, medial rectus muscle recession still can be effective even in cases in which there appears to be no abduction at all. There are two reasons for this. First, medial rectus muscle restriction is frequently an important factor in Duane's syndrome; some abduction ability is unmasked when the restriction is relieved. When marked restriction is present, medial rectus muscle recession must be at least part of the surgical plan. Second, aberrant innervation of the lateral rectus muscle in Duane's syndrome often provides some tonic abducting force to counterbalance the weakened medial rectus muscle in primary position, even when there is no active contraction of the lateral rectus muscle on attempted abduction. Of course, absent any component of normal lateral rectus muscle innervation, a transposition is still needed to provide abduction much beyond midline; however, it should be remembered that the primary goal in sixth nerve palsy and Duane's syndrome are different. Good abduction with a large field of single binocular vision is needed for relief of diplopia in acquired sixth nerve palsy, but abduction to just beyond midline is often sufficient to eliminate the face turn in Duane's syndrome.

To effectively reduce or eliminate the face turn, a large recession—in the range of 8 to 10 mm as measured from the original insertion—often is needed. Standard recession techniques or a “hang-back” technique may be used, which can make the procedure easier to perform when a large recession is needed. Large recessions can cripple adduction and are contraindicated when there is strong co-contraction or restriction of the lateral rectus muscle, as indicated by exotropia in the contralateral field of gaze. When significant mechanical restriction owing to contracture or fibrosis of the medial rectus muscle is present, smaller recessions usually are sufficient.⁵ A large recession easily can cripple a fibrotic muscle, which often has contractile properties that are almost as severely compromised as its elastic properties. For adult patients, adjustable suture recession techniques are helpful for finding the best compromise between adequate relief of the face turn and crippled adduction.

Surgery on the Contralateral Eye

Recession of the contralateral medial rectus muscle in esotropic Duane's syndrome can reduce esotropia in primary position by reducing innervation to the ipsilateral medial rectus muscle and recruiting any minimal normal innervation of the ipsilateral lateral rectus muscle that may exist, without any detrimental effect on adduction of the Duane's syndrome eye. Possible improvement in the field of single binocular vision by matching the duction deficit of the ipsilateral eye provides another rationale for this procedure. A contralateral medial rectus muscle recession needs to be large (often larger than an ipsilateral recession) because it, in effect, is treating the much larger secondary deviation. However, contralateral medial rectus recession by itself, regardless of size, cannot correct a face turn if the ipsilateral eye is unable to abduct at least a little past midline. This should be tested prior to performing contralateral surgery. If the patient cannot abduct the ipsilateral eye well enough to comfortably maintain a straight head posture with the contralateral eye patched, contralateral eye surgery alone cannot correct the face turn. The adequacy of contralateral surgery also can be evaluated, with both eyes open, by testing whether the face turn can be eliminated by holding base-out prism in front of the contralateral eye. Even when not adequate by itself, contralateral medial rectus recession may allow a smaller recession of the ipsilateral medial rectus muscle to be effective. Recession of the contralateral medial rectus muscle is a particularly useful adjunct to ipsilateral medial rectus muscle recession when primary position esotropia exceeds 20 Δ or a large recession is contraindicated because of poor adduction or marked retraction.¹ Recession of the contralateral medial rectus muscle also may reduce the long-term risk of recurrent face turn owing to development of ipsilateral medial rectus muscle contracture by reducing its tonic innervation. Similar arguments have been made for recession of the contralateral lateral rectus muscle to improve the abnormal head posture and expand the field of single binocular vision in exotropic Duane's syndrome.⁶

Unlike other forms of incomitant strabismus, contralateral surgery in Duane's syndrome can interact with the aberrant innervation in unexpected ways. Particularly in patients with small-angle esotropia, the lateral rectus muscle may have significant tonic aberrant innervation in primary position that aids in holding the eye close to straight. Contralateral medial rectus muscle recession reduces innervation to the ipsilateral medial rectus muscle but also reduces the constructive aberrant innervation to the ipsilateral lateral rectus muscle in primary position.⁷ The proportion of aberrant and normal innervation to the lateral and medial rectus muscles determines the net change in balance between adducting and abducting forces in a specific patient. Usually, the reduction in aberrant lateral rectus muscle innervation does not quite fully negate the reduction in medial rectus muscle innervation in primary position; however, even a paradoxical worsening of the face turn has been described in some patients.⁸ Conversely, recessing the contralateral lateral rectus muscle in exotropic Duane's syndrome increases innervation to the ipsilateral medial rectus muscle and consequently the aberrant innervation to the ipsilateral lateral rectus muscle. Again, the results depend upon how the balance between abducting and adducting forces is changed.

Posterior fixation of the contralateral medial rectus muscle has been advocated as a way of equalizing rotations of the two eyes and improving the field of binocular single vision.⁶ It can have applications in esotropic and exotropic Duane's syndrome, whenever there is limitation of abduction in the Duane's syndrome eye preoperatively or expected after the planned surgery. Unlike recession, posterior fixation does not reduce the constructive primary position aberrant innervation to the ipsilateral lateral rectus muscle; however, it is unlikely to have much effect unless there is at least some component of normal innervation to the ipsilateral lateral rectus muscle.

Recess-Resect Procedure

Resection procedures generally are regarded as contraindicated in Duane's syndrome because of concern for making the retraction, adduction deficit, enophthalmos, or upshoot and downshoot worse. However, a small recess-resect procedure on the eye with Duane's syndrome can provide correction of the head posture comparable to that from a single large medial rectus recession, but with less limitation of adduction and better abduction in favorable cases.^9,10 To minimize the risk of making the retraction significantly worse, this approach should only be considered in eyes with modest signs of co-contraction and even then resection of the lateral rectus muscle should be limited to less than 4 mm.

Vertical Rectus Muscle Transposition

In esotropic Duane's syndrome, transposition of the vertical rectus muscles to a position adjacent to the lateral rectus, with or without recession of the ipsilateral medial rectus muscle, has been suggested as a means of correcting the primary position esotropia, improving abduction and enlarging the field of single binocular vision.^11–13 This procedure may provide better abduction than medial rectus recession alone. However, it is more difficult to perform; may exacerbate retraction, upshoot, or downshoot; and can create new vertical deviations.¹¹ Transposition is best suited for cases in which abduction is severely limited and there is no pre-existing vertical deviation. If there are significant signs of co-contraction, transposition should be done in conjunction with a procedure to weaken or eliminate lateral rectus muscle function (e.g., a large recession or disinsertion with terminal myectomy or fixation to the lateral orbital periosteum).

Particularly in adult patients on whom a medial rectus recession is performed concurrently, there is some risk of anterior segment ischemia.¹⁴ If medial rectus muscle restriction is absent or mild, transposition with lateral posterior augmentation sutures can reduce the likelihood that associated medial rectus muscle surgery will be needed.¹⁵ However, augmented transposition procedures produce more limitation of adduction, increase the risk of a secondary vertical deviation, and anterior segment ischemia can still occur.¹⁶ The risk of anterior segment ischemia can be further reduced by performing an augmented partial tendon transposition that spares one ciliary vessel in each vertical rectus muscle.¹⁷

Lateral Rectus Muscle Recession for Exotropic Duane's Syndrome

An exodeviation in primary position with a face turn away from the side of the affected eye is a less common presentation of Duane's syndrome. In these cases, the face turn usually is treated with recession of the ipsilateral lateral rectus muscle. In patients who have both primary position exotropia and a marked upshoot or downshoot, a lateral rectus recession usually is combined with an additional measure to minimize sideslip of the lateral rectus muscle across the globe—either a Y-splitting procedure or a posterior fixation suture. A large recession of the lateral rectus muscle also provides some improvement in the retraction with adduction. As with medial rectus muscle recessions, recessions of the lateral rectus muscles should be smaller when there is significant restriction.

Bilateral Lateral Rectus Muscle Shift for V-Pattern

Dramatic abduction of both eyes in elevation can occur in what is probably a variant of Duane's syndrome with aberrant innervation of the lateral rectus muscles by fibers from the superior branch of the oculomotor nerves.¹⁸ A similar picture can develop in bilateral cases of more typical Duane's syndrome after bilateral medial rectus muscle recessions. It is treated by small recession and full tendon width up-shift of the lateral rectus muscles bilaterally.

