Journal of Neuro-Ophthalmology, June 2003, Volume 23, Issue 2

Surgical Management of Skew Deviation

There are no published data on the outcomes of realignment surgery for skew deviation. A retrospective chart review disclosed 10 patients who had undergone surgical correction of skew deviation by three surgeons at a single institution between 1991 and 2002. Nine of 10 patients had satisfactory relief of diplopia with an acceptable field of single binocular vision. Vertical rectus recession or resection was the most common procedure. Four patients required more than one procedure. For nonalternating hypertropias, resection of the inferior rectus muscle or recession of the superior rectus muscle of the hypertropic eye was successful. For alternating hypertropia, resection of both inferior rectus muscles was successful. Oblique muscle surgery was not associated with good outcomes.

ORIGINAL CONTRIBUTION Surgical Management of Skew Deviation R. Michael Siatkowski, MD, Robert F. Sanke, MD, and Bradley K. Farris, MD Abstract: There are no published data on the outcomes of realignment surgery for skew deviation. A retrospective chart review disclosed 10 patients who had undergone sur­gical correction of skew deviation by three surgeons at a single institution between 1991 and 2002. Nine of 10 pa­tients had satisfactory relief of diplopia with an acceptable field of single binocular vision. Vertical rectus recession or resection was the most common procedure. Four patients required more than one procedure. For nonalternating hy-pertropias, resection of the inferior rectus muscle or reces­sion of the superior rectus muscle of the hypertropic eye was successful. For alternating hypertropia, resection of both inferior rectus muscles was successful. Oblique muscle surgery was not associated with good outcomes. ( JNeuro- Ophthalmol 2003; 23: 136- 141) Skew deviation is a vertical ocular misalignment result­ing from damage to supranuclear areas controlling eye movements ( 1,2). Varying amounts of torsion and the ocu­lar tilt reaction may be present ( 3,4). Prisms may be useful in alleviating diplopia in some cases, but generally only if the misalignment is small and fairly comitant, and torsional diplopia is minimal or absent. If prisms are not effective, extraocular muscle surgery may provide symptomatic im­provement. However, a review of the literature yields no studies evaluating the outcomes of surgical treatment of skew deviation. We performed a retrospective review of 10 cases of skew deviation in which surgery was performed at a single institution by three surgeons over the past decade. METHODS After appropriate University of Oklahoma Health Sciences Center Institutional Review Board approval, charts from 30 patients who were diagnosed with skew Dean A. McGee Eye Institute, Department of Ophthalmology, Univer­sity of Oklahoma College of Medicine, Oklahoma City, Oklahoma. Address correspondence to R. Michael Siatkowski, MD, Dean A. McGee Eye Institute, 608 Stanton L. Young Blvd., Oklahoma City, OK 73104, USA; E- mail: Rmichael- siatkowski@ ouhsc. edu Supported by an unrestricted grant from Research to Prevent Blind­ness, New York, New York. Presented in part as a poster at the 2002 North American Neuro- Ophfhalmology Society Annual Meeting at Copper Mountain, Colorado, February 9- 14, 2002. deviation between January 1991 and January 2002 were ret­rospectively reviewed. The diagnosis of skew deviation was made based on the presence of vertical diplopia follow­ing an ischemic, inflammatory, traumatic, degenerative, or neoplastic disorder affecting the supranuclear vertical eye movement centers in the brain stem. Cases with concomi­tant oculomotor or trochlear nerve paresis or orbital pathol­ogy ( n = 10) were excluded. Ten candidates were excluded because fusion could not be achieved in primary gaze with prism correction. Thus, surgical subjects had no preopera­tive evidence of symptomatic subjective torsional diplopia or central disruption of fusion. After exclusions, there were 10 patients who met study criteria ( Table 1). Their charts were reviewed for age and gender of the patient, initial stra­bismic measurements, cause of the misalignment, surgical procedure or procedures performed, final ocular alignment, and the patients' assessment of whether they were satisfied with the surgical results. All patients underwent surgery only after the mis­alignment had been stable for at least 6 months. Two au­thors performed procedures in nine patients ( BKF, six cases; RMS, three cases); a third surgeon