Improvement of Compensatory Head Postures after Strabismus Surgery Stephen P. Kraft, MD, FRCSC, Eamonn P. O'Donoghue, FRCS (Edin), FCOphth, John D. Roarty, MD Purpose: To determine the incidence and relative frequencies of orientations of compensatory head postures (CHP) and success in surgically treating the CHP seen in patients with pure lateral rectus paresis (group I), superior oblique paresis (group II), Duane syndrome (group III), Brown syndrome (group IV), and congenital motor nystagmus (group V). Methods: The authors retrospectively reviewed all patients assessed and treated between 1985 and 1991 with these five conditions. Frequencies and orientations of the CHP were determined in all patients in each group. The success of surgery in eliminating the CHP in the subgroups of patients who underwent surgery to treat the CHP also was determined. Results: The frequencies of CHP were 29.0% of 93 group I cases, 71.2% of 139 group II cases, 68.1% of 91 group III cases, 17.4% of 35 group IV cases, and all 23 group V cases, for an overall incidence of 56.7% of 381 patients. For patients who underwent surgery to eliminate a CHP, success rates were 85.7% of 21 group I cases, 75.6% of 41 group II cases, 87.2% of 47 group III cases, 100% of 5 group IV cases, and 78.3% of 23 group V patients, for an overall success rate of 82.5% of 137 surgical cases. Conclusion: Because CHP is seen frequently in strabismus and nystagmus disorders, ocular causes must be ruled out in any case of an anomalous head posture. Appropriately planned surgery for CHP caused by incomitant strabismus or eccentric nystagmus null zones has a high rate of success in eliminating CHP. Ophthalmology 1992;99: 130 1-1308
Compensatory head postures (CHP) can be caused by ocular or nonocular causes. The ocular causes include nystagmus, strabismus, blepharoptosis, and uncorrected refractive errors. I - 3 Nonocular causes include musculoskeletal torticollis and vestibular and auditory disorders. 2 ,3 Congenital motor nystagmus with a null zone away from the primary position, whether caused by conjugate Originally received: October 14, 1991. Revision accepted: March 5, 1992. From the Department of Ophthalmology, The Hospital for Sick Children, Toronto, Ontario, Canada. Presented in part at a symposium at the American Academy of Oph· thalmology Annual Meeting, Anaheim, October 1991. Reprint requests to Stephen P. Kraft, MD, Department of Ophthal· mology, The Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada M5G IX8.
null zones (without strabismus) or by disconjugate adduction null zones associated with infantile esotropia (Ciancia syndrome), can induce a patient to adopt a CHP to minimize the intensity of the nystagmus and thereby gain optimal visual acuity.4-7 Incomitant strabismus, due to innervational or mechanical causes, may preclude fusion in the primary position and force a patient to adopt a CHP to place the eyes in the field of gaze in which binocular single vision (BSV) can be regained. Severe mechanical restrictions, which can prevent patients from using their preferred fixating eyes in primary position, frequently lead to CHP?·3 In addition, patients with diplopia occasionally will adopt a paradoxic CHP to maximize the separation of the diplopic images. 2 ,8.9 We retrospectively reviewed the frequency ofCHP and the orientations of these CHP in patients with lateral rectus paresis, superior oblique paresis, Duane syndrome, and
1301
Ophthalmology
Volume 99, Number 8, August 1992
Brown syndrome. We also analyzed our success in surgically treating the CHP in patients with these four strabismus entities and patients with CHP caused by congenital motor nystagmus. Prior studies have not systematically addressed the variety of orientations of CHP that can be caused by these various disorders, both unilateral and bilateral, nor have they assessed the success of surgical treatment for the various combinations of possible CHP.
Patients and Methods We reviewed the charts of all patients seen in the strabismus clinic at The Hospital for Sick Children, Toronto, Ontario, from January 1985 to June 1991 who had pure unilateral or bilateral lateral rectus paresis (group I), superior oblique paresis (group II), Duane syndrome type I or II (group III), Brown syndrome (group IV), and congenital motor nystagmus with conjugate null zones (group V). For each of the five entities, we extracted a list of all patients in our data base, their male/female distribution and ages, and the side affected in unilateral cases. We noted the number of patients affected by a clinically significant CHP and the orientation(s) of their head postures. In this study, we considered a CHP clinically significant if the following criterion for each orientation was met: a face turn of 15° or more, chin up or chin down of 10° or more, and head tilt of 10° or more. All patients underwent thorough orthoptic and ophthalmologic examinations. Whenever possible, the component directions of the head position were measured with an arc perimeter or surgical protractor with graded markings. 6 In some patients, especially children, we estimated the angles of the head positions using known reference angles on horizontal or vertical structures within the examining lane. 5,6 From each of the four strabismus groups (groups I to IV), we extracted data on the subgroup of patients with a clinically significant CHP who had eye muscle surgery and who had a minimum of 6 months of postoperative follow-up. For each of these subgroups, we recorded the preoperative alignment, the procedures performed, and
the ocular alignment and size of any clinically significant CHP at the last follow-up visit. We categorized the postoperative results as no clinically significant CHP, partial reduction of CHP from the preoperative angle but still clinically significant, no improvement from the preoperative angle, or a habit CHP. 2 We applied this last category to a strabismus patient who had complete correction of the strabismus angle or diplopia or both in primary position yet continued to adopt a clinically significant CHP under both monocular and binocular conditions. We analyzed the unilateral and bilateral cases separately within each group and subgroup. In group V, we included patients with congenital motor nystagmus and no strabismus who had surgery for a clinically significant CHP and at least 6 months of postoperative follow-up. For these patients, we noted the procedures performed and categorized the results at the latest visit as no clinically significant CHP, partial reduction of CHP but still clinically significant, no change in CHP, or overcorrection of the CHP by more than 10° to the opposite direction. There were several patients in the strabismus groups who had minimal CHP of a few degrees in one or more orientations, but the magnitudes of the postures were less than our minimal criteria for clinical significance in this study. In addition, some patients in each group who did not have clinically significant CHP had surgery to correct other symptomatic aspects of their strabismus problems. We restricted our analysis to patients who underwent surgery specifically to correct a clinically significant CHP.
Results Overall Results Overall results are presented in Table 1. Our study included 358 patients in the 4 strabismus groups (groups I to IV), of whom 193 (53.9%) had a clinically significant CHP. Surgery was performed on 137 patients with CHP: 114 with strabismus (groups I to IV) and all 23 patients with congenital motor nystagmus (group V). Our surgery re-
Table 1. Overall Frequencies of Compensatory Head Posture and Surgical Results in Five Groups Group No. II III
IV V Total
Diagnosis Lateral Rectus Paresis Superior Oblique Paresis Duane Syndrome Brown Syndrome Congenital Nystagmus
Total No.
93 139 91 35 23 381
Compensatory Head Posture No. (%)
No. who had Surgery
No. with 1 Surgery
No. with Reoperations
Total No. (%)
26 (29.0) 99 (71.2) 62 (68.1) 6 (17.4) 23 (100*) 216 (56.7)
21 41 47 5 23 137
17 28 36 5 17 103
1 3 5 0
18 (85.7) 31 (75.6) 41 (87.2) 5 (100.0) 18 (78.3) 113 (82.5)
• All nystagmus patients in the study had compensatory head posture.