Ipsilateral Recessions of Both the Medial and Lateral Rectus Muscles

Enophthalmos in Duane's syndrome is occasionally severe enough to warrant specific remediation. Large recessions of both the medial and lateral rectus muscles can reduce enophthalmos as well as upshoot, downshoot, and retraction in adduction.¹⁹ The medial rectus muscle can be recessed 5.5 to 6.5 mm and the lateral rectus muscle can be recessed 7 to 8 mm, with some adjustment for any primary position heterotropia.²⁰

Y-Splitting Procedure

Co-contraction of the medial and lateral rectus muscles on attempted adduction can cause a striking upshoot or downshoot of the eye. This effect is attributed to sideslip of a tight lateral rectus muscle over the globe (mechanical factors) in most cases. The Y-splitting procedure effectively results in a broad lateral rectus muscle insertion that stabilizes its position and prevents it from flipping superiorly or inferiorly over the globe, thus eliminating or greatly reducing the upshoot or downshoot of the affected eye in attempted adduction.²¹ The Y-splitting procedure may be combined with a moderate recession if there is associated primary position exotropia.

  Step 1. The lateral rectus muscle is isolated through either a limbal (Fig. 1) or fornix incision. The muscle is dissected free of intermuscular septum and check ligaments for a distance of approximately 15 mm posterior to its insertion. When a fornix approach is used, the procedure can be performed through a single inferior temporal incision; however, making a second superior temporal fornix incision facilitates reattachment of the superior arm of the muscle.

Fig. 1. The lateral rectus muscle is isolated through a limbal incision and dissected free of intermuscular septum and check ligaments for a distance of approximately 15 mm posterior to its insertion.

  Step 2. A Stevens hook is brought from beneath the muscle to perforate it at its midpoint. Then, the hook is used to split the muscle into equal halves for a distance of 15 mm posterior to its insertion (Fig. 2). Blunt Westcott scissors may be used to complete the ends of the split, particularly at the muscle insertion (Fig. 3).

Fig. 2. A Stevens hook is brought from beneath the muscle to perforate it at its midpoint. Then, the hook is used to split the muscle into equal halves for a distance of 15 mm posterior to its insertion.

Fig. 3. Blunt Westcott scissors may be used to complete the ends of the split, particularly at the muscle insertion.

  Step 3. Two double-armed 6-0 Vicryl sutures are used to secure each half of the muscle separately (Fig. 4). Both halves of the muscle are disinserted from the sclera (Fig. 5).

Fig. 4. Two double-armed 6-0 Vicryl sutures are used to secure each half of the muscle separately.

Fig. 5. Both halves of the muscle are disinserted from the sclera.

  Step 4. The two halves of the muscle are separated into a Y configuration and reattached to the sclera as shown in Figure 6A. The inferior edge of the upper arm of the Y should be attached at the superior border of the original muscle insertion, and the superior edge of the inferior half of the muscle should be attached at the inferior border of the previous muscle insertion. If a recession is combined with the Y-splitting procedure, then these insertion points are relocated the desired amount posterior to the superior and inferior poles of the original insertion (Fig. 6B).

Fig. 6. The two halves of the muscle are separated into a Y configuration and reattached to the sclera. A. The inferior edge of the upper arm of the Y should be attached at the superior border of the original muscle insertion, and the superior edge of the inferior half of the muscle should be attached at the inferior border of the previous muscle insertion. B. If a recession is combined with the Y-splitting procedure, then these insertion points are relocated the desired amount posterior to the superior and inferior poles of the original insertion.

Lateral Rectus Muscle Posterior Fixation Suture

A posterior fixation suture on the lateral rectus muscle can effectively prevent slippage of the muscle belly over the globe. It may be used as an alternative procedure to treat upshoots and downshoots.²² As with the Y-splitting procedure, a posterior fixation suture can be combined with a lateral rectus recession when appropriate.

Vertical Rectus and Inferior Oblique Muscle Weakening Procedures

Besides the mechanical factors discussed, innervational factors, presumably aberrant co-contraction of the vertical rectus or inferior oblique muscles, may contribute to an upshoot or downshoot in some patients with Duane's syndrome.²³ When there is a significant vertical deviation in primary position, horizontal rectus muscle surgery alone generally does not correct the problem adequately, and recession of the appropriate vertical rectus muscle is needed.^1,24 The upshoot in adduction that is often seen in Duane's syndrome bears some resemblance to inferior oblique overaction, but inferior oblique weakening procedures usually are ineffectual in correcting the problem.^21,23,25

EXPECTED RESULTS

Horizontal muscle recession reportedly eliminates the face turn in 79% of cases and significantly reduces the face turn in most of the remaining cases.^4,26 In a patient with esotropic Duane's syndrome, recession of the ipsilateral medial rectus muscle usually provides a modest improvement in abduction; however, this improvement often comes at the expense of reduction of adduction. The field of single binocular vision is shifted to include primary position but remains relatively unchanged in size. Vertical rectus transposition procedures may provide better abduction and a larger field of single binocular vision, but no studies have directly compared the efficacy of transposition with unilateral or bilateral medial rectus recession.

When they are due to mechanical factors, upshoots and downshoots usually can be satisfactorily reduced or eliminated by the Y-splitting procedure, posterior fixation, or large recessions of both ipsilateral horizontal rectus muscles. On the other hand, when the upshoot or downshoot results from innervational factors, an appropriate recession of a vertical rectus muscle eliminates the vertical deviation in primary position, but some vertical deviation usually remains when the eye is adducted. Large recessions of both horizontal rectus muscles usually provide satisfactory correction of enophthalmos in primary position, but generally, some retraction in adduction remains.

COMPLICATIONS AND THEIR MANAGEMENT

Undercorrection of primary position esotropia and the face turn probably is the most common adverse outcome after surgical treatment of Duane's syndrome, especially when surgeons use amounts of recession typical of their experience in treating concomitant esotropia. Undercorrection may be obvious in the immediate postoperative period, or the face turn may reappear over a period of years after what initially appeared to be a good result. Late recurrence of the face turn usually results from the development of contracture of the ipsilateral medial rectus. Undercorrection can be managed by re-recession of the medial rectus muscle, if the initial recession was relatively small (less than 8 mm),²⁷ or by vertical rectus transposition. Re-recession of the medial rectus is necessary if passive duction testing still indicates restriction.

With the large recessions necessary for treatment of Duane's syndrome, overcorrection occurs occasionally.^28–30 Management of this secondary exotropia is often problematic and usually requires both advancement of the recessed medial rectus muscle and recession of the lateral rectus muscle, which is often found to be restricted on passive duction testing.

New vertical deviations may result from vertical rectus muscle transposition procedures. This condition is managed by dissection, which involves considerable scar tissue, and recession of the appropriate transposed vertical rectus muscle in the ipsilateral eye, particularly if vertical passive ductions are positive, or recession of the appropriate vertical rectus muscle in the contralateral eye.

True Brown's syndrome must be distinguished from other simulating conditions, such as inferior oblique palsy, monocular elevation deficiency (double elevator palsy), paratrochlear trauma or inflammation with resulting scarring and fibrosis, congenital periocular fibrous bands, orbital floor fracture with entrapment of orbital contents, inferior rectus fibrosis syndrome, mass effect from a glaucoma Seton implant,^34,35 and inferotemporal or posterior superonasal orbital fibrosis secondary to the fat adherence syndrome that is associated with rupture of Tenon's capsule and prolapse of fat during inferior oblique surgery. Exclusion of these simulating conditions (pseudo-Brown's syndrome) is important because most require a different management approach. The diagnosis of true Brown's syndrome requires: (a) the presence of typical clinical features described in the preceding; (b) the presence of mechanical restriction to elevation of the eye in adduction as demonstrated by forced duction testing either before or during surgery; and (c) complete resolution of all restriction after transection of the superior oblique tendon. Any patient with suspected Brown's syndrome who does not meet all three criteria should be evaluated thoroughly for alternative etiologies.

INDICATIONS

In true Brown's syndrome, surgical therapy is indicated for patients with abnormal head posture, a hyperdeviation in primary position, or diplopia. Most patients with Brown's syndrome have fusional ability; so abnormal head posture is the most common indication for surgery. The patient may have a chin-up posture to eliminate hypodeviation in primary position or a face turn away from the affected eye, which cannot adduct without depressing. The chronic head posturing can result in musculoskeletal problems, difficulty with balance (especially in young children), and cosmetic disfigurement. Diplopia is an uncommon complaint in children who generally suppress the non-fixating eye when a deviation is present, but it can be a problem in previously well-compensated adults who later decompensate when driving or when they become tired. Occasionally, adults with Brown's syndrome complain of vague ocular discomfort characterized as a pulling sensation or a sensation of ocular fatigue. However, by themselves, these symptoms rarely are severe enough to indicate a need for surgery. In addition, surgical correction sometimes should be considered for occupational reasons in patients whose occupations (e.g., librarian, airline pilot, painter, or professional basketball player) have significant upgaze requirements.