operated on one patient ( patient # 3). The selection of the procedure was at the discretion of each physician. In general, amounts of ver­tical rectus surgery were based on achieving 2 to 3 prism diopters of realignment for each millimeter of muscle re­cession or resection. Adjustable sutures were used in three cases ( patients # 5, 7, and 8). Surgical success was defined as complying with at least one of the following three crite­ria: ( 1) elimination of diplopia in primary and reading gaze; ( 2) conversion of the preoperative misalignment to a smaller and more comitant postoperative misalignment such that prisms could achieve single vision in primary and reading positions; and ( 3) patient report of satisfaction on the following scale: very satisfied, satisfied, neutral, or sur­gical failure. RESULTS The age range for the 10 study patients was 31 to 83 years, with a mean of 57 years ( Table 1). Six patients were female. The causes of skew deviation were brain stem stroke ( n = 6), mesencephalic mass lesion ( n = 1), spino­cerebellar degeneration ( n = 1), lithium toxicity ( n = 1), and ethanol toxicity ( n = 1). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 136 J Neuro- Ophthalmol, Vol. 23, No. 2, 2003 SKEW DEVIATION JNeuro- Ophthalmol, Vol. 23, No. 2, 2003 Nine of the 10 patients had a successful surgical out­come, six after one procedure, two after two procedures, and one after three procedures. There was one surgical fail­ure, in a patient who underwent two procedures. Most patients (# 1- 7, 10) had nonalternating hyper-tropias; three of them (# 1, 5, 10) had relatively comitant misalignment, and five (# 2, 3, 4, 6, 7) had misalignment patterns that mimicked a unilateral inferior rectus paresis, with a nonalternating hypertropia worsening on ipsilateral gaze. Two patients (# 8, 9) had alternating hypertropias mimicking bilateral inferior rectus pareses. Surgical results for these three patterns are discussed separately. Relatively Comitant Nonalternating Hypertropia ( Patients # 1,5,10) Patient # 1 had a strictly comitant vertical deviation successfully treated with resection of the inferior rectus of the hypertropic eye. A small preoperative exodeviation dis­appeared after surgery, perhaps because of increased hori­zontal fusion ability after the vertical deviation had re­solved, or perhaps because of the improved tertiary adduc-tive effect of the strengthened inferior rectus. Inferior rectus resection was chosen over superior rectus recession be­cause the largest misalignment was in downgaze and ab­duction, the field of action of the inferior rectus. Patients # 5 and # 10 had large concomitant horizontal misalignments in addition to hypertropia. Patient # 5 had bi­lateral sixth nerve pareses, and patient # 10 had a comitant exotropia. Patient # 10 had resolution of the vertical devia­tion after a horizontal recess- resect procedure with superior transposition of the horizontal recti to correct the vertical misalignment. The transposition was performed to limit surgery to one eye and to avoid the potential of anterior segment ischemia. Patient # 5 required three procedures. The first con­sisted of a unilateral horizontal recess- resect procedure for esotropia, combined with an ipsilateral inferior oblique re­cession to correct 10 prism- diopters of hypertropia. The choice of inferior oblique surgery was made to attempt to achieve a successful result with surgery by operating on only one eye, and to avoid the potential of anterior segment ischemia. This resulted in a significant vertical overcorrec­tion, however, requiring two recessions of the contralateral superior rectus. Two additional procedures ( including chemodenervation of a medial rectus) were also required for the esotropia caused by sixth nerve pareses, but the final postoperative outcome was satisfactory. Nonalternating Hypertropia Mimicking Ipsilateral Inferior Rectus Paresis ( Patients # 2, 3, 4, 6, 7) Patients # 2, 4, 6, and 7 had successful results after a single surgery. Three patients had resections of the inferior rectus of the hypertropic eye, and one had a recession of the superior rectus of the hypertropic eye ( patient # 7). Resec­tions of the interior rectus were performed more frequently than superior rectus recessions because the misalignment was greatest in downgaze and abduction, and surgery on the muscle working in this field of action was desired. How­ever, both strategies had equal success. Patient # 3 had a postoperative vertical overcorrection after inferior rectus resection and also reported