1302
Surgical Successes
10
Kraft et al . Surgery for Compensatory Head Postures duced the CHP to clinically insignificant in 113 patients (82.5%): 103 after one surgical procedure and 10 after reoperations. Two further patients with residual CHP, both with lateral rectus paresis, were treated with prisms that eliminated their CHP, so the number of successfully treated patients actually totalled 115 (83.9%). Group I: Lateral Rectus Paresis Our review of the records indicated that 93 patients had lateral rectus paresis. They ranged in age from 1 to 88 years (mean, 42.7 years); 51 (54.8%) were male. We considered the unilateral and bilateral cases separately. Within each group, we classified the cases into those with total palsies, in which there was no abduction beyond the midline, and those with partial paresis. Unilateral Lateral Rectus Paresis. The group of patients with unilateral lateral rectus paresis included 71 patients (Table 2). The left eye was affected in 42 (59.1 %) of the cases. Among the 51 patients with partial paresis, 11 (21.6%) showed a clinically significant CHP: a face turn toward the involved eye in all cases. The esotropia (ET) in primary position ranged from 10 to 40 prism diopters (PO). Surgery eliminated the CHP in all 8 of these 11 patients who had operations. A medial rectus recession and lateral rectus resection was used in 7 patients, and a medial rectus recession along with a Hummelsheim transposition was done in 1. Adjustable sutures were used in 6.10 Among the 20 complete palsies, 8 (40.0%) showed a clinically significant CHP that was consistently associated with a face turn in the direction of the involved eye. The ET in primary position ranged from 30 to 75 PD. Surgery successfully eliminated CHP in 5 of 7 patients who had operations. Four surgeries involved a botulinum toxin injection to the ipsilateral contractured medial rectus muscle followed 2 or 3 weeks later by a vertical rectus muscle transposition. 11-14 Two patients had muscle transpositions at the same time as ipsilateral medial rectus recessions on adjustable sutures. One patient had a large recess-resect procedure with an adjustable suture. Surgical follow-up ranged from 6 months to 4 years. Overall, the CHP was reduced to clinically insignificant in 13 (86.7%) of the 15 surgical cases (Table 2). The residual face turn in one of the other two patients responded to a base-out prism for the affected eye. Bilateral Lateral Rectus Paresis. The group of patients with bilateral lateral rectus paresis included 22 pa-
tients: 17 (77.3%) had bilateral partial paresis, and 5 (22.7%) total palsies. Among the 17 patients with partial paresis, only 2 (11.8%) had a CHP, and each had a face turn of30° toward the eye with more limited abduction. Surgery in the one patient consisted of bilateral medial rectus recessions; in the other, it involved a unilateral recess-resect procedure on the dominant eye that was affected by the more severe paresis. Both surgeries included adjustable sutures. 10 Both patients regained BSV in primary position and resolution of their CHP. Follow-up in these 2 patients was 6 months and 1 year after surgery. All 5 patients with bilateral total palsies showed large (45°) face turns in the directions of their dominant eyes. Four patients had surgery, of whom 3 underwent treatment consisting of bilateral botulinum injections followed by muscle transpositions.II,l3,14 One patient had no CHP after follow-up of 1 year. The second patient had a residual CHP due to a residual esotropia that was successfully treated with a base-out prism; the result has been stable for 2 years. The third patient required a reoperation involving bilateral medial rectus recessions to eliminate the primary position misalignment; this result has been stable for 2 years. The fourth patient had a large recess-resect procedure that eliminated a face turn; there has been no recurrence over 6 months. Thus, we have eliminated the CHP with surgery in 5 (83.3%) of 6 patients with bilateral palsies; the sixth responded to a prism after surgery. Group II: Superior Oblique Paresis Our review of the records identified 139 patients with superior oblique paresis, They ranged in age from 2 to 78 years (mean, 28.2 years); 88 (63.3%) were male. We considered the unilateral and bilateral cases separately. Unilateral Superior Oblique Paresis. One hundred and twenty-one of the 139 patients in group II had unilateral superior oblique paresis (Table 3). The right eye was affected in 62 (51.2%) of these cases. Of these 121 patients, 90 (74.4%) had a clinically significant CHP. The frequency of the CHP orientations are listed in Table 3. Of the 41 patients with combination CHP, 36 included a head tilt as part of the orientation. Of those patients who had head tilts as part of their CHP, the tilt was almost always to the shoulder contralateral to the involved eye, a common finding.1. 9 ,15 However, 2 patients tilted to the side ipsilateral to the paretic
Table 2. Data for Patients with Unilateral Lateral Rectus Paresis
Subgroup Partial paresis Complete palsy CHP
=
Total No.
Patients with Face Turns No. (%)
Range of Face Turns (Degrees)
51
11 (21.6)
20
8 (40,0)
15-30 30-45
No. with CHP who had Surgery
No. with Insignificant CHP after Surgery
8
8
7
5
compensatory head posture,
1303
Ophthalmology
Volume 99, Number 8, August 1992
Table 3. Orientations of CHP for Patients with Unilateral Superior Oblique Paresis
CHP
=
Orientation
Patients No. (%)
Head tilt alone Face turn alone Chin up alone Chin down alone Combinations No significant CHP Total
37 (30.6) 6 (5.0) 4 (3.3) 2 (1.6) 41 (33.9)* 31 (25.6) 121 (100.0)
compensatory head posture.
* Thirty six of these 41 combinations included a tilt as a component of
CHP.
eye to produce large separation of their diplopic images. 8,9,15 Patients with face turns as part of the CHP orientation showed turns to the side away from the involved eye, except for one patient who turned toward the affected side to separate the diplopic images. Thus, we found 3 (3.3%) of the 90 patients with CHP whose CHP was oriented opposite to that expected. These head postures are termed "paradoxic.,,8 We performed surgery on 32 of the patients with CHP. 16 The procedures were distributed as follows: inferior oblique weakening alone, 15; inferior oblique weakening and 1 vertical rectus recession, 13; inferior oblique weakening and a Harada-Ito transposition, 2; inferior oblique weakening and 2 vertical rectus recessions, 1; superior rectus recession, 1. Adjustable sutures were used in 10 patients who had vertical rectus recessions. In 10 patients, horizontal rectus surgery was added to the plan to treat a coexisting horizontal deviation. Follow-up ranged from 6 months to 3 years. Table 4 lists the surgical results for the 32 cases. Among the 23 (71.9%) surgical successes, 3 required reoperations. Two ofthese reoperations were required to treat unmasked contralateral superior oblique paresis I7 ,'8 that induced head tilts to the sides contralateral to the eyes with the newly uncovered paresis. The third patient had a reoperation to eliminate a vertical overcorrection caused by the first surgery. There was a residual CHP in 5 (15.6%) surgical patients, despite reoperations in 2. All 5 had residual deviations in primary position. Finally, we assigned 4 (12.5%) of the patients to the category of habit CHP because they demonstrated clinically significant residual CHP that persisted under monocular and binocular conditions despite successful elimination of their deviations in primary position. None of these four patients had subjective or objective excyclotorsion postoperatively. 19,20 Bilateral Superior Oblique Paresis. The group of patients with bilateral superior oblique paresis included 18 patients, of whom 9 (50.0%) showed a clinically significant CHP. All 9 patients had surgery. Their data, including CHP orientations and surgical procedures, are
1304
listed in Table 5. 21 Three of the patients had asymmetric bilateral palsies. Overall, the CHP was reduced to clinically insignificant in 8 (88.9%) of the 9 patients. Follow-up ranged from 6 months to 2 years. Group III: Duane Syndrome Types I and II Ninety-one patients with Duane syndrome types I and II were included in group III. They ranged in age from 8 months to 52 years (mean, 8.2 years); 53 (58.2%) were male. We analyzed the unilateral and bilateral cases separately. Unilateral Duane Syndrome. The group with unilateral Duane syndrome included 81 patients; 74 were type I and 7 were type 11.22-25 The data for these patients are recorded in Table 6. Overall, 55 (67.9%) of the 81 patients presented with a clinically significant CHP, all face turns. In almost all type I cases, a face turn was associated with a severe limitation of abduction resulting in an ET over 10 PO in primary position. Most cases with partial limitations of ductions in the involved eye did not adopt a clinically significant CHP. Six (85.7%) of the 7 type II patients had severe adduction limitations and large exotropias (XT) and adopted clinically significant CHP, all of which were face turns to the side contralateral to the involved eye. We performed surgery on 42 of the 55 patients with CHP; 37 were type I and 5 were type II. Follow-up ranged from 6 months to 3 years. Among the 37 type I patients, preoperative ET ranged from 10 to 35 PD. Initial surgery involved one medial rectus recession in 21 patients, bilateral medial recessions in 10, and bilateral medial recessions and a large lateral rectus recession (to treat retraction) in 6. 26 The postoperative results are listed in Table 7. Five patients required reoperations to obtain successful results: 4 required transposition procedures and 1 had bilateral medial rectus rerecessions. One patient with a residual CHP had a reoperation, a unilateral medial re-recession, without complete resolution of his CHP. The one patient listed as having an overcorrection developed an XT after are-recession of the medial rectus of his involved eye, creating a face turn of 20° to the side opposite the involved eye. Table 4. Results of Surgery for CHP in Patients with Unilateral Superior Oblique Paresis
CHP
=
Result
Patients No. (%)
No significant CHP Residual CHP HabitCHP Total
23 (71.9)* 5 (15.6) 4 (12.5) 32 (100.0)
compensatory head posture.
• Reoperations needed in 3 patients.
Kraft et al . Surgery for Compensatory Head Postures Table 5. Data for Patients with Bilateral Superior Oblique Paresis Patient No.
Age (yrs)
Sex
1 2 3 4 5
16 56 46 33 47
F M M F M
6
40
F
7
5
F
8 9
4 26
M F
Orientation of CHP Preoperatively Chin down Chin down Chin down Chin down Chin down Right turn Chin down Left tilt Left tilt Right tilt Right tilt Right turn
Surgery Performed
Postoperative CHP
Bilateral Harada-Ito Bilateral Harada-Ito Bilateral Harada-Ito Bilateral Harada-Ito Bilateral inferior oblique myectomies Right Harada-Ito
None None None Chin up, left turn' None
Bilateral inferior oblique myectomies Left inferior oblique myectomy Left superior rectus recess; right medial rectus recess-lateral rectus resect
None
None
None None
CHP = compensatory head posture . • Reoperation (bilateral inferior rectus recessions) failed to eliminate CHP.