Although immediate surgery is not necessary, children with mild to moderate involvement without compensatory head posturing or vertical strabismus in primary position need close observation and careful follow-up for later development of the preceding surgical indications. Particular attention should be given to documentation of the visual acuity in each eye, the fusional and binocular status, and the presence of even mild or intermittent degrees of head posturing. Evidence of loss of binocularity, development of amblyopia, or loss of head posture suggests decompensation with the development of suppression and a need for immediate surgical correction.

CONTRAINDICATIONS

Many patients with Brown's syndrome have none of the preceding indications and do not require surgery. They may have only minimal symptoms or symptoms that are present only in extreme upgaze without head posture, or they may report only unusual disconjugate eye movements in upgaze. Children spend much of their time in upgaze looking at taller adults, which accentuates even mild to moderate head posturing or disconjugate eye movements. The marked secondary over-elevation of the unaffected eye during upgaze may be the feature most readily observed by parents, leading them to suspect an abnormality of the contralateral normal eye. As a small child grows, the need for upgaze decreases, and this sign becomes less noticeable.

True Brown's syndrome is congenital; therefore, patients with unusual features such as acute or acquired onset, the presence of pain or active inflammation around the trochlea, or signs of any of the simulating conditions listed in the preceding, require further evaluation before surgery is considered. Some forms of acquired Brown's syndrome, such as those with associated pain, other signs of inflammation, or intermittency, respond to local corticosteroid injection. Other forms may be an early manifestation of orbital tumors or localized subperiosteal abscesses, and radiologic or sonographic imaging is needed for proper diagnosis and treatment. In patients with acute traumatic Brown's syndrome, which may occur after blunt orbital trauma or in canine tooth syndrome, surgery should be deferred for at least 4 to 6 months to determine whether spontaneous recovery occurs.

PROCEDURES

Superior Oblique Forced Duction Testing

When any surgical procedure for suspected Brown's syndrome is performed, superior oblique forced duction testing is required both preoperatively and intraoperatively to confirm the diagnosis of true Brown's syndrome and assess the adequacy of the surgical treatment. A simple forced duction test is useful before surgery to confirm the diagnosis. The globe is grasped with a toothed forceps at the limbus in the inferotemporal quadrant. Taking care to maintain the globe in its proper anterior-posterior position, the surgeon attempts to adduct and elevate the eye into the superonasal quadrant. Tightness of the abnormally short superior oblique muscle–tendon complex is easily appreciated as the cause of the elevation limitation. Further confirmation of the diagnosis is accomplished by demonstrating relatively free range of motion into the superior temporal, inferior temporal, and inferior nasal quadrants. A more sensitive technique, described by Guyton,³⁶ is the exaggerated forced duction test. With this technique, the globe is grasped near the limbus with a pair of forceps, at 3 and 9 o'clock, and then maximally retro pulsed, extorted, elevated, and adducted to place the superior oblique tendon on maximum stretch. Its tension then can be felt by rocking the temporal forceps from inferior nasally to superior temporally. As this maneuver is performed, the hand holding the temporal forceps feels a tightening and loosening force as the superior oblique tendon flips over the surface of the globe. With practice, the degree of tightness of the superior oblique tendon can be graded from normal to markedly increased, as in severe Brown's syndrome. The exaggerated superior oblique forced duction test is valuable immediately after every superior oblique tenotomy or other lengthening procedure to verify that the entire tendon has been isolated and cut. The presence of any residual flip requires a thorough search for residual uncut tendon fibers. Alternatively, this residual action may indicate the presence of posterior fibrous bands that restrict the globe and must be sought and transected. It is likely that many cases of unsuccessful surgery to correct Brown's syndrome are related to incomplete transsection of all superior oblique tendon fibers, particularly the most posterior ones that can be especially difficult to isolate. The exaggerated superior oblique forced duction test should be performed immediately after tenotomy to prevent this problem.

Superior Oblique Weakening Procedures

The abnormally short; therefore, a tight, superior oblique muscle–tendon complex in Brown's syndrome can be lengthened (weakened) with a number of techniques. As with other extraocular muscles, a superior oblique tendon recession operation has been described.³⁷ Theoretically this procedure can be performed in a graded fashion, as shown in Figure 7, with larger amounts of recession for more severe restriction. In practice, however, several technical problems compromise the effectiveness and predictability of this operation. First, the broadly fanned, almost diaphanous superior oblique insertion must be narrowed at the time of recession so that it will hold a suture adequately. As shown in Figure 7, this narrowed insertion is recessed a measured amount from the original insertion (for small recessions) or reattached to the sclera a measured distance from another reference point, such as the nasal border of the superior rectus insertion, for larger recessions. Conversion of the normal broad posterior lateral insertion, whose proximal tendon slides easily over the nasal sclera, as the eye moves from elevation to depression, to a narrow anterior nasal insertion fixed to the sclera may convert the superior oblique from a strong depressor and abductor to a weak elevator and adductor. The alteration in the force vector thus created sometimes results in clinically significant limitation of depression in the operated eye.^32,36 This complication can be reduced by reattaching the recessed tendon more posteriorly on the globe.³⁶ Because of these problems, the recession procedure is not widely used.

Various other superior oblique weakening procedures have been proposed to produce a weakening effect without complete tenotomy, with its potential for late overcorrection and superior oblique palsy. Split tendon lengthening at the insertion of the superior oblique tendon has been described.³⁸ With this procedure, shown in Figure 8, the distal superior oblique tendon is separated into an anterior and a posterior half for a specified length using a small muscle hook. The tendon then is transected through the posterior half distally at the insertion and through the anterior half at the proximal end of the split (Fig. 8A). The distal posterior hemitendon is sutured to the proximal anterior hemitendon, as shown in Figure 8B, to lengthen the tendon by an amount equal to the length of the anterior hemitendon. However, the procedure requires extensive dissection of the intermuscular septum surrounding the distal tendon, and it is technically difficult to perform because of the wispy, diaphanous nature of the tendon in this region.³² Some strabismus surgeons have found that extensive scarring of the floppy distal tendon may make the procedure unpredictable in its initial effect and variable in its long-term stability. However, recent reports by Stolovitch and co-workers³⁹ and Bardorf and Baker⁴⁰ have suggested that this procedure can be used safely and effectively for lengthening the superior oblique in selected patients with Brown's syndrome.

Another procedure, the Z-tenotomy of the distal superior oblique tendon, is even less valuable as a superior oblique weakening procedure. The Z-tenotomy, useful in weakening rectus muscles, lengthens the tendon by adjacent two-thirds marginal tenotomies, as shown in Figure 9A, one from the anterior tendon border and the other from the posterior tendon border, that overlap in the center of the tendon. Theoretically, spreading of these adjacent tenotomies elongates and loosens the tight superior oblique tendon (Fig. 9B). However, unlike a Z-tenotomy performed in muscle tissue, there is little cross-linking force between longitudinal tendon fibers in the superior oblique tendon. Therefore, the lengthened tendon tends to pull apart, converting the Z-tenotomy into an inadvertent complete distal tenotomy. On the other hand, if neither marginal tenotomy transects the central longitudinal tendon fibers as shown in Figure 9C, no tendon lengthening or weakening effect results.

To reduce the risk of late overcorrection, two additional superior oblique weakening procedures have been studied: superior oblique weakening by a peripheral tenotomy near the tendon insertion and superior oblique weakening by posterior two-thirds or seven-eighths tenectomy at the insertion, as shown in Figure 10.⁴¹ However, in the authors' experience, these procedures (especially the posterior seven-eighths tenectomy), although sometimes effective in mild superior oblique overaction with A-pattern, often produced unreliable and sometimes unstable results, especially undercorrection or early recurrence.