cyclotor-sional diplopia after the first procedure, with excyclotorsion of the hypertropic eye. Strengthening the anterior fibers of the superior oblique tendon was attempted for this but re­sulted in worsening vertical and torsional misalignment. It was speculated that the superior oblique tendon had slipped after the second procedure, but the patient refused further surgery. This case constituted the only surgical failure. Alternating Hypertropia Mimicking Bilateral Inferior Rectus Pareses ( Patients # 8, 9) Both patients (# 8, 9) initially underwent bilateral in­ferior rectus resections and each remained undercorrected with symptomatic vertical diplopia on side gaze after the first procedure. Patient # 8 had a second procedure consist­ing of bilateral superior rectus recessions, and patient # 9 had a re- resection of one inferior rectus muscle. Both had an improved field of single binocular vision and notable sub­jective improvement after the second surgery. Despite surgical manipulation of the vertical recti, no patients had a significant A- or V- pattern strabismus post­operatively. Additionally, none had subjective cosmetic or functional problems regarding lid position following sur­gery, presumably because of intraoperative dissection of the inferior rectus from the lower lid retractors and dissec­tion of the attachments of the superior rectus from the leva­tor palpebrae superioris. Adjustable sutures were used in three cases, twice during the first procedure ( patients # 5, 7) and once during a second procedure ( patient # 8). This technique was associ­ated with a satisfactory outcome in each case. Further details of surgical procedures and outcomes are presented in Table 1. DISCUSSION Vertical misalignment of the visual axes following damage to the supranuclear control areas for ocular move­ment is termed the Magendie- Hertwig syndrome, or skew deviation ( 1). Coordinated ocular movements are the result of inputs to the ocular motor nuclei from the vestibular sys­tem, brain stem, and cerebellum, with the vestibulo- ocular reflex playing a crucial role in these activities ( 5- 8). Skew deviation results from unilateral or asymmetric disruption Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 137 TABLE 1. Clinical features often patients who underwent ocular alignment surgery for skew deviation Patient Age/ gender Etiology Initial alignment First procedure 31/ M 53/ M 83/ F Cerebellar AVM Parinaud syndrome Stroke 12LHT 6 XT 12LHT 6 XT 12LHT 6 XT 20RHT 4ET 12RHT 2ET 2RHT OH 4LHT 6ET 6LHT 4ET 9LHT 6ET LIR Resect 5.0 RIR Resect 5.0 LIR Resect 3.5 55/ M Stroke 4LHT 4 XT 10LHT 5 XT IILHT 4 XT LIR Resect 5.0 67/ M 53/ F Stroke Spinocerebellar degeneration 10RHT 35 ET 10RHT 30 ET 16RHT 35 ET 3LHT 2ET 3LHT 2ET IILHT 8ET RMR Recess 4.0* RLR Resect 7.0 RIO Recess 4.0 LIR Resect 3.0 63/ F 33/ F Stroke Lithium toxicity 2LHT 4ET 10LHT 7ET 12LHT 10 ET 12RHT OH 3RHT 2ET 4LHT 1ET LSR Recess* RIR Resect 4.5 LIR Resect 2.0 10 77/ F 58/ F Stroke Stroke, alcohol abuse 5RHT 4LHT 6LHT 25 XT 4LHT 30 XT 8LHT 25 XT LIR Resect 3.5 RIR Resect 3.5 RLR Recess 7.0f RMR Resect 7.0f ( continues) Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 138 © 2003 Lippincott Williams & Wilkins TABLE 1. Continued Second procedure Third procedure Final alignment Patient satisfaction Harada- Ito procedure RSO 2LHT 2LHT 2LHT IRHT 8RHT 8ET 10RHT Satisfied Satisfied Unsatisfied 4 XT 2 XT 4 XT IRHT IXT Satisfied LSR Recess 5.0 LLR Resect 7.0 LLR Re- resect 5.0 LSR Recess 2.0 Botulinum toxin LMR 12 ET 2LHT 3LHT 2ET 10 ET Satisfied 2ET 2ET ILHT 2ET 3LHT 5ET Satisfied 2RHT 2ET 3RHT 6ET 2ET Very satisfied LSR Recess 5.0* RSR Recess 10.0* RIR Reresect 3.5 8RHT 6ET IRHT 4ET ILHT IRHT 2ET 2ET 4LHT 4ET ILHT Satisfied Very satisfied 2LHT 2ET 4 XT Satisfied Misalignment measured in prism diopters; extraocular muscle procedures measured in millimeters. * Adjustable suture; fsuperior transposition. A, aligned. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 139 JNeuro- Ophthalmol, Vol. 23, No. 2, 2003 Siatkowski et al. of the central graviceptive input in the roll plane as it as­cends from the utricles to the interstitial nucleus of Cajal in the mesencephalon ( 9,10). This nucleus maintains tonic in­nervation of motor neurons that supply the extraocular muscles controlling vertical and cyclotorsional move­ments. Skew deviation may be a component of the triad known as the ocular tilt reaction, which consists of vertical deviation, head tilt, and conjugate torsion of the globes to the same side ( 11- 14). Skew deviation may manifest as a comitant or fairly comitant nonalternating vertical misalignment with varying amounts of torsion. When isolated and not accompanied by significant horizontal misalignment, conventional vertical rectus muscle recession or resection using standard surgical doses ( 2- 3 prism diopters of realignment per millimeter of muscle recession or resection) sufficed in 9 of our 10 cases. Surgical planning is significantly complicated when large horizontal deviations are also present. We had one success with simple vertical transposition of the horizontal recti for a small hypertropia ( patient # 10), but in the au­thors' experience, this technique is not likely be effective in correcting vertical misalignments of greater than 8 to 10 prism- diopters. In the case with bilateral abducens pareses ( patient # 5), in which inferior oblique recession was per­formed, there was a large vertical overcorrection that we cannot fully explain. This occurrence, combined with the comparative success of vertical rectus muscle surgery, however, has significantly lessened our enthusiasm for ob­lique muscle surgery in the patterns of skew deviation dis­cussed in this paper. For patients with nonalternating, incomitant hyper-tropias mimicking a unilateral inferior rectus paresis, stan­dard- dose vertical rectus muscle surgery was generally ef­fective. We preferred inferior rectus resection to superior rectus recession because the deviation was greater in down-gaze, but in one case ( patient # 7), the surgical procedure was a superior rectus recession, after which the patient did well. Another potential advantage of inferior rectus resec­tion over superior rectus recession is the effect on postop­erative eyelid position. When the inferior rectus is dissected from the lower lid retractors and subsequently resected in the amounts used in our patients ( all 5 mm or less), there is no significant superior displacement of the lower eyelid. However, even with dissection of the attachments between the superior rectus and the levator, lid retraction may occur, and in a patient who also now has decreased upgaze, expo­sure keratopathy may result. For patients with alternating hypertropias mimicking bilateral inferior rectus pareses ( 15,16), bilateral vertical rectus muscle surgery is required. However, our two pa­tients with this pattern ( patients # 8, 9) were initially under-corrected and required a second procedure, even though the degree of hypertropia was not significantly different from that of the patients with nonalternating hypertropias ( pa­tients # 2,3,4,6, and 7), who generally did well with a single surgery. This may support augmentation of initial surgical doses for correction of this type of skew deviation. For skew deviation mimicking unilateral or bilateral inferior rectus paresis, one could argue that weakening of the superior oblique muscle of the hypotropic eye could be just as effective as inferior rectus resection of the hyper-tropic eye. In fact, Hamed ( 17) has argued that many pa­tients with A- pattern strabismus originally believed to be due to primary superior oblique overaction actually have skew deviation. However, superior oblique weakening pro­cedures such as tenotomy or tenectomy are less predictable and more difficult to quantify than inferior rectus surgery. They are also likely to produce symptomatic excyclotor-sional diplopia in patients who have no symptomatic pre­operative torsional complaints. In addition, superior ob­lique procedures are generally more technically difficult than inferior rectus procedures. Even the use of a superior oblique tendon expander, which may be more quantifiable than tenectomy or tenotomy, can be unpredictable and pro­duce restrictive motility deficits postoperatively. A weakness of this report is lack of consistent pre-and postoperative information on torsional status and head tilt. However, we selected for surgery only those patients who could fuse with prisms and had no symptomatic tor­sional diplopia preoperatively. When encountering patients with notable torsional complaints, it is likely that oblique muscle surgery would be required, as these muscles are the major cyclotorters of the eye. However, smaller amounts of torsion, perhaps up to 10 degrees, may be treated with ver­tical rectus muscle surgery, since inferior rectus resection or superior rectus recession will produce relative postopera­tive excyclotorsion, and inferior rectus recession and/ or superior rectus resection will produce relative postopera­tive incyclotorsion ( 18). In addition, with time, both