Among the 5 type II patients, the preoperative XT ranged from 25 to 50 PD. Initial surgery involved bilateral lateral rectus recessions in three patients who had XT in primary position, and unilateral combined lateral rectus and superior rectus recessions in two patients who had XT and hypertropias in primary position. 26 The results are listed in Table 7. One patient had a residual 15 0 face turn due to a residual XT in primary position. Overall, the CHP was reduced to clinically insignificant in 37 (88.1 %) of the 42 unilateral cases. Bilateral Duane Syndrome. Our review of the records included 10 patients with bilateral Duane syndrome (5 females): 9 had bilateral type I syndromes, and 1 a type I in 1 eye and type II in the other. Seven (70%) of the patients, all with bilateral type I, had a clinically significant face turn, ranging from 15 0 to 35 0 • They all had severe limitations of abduction in the dominant eye. We performed surgery on 5 of the 7 patients, and in all cases we recessed both medial rectus muscles. 26 Three involved adjustable sutures. In 4 patients (80%), the CHP was reduced to clinically insignificant. One patient had a
clinically significant residual face turn postoperatively, with an ET of 20 PD in primary position.
Group IV: Brown Syndrome Thirty-five patients with Brown syndrome were included in group IV. They ranged in age from 9 months to 16 years (mean age, 5.3 years); 19 (54.3%) were female. Two patients had a bilateral syndrome, one of which was transient in association with rheumatoid arthritis. In 19 (57.6%) of the 33 patients with unilateral cases, the right eye was affected. Only 5 (15.2%) of the unilateral cases had a clinically significant CHP according to our criteria. They were distributed as follows: face turn alone, 2; chin-up alone, 1; combination of face turn and chin-up, 2. All 5 patients had surgery that consisted of superior oblique tenotomies. 27 The CHP was reduced to clinically insignificant in all 5 patients, with follow-up ranging from 6 months to 2 years. No patient has developed a postoperative superior oblique palsy.27-29
Table 6. Data on Patients with Unilateral Duane Syndrome Type of Duane Syndrome II Total
Total No. of Eyes
Left Eyes Affected No. (%)
Patients with Significant Face Turns' No. (%)
Range of Face Turns (degree)
74
56 (75.7) 2 (28.6) 58 (71.6)
49 (66.2) 6 (85.7) 55 (67.9)
15-35 15-45 15-45
7 81
• Minimum face turn of 15 0 •
1305
Volume 99, Number 8, August 1992
Ophthalmology
The patient with bilateral Brown syndrome associated with arthritis had a chin-up posture, whereas the patient with the congenital bilateral syndrome did not adopt any CHP. Neither patient underwent surgery. Group V: Congenital Motor Nystagmus Group V consisted of 23 patients with a CHP who had congenital motor nystagmus with conjugate null zones and no strabismus. 4- 7 Their age ranges before surgery were 4 to 22 years (mean, 9.5 years); 17 (73.9%) were male. Face Turns. Twenty patients presented with face turns as their CHP, 12 (60%) directed to their left sides. Of these, 6 measured between 15° and 30° (moderate), while 14 measured from 35° to 50° (large). These patients all underwent Kestenbaum-Anderson procedures based on the Parks 5-6-7-8 rule for 15° to 20° of face turn. 5 ,6,30 For turns greater than 20°, these numbers all were increased in proportion to the size of face tum, up to a maximum of 60% increase for a 50° turn. 5- 7 The results of surgery in these 20 patients with face turns are listed in Table 8. Follow-up ranged from 6 months to 4 years. In the subgroup of moderate turns, 2 patients were left with residual turns, 1 of 15 ° and the other of 20°. In the subgroup of large turns, 1 patient required a reoperation to eliminate a residual 15 ° face tum after initially having an augmented KestenbaumAnderson procedure. The 3 patients with residual CHP had turns ranging from 15 ° to 25 0. Overall, we successfully reduced CHP to clinically insignificant in 15 (75.0%) of the 20 cases with face turns. Miscellaneous CHP. We had 3 nystagmus patients with CHP other than pure face turns who underwent surgery. One had a chin-up posture of 10° to 15°. He was successfully treated with bilateral inferior rectus recessions of6 mm. A second patient presented with a combined left face tum of 30° and chin-up posture of 10°. He underwent a combination of horizontal rectus recessions (right lateral recession 10.0 mm, left medial recession 8.0 mm) with
Table
7. Results of Surgery for Unilateral Duane Syndrome with CHP
Postoperative CHP Compared with Preoperative
No.
(%)
None or clinically insignificant Residual but improved Not improved Total
33 3 1 37
89.2" 8.1t
CHP
=
Type I
2·n
100.0
Type II
. No. 4 1 0
5
(%)
80.0 20.0 0.0 100.0
compensatory head posture.
• Five patients required reoperations to eliminate CHP.
t t
One patient had a residual CHP despite reoperation. Patient was overcorrected after reoperation for residual face turn.
1306
Table 8. Results of Surgery for Congenital Nystagmus with Face Turn Size of Preoperative Turn Postoperative Face Turn Compared with Preoperative
No.
(%)
No.
(%)
None or clinically insignificant Residual but improved Overcorrected Total
4 2 0 6
66.7 33.3 0.0 100.0
11 3 0
78.6" 21.4 0.0 100.0
15°
to
30°
35° to 50°
14
• One patient required bilateral reoperation to eliminate residual face turn.
bilateral 5.0 mm inferior rectus recessions. He had a residualleft tum of 10°, which was cosmetically acceptable. Finally, we had one patient with a right head tilt of 15 ° with a left tum of 5 ° to 10°. We weakened the right superior oblique with a 6-mm tendon expander31 and myectomized the left inferior oblique 10 mm. The head tilt was eliminated, but he has a residual 10° left face tum that has not increased over 9 months of follow-up.
Discussion Before treating a patient with CHP, the physician must first determine if it is caused by an ocular or nonocular problem. The ophthalmologist's obligation is to rule out all possible ocular causes in the differential diagnosis. If no ocular cause is found, then nonocular possibilities remain as the diagnoses of exclusion. If an ocular cause is diagnosed, then therapy to eliminate the cause also should lead to resolution of the CHP. However, some patients with long-standing head postures will, despite successful treatment of the underlying condition, continue to demonstrate a habit CHP. 2 These are usually attributed to alterations in the tone of the neck muscles that may not reverse even if the ocular causes are rectified. We found four such patients, all in our group of unilateral superior oblique paresis corrected with surgery. Thus, it is important to educate primary care physicians to refer any child with a CHP for ophthalmologic assessment so that ocular causes can be ruled out. 2 •3 Compensatory head postures yield a number of possible advantages for patients when there are ocular causes. A patient may gain improved visual acuity, as in cases of nystagmus or refractive error, or avoid diplopia and maintain fusion, as seen with incomitant strabismus. A CHP can be a consequence of a patient's need to fixate with a dominant eye affected by a muscle imbalance, or simply a necessity for vision, as in cases ofblepharoptosis . Rarely, patients can intentionally maximize the separation of diplopic images if they have incomitant strabismus with intractable diplopia.