Most strabismus surgeons now believe that the optimal surgical procedure for correcting true Brown's syndrome involves complete tenotomy or tenectomy of the superior oblique tendon nasal to the superior rectus muscle border, either in an uncontrolled fashion or with the use of a mechanical spacer to lengthen the cut ends of the tendon in a controlled fashion. In 1946, Berke⁴² described a technique for complete tenotomy of the superior oblique tendon nasal to the superior rectus muscle. He performed the procedure, which consisted of blindly sweeping a muscle hook 10 to 12 mm posteriorly through a superonasal conjunctival incision directly overlying the nasal portion of the tendon. The hook then was swept superiorly to catch the superior oblique tendon, but also surrounding Tenon's capsule and orbital fat, which were later lifted off of the hook to identify the tendon. A second report described three variations of this technique to control the weakening effect.⁴³ Berke recommended cutting the tendon just medial to the superior rectus muscle if a minimal effect was desired.⁴² If a larger effect was desired, he performed a small tenectomy, but left the surrounding “sheath” intact to allow some connection between the cut end of the tendon and the globe. Complete superior oblique palsy occurred if a portion of the tendon with its entire sheath was excised close to the trochlea. Although isolated reports of superior oblique tenotomy or tenectomy appeared after Berke's initial description,^44,45 Crawford was the first to report a large series comparing complete tenotomy by the Berke technique with Z-tenotomy, split tendon lengthening, and tenectomy. In his study, only tenotomy or tenectomy by the Berke technique was completely effective.³⁸ However, incomplete tenotomy or tenectomy, possibly owing to inadequate visualization of the tendon before it was isolated with a muscle hook, resulted in surgical failure on several occasions. Others also experienced this difficulty with the use of the Berke technique.⁴⁶

Parks^47,48 subsequently developed an alternate technique to address some of the technical problems he experienced with the Berke technique. A superior temporal fornix incision is made through conjunctiva and Tenon's capsule down to bare sclera, and the superior rectus is hooked through this incision. The superior rectus then is exposed by reflecting the intermuscular septum, Tenon's capsule, and conjunctiva over the toe of the large hook. The nasal border of the superior rectus is exposed by temporal traction on the muscle hook, securing the superior rectus insertion while the conjunctival wound is further displaced nasally with two Stevens muscle hooks or a Desmarres retractor. The superior oblique tendon is identified visually as a cord of parallel fibers seen coursing from beneath the nasal border of the superior rectus approximately 8 to 12 mm posterior to its insertion. This technique allows isolation of the superior oblique tendon under direct visualization. Once isolated in this way, tenotomy or tenectomy of the tendon can be completed with almost no violation of surrounding intermuscular septum or Tenon's capsule. Therefore, once the tendon is cut, the cut ends separate but maintain their normal anatomic relationships. Because they are still embedded in intermuscular septum and surrounding Tenon's tissue, they continue to provide a weakened but vectorially normal force on the globe, thereby reducing the incidence of consecutive superior oblique palsy.⁴⁹ At the completion of the procedure, all instruments are removed from the eye. The exaggerated forced duction test is repeated to verify that all superior oblique tendon fibers have been transected.

The advantages of the Parks technique over others described include: (a) direct visualization of the entire tendon before isolation with the muscle hook, thereby eliminating the riskier blind hooking of the tendon, with its increased risk of incomplete tendon isolation, violation of the orbital fat, or inadvertent transection of the superior rectus muscle;⁵⁰ (b) transection of the cordlike tendon nasal to the superior rectus, where the weakening effect is more predictable; and (c) transection with minimal or no dissection of surrounding intermuscular septum, thereby preventing uncontrolled scarring with reattachment of the superior oblique tendon to the sclera in an unwanted location, or alteration of the tendon's course, which might change its vector of action on the globe. A modification of the Park's technique, preferred by the authors, is described in the following.⁴⁹ Additional information about the superior oblique tenotomy or tenectomy technique is provided in another chapter.

Superior Oblique Tenotomy Performed on the Right Superior Oblique Tendon

  Step 1. With the eye depressed and adducted by the assistant, a superior-temporal, two-plane fornix incision is made approximately 8 mm posterior to the limbus. It is important to keep the incision parallel to the upper lid so that the upper lid will cover the closed wound at the end of the procedure.
  Step 2. The temporal border of the superior rectus is then isolated through the incision, first with a small Stevens muscle hook and then with a larger Green or Jameson muscle hook to provide adequate control of the globe.
  Step 3. Although traction is applied on the superior rectus with the large muscle hook to depress the globe, a Stevens hook held in the other hand is used to expose the superior rectus insertion by reflecting the intermuscular septum, Tenon's capsule, and conjunctiva over the toe of the large hook as shown in Figures 11A (before reflection) and 11B (after reflection).

Fig. 11. A. Traction on the superior rectus muscle with the large muscle hook is applied to depress the globe. B. A Stevens hook held in the other hand is used to expose the superior rectus insertion by reflecting the intermuscular septum, Tenon's capsule, and conjunctiva over the toe of the hook. (Del Monte MA, Archer SM: Atlas of Pediatric Ophthalmology and Strabismus Surgery. New York: Churchill Livingstone, 1993)

  Step 4. With the eye maximally depressed with the Green hook beneath the superior rectus, the superior surface of the superior rectus is exposed with two Stevens hooks, as shown in Figure 12. This maneuver exposes the broad, sheetlike white falciform check ligament that fuses to the superior rectus diagonally 8 to 12 mm posterior to its insertion. The ligament is opened centrally at its insertion with blunt Westcott scissors.

Fig. 12. With the eye maximally depressed with the Green hook beneath the superior rectus muscle, the superior surface of the muscle is exposed with two Stevens hooks. The broad, sheet-like white falciform check ligament is exposed. This ligament fuses to the superior rectus muscle diagonally 8 to 12 mm posterior to its insertion. The ligament is opened centrally at its insertion with blunt Westcott scissors. (Del Monte MA, Archer SM: Atlas of Pediatric Ophthalmology and Strabismus Surgery. New York: Churchill Livingstone, 1993)

  Step 5. Two Stevens hooks are used to enlarge the opening in the check ligament and expose the bare superior surface of the superior rectus muscle posteriorly. A Desmarres retractor is inserted through this opening to further retract Tenon's capsule exposing the nasal border of the superior rectus.
  Step 6. To improve visualization of the nasal border of the superior rectus, the Green muscle hook under the superior rectus is pulled temporally and the Desmarres retractor moved nasally. This maneuver exposes the untouched wispy nasal intermuscular septum that covers and encases the parallel fibers of the superior oblique tendon (Fig. 13A) as they pass beneath the superior rectus muscle. In some patients, exposure of the tendon can be improved at this time by removing the lid speculum, allowing greater retraction with the Desmarres retractor.

Fig. 13. A. The Green muscle hook under the superior rectus muscle is pulled temporally and a Desmarres retractor moved nasally to improve visualization of the nasal border of the muscle. This maneuver exposes the nasal intermuscular septum, which covers and encases the parallel fibers of the superior oblique tendon as they pass beneath the superior rectus muscle. B. A Stevens muscle hook, with its tip pointing nasally, is moved posteriorly along the nasal border of the superior rectus muscle until the posterior margin of the cordlike superior oblique tendon is clearly identified. The tip of the hook is rotated posteriorly behind the tendon and passed toward the sclera anteriorly beneath the superior oblique tendon. (Del Monte MA, Archer SM: Atlas of Pediatric Ophthalmology and Strabismus Surgery. New York: Churchill Livingstone, 1993)

  Step 7. A Stevens muscle hook, with its tip pointing nasally, is moved posteriorly along the nasal border of the superior rectus muscle until the posterior margin of the cordlike superior oblique tendon is clearly identified. The tip of the hook is then rotated posteriorly behind the tendon and passed toward the sclera and anteriorly beneath the superior oblique tendon, pushing the intermuscular septum behind the tendon to surround it completely, as shown in Figure 13B.
  Step 8. The two layers of intermuscular septum on the toe of the Stevens hook are opened anterior to the tendon with a blunt Westcott scissors, as shown in Figure 14A. A second Stevens hook is passed through the anterior opening in intermuscular septum and beneath the tendon from anterior to posterior. These two Stevens hooks, passing beneath the tendon in opposite directions, securely hold the tendon in place until the tenotomy is completed. The tenotomy is performed between them with blunt Westcott scissors, as shown in Figure 14B, with care taken not to damage adjacent intermuscular septum. The opposite pointing Stevens hooks hold the tendon securely until it is completely transected and the hooks are released.