tor­sional and vertical fusion amplitudes, as well as head tilt, may improve because of central nervous system adaptation ( 19). Nine of our patients were subjectively satisfied post­operatively, indicating that symptomatic torsional diplopia was not present. Our single clear surgical failure was, in our opinion, most likely due to a slipped muscle rather than misjudgment in patient or procedure selection. In general, we believe that vertical rectus muscle surgery is effective in patients with skew deviation who lack substantial torsional complaints. When diplopia was not significantly relieved with surgery alone, the misalignment had been reduced sufficiently and made more comitant such that small amounts of prism could alleviate diplopia in primary and reading positions. Although a 40% reoperation rate may seem high, this is comparable to a reoperation rate of approximately one third in infantile esotropia ( 20). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 140 © 2003 Lippincott Williams & Wilkins SKEW DEVIATION JNeuro- Ophthalmol, Vol. 23, No. 2, 2003 Skew deviation is frequently associated with other, often severely compromising, neurologic changes following damage to the vestibular, cerebellar, and brain stem areas. Improving the patient's vision to its maximal potential can be quite significant, particularly for patients with balance or coordination difficulties. For these patients, even partial improvement of diplopia can provide a major functional ad­vantage. Our experience shows that strabismus surgery can greatly improve the patient's ocular alignment and quality of life. We encourage physicians to consider such surgery in the management of appropriately selected patients with skew deviation. REFERENCES 1. Smith JL, David NJ, Klintworth G. Skew deviation. Neurology 1964; 14: 96- 105. 2. Keane JR. Ocular skew deviation. Analysis of 100 cases. Arch Neu­rol 1975; 32: 185- 90. 3. Galetta SL, Liu GT, Raps EC, et al. Cyclodeviation in skew devia­tion. Am J Ophthalmol 1994; 18: 509- 14. 4. Brandt T, Dieterich M. Skew deviation with ocular torsion: a ves­tibular brainstem sign of topographic diagnostic value. Ann Neurol 1993; 33: 528- 34. 5. Leigh RJ, Brandt T. A reevaluation of the vestibulo- ocular reflex: new ideas of its purpose, properties, neural substrate, and disorders. Neurology 1993; 43: 1288- 95. 6. Brandt T. Man in motion: historical and clinical aspects of vestib­ular function. A review. Brain 1991; 114: 2159- 74. 7. Liu GT, Volpe NJ, Galetta SL. Eye movement disorders: conjugate gaze abnormalities. In AJeuro- Ophthalmology: Diagnosis and Man­agement Philadelphia: WB Saunders, 2001: 584- 626. 8. Averbuch- Heller L. Supranuclear control of ocular motility. Oph­thalmol Clin North Am 2001; 14: 187- 204. 9. Zee DS. Considerations on the mechanisms of alternating skew de­viation in patients with cerebellar lesions. J Vestib Res 1996; 6: 395- 401. 10. Brandt T, Dieterich M. Vestibular syndromes in the roll plane: to­pographic diagnosis from brainstem to cortex. AnnNeurol 1994; 36: 337^ 17. 11. Wolfe GI, Taylor CL, FlammES, etal. Ocular tilt reaction resulting from vestibuloacoustic nerve injury. Neurosurg 1993; 32: 417- 21. 12. Halmagyi GM, Brandt T, Dieterich M, et al. Tonic contraversive ocular tilt reaction due to unilateral meso- diencephalic lesion. Neu­rology 1990; 40: 1503- 09. 13. Brandt T, Dieterich M. Pathological eye- head coordination in roll: tonic ocular tilt reaction in mesencephalic and medullary lesions. Brain 1987; 110: 649- 66. 14. Brandt T, Dieterich M. Different types of skew deviation. J Neurol Neurosurg Psychiatr 1991; 54: 549- 50. 15. Keane JR. Alternating skew deviation: 47 patients. Neurology 1985; 35: 725- 28. 16. Moster ML, Schatz NJ, Savino PJ, et al. Alternating skew on lateral gaze ( bilateral abducting hypertropia). Ann Neurol 1988; 23: 190- 92. 17. Hamed LM, Maria BL, Quisling RG, et al. Alternating skew on lateral gaze: neuroanatomic pathway and relationship to superior oblique overaction. Ophthalmology 1993; 100: 281- 86. 18. Surgery of the extra- ocular muscles. In Pediatric Ophthalmology and Strabismus, Basic and Clinical Science Course Section 6. San Francisco: American Academy of Ophthalmology, 2000: 136- 56. 19. Maxwell JS, Schor CM. Adaptation of vertical eye alignment in relation to head tilt. Vision Res 1996; 36: 1195- 1205. 20. Trigler L, Siatkowski RM. Factors associated with horizontal reop­eration in infantile esotropia. J Am Assoc Ped Ophthalmol Strab 2002; 6: 15- 20. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 141