Kraft et al . Surgery for Compensatory Head Postures However, there are many reasons patients with these diagnoses do not universally demonstrate CHP. In cases of congenital nystagmus, the null zone may be close enough to primary position that no significant head posturing is required. Patients with unilateral ptosis may fixate with the nonaffected eye. This is a concern in children with early-onset blepharoptosis because of the risk of amblyopia in the affected eye. In patients with strabismus, there are various possible explanations why they may not adopt a CHP, and all applied to at least some patients in groups I to IV in our study. First, the most common reason is the presence of amblyopia or suppression, which causes the patient to fixate with the preferred eye and have no stimulus for fusion that would promote a CHP. 9 Second, a patient may develop larger-than-normal fusional amplitudes so that BSV can be maintained in primary position. This also may allow a good range of BSV field around primary position. This is usually seen in patients with long-standing paretic strabismus or muscle paresis that is very slowly progressive, as seen in cases of superior oblique or lateral rectus paresis. 8 .9 Third, an eye affected by a unilateral muscle imbalance, either mechanical or paretic, may have some range of movement into the affected field of gaze causing minimal or no deviation in primary position and a range of BSV field to the affected side. This is frequent in patients with mild Duane syndromes or partial sixth nerve paresis, for example. Finally, a patient with diplopia because of an acquired strabismus may ignore the displaced image and maintain a normal head posture without symptoms. 1,9 Among our patients with unilateral superior oblique palsies, we found that 3 (3.3%) of the 90 patients with CHP had paradoxic postures, a term we use for a CHP opposite to that which we would expect a patient to adopt to maximize separation of diplopic images. 8 ,9 Our incidence is similar to the 4.9% (7 of 144 cases) reported by von Noorden et al 8 in their series of patients with unilateral superior oblique palsies with CHP. Only 21 of our 93 cases oflateral rectus paresis (group I) had surgery. A number offactors explain why a limited number of patients had surgery. Partial paresis in many patients resolved spontaneously with time, especially when associated with head trauma or vascular causes. Eight of these received botulinum injections into the medial rectus muscles of the involved eyes. I I - 14 Some patients had sixth nerve palsies that were caused by intracranial tumors, and eye muscle surgery was not pursued because of poor prognosis for the underlying conditions. Many patients had small CHP, smaller than our minimal criteria for clinical significance, and refused surgical intervention since their range ofBSV field was large enough for them to continue their daily activities. We treated some patients with partial paresis with prisms to eliminate diplopia in primary position. Finally, in some patients with unilateral cases, the underlying condition led to such reduced vision in the affected eye that the patients did not experience diplopia and, consequently, had no reason to adopt a CHP. Our incidence of CHP in superior oblique paresis
(group II) was 71.2%, almost exactly the same as the 67.8% reported by von Noorden et al. 8 Our surgical success rate in eliminating the CHP was 75.6%, or slightly less than their reported 89% in 96 surgeries. However, they recorded data only on head tilts for their CHP, whereas we included all combinations of tilts, chin-up or chin-down, or face turns, and combinations of these among our patients with CHP for both our preoperative and postoperative analyses. In addition, we had four patients with habit tilts after surgery, which slightly reduced our success rate. We did not find any patients in the subgroup of unilateral superior oblique palsies in whom we believed that the cyclotorsion alone was responsible for a head tilt posture. Guyton,19 von Noorden and Ruttum,15 and von Noorden 20 have suggested that, occasionally, cyclotorsion of one eye may induce a compensatory tilt to the opposite shoulder. In such a case, a patient would have a head tilt when the involved eye fixates and would have no CHP when the uninvolved eye fixates. This contrasts with the much more common situation in which the head tilt is caused by an incomitant vertical tropia, with a forced head tilt difference between the vertical deviation on right versus left head tilt. 8,15-19 In this more common scenario, the head tilt would manifest under binocular conditions as the patient adopts a CHP to reduce the size of the vertical deviation, but it would vanish when either eye fixates as the patient is rendered diplopia-free. Our 68.1 % incidence of CHP in patients with Duane syndrome (group III) is comparable with the 77.4% reported by Raab 23 and higher than the 51 % reported by O'Malley et al. 25 Our 87.2% surgical success is slightly higher than the 79% reported by Pressman and Scott.24 Duane syndrome is a spectrum of motility disorders with variable degrees oflimited ductions, so affected eyes may have muscle forces that are balanced at or near the primary position. 26 Thus, a large proportion of patients with Duane syndrome do not have a deviation in primary position large enough to cause a CHP. We had a relatively low incidence ofCHP in our Brown syndrome group (group IV), only 16 (17.4%) of 35 patients in contrast to an incidence of 35% reported by Wilson et al. 27 However, they did not indicate the magnitudes of the CHP in their patients, so some patients may have had mild postures that would not have been labeled clinically significant according to the criteria we used in this study. We performed surgeries on all five patients with unilateral Brown syndrome who also had clinically significant CHP, all of which were successful. Sprunger et al29 eliminated a clinically significant CHP in 17 (70.8%) of 24 patients with Brown syndrome whom they treated with surgery. A few other patients among our unilateral cases had surgery to treat a cosmetically noticeable downshoot in adduction of the affected eye, but we did not address the surgical results in these patients in this study since they had clinically insignificant or no CHP preoperatively. Finally, we successfully reduced the CHP to clinically insignificant in 18 (78 .3%) of 23 patients with congenital motor nystagmus with CHP and no strabismus. This compares to success rates reported by other authors 5- 7 in
1307
Ophthalmology
Volume 99, Number 8, August 1992
recent years since augmented surgery has been increasingly used for face turns over 25 0 to 30 0 • Like these authors, we have found that very large amounts of surgery used to treat large face turns inevitably lead to limited ductions in one direction of gaze. In our experience this is a necessary trade-off to ensure a stable result. None of our patients had their CHP overcorrected by the surgery, even with augmented amounts of 50% or 60%. It should be emphasized that not all patients with CHP due to ocular causes require surgery. In our study, 193 (53.9%) of 358 patients in the four strabismus groups (groups I to IV) presented with clinically significant CHP. Of these 186 patients, only 114 (61.3%) underwent surgery. There are several reasons a large number of patients with CHP caused by strabismus do not undergo surgery. The CHP may be transient and vanish as a transient strabismus resolves. Prisms, patching of one eye, and botulinum toxin injections can be useful treatments in selected instances to help patients cope with the conditions causing their CHP. Finally, some patients are only minimally bothered by their CHP, or their CHP can be small at the outset and require no treatment. In our entire series of 137 surgeries to correct CHP in our 5 patient groups (i.e., including the 23 patients with congenital nystagmus), we reduced the CHP to clinically insignificant in 113 (82.5%) patients. Because CHP can be caused by several ocular disorders, the ophthalmologist must play an integral role in the work-up of a patient who presents with an abnormal head position. From the results of our study, we conclude that if an ocular cause that is amenable to surgical correction is found, then an appropriate surgical plan will have a high chance of eliminating the CHP.
References 1. von Noorden GK. Atlas of Strabismus, 4th ed. St. Louis: CV Mosby, 1983;128-41. 2. Morris JE. Head posturing in horizontal strabismus. Am Orthopt J 1983;33:2-6. 3. Helveston EM. Strabismus surgical treatment for abnormal head posturing. Am Orthopt J 1983;33:32-5. 4. Dell'Osso LF, Flynn JT. Congenital nystagmus surgery: a quantitative evaluation of the effects. Arch Ophthalmol 1Q79;97:462-9. 5. Scott WE, Kraft SP. Surgical treatment of compensatory head position in congenital nystagmus. J Pediatr Ophthalmol Strabismus 1984;21 :85-95. 6. Mitchell PR, Wheeler MB, Parks MM. Kestenbaum surgical procedure for torticollis secondary to congenital nystagmus. J Pediatr Ophthalmol Strabismus 1987;24:87-92. 7. Nelson LB, Ervin-Mulvey LD, Calhoun JH, et al. Surgical management for abnormal head position in nystagmus: the augmented modified Kestenbaum procedure. Br J Ophthalmol 1984;68:796-800. 8. von Noorden GK, Murray E, Wong SY. Superior oblique paralysis. A review of270 cases. Arch OphthalmoI1986;104: 1771-6.
1308
9. Burian HM, Rowan PJ, Sullivan MS. Absence of spontaneous head tilt in superior oblique palsy. Am J Ophthalmol 1975;79:972-7. 10. Kraft SP, Jacobson ME. Techniques of adjustable suture strabismus surgery. Ophthalmic Surg 1990;21 :633-40. 11. Kraft SP, Allan K. The use of botulinum toxin in the management of sixth nerve palsy. Am Orthopt J 1989;39:8997. 12. Scott AB, Kraft SP. Botulinum toxin injection in the management of lateral rectus paresis. Ophthalmology 1985;92: 676-83. 13. Fitzsimons R, Lee JP, Elston J. Treatment of sixth nerve palsy in adults with combined botulinum toxin chemodenervation and surgery. Ophthalmology 1988;95: 1535-42. 14. Rosenbaum AL, Kushner BJ, Kirschen D. Vertical rectus muscle transposition and botulinum toxin (oculinum) to medial rectus for abducens palsy. Arch Ophthalmol 1989; 107:820-3. 15. von Noorden GK, Ruttum M. Torticollis in paralysis of the trochlear nerve. Am Orthopt J 1983;33:16-20. 16. Scott WE, Kraft SP. Classification and surgical treatment of superior oblique palsies: I. Unilateral superior oblique palsies. In: Pediatric Ophthalmology and Strabismus: Trans New Orleans Acad Ophthalmol. New York: Raven Press, 1986;15-38. 17. Kraft SP, Scott WE. Masked bilateral superior oblique palsy: clinical features and diagnosis. J Pediatr Ophthalmol Strabismus 1986;23:264-72. 18. Kushner BJ. The diagnosis and treatment of bilateral masked superior oblique palsy. Am J Ophtha1mol 1988;105:18694. 19. Guyton DL. Clinical assessment of ocular torsion. Am Orthopt J 1983;33:7-15. 20. von Noorden GK. Clinical observations in cyclodeviations. Ophthalmology 1979;86: 1451-61. 21. Scott WE, Kraft SP. Classification and treatment of superior oblique palsies: II. Bilateral superior oblique palsies. In: Pediatric Ophthalmology and Strabismus: Trans New Orleans Acad Ophthalmol. New York: Raven Press, 1986;265-91. 22. Huber A. Electrophysiology of the retraction syndromes. Br J Ophthalmol 1974;58:293-300. 23. Raab EL. Clinical features of Duane's syndrome. J Pediatr Ophthalmol Strabismus 1986;23:64-8. 24. Pressman SH, Scott WE. Surgical treatment for Duane's syndrome. Ophthalmology 1989;93:29-38. 25. O'Malley ER, Helveston EM, Ellis FD. Duane's retraction syndrome-plus. J Pediatr Ophthalmol Strabismus 1982; 19: 161-5. 26. Kraft SP. A surgical approach for Duane syndrome. J Pediatr Ophthalmol Strabismus 1988;25: 119-30. 27. Wilson ME, Eustis HS, Parks MM. Brown's syndrome. Surv Ophthalmol 1989;34: 153-72. 28. Eustis HS, O'Reilly C, Crawford JS. Management of superior oblique palsy after surgery for true Brown's syndrome. J Pediatr Ophthalmol Strabismus 1987;24: 10-16. 29. Sprunger DT, von Noorden GK, Helveston EM. Surgical results in Brown syndrome. J Pediatr Ophthalmol Strabismus 1991;28:164-7. 30. Parks MM. Congenital nystagmus surgery. Am Orthopt J 1973;23:35-9. 31. Wright KW. Superior oblique silicone expander for Brown syndrome and superior oblique overaction. J Pediatr Ophthalmol Strabismus 1991;28:101-7.