Fig. 14. A. The two layers of intermuscular septum on the toe of the Stevens hook are opened anterior to the tendon with blunt Westcott scissors. B. A second Stevens hook is passed through the anterior opening of the intermuscular septum and beneath the tendon from anterior to posterior. The tenotomy is performed between them with blunt Westcott scissors. Care is taken to avoid damaging the adjacent intermuscular septum. C. The cut ends of the tendon are visualized as they lie separated within the small rent in the intermuscular septum nasal to the superior rectus muscle. (Del Monte MA, Archer SM: Atlas of Pediatric Ophthalmology and Strabismus Surgery. New York: Churchill Livingstone, 1993)

  Step 9. The cut ends of the tendon are visualized as they lie separated, exposed within the rent in the intermuscular septum nasal to the superior rectus muscle, as shown in Figure 14C. If the tendon is particularly tight, the nasal end may retract out of sight toward the trochlea, along its normal course through intermuscular septum.
  Step 10. All instruments are removed from the incision, and the superior temporal conjunctival wound is massaged closed into the fornix. The surgeon must then verify that the tenotomy is complete by repeating the exaggerated forced duction test described in the preceding. In a complete tenotomy, no superior oblique tightness should be felt. Any residual tension, even mild, suggests that the tenotomy is incomplete. In that case, the Green hook must be replaced beneath the superior rectus and a Desmarres retractor repositioned to expose the nasal border of the superior rectus muscle. A thorough visual and physical search for residual uncut posterior tendon fibers can then be made with a Stevens hook. Only when the forced ductions are completely negative can the surgeon be certain that complete tenotomy has been accomplished. This confirmation is important because even a few remaining uncut fibers will prevent any surgical weakening effect.

Superior Oblique Tendon Silicone Expander Procedure

Recently, Wright and associates^51–53 described a new technique for correcting superior oblique overaction and Brown's syndrome by performing a superior oblique tenotomy and inserting a measured spacer of #240 or #40 silicone retinal band securely between the cut ends of the tendon. The degree of lengthening can be controlled by the length of silicone band inserted, with a longer band inserted to correct a greater restriction. Forced duction testing after insertion will verify the proper weakening effect. The presumed advantage of the silicone expander technique over uncontrolled tenotomy is that the expander prevents the excessive lengthening of the tendon that can result in superior oblique palsy.⁵¹ In addition, the expander lengthens the tendon without altering the location or functional characteristics of the broadly fanned posterior insertion. The use of a silicone band has a theoretical advantage over the use of sutures or other flexible spacers because the silicone band retains its length after healing and prevent scar contraction from reducing the tendon separation and causing recurrence of the Brown's syndrome.⁵¹ When performed properly, with exposure of the tendon using the Parks superior oblique tenotomy technique and careful removal of the tendon from its capsule before transection, the silicone expander retracts into the tendon capsule at the end of the procedure without scarring surrounding tissues or underlying sclera.

Superior Oblique Silicone Tendon Expander Procedure Performed on the Right Superior Oblique Tendon

  Step 1. A Jameson hook is placed beneath the superior rectus muscle through a superior-temporal fornix incision, and the conjunctiva and Tenon's capsule are reflected over the superior rectus insertion with a Stevens hook. This maneuver exposes the distal superior rectus with the adherent broad sheetlike white check ligament that fuses to its superior surface diagonally 8 to 12 mm posterior to its insertion. The ligament is opened centrally at its insertion on the surface of the superior rectus with blunt Westcott scissors, as shown in Figure 15.

Fig. 15. A Jameson hook is placed beneath the superior rectus muscle through a superior temporal fornix incision. The conjunctiva and Tenon's capsule are reflected over the superior rectus insertion with a Stevens hook. This maneuver exposes the distal superior rectus muscle, with its adherent broad sheet like check ligament fusing to its superior surface diagonally 8 to 12 mm posterior to its insertion. The ligament is opened centrally at its insertion on the surface of the superior rectus muscle with blunt Westcott scissors.

  Step 2. Two Stevens hooks are used to enlarge the opening in the check ligament, exposing the bare superior surface of the superior rectus muscle posteriorly. A Desmarres retractor is inserted through this opening to further retract Tenon's capsule and expose the nasal border of the superior rectus muscle, allowing direct visualization of the underlying parallel cordlike fibers of the superior oblique tendon (Fig. 16). Visualization can be improved by removing the lid speculum and by moving the Jameson hook beneath the superior rectus temporally and the Desmarres retractor nasally, as shown in Figure 16.

Fig. 16. Two Stevens hooks are used to enlarge the opening in the check ligament, exposing the bare superior surface of the superior rectus muscle posteriorly. A Desmarres retractor is inserted through this opening to further attract Tenon's capsule and expose the nasal border of the superior rectus muscle, allowing direct visualization of the underlying parallel cordlike fibers of the superior oblique tendon (arrow). A small incision is made through the wispy capsule overlying the superior oblique tendon to bare the tendon fibers.

  Step 3. A small incision is made through the wispy capsule overlying the superior oblique tendon to bare the tendon fibers. It is important to avoid incising surrounding intermuscular septum, as shown in Figure 16.
  Step 4. The exposed superior oblique tendon is engaged with a Stevens hook through the small hole in the superior oblique tendon capsule (Fig. 17A), with care taken not to rupture the posterior capsule and expose bare sclera (Fig. 17B). Any remaining adherent fascia or tendon capsule is removed carefully with a 0.3 Castroviejo forceps, as shown.

Fig. 17. A. The exposed superior oblique tendon is engaged with a Stevens hook through the small hole in the superior oblique tendon capsule. B. The entire tendon is secured on the Stevens hook, where adherent fascia and intermuscular septum are removed with a 0.3-mm Castroviejo forceps.

  Step 5. A second Stevens hook is placed beneath the tendon, and the exposed segment is lifted out of its capsule. The tendon is secured with two double-armed 5-0 or 6-0 Mersilene (Ethicon, Somerville, NJ) sutures with spatulated needles or 6-0 Ethibond sutures with tapered vascular needles, one 2 to 3 mm nasal to the superior rectus border and the other 2 to 3 mm nasal to the first suture, as shown in Figure 18. Each suture is secured by a full-width, half-thickness pass through the tendon, with locking bites placed on each edge and both arms tied together in a square knot on the tendon surface, as shown in Figure 18.

Fig. 18. A second Stevens hook is placed beneath the tendon, and the exposed segment is lifted out of its capsule. The tendon is secured with two double-armed 5-0 or 6-0 Mersilene sutures with spatulated needles, one 1- to 3-mm nasal to the superior rectus muscle border and the other 2- to 3-mm nasal to the first suture. Each is secured by a full-tendon, half-thickness pass through the tendon, with locking bites placed on each edge and both arms tied together in a square knot on the tendon surface.

  Step 6. The tendon is transected between the two pre-placed sutures, as shown in Figure 19. The cut proximal tendon end will retract into its capsule, allowing inspection of the opened tendon capsule for any defects. Large defects discovered at this time can be closed with 6-0 Vicryl (Ethicon, Somerville, NJ) suture to prevent adherence of the tendon or expander to the sclera postoperatively.

Fig. 19. The superior oblique tendon is transected between the two preplaced sutures.

  Step 7. A segment of a 240 silicone retinal band that has been previously soaked in antibiotic solution is cut to the proper length with a #11 Beaver blade and surgical ruler, as shown in Figure 20. The proper length that is needed to release the restriction without compromise of superior oblique function can be determined by experience based on the amount of forced duction tightness. A 7- to 8-mm segment of band can be used initially in moderate to severe Brown's syndrome, with the proper length of expander verified by forced duction testing after insertion.

Fig. 20. A segment of #240 silicone retinal band is cut to the proper length with a #11 Beaver blade. The proper length can be determined by experience based on the tightness on forced duction testing. A 7- to 8-mm segment of band can be used initially in moderate to severe Brown's syndrome, with the proper length of expander verified by forced duction testing after insertion.

  Step 8. The double-armed Mersilene sutures from the cut ends of the tendon are then secured to the ends of the silicone band in mattress fashion, as shown in Figure 21. The sutures are pulled up to position the silicone band tightly against the cut ends of the tendon. The double-armed sutures are then tied together, and the suture ends trimmed (Fig. 22).

Fig. 21. The double-armed Mersilene sutures from the cut ends of the tendon are secured to the ends of the silicone band in mattress fashion.

Fig. 22. The sutures are pulled up to position the silicone band tightly against the ends of the tendon. Then, the double-armed sutures are tied together, and the suture ends trimmed.

  Step 9. When the sutures and expander are released, the proximal tendon retracts and frequently pulls most, if not all, of the expander into the intact proximal tendon capsule, as shown in Figure 23, leaving little, if any, expander exposed. The small opening in the tendon capsule is closed with 6-0 plain gut suture to completely isolate the cut tendon ends and silicone expander within the tendon capsule to prevent scarring to surrounding sclera, orbital fat, superior rectus muscle, or overlying conjunctiva. All instruments are removed, and the conjunctival incision is massaged closed in the superior temporal quadrant.

Fig. 23. When the sutures and expander are released, the proximal tendon retracts. Frequently, it pulls most, if not all, of the expander into the intact proximal tendon capsule, as shown. A short segment of expander is still exposed (arrow). Then, the small opening in the tendon capsule is carefully closed with 6-0 plain gut sutures to completely isolate the cut tendon ends and the expander within the tendon capsule.

  Step 10. Finally, standard (not exaggerated) superior oblique forced duction testing should be performed, as shown in Figure 24. If the superior oblique restriction has not been relieved, the silicone expander should be replaced with a longer one, and forced duction testing repeated.

Fig. 24. Standard (not exaggerated) superior oblique forced duction testing is performed. The globe is grasped near the limbus in the inferotemporal quadrant and rotated superiorly and nasally to verify that superior oblique restriction has been relieved. The inferior limbus should move well above an imaginary line connecting the medial and lateral canthus.

Suture Guarded Superior Oblique Tenotomy via the Temporal Approach

An alternative procedure to the superior oblique expander procedure is the suture guarded superior oblique tenotomy. This procedure shares many advantages with the silicone expander over uncontrolled tenotomy in that it is a graded procedure so the degree of tendon lengthening can be controlled by the length of the suture separating the cut ends of the tendon. In addition, a potential advantage of the suture-guarded tenotomy over the silicone spacer procedure is the smaller size of the suture material and the lower likelihood for extrusion or encapsulation in dense scar tissue.

The following steps outline the authors' approach to a temporal guarded tenotomy of the left superior oblique from the surgeon's view (feet toward the top and head toward the bottom):

  Step 1. After appropriate forced duction testing, the eye is depressed and adducted by the assistant and a superior temporal, two-plane fornix incision is made through conjunctiva and Tenon's capsule approximately 8 mm posterior to the limbus.
  Step 2. The superior rectus muscle is isolated through the incision first with a small Stevens hook, and then with a larger Green or Jameson hook.
  Step 3. While inferior traction is applied to the superior rectus with the large muscle hook, the lid speculum is removed and another Stevens hook is inserted into the wound to aid in exposure and visualization, exposing the parallel white fibers of the superior oblique tendon running beneath the superior rectus muscle, approximately 8 mm posterior to the superior rectus insertion (Fig. 25).

Fig. 25. Although inferior traction is applied to the superior rectus with the large muscle hook, the lid speculum is removed and another Stevens hook is inserted into the wound to aid in exposure and visualization, exposing the parallel white fibers of the superior oblique tendon running beneath the superior rectus muscle, approximately 8 mm posterior to the superior rectus insertion.

  Step 4. Once the superior oblique tendon is visualized, a second Stevens hook, with its tip pointed posteriorly, is used to engage the anterior fibers of the superior oblique tendon (Fig. 26). The engaged fibers on the Stevens hook then are pulled temporally, exposing the more fusiform portion of the tendon that normally lies nasal to the superior rectus. This maneuver allows direct visualization of the entire anterior and posterior borders of the tendon. Once this fusiform portion of the tendon is directly visualized, it can be safely isolated with a Stevens hook. The Stevens hook, with its tip nasally, is moved posteriorly along the temporal border of the superior rectus muscle until the posterior margin of the superior oblique tendon is identified. The tip of the hook then is rotated inferiorly behind the tendon, and then passed anteriorly beneath the tendon.

Fig. 26. Once the superior oblique tendon is visualized more fully with the help of another Stevens hook to pull the superior rectus nasally, a second Stevens hook, with its tip pointed posteriorly, is used to engage the anterior fibers of the superior oblique tendon.

  Step 5. A second Stevens hook is placed beneath the tendon, and the exposed segment is lifted out of the capsule (Fig. 27). The tendon is secured with two doubled armed 6-0 Ethibond sutures with tapered vascular needles, one 2 to 3 mm temporal to the temporal border of the superior rectus, and the other 2 to 3 mm temporal to the first. Each suture is secured by a full width, half-thickness pass through the tendon, with locking bites at either end (Fig. 28).

Fig. 27. With infraduction maintained by a Green Hook beneath the superior rectus, a second Stevens hook is placed beneath the tendon, and the exposed segment is lifted out of the capsule and away from the sclera.

Fig. 28. The tendon is secured with two doubled-armed 6-0 Ethibond sutures with tapered vascular needles, one 2 to 3 mm temporal to the temporal border of the superior rectus, and the other 2 to 3 mm temporal to the first. Each suture is secured by a full width, half-thickness pass through the tendon, with locking bites at either end.

  Step 6. The tendon is transected between the two pre-placed sutures (Fig. 29).

Fig. 29. The tendon is transected between the two preplaced sutures with a blunt Westcott scissors.

  Step 7. Calipers set at desired amount of tendon lengthening (initially try 7 to 8 mm for a moderate to severe Brown's syndrome) are used to measure this distance from the proximally cut end of the tendon along the double-armed suture. A straight or curved needle driver is used to cross-clamp the two ends of the suture at this measured point (Fig. 30).

Fig. 30. Calipers set at desired amount of tendon lengthening (initially try 7 to 8 mm for a moderate to severe Brown syndrome) are used to measure this distance from the proximally cut end of the tendon along the double-armed suture. A curved needle driver is used to clamp the two ends of the suture at this measured point.

  Step 8. The two double-armed sutures are then sutured together (superior arm to superior arm; inferior arm to inferior arm) at the point marked by the needle driver, taking care to bring the knots as close to the needle driver as possible (Fig. 31).

Fig. 31. The two double-armed sutures are then sutured together (superior arm to superior arm; inferior arm to inferior arm) at the point marked by the needle driver, taking care to bring the knots as close to the needle driver as possible.

  Step 9. The suture ends are cut, and the needle driver is released, allowing the cut tendon ends to separate by the measured suture length. When the needle driver is released, the proximal tendon retracts beneath the temporal border of the superior rectus.
  Step 10. All instruments are removed from the eye and standard forced duction testing is performed. If the superior oblique restriction has not been relieved, the sutures can be replaced with a longer measured segment of suture in the same manner.
  Step 11. Once adequate forced ductions are confirmed, the superior-temporal incision is massaged closed with a Stevens hook.

EXPECTED RESULTS

The goals of therapy for Brown's syndrome include: (a) elimination of abnormal head posturing, (b) normalization of ductions and versions, and (c) functional improvement in fusional status in important fields of gaze. Initial therapy, based on the incorrect assumption that Brown's syndrome was caused by a shortening or contracting of the sheath surrounding the superior oblique tendon, consisted of lysis or stripping of the sheath. The results of studies in which this technique was used were poor.^54,55 Brown⁵⁶ reported poor results in 36 patients treated with sheathotomy, sometimes associated with an ipsilateral inferior oblique tuck or a superior rectus recession. Only five of 36 showed significant improvement. Results reported by others were even more disappointing.^32,38

In 1955, Nutt⁴³ reported the first good result in treating a patient with Brown's syndrome using superior oblique tenectomy. However, it was not until 1976 that a large study by Crawford³⁸ showed the superiority of superior oblique tenotomy for correction of this condition. In his retrospective review of 30 patients, one underwent sheathotomy, nine underwent Z-tenotomy, two underwent split tendon lengthening, 16 underwent complete tenotomy, and two were treated with tenectomy. From this study, he concluded that tenotomy of the superior oblique tendon gave the best results, with nine of 16 patients obtaining normal or almost normal ocular movement. Eustis and associates,⁵⁷ in a retrospective review of 30 patients, confirmed these relatively good results, 26 of which had undergone either tenotomy or tenectomy. Again, division of the superior oblique tendon nasal to the superior rectus muscle was effective in eliminating the restriction of Brown's syndrome. However, the authors also reported a high incidence of postoperative decompensated superior oblique palsy. This condition was found in 14 (54%) patients, and 10 (38%) required further surgical treatment.