Compensatory head postures (CHP) can be caused by ocular or nonocular causes. The ocular causes include nystagmus, strabismus, blepharoptosis, and uncorrected refractive errors. I - 3 Nonocular causes include musculoskeletal torticollis and vestibular and auditory disorders. 2 ,3 Congenital motor nystagmus with a null zone away from the primary position, whether caused by conjugate Originally received: October 14, 1991. Revision accepted: March 5, 1992. From the Department of Ophthalmology, The Hospital for Sick Children, Toronto, Ontario, Canada. Presented in part at a symposium at the American Academy of Oph· thalmology Annual Meeting, Anaheim, October 1991. Reprint requests to Stephen P. Kraft, MD, Department of Ophthal· mology, The Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada M5G IX8.
null zones (without strabismus) or by disconjugate adduction null zones associated with infantile esotropia (Ciancia syndrome), can induce a patient to adopt a CHP to minimize the intensity of the nystagmus and thereby gain optimal visual acuity.4-7 Incomitant strabismus, due to innervational or mechanical causes, may preclude fusion in the primary position and force a patient to adopt a CHP to place the eyes in the field of gaze in which binocular single vision (BSV) can be regained. Severe mechanical restrictions, which can prevent patients from using their preferred fixating eyes in primary position, frequently lead to CHP?·3 In addition, patients with diplopia occasionally will adopt a paradoxic CHP to maximize the separation of the diplopic images. 2 ,8.9 We retrospectively reviewed the frequency ofCHP and the orientations of these CHP in patients with lateral rectus paresis, superior oblique paresis, Duane syndrome, and
1301
Ophthalmology
Volume 99, Number 8, August 1992
Brown syndrome. We also analyzed our success in surgically treating the CHP in patients with these four strabismus entities and patients with CHP caused by congenital motor nystagmus. Prior studies have not systematically addressed the variety of orientations of CHP that can be caused by these various disorders, both unilateral and bilateral, nor have they assessed the success of surgical treatment for the various combinations of possible CHP.
Patients and Methods We reviewed the charts of all patients seen in the strabismus clinic at The Hospital for Sick Children, Toronto, Ontario, from January 1985 to June 1991 who had pure unilateral or bilateral lateral rectus paresis (group I), superior oblique paresis (group II), Duane syndrome type I or II (group III), Brown syndrome (group IV), and congenital motor nystagmus with conjugate null zones (group V). For each of the five entities, we extracted a list of all patients in our data base, their male/female distribution and ages, and the side affected in unilateral cases. We noted the number of patients affected by a clinically significant CHP and the orientation(s) of their head postures. In this study, we considered a CHP clinically significant if the following criterion for each orientation was met: a face turn of 15° or more, chin up or chin down of 10° or more, and head tilt of 10° or more. All patients underwent thorough orthoptic and ophthalmologic examinations. Whenever possible, the component directions of the head position were measured with an arc perimeter or surgical protractor with graded markings. 6 In some patients, especially children, we estimated the angles of the head positions using known reference angles on horizontal or vertical structures within the examining lane. 5,6 From each of the four strabismus groups (groups I to IV), we extracted data on the subgroup of patients with a clinically significant CHP who had eye muscle surgery and who had a minimum of 6 months of postoperative follow-up. For each of these subgroups, we recorded the preoperative alignment, the procedures performed, and
the ocular alignment and size of any clinically significant CHP at the last follow-up visit. We categorized the postoperative results as no clinically significant CHP, partial reduction of CHP from the preoperative angle but still clinically significant, no improvement from the preoperative angle, or a habit CHP. 2 We applied this last category to a strabismus patient who had complete correction of the strabismus angle or diplopia or both in primary position yet continued to adopt a clinically significant CHP under both monocular and binocular conditions. We analyzed the unilateral and bilateral cases separately within each group and subgroup. In group V, we included patients with congenital motor nystagmus and no strabismus who had surgery for a clinically significant CHP and at least 6 months of postoperative follow-up. For these patients, we noted the procedures performed and categorized the results at the latest visit as no clinically significant CHP, partial reduction of CHP but still clinically significant, no change in CHP, or overcorrection of the CHP by more than 10° to the opposite direction. There were several patients in the strabismus groups who had minimal CHP of a few degrees in one or more orientations, but the magnitudes of the postures were less than our minimal criteria for clinical significance in this study. In addition, some patients in each group who did not have clinically significant CHP had surgery to correct other symptomatic aspects of their strabismus problems. We restricted our analysis to patients who underwent surgery specifically to correct a clinically significant CHP.
Results Overall Results Overall results are presented in Table 1. Our study included 358 patients in the 4 strabismus groups (groups I to IV), of whom 193 (53.9%) had a clinically significant CHP. Surgery was performed on 137 patients with CHP: 114 with strabismus (groups I to IV) and all 23 patients with congenital motor nystagmus (group V). Our surgery re-
Table 1. Overall Frequencies of Compensatory Head Posture and Surgical Results in Five Groups Group No. II III
IV V Total
Diagnosis Lateral Rectus Paresis Superior Oblique Paresis Duane Syndrome Brown Syndrome Congenital Nystagmus
Total No.
93 139 91 35 23 381
Compensatory Head Posture No. (%)
No. who had Surgery
No. with 1 Surgery
No. with Reoperations
Total No. (%)
26 (29.0) 99 (71.2) 62 (68.1) 6 (17.4) 23 (100*) 216 (56.7)
21 41 47 5 23 137
17 28 36 5 17 103
1 3 5 0
18 (85.7) 31 (75.6) 41 (87.2) 5 (100.0) 18 (78.3) 113 (82.5)
• All nystagmus patients in the study had compensatory head posture.
1302
Surgical Successes
10
Kraft et al . Surgery for Compensatory Head Postures duced the CHP to clinically insignificant in 113 patients (82.5%): 103 after one surgical procedure and 10 after reoperations. Two further patients with residual CHP, both with lateral rectus paresis, were treated with prisms that eliminated their CHP, so the number of successfully treated patients actually totalled 115 (83.9%). Group I: Lateral Rectus Paresis Our review of the records indicated that 93 patients had lateral rectus paresis. They ranged in age from 1 to 88 years (mean, 42.7 years); 51 (54.8%) were male. We considered the unilateral and bilateral cases separately. Within each group, we classified the cases into those with total palsies, in which there was no abduction beyond the midline, and those with partial paresis. Unilateral Lateral Rectus Paresis. The group of patients with unilateral lateral rectus paresis included 71 patients (Table 2). The left eye was affected in 42 (59.1 %) of the cases. Among the 51 patients with partial paresis, 11 (21.6%) showed a clinically significant CHP: a face turn toward the involved eye in all cases. The esotropia (ET) in primary position ranged from 10 to 40 prism diopters (PO). Surgery eliminated the CHP in all 8 of these 11 patients who had operations. A medial rectus recession and lateral rectus resection was used in 7 patients, and a medial rectus recession along with a Hummelsheim transposition was done in 1. Adjustable sutures were used in 6.10 Among the 20 complete palsies, 8 (40.0%) showed a clinically significant CHP that was consistently associated with a face turn in the direction of the involved eye. The ET in primary position ranged from 30 to 75 PD. Surgery successfully eliminated CHP in 5 of 7 patients who had operations. Four surgeries involved a botulinum toxin injection to the ipsilateral contractured medial rectus muscle followed 2 or 3 weeks later by a vertical rectus muscle transposition. 11-14 Two patients had muscle transpositions at the same time as ipsilateral medial rectus recessions on adjustable sutures. One patient had a large recess-resect procedure with an adjustable suture. Surgical follow-up ranged from 6 months to 4 years. Overall, the CHP was reduced to clinically insignificant in 13 (86.7%) of the 15 surgical cases (Table 2). The residual face turn in one of the other two patients responded to a base-out prism for the affected eye. Bilateral Lateral Rectus Paresis. The group of patients with bilateral lateral rectus paresis included 22 pa-
tients: 17 (77.3%) had bilateral partial paresis, and 5 (22.7%) total palsies. Among the 17 patients with partial paresis, only 2 (11.8%) had a CHP, and each had a face turn of30° toward the eye with more limited abduction. Surgery in the one patient consisted of bilateral medial rectus recessions; in the other, it involved a unilateral recess-resect procedure on the dominant eye that was affected by the more severe paresis. Both surgeries included adjustable sutures. 10 Both patients regained BSV in primary position and resolution of their CHP. Follow-up in these 2 patients was 6 months and 1 year after surgery. All 5 patients with bilateral total palsies showed large (45°) face turns in the directions of their dominant eyes. Four patients had surgery, of whom 3 underwent treatment consisting of bilateral botulinum injections followed by muscle transpositions.II,l3,14 One patient had no CHP after follow-up of 1 year. The second patient had a residual CHP due to a residual esotropia that was successfully treated with a base-out prism; the result has been stable for 2 years. The third patient required a reoperation involving bilateral medial rectus recessions to eliminate the primary position misalignment; this result has been stable for 2 years. The fourth patient had a large recess-resect procedure that eliminated a face turn; there has been no recurrence over 6 months. Thus, we have eliminated the CHP with surgery in 5 (83.3%) of 6 patients with bilateral palsies; the sixth responded to a prism after surgery. Group II: Superior Oblique Paresis Our review of the records identified 139 patients with superior oblique paresis, They ranged in age from 2 to 78 years (mean, 28.2 years); 88 (63.3%) were male. We considered the unilateral and bilateral cases separately. Unilateral Superior Oblique Paresis. One hundred and twenty-one of the 139 patients in group II had unilateral superior oblique paresis (Table 3). The right eye was affected in 62 (51.2%) of these cases. Of these 121 patients, 90 (74.4%) had a clinically significant CHP. The frequency of the CHP orientations are listed in Table 3. Of the 41 patients with combination CHP, 36 included a head tilt as part of the orientation. Of those patients who had head tilts as part of their CHP, the tilt was almost always to the shoulder contralateral to the involved eye, a common finding.1. 9 ,15 However, 2 patients tilted to the side ipsilateral to the paretic
Table 2. Data for Patients with Unilateral Lateral Rectus Paresis
Subgroup Partial paresis Complete palsy CHP
=
Total No.
Patients with Face Turns No. (%)
Range of Face Turns (Degrees)
51
11 (21.6)
20
8 (40,0)
15-30 30-45
No. with CHP who had Surgery
No. with Insignificant CHP after Surgery
8
8
7
5
compensatory head posture,
1303
Ophthalmology
Volume 99, Number 8, August 1992
Table 3. Orientations of CHP for Patients with Unilateral Superior Oblique Paresis
CHP
=
Orientation
Patients No. (%)
Head tilt alone Face turn alone Chin up alone Chin down alone Combinations No significant CHP Total
37 (30.6) 6 (5.0) 4 (3.3) 2 (1.6) 41 (33.9)* 31 (25.6) 121 (100.0)
compensatory head posture.
* Thirty six of these 41 combinations included a tilt as a component of
CHP.
eye to produce large separation of their diplopic images. 8,9,15 Patients with face turns as part of the CHP orientation showed turns to the side away from the involved eye, except for one patient who turned toward the affected side to separate the diplopic images. Thus, we found 3 (3.3%) of the 90 patients with CHP whose CHP was oriented opposite to that expected. These head postures are termed "paradoxic.,,8 We performed surgery on 32 of the patients with CHP. 16 The procedures were distributed as follows: inferior oblique weakening alone, 15; inferior oblique weakening and 1 vertical rectus recession, 13; inferior oblique weakening and a Harada-Ito transposition, 2; inferior oblique weakening and 2 vertical rectus recessions, 1; superior rectus recession, 1. Adjustable sutures were used in 10 patients who had vertical rectus recessions. In 10 patients, horizontal rectus surgery was added to the plan to treat a coexisting horizontal deviation. Follow-up ranged from 6 months to 3 years. Table 4 lists the surgical results for the 32 cases. Among the 23 (71.9%) surgical successes, 3 required reoperations. Two ofthese reoperations were required to treat unmasked contralateral superior oblique paresis I7 ,'8 that induced head tilts to the sides contralateral to the eyes with the newly uncovered paresis. The third patient had a reoperation to eliminate a vertical overcorrection caused by the first surgery. There was a residual CHP in 5 (15.6%) surgical patients, despite reoperations in 2. All 5 had residual deviations in primary position. Finally, we assigned 4 (12.5%) of the patients to the category of habit CHP because they demonstrated clinically significant residual CHP that persisted under monocular and binocular conditions despite successful elimination of their deviations in primary position. None of these four patients had subjective or objective excyclotorsion postoperatively. 19,20 Bilateral Superior Oblique Paresis. The group of patients with bilateral superior oblique paresis included 18 patients, of whom 9 (50.0%) showed a clinically significant CHP. All 9 patients had surgery. Their data, including CHP orientations and surgical procedures, are
1304
listed in Table 5. 21 Three of the patients had asymmetric bilateral palsies. Overall, the CHP was reduced to clinically insignificant in 8 (88.9%) of the 9 patients. Follow-up ranged from 6 months to 2 years. Group III: Duane Syndrome Types I and II Ninety-one patients with Duane syndrome types I and II were included in group III. They ranged in age from 8 months to 52 years (mean, 8.2 years); 53 (58.2%) were male. We analyzed the unilateral and bilateral cases separately. Unilateral Duane Syndrome. The group with unilateral Duane syndrome included 81 patients; 74 were type I and 7 were type 11.22-25 The data for these patients are recorded in Table 6. Overall, 55 (67.9%) of the 81 patients presented with a clinically significant CHP, all face turns. In almost all type I cases, a face turn was associated with a severe limitation of abduction resulting in an ET over 10 PO in primary position. Most cases with partial limitations of ductions in the involved eye did not adopt a clinically significant CHP. Six (85.7%) of the 7 type II patients had severe adduction limitations and large exotropias (XT) and adopted clinically significant CHP, all of which were face turns to the side contralateral to the involved eye. We performed surgery on 42 of the 55 patients with CHP; 37 were type I and 5 were type II. Follow-up ranged from 6 months to 3 years. Among the 37 type I patients, preoperative ET ranged from 10 to 35 PD. Initial surgery involved one medial rectus recession in 21 patients, bilateral medial recessions in 10, and bilateral medial recessions and a large lateral rectus recession (to treat retraction) in 6. 26 The postoperative results are listed in Table 7. Five patients required reoperations to obtain successful results: 4 required transposition procedures and 1 had bilateral medial rectus rerecessions. One patient with a residual CHP had a reoperation, a unilateral medial re-recession, without complete resolution of his CHP. The one patient listed as having an overcorrection developed an XT after are-recession of the medial rectus of his involved eye, creating a face turn of 20° to the side opposite the involved eye. Table 4. Results of Surgery for CHP in Patients with Unilateral Superior Oblique Paresis
CHP
=
Result
Patients No. (%)
No significant CHP Residual CHP HabitCHP Total
23 (71.9)* 5 (15.6) 4 (12.5) 32 (100.0)
compensatory head posture.
• Reoperations needed in 3 patients.
Kraft et al . Surgery for Compensatory Head Postures Table 5. Data for Patients with Bilateral Superior Oblique Paresis Patient No.
Age (yrs)
Sex
1 2 3 4 5
16 56 46 33 47
F M M F M
6
40
F
7
5
F
8 9
4 26
M F
Orientation of CHP Preoperatively Chin down Chin down Chin down Chin down Chin down Right turn Chin down Left tilt Left tilt Right tilt Right tilt Right turn
Surgery Performed
Postoperative CHP
Bilateral Harada-Ito Bilateral Harada-Ito Bilateral Harada-Ito Bilateral Harada-Ito Bilateral inferior oblique myectomies Right Harada-Ito
None None None Chin up, left turn' None
Bilateral inferior oblique myectomies Left inferior oblique myectomy Left superior rectus recess; right medial rectus recess-lateral rectus resect
None
None
None None
CHP = compensatory head posture . • Reoperation (bilateral inferior rectus recessions) failed to eliminate CHP.
Among the 5 type II patients, the preoperative XT ranged from 25 to 50 PD. Initial surgery involved bilateral lateral rectus recessions in three patients who had XT in primary position, and unilateral combined lateral rectus and superior rectus recessions in two patients who had XT and hypertropias in primary position. 26 The results are listed in Table 7. One patient had a residual 15 0 face turn due to a residual XT in primary position. Overall, the CHP was reduced to clinically insignificant in 37 (88.1 %) of the 42 unilateral cases. Bilateral Duane Syndrome. Our review of the records included 10 patients with bilateral Duane syndrome (5 females): 9 had bilateral type I syndromes, and 1 a type I in 1 eye and type II in the other. Seven (70%) of the patients, all with bilateral type I, had a clinically significant face turn, ranging from 15 0 to 35 0 • They all had severe limitations of abduction in the dominant eye. We performed surgery on 5 of the 7 patients, and in all cases we recessed both medial rectus muscles. 26 Three involved adjustable sutures. In 4 patients (80%), the CHP was reduced to clinically insignificant. One patient had a
clinically significant residual face turn postoperatively, with an ET of 20 PD in primary position.