These studies and others clearly show that complete tenotomy or tenectomy of the superior oblique tendon is effective in reducing or eliminating the motility disturbance and anomalous head posture in true Brown's syndrome.^32,62 However, careful attention to proper detail in surgical technique, especially careful confirmation of the completeness of tenotomy with the use of exaggerated force duction test, is necessary to maximize success while minimizing the complication rate. Several studies have looked at the short- and long-term results of silicone expander for Brown's syndrome.^59,60,62 Seawright and co-workers⁵⁹ reported a subset of 13 patients with Brown's syndrome with preoperative hypotropia in primary position who underwent the superior oblique tendon expander procedure. The mean follow-up period for all patients in the study was 28 months. Thirty-eight percent (five patients) had complete resolution of vertical deviation if primary gaze, and 54% (seven patients) had residual but reduced vertical deviation in primary position (mean reduction from 11 to four prism diopters). No patient manifested superior oblique palsy at the last visit. Stager and co-workers⁶⁰ reported similarly favorable results with their long-term study on the silicone tendon expander. This retrospective study looked at 20 eyes of 19 consecutive patients who underwent silicone expander procedure for moderate to severe Brown's syndrome. A 6 to 10 mm expander was placed in all patients. One hundred percent of patients had resolution of their downshoot in adduction with some or full ability to elevate the eye in adduction. The complication rate was relatively low with 20% of patients requiring re-operation (consisting of inferior oblique surgery) for significant overcorrection. One patient extruded his silicone implant. Although there was a relatively high incident of residual Brown's syndrome postoperatively (13 of 19 patients), all were converted from a severe to mild Brown's syndrome, and the majority of these continued to improve with time (up to 3 years postoperatively). Wright⁶¹ also has reported excellent short- (14/15, 93%) and long-term (4/5, 80%) success using the superior oblique silicone expander technique.

No studies have looked at the long-term results of guarded tenotomy, or compared them to the results of the silicone expander technique. Such a study would be helpful, because its authors' opinion that the guarded tenotomy may offer many of the advantages of the silicone expander technique, without being as technically difficult, especially in children, and without the added risk of implant extrusion.

COMPLICATIONS AND THEIR MANAGEMENT

As with other types of strabismus surgery, the most frequent adverse results of the surgical treatment of true Brown's syndrome are under-corrections or over-corrections. Under corrections, with residual limitation of elevation in adduction, may be more common after modified tendon-weakening procedures such as Z-tenotomy of the tendon near its insertion, tenectomy of the posterior tendon at its insertion, and superior oblique recession or split tendon-lengthening procedures. However, under correction can also be seen after superior oblique tenotomy or tenectomy, generally because of uncut posterior tendon fibers that are missed at the time of surgery. This complication can be minimized by careful attention to surgical detail, isolation of the tendon under direct visualization, and especially, using Guyton's exaggerated forced duction test to confirm that a complete tenotomy has been performed. In addition, a patient sometimes is seen who has an apparently complete tenotomy (documented by exaggerated forced duction test) with an excellent initial result. Then, weeks or months later, the patient develops a recurrence of Brown's syndrome, presumably from reanastomosis or linking of the severed tendon ends with fibrous scarring and then secondary scar contraction. Therefore, patients must be warned of possible late recurrence of Brown's syndrome, even after an excellent initial surgical result.

Probably more common than under-correction is early or delayed over-correction, with development of a decompensated superior oblique palsy, head tilt, and inferior oblique overaction. The incidence of this complication has ranged from 20% to 85% in recent studies.^32,57,59,63 In these studies, 20% to 82% of patients required additional surgery for treatment of iatrogenic superior oblique palsy.^32,57 To reduce the incidence of superior oblique palsy and the need for secondary surgery, some surgeons combine an inferior oblique weakening procedure with the superior oblique tenotomy for Brown's syndrome. Parks and Eustis⁵⁸ reported the results of combining 14-mm recession of the inferior oblique with tenotomy of the superior oblique in 16 eyes of patients with Brown's syndrome. Good or excellent results were obtained in 94% of eyes, and further surgery was not needed. However, inferior oblique under action, with many features similar to under-correction of Brown's syndrome, was seen in 75% of eyes immediately after surgery. Although this elevation deficit in adduction improved over time, Parks and Eustis⁵⁸ recommended reducing the amount of inferior oblique recession from 14 to 10 mm.

Alternatively, a controlled tenotomy, such as the superior oblique silicone tendon expander operation, may yield as good an initial surgical result without later overcorrection.^57,60 Further controlled studies are needed to show superiority of this technique over the simpler tenotomy, with or without simultaneous inferior oblique weakening.

Care must be taken to monitor the binocular status of all patients before and after surgery for Brown's syndrome.⁶⁴, ⁶⁵ Although Sprunger co-worker⁶⁴ reported improvement in binocular vision with superior oblique weakening of various types, Eustis and colleagues⁶⁵ reported a loss of binocularity in 11% of 17 patients who underwent tendon surgery for Brown's syndrome. Altered alignment after surgery, especially in very young patients, may result in suppression with eventual loss of binocularity. Postponing correction in young infants with evidence of fusion may reduce this risk.

With respect to complications associated with the silicone expander, Wilson and co-workers⁶⁶ reported two cases of late down gaze restriction following the silicone tendon expander procedure, both which were secondary to adhesions between the superior oblique tendon and nasal border of the superior rectus muscle. Stager and co-workers⁶⁰ did not observe this complication in any of their cases, and advised the use of a narrower #40 silicone implant (rather than the #240) as well as careful attention to preservation of the inner layer of the intermuscular septum, as means of avoiding this problem.

Many other complications that have been reported in patients treated surgically for Brown's syndrome are related to poor surgical judgment or technique. Tendon isolation for superior oblique tenotomy by blindly sweeping 10 to 12 mm posteriorly through a superonasal fornix incision, as described by Berke,⁴¹ has been associated with paresis or inadvertent transection of the superior rectus,^45,49 permanent blepharoptosis, massive hemorrhage from rupture of a vortex vein, and rupture of Tenon's capsule with resultant prolapse of orbital fat and possible development of fat adherence syndrome. These complications should be eliminated by isolation of the superior oblique tendon by direct visualization, as described by Parks and Helveston.^46,47 With careful attention to surgical anatomy and technique, as described, superior oblique tenotomy, with or without inferior oblique recession, can safely and effectively eliminate the clinically significant motility disorder in Brown's syndrome, with minimal complications.

1. Kraft SP: A surgical approach for Duane syndrome. J Pediatr Ophthalmol Strabismus 25:119, 1988

2. Raab EL: Clinical features of Duane's syndrome. J Pediatr Ophthalmol Strabismus 23:64, 1966

3. Isenberg S, Urist MJ: Clinical observations in 101 consecutive patients with Duane's retraction syndrome. Am J Ophthalmol 84:419, 1977

4. Pressman SH, Scott WE: Surgical treatment of Duane's syndrome. Ophthalmology 93:29, 1986

5. Kubota N, Takahashi H, Hayashi T, Sakaue T, Maruo T: Outcome of surgery in 124 cases of Duane's retraction syndrome (DRS) treated by intraoperatively graduated recession of the medial rectus for esotropic DRS, and of the lateral rectus for exotropic DRS. Bin Vis Str Qty 16:15, 2001

6. Saunders RA, Wilson ME, Bluestein EC, Sinatra RB: Surgery in the normal eye in Duane retraction syndrome. J Pediatr Ophthalmol Strabismus 31:162, 1994

7. Archer SM: Contralateral medial rectus muscle recession in Duane syndrome. J AAPOS 8:296, 2004

8. Greenberg MF, Pollard ZF: Poor results after recession of both medial rectus muscles in unilateral small-angle Duane's syndrome, type I. J AAPOS 7:142, 2003

9. de Decker W: Kestenbaum transposition operation for treatment of the Duane I retraction syndrome. Trans Ophthalmol Soc UK 100:479, 1980

10. Morad Y, Kraft SP, Mims JL 3rd: Unilateral recession and resection in Duane syndrome. J AAPOS 5:158, 2001

11. Molarte AB, Rosenbaum AL: Vertical rectus muscle transposition surgery for Duane's syndrome. J Pediatr Ophthalmol Strabismus 27:171, 1990

12. Carlson MR, Jampolsky A: An adjustable transposition procedure for abduction deficiencies. Am J Ophthalmol 87:382, 1979