Group IV: Brown Syndrome Thirty-five patients with Brown syndrome were included in group IV. They ranged in age from 9 months to 16 years (mean age, 5.3 years); 19 (54.3%) were female. Two patients had a bilateral syndrome, one of which was transient in association with rheumatoid arthritis. In 19 (57.6%) of the 33 patients with unilateral cases, the right eye was affected. Only 5 (15.2%) of the unilateral cases had a clinically significant CHP according to our criteria. They were distributed as follows: face turn alone, 2; chin-up alone, 1; combination of face turn and chin-up, 2. All 5 patients had surgery that consisted of superior oblique tenotomies. 27 The CHP was reduced to clinically insignificant in all 5 patients, with follow-up ranging from 6 months to 2 years. No patient has developed a postoperative superior oblique palsy.27-29
Table 6. Data on Patients with Unilateral Duane Syndrome Type of Duane Syndrome II Total
Total No. of Eyes
Left Eyes Affected No. (%)
Patients with Significant Face Turns' No. (%)
Range of Face Turns (degree)
74
56 (75.7) 2 (28.6) 58 (71.6)
49 (66.2) 6 (85.7) 55 (67.9)
15-35 15-45 15-45
7 81
• Minimum face turn of 15 0 •
1305
Volume 99, Number 8, August 1992
Ophthalmology
The patient with bilateral Brown syndrome associated with arthritis had a chin-up posture, whereas the patient with the congenital bilateral syndrome did not adopt any CHP. Neither patient underwent surgery. Group V: Congenital Motor Nystagmus Group V consisted of 23 patients with a CHP who had congenital motor nystagmus with conjugate null zones and no strabismus. 4- 7 Their age ranges before surgery were 4 to 22 years (mean, 9.5 years); 17 (73.9%) were male. Face Turns. Twenty patients presented with face turns as their CHP, 12 (60%) directed to their left sides. Of these, 6 measured between 15° and 30° (moderate), while 14 measured from 35° to 50° (large). These patients all underwent Kestenbaum-Anderson procedures based on the Parks 5-6-7-8 rule for 15° to 20° of face turn. 5 ,6,30 For turns greater than 20°, these numbers all were increased in proportion to the size of face tum, up to a maximum of 60% increase for a 50° turn. 5- 7 The results of surgery in these 20 patients with face turns are listed in Table 8. Follow-up ranged from 6 months to 4 years. In the subgroup of moderate turns, 2 patients were left with residual turns, 1 of 15 ° and the other of 20°. In the subgroup of large turns, 1 patient required a reoperation to eliminate a residual 15 ° face tum after initially having an augmented KestenbaumAnderson procedure. The 3 patients with residual CHP had turns ranging from 15 ° to 25 0. Overall, we successfully reduced CHP to clinically insignificant in 15 (75.0%) of the 20 cases with face turns. Miscellaneous CHP. We had 3 nystagmus patients with CHP other than pure face turns who underwent surgery. One had a chin-up posture of 10° to 15°. He was successfully treated with bilateral inferior rectus recessions of6 mm. A second patient presented with a combined left face tum of 30° and chin-up posture of 10°. He underwent a combination of horizontal rectus recessions (right lateral recession 10.0 mm, left medial recession 8.0 mm) with
Table
7. Results of Surgery for Unilateral Duane Syndrome with CHP
Postoperative CHP Compared with Preoperative
No.
(%)
None or clinically insignificant Residual but improved Not improved Total
33 3 1 37
89.2" 8.1t
CHP
=
Type I
2·n
100.0
Type II
. No. 4 1 0
5
(%)
80.0 20.0 0.0 100.0
compensatory head posture.
• Five patients required reoperations to eliminate CHP.
t t
One patient had a residual CHP despite reoperation. Patient was overcorrected after reoperation for residual face turn.
1306
Table 8. Results of Surgery for Congenital Nystagmus with Face Turn Size of Preoperative Turn Postoperative Face Turn Compared with Preoperative
No.
(%)
No.
(%)
None or clinically insignificant Residual but improved Overcorrected Total
4 2 0 6
66.7 33.3 0.0 100.0
11 3 0
78.6" 21.4 0.0 100.0
15°
to
30°
35° to 50°
14
• One patient required bilateral reoperation to eliminate residual face turn.
bilateral 5.0 mm inferior rectus recessions. He had a residualleft tum of 10°, which was cosmetically acceptable. Finally, we had one patient with a right head tilt of 15 ° with a left tum of 5 ° to 10°. We weakened the right superior oblique with a 6-mm tendon expander31 and myectomized the left inferior oblique 10 mm. The head tilt was eliminated, but he has a residual 10° left face tum that has not increased over 9 months of follow-up.
Discussion Before treating a patient with CHP, the physician must first determine if it is caused by an ocular or nonocular problem. The ophthalmologist's obligation is to rule out all possible ocular causes in the differential diagnosis. If no ocular cause is found, then nonocular possibilities remain as the diagnoses of exclusion. If an ocular cause is diagnosed, then therapy to eliminate the cause also should lead to resolution of the CHP. However, some patients with long-standing head postures will, despite successful treatment of the underlying condition, continue to demonstrate a habit CHP. 2 These are usually attributed to alterations in the tone of the neck muscles that may not reverse even if the ocular causes are rectified. We found four such patients, all in our group of unilateral superior oblique paresis corrected with surgery. Thus, it is important to educate primary care physicians to refer any child with a CHP for ophthalmologic assessment so that ocular causes can be ruled out. 2 •3 Compensatory head postures yield a number of possible advantages for patients when there are ocular causes. A patient may gain improved visual acuity, as in cases of nystagmus or refractive error, or avoid diplopia and maintain fusion, as seen with incomitant strabismus. A CHP can be a consequence of a patient's need to fixate with a dominant eye affected by a muscle imbalance, or simply a necessity for vision, as in cases ofblepharoptosis . Rarely, patients can intentionally maximize the separation of diplopic images if they have incomitant strabismus with intractable diplopia.
Kraft et al . Surgery for Compensatory Head Postures However, there are many reasons patients with these diagnoses do not universally demonstrate CHP. In cases of congenital nystagmus, the null zone may be close enough to primary position that no significant head posturing is required. Patients with unilateral ptosis may fixate with the nonaffected eye. This is a concern in children with early-onset blepharoptosis because of the risk of amblyopia in the affected eye. In patients with strabismus, there are various possible explanations why they may not adopt a CHP, and all applied to at least some patients in groups I to IV in our study. First, the most common reason is the presence of amblyopia or suppression, which causes the patient to fixate with the preferred eye and have no stimulus for fusion that would promote a CHP. 9 Second, a patient may develop larger-than-normal fusional amplitudes so that BSV can be maintained in primary position. This also may allow a good range of BSV field around primary position. This is usually seen in patients with long-standing paretic strabismus or muscle paresis that is very slowly progressive, as seen in cases of superior oblique or lateral rectus paresis. 8 .9 Third, an eye affected by a unilateral muscle imbalance, either mechanical or paretic, may have some range of movement into the affected field of gaze causing minimal or no deviation in primary position and a range of BSV field to the affected side. This is frequent in patients with mild Duane syndromes or partial sixth nerve paresis, for example. Finally, a patient with diplopia because of an acquired strabismus may ignore the displaced image and maintain a normal head posture without symptoms. 1,9 Among our patients with unilateral superior oblique palsies, we found that 3 (3.3%) of the 90 patients with CHP had paradoxic postures, a term we use for a CHP opposite to that which we would expect a patient to adopt to maximize separation of diplopic images. 8 ,9 Our incidence is similar to the 4.9% (7 of 144 cases) reported by von Noorden et al 8 in their series of patients with unilateral superior oblique palsies with CHP. Only 21 of our 93 cases oflateral rectus paresis (group I) had surgery. A number offactors explain why a limited number of patients had surgery. Partial paresis in many patients resolved spontaneously with time, especially when associated with head trauma or vascular causes. Eight of these received botulinum injections into the medial rectus muscles of the involved eyes. I I - 14 Some patients had sixth nerve palsies that were caused by intracranial tumors, and eye muscle surgery was not pursued because of poor prognosis for the underlying conditions. Many patients had small CHP, smaller than our minimal criteria for clinical significance, and refused surgical intervention since their range ofBSV field was large enough for them to continue their daily activities. We treated some patients with partial paresis with prisms to eliminate diplopia in primary position. Finally, in some patients with unilateral cases, the underlying condition led to such reduced vision in the affected eye that the patients did not experience diplopia and, consequently, had no reason to adopt a CHP. Our incidence of CHP in superior oblique paresis