13. Gobin MH: Surgical management of Duane's syndrome. Br J Ophthalmol 58:301, 1974

14. Saunders RA, Phillips MS: Anterior segment ischemia after three rectus muscle surgery. Ophthalmology 95:533, 1988

15. Velez FG, Foster RS, Rosenbaum AL: Vertical rectus muscle augmented transposition in Duane syndrome. J AAPOS 5:105, 2001

16. Murdock TJ, Kushner BJ. Anterior segment ischemia after surgery on 2 vertical rectus muscles augmented with lateral fixation sutures. J AAPOS 5:323, 2001

17. Paysse EA, McCreery KM, Ross A, Coats DK: Use of augmented rectus muscle transposition surgery for complex strabismus. Ophthalmology 109:1309, 2002

18. Freedman HL, Kushner BJ: Congenital ocular aberrant innervation—new concepts. J Pediatr Ophthalmol Strabismus. 34:10, 1997

19. von Noorden GK: Recession of both horizontal rectus muscles in Duane's retraction syndrome with elevation and depression of the adducted eye. Am J Ophthalmol 114:311, 1992

20. Sprunger DT: Recession of both horizontal rectus muscles in Duane syndrome with globe retraction in primary position. J AAPOS 1:31, 1997

21. Rogers GL, Bremer DL: Surgical treatment of the upshoot and downshoot in Duane's retraction syndrome. Ophthalmology 91:1380, 1984

22. von Noorden GK, Murray E: Up- and downshoot in Duane's retraction syndrome. J Pediatr Ophthalmol Strabismus 23:212, 1986

23. Scott AB, Wong GY: Duane's syndrome: An electromyographic study. Arch Ophthalmol 87:140, 1972

24. Mohan K, Saroha V: Vertical rectus recession for the innervational upshoot and downshoot in Duane's retraction syndrome. J Pediatr Ophthalmol Strabismus 39:94, 2002

25. Pfaffenbach DD, Cross HE, Kearns TP: Congenital anomalies in Duane's retraction syndrome. Arch Ophthalmol 88:635, 1972

26. Kraft SP, O'Donoghue EP, Roarty JD: Improvement of compensatory head postures after strabismus surgery. Ophthalmology 99:1301, 1992

27. Biedner BZ: Medial rectus re-recession for unacceptable head-turn and esotropia in Duane's retraction syndrome. Ophthal Surg Lasers 28:762, 1997

28. Goldstein JH, Sacks DB: Bilateral Duane's syndrome. J Pediatr Ophthalmol 14:12, 1977

29. Nelson LB: Severe adduction deficiency following a large medial rectus recession in Duane's retraction syndrome. Arch Ophthalmol 104:859, 1986

30. Noel LP, Clarke WN: Adduction deficiency following medial recti recession in Duane's retraction syndrome: Case report. Arch Ophthalmol 105:465, 1987

31. Brown HW: Congenital structural muscle anomalies. In Strabismus ophthalmic symposium I. St. Louis: Mosby-Year Book, 1950:205

32. Parks MM: Surgery for Brown's syndrome. In Symposium on Strabismus. St. Louis: Mosby, 1978:157

33. Wilson ME, Eustis HS Jr, Parks MM: Brown's syndrome. Surv Ophthalmol 34:153, 1989

34. Coats DK, Paysse EA, Orenga-Nania S, Acquired Pseudo-Brown's syndrome immediately following Ahmed valve glaucoma implant. Ophthal Surg Lasers 30:396, 1999

35. Dobler-Dixon AA, Cantor LB, Sondhi N, Ku WS, Hoop J: Prospective evaluation of extraocular motility following double-plate Molteno implantation. Arch Ophthalmol 117:1155, 1999

36. Guyton DL: Exaggerated traction test for the oblique muscles. Ophthalmology 88:1035, 1981

37. Romano P, Roholt P: Measured graduated recession of the superior oblique muscle. J Pediatr Ophthalmol Strabismus 20:134, 1983

38. Crawford JS: Surgical treatment of true Brown's syndrome. Am J Ophthalmol 81:289, 1976

39. Stolovitch C, Leibovitch I, Loewenstein A: Long-term results of superior oblique tendon elongation for Brown's syndrome. J Pediatr Ophthalmol Strabismus 39:90, 2002

39A. Bardorf CM, Baker JD: The efficacy of superior oblique split Z-tendon lengthening for superior oblique overaction. Am Assoc Pediatr Ophthalmol Strabismus 7:96, 2003

40. Prieto-Diaz J: Posterior tenectomy of the superior oblique. J Pediatr Ophthalmol Strabismus 16:321, 1979

41. Berke RN: Tenotomy of the superior oblique for hypertropia: Preliminary report. Trans Am Ophthalmol Soc 44:304, 1946

42. Berke RN: Tenotomy of the superior oblique muscle for hypertropia. Arch Ophthalmol 38:605, 1947

43. Nutt AB: Observations on the aetiology and treatment of the vertical congenital ocular palsies. Br Orthoptic J 12:4, 1955

44. Jacobi KW: Tenektomie des Obliquus superior bei Sehnenscheiden syndrome (Brown). Klin Monatsbl Augenheilkd 160:669, 1972

45. McNeer KW: Untoward effects of superior oblique tenotomy. Ann Ophthalmol 4:747, 1972

46. Parks MM: Superior oblique tendon surgery. In Atlas of Strabismus Surgery. Philadelphia: Harper & Row, 1983:189

47. Parks MM, Helveston EM: Direct visualization of the superior oblique tendon. Arch Ophthalmol 84:491, 1970

48. Del Monte M, Archer S: Atlas of Pediatric Ophthalmology and Strabismus Surgery. New York: Churchill Livingstone, 1993:95

49. Raymond MW, Parks MM: Transection of the superior rectus muscle during intended superior oblique tenotomy: A report of three cases. Ophthal Surg 26:244, 1995

50. Wright KW: Superior oblique silicone expander for Brown syndrome and superior oblique overaction. J Pediatr Ophthalmol Strabismus 28:101, 1991

51. Wright KW: Superior oblique weakening procedures. In Color Atlas of Ophthalmic Surgery: Strabismus. Philadelphia: JB Lippincott, 1991:201

52. Wright KW, Min BM, Park C: Comparison of superior oblique tendon expander to superior oblique tenotomy for the management of superior oblique overaction and Brown syndrome. J Pediatr Ophthalmol Strabismus 29:92, 1992

53. Wright KW: Surgical management of superior oblique overaction and Brown's syndrome. Ophthalmol Clin North Am 5(1):67, 1992

54. Folk ER: Superior oblique tendon sheath syndrome. Arch Ophthalmol 57:39, 1957

55. Brown HW: True and simulated superior oblique tendon sheath syndromes. Doc Ophthalmol 34:123, 1973

56. Brown HW: True and simulated superior oblique tendon sheath syndromes. Aust J Ophthalmol 2:12, 1974

57. Eustis HS, O'Reilly C, Crawford JS: Management of superior oblique palsy after surgery for true Brown's syndrome. J Pediatr Ophthalmol Strabismus 24:10, 1987

58. Parks MM, Eustis HS: Simultaneous superior oblique tenotomy and inferior oblique recession in Brown's syndrome. Ophthalmology 94:1043, 1987

59. Seawright AA, Gole GA: Results of treatment of superior oblique overaction by silicone tendon-expander technique. J Pediatr Ophthalmol Strabismus 35:33, 1998

60. Stager DR, Parks MM, Stager Sr DR: Long-term results of silicone expander for moderate and severe Brown syndrome (Brown syndrome “plus”). J AAPOS 3(6):328, 1999

61. Wright KW. Brown's syndrome: diagnosis and management. Trans Am Ophthalmol Soc 97:1023, 1999

62. Clarke MP. Bray LC. Manners T: Superior oblique tendon expansion in the management of superior oblique dysfunction. Br J Ophthalmol 79:661, 1995

63. von Noorden GK, Olivier P: Superior oblique tenectomy in Brown's syndrome. Ophthalmology 89:303, 1982

64. Sprunger DT, von Noorden GK, Helveston, EM: Surgical results in Brown syndrome. J Pediatr Ophthalmol Strabismus 28:164, 1991

65. Eustis HS., O'Reilly C, Crawford JS: Management of superior oblique palsy after surgery for true Brown's syndrome. J Ophthalmic Nurs Technol 6:112, 1987

66. Wilson ME, Sinatra RB, Saunders RA: Downgaze restriction after placement of superior oblique tendon spacer for Brown syndrome. J Pediatr Ophthalmol Strabismus 32:29, 1995