(group II) was 71.2%, almost exactly the same as the 67.8% reported by von Noorden et al. 8 Our surgical success rate in eliminating the CHP was 75.6%, or slightly less than their reported 89% in 96 surgeries. However, they recorded data only on head tilts for their CHP, whereas we included all combinations of tilts, chin-up or chin-down, or face turns, and combinations of these among our patients with CHP for both our preoperative and postoperative analyses. In addition, we had four patients with habit tilts after surgery, which slightly reduced our success rate. We did not find any patients in the subgroup of unilateral superior oblique palsies in whom we believed that the cyclotorsion alone was responsible for a head tilt posture. Guyton,19 von Noorden and Ruttum,15 and von Noorden 20 have suggested that, occasionally, cyclotorsion of one eye may induce a compensatory tilt to the opposite shoulder. In such a case, a patient would have a head tilt when the involved eye fixates and would have no CHP when the uninvolved eye fixates. This contrasts with the much more common situation in which the head tilt is caused by an incomitant vertical tropia, with a forced head tilt difference between the vertical deviation on right versus left head tilt. 8,15-19 In this more common scenario, the head tilt would manifest under binocular conditions as the patient adopts a CHP to reduce the size of the vertical deviation, but it would vanish when either eye fixates as the patient is rendered diplopia-free. Our 68.1 % incidence of CHP in patients with Duane syndrome (group III) is comparable with the 77.4% reported by Raab 23 and higher than the 51 % reported by O'Malley et al. 25 Our 87.2% surgical success is slightly higher than the 79% reported by Pressman and Scott.24 Duane syndrome is a spectrum of motility disorders with variable degrees oflimited ductions, so affected eyes may have muscle forces that are balanced at or near the primary position. 26 Thus, a large proportion of patients with Duane syndrome do not have a deviation in primary position large enough to cause a CHP. We had a relatively low incidence ofCHP in our Brown syndrome group (group IV), only 16 (17.4%) of 35 patients in contrast to an incidence of 35% reported by Wilson et al. 27 However, they did not indicate the magnitudes of the CHP in their patients, so some patients may have had mild postures that would not have been labeled clinically significant according to the criteria we used in this study. We performed surgeries on all five patients with unilateral Brown syndrome who also had clinically significant CHP, all of which were successful. Sprunger et al29 eliminated a clinically significant CHP in 17 (70.8%) of 24 patients with Brown syndrome whom they treated with surgery. A few other patients among our unilateral cases had surgery to treat a cosmetically noticeable downshoot in adduction of the affected eye, but we did not address the surgical results in these patients in this study since they had clinically insignificant or no CHP preoperatively. Finally, we successfully reduced the CHP to clinically insignificant in 18 (78 .3%) of 23 patients with congenital motor nystagmus with CHP and no strabismus. This compares to success rates reported by other authors 5- 7 in
1307
Ophthalmology
Volume 99, Number 8, August 1992
recent years since augmented surgery has been increasingly used for face turns over 25 0 to 30 0 • Like these authors, we have found that very large amounts of surgery used to treat large face turns inevitably lead to limited ductions in one direction of gaze. In our experience this is a necessary trade-off to ensure a stable result. None of our patients had their CHP overcorrected by the surgery, even with augmented amounts of 50% or 60%. It should be emphasized that not all patients with CHP due to ocular causes require surgery. In our study, 193 (53.9%) of 358 patients in the four strabismus groups (groups I to IV) presented with clinically significant CHP. Of these 186 patients, only 114 (61.3%) underwent surgery. There are several reasons a large number of patients with CHP caused by strabismus do not undergo surgery. The CHP may be transient and vanish as a transient strabismus resolves. Prisms, patching of one eye, and botulinum toxin injections can be useful treatments in selected instances to help patients cope with the conditions causing their CHP. Finally, some patients are only minimally bothered by their CHP, or their CHP can be small at the outset and require no treatment. In our entire series of 137 surgeries to correct CHP in our 5 patient groups (i.e., including the 23 patients with congenital nystagmus), we reduced the CHP to clinically insignificant in 113 (82.5%) patients. Because CHP can be caused by several ocular disorders, the ophthalmologist must play an integral role in the work-up of a patient who presents with an abnormal head position. From the results of our study, we conclude that if an ocular cause that is amenable to surgical correction is found, then an appropriate surgical plan will have a high chance of eliminating the CHP.
References 1. von Noorden GK. Atlas of Strabismus, 4th ed. St. Louis: CV Mosby, 1983;128-41. 2. Morris JE. Head posturing in horizontal strabismus. Am Orthopt J 1983;33:2-6. 3. Helveston EM. Strabismus surgical treatment for abnormal head posturing. Am Orthopt J 1983;33:32-5. 4. Dell'Osso LF, Flynn JT. Congenital nystagmus surgery: a quantitative evaluation of the effects. Arch Ophthalmol 1Q79;97:462-9. 5. Scott WE, Kraft SP. Surgical treatment of compensatory head position in congenital nystagmus. J Pediatr Ophthalmol Strabismus 1984;21 :85-95. 6. Mitchell PR, Wheeler MB, Parks MM. Kestenbaum surgical procedure for torticollis secondary to congenital nystagmus. J Pediatr Ophthalmol Strabismus 1987;24:87-92. 7. Nelson LB, Ervin-Mulvey LD, Calhoun JH, et al. Surgical management for abnormal head position in nystagmus: the augmented modified Kestenbaum procedure. Br J Ophthalmol 1984;68:796-800. 8. von Noorden GK, Murray E, Wong SY. Superior oblique paralysis. A review of270 cases. Arch OphthalmoI1986;104: 1771-6.
1308
9. Burian HM, Rowan PJ, Sullivan MS. Absence of spontaneous head tilt in superior oblique palsy. Am J Ophthalmol 1975;79:972-7. 10. Kraft SP, Jacobson ME. Techniques of adjustable suture strabismus surgery. Ophthalmic Surg 1990;21 :633-40. 11. Kraft SP, Allan K. The use of botulinum toxin in the management of sixth nerve palsy. Am Orthopt J 1989;39:8997. 12. Scott AB, Kraft SP. Botulinum toxin injection in the management of lateral rectus paresis. Ophthalmology 1985;92: 676-83. 13. Fitzsimons R, Lee JP, Elston J. Treatment of sixth nerve palsy in adults with combined botulinum toxin chemodenervation and surgery. Ophthalmology 1988;95: 1535-42. 14. Rosenbaum AL, Kushner BJ, Kirschen D. Vertical rectus muscle transposition and botulinum toxin (oculinum) to medial rectus for abducens palsy. Arch Ophthalmol 1989; 107:820-3. 15. von Noorden GK, Ruttum M. Torticollis in paralysis of the trochlear nerve. Am Orthopt J 1983;33:16-20. 16. Scott WE, Kraft SP. Classification and surgical treatment of superior oblique palsies: I. Unilateral superior oblique palsies. In: Pediatric Ophthalmology and Strabismus: Trans New Orleans Acad Ophthalmol. New York: Raven Press, 1986;15-38. 17. Kraft SP, Scott WE. Masked bilateral superior oblique palsy: clinical features and diagnosis. J Pediatr Ophthalmol Strabismus 1986;23:264-72. 18. Kushner BJ. The diagnosis and treatment of bilateral masked superior oblique palsy. Am J Ophtha1mol 1988;105:18694. 19. Guyton DL. Clinical assessment of ocular torsion. Am Orthopt J 1983;33:7-15. 20. von Noorden GK. Clinical observations in cyclodeviations. Ophthalmology 1979;86: 1451-61. 21. Scott WE, Kraft SP. Classification and treatment of superior oblique palsies: II. Bilateral superior oblique palsies. In: Pediatric Ophthalmology and Strabismus: Trans New Orleans Acad Ophthalmol. New York: Raven Press, 1986;265-91. 22. Huber A. Electrophysiology of the retraction syndromes. Br J Ophthalmol 1974;58:293-300. 23. Raab EL. Clinical features of Duane's syndrome. J Pediatr Ophthalmol Strabismus 1986;23:64-8. 24. Pressman SH, Scott WE. Surgical treatment for Duane's syndrome. Ophthalmology 1989;93:29-38. 25. O'Malley ER, Helveston EM, Ellis FD. Duane's retraction syndrome-plus. J Pediatr Ophthalmol Strabismus 1982; 19: 161-5. 26. Kraft SP. A surgical approach for Duane syndrome. J Pediatr Ophthalmol Strabismus 1988;25: 119-30. 27. Wilson ME, Eustis HS, Parks MM. Brown's syndrome. Surv Ophthalmol 1989;34: 153-72. 28. Eustis HS, O'Reilly C, Crawford JS. Management of superior oblique palsy after surgery for true Brown's syndrome. J Pediatr Ophthalmol Strabismus 1987;24: 10-16. 29. Sprunger DT, von Noorden GK, Helveston EM. Surgical results in Brown syndrome. J Pediatr Ophthalmol Strabismus 1991;28:164-7. 30. Parks MM. Congenital nystagmus surgery. Am Orthopt J 1973;23:35-9. 31. Wright KW. Superior oblique silicone expander for Brown syndrome and superior oblique overaction. J Pediatr Ophthalmol Strabismus 1991;28:101-7.
