The surgical treatment of total sixth cranial nerve palsy can be challenging. Affected individuals have severe limitation of abduction in the involved eye associated with esotropia, diplopia, and absence of fusion.1–4 Strabismus surgery is performed to restore ocular alignment in primary position and alleviate diplopia.5 The preferred surgical procedures include a variety of partial- or full-tendon vertical rectus muscle transpositions (VRTs) to the lateral rectus muscle insertion.1–3,5–11 These can be augmented by either posterior fixation sutures or resection of the transposed muscles.1,2,10–14 In addition, VRT may be combined with weakening of the ipsilateral medial rectus muscle by recession or injection of botulinum toxin.1,3,5,6
Although prior studies have assessed the surgical correction achieved by various VRT procedures, little is known about the stability of the immediate postoperative alignment. The purpose of our study was to evaluate both the amount of correction obtained and the postoperative drift that occurred with two frequently used VRT procedures to determine the ideal immediate postoperative alignment.
Patients and Methods
A retrospective review of medical records was performed for patients with total sixth cranial nerve palsy who underwent VRT by three different surgeons at the University of Miami. This study was approved by the institutional review board of the University of Miami, was performed in compliance with the Health Insurance Portability and Accountability Act, and adhered to the tenets of the Declaration of Helsinki. Financial claims data were searched to identify patients with total sixth cranial nerve palsy who underwent VRT procedures. Total sixth cranial nerve palsy was defined by an inability to abduct past midline with −4 or greater abduction limitation. In patients with bilateral sixth cranial nerve palsy, only the first operated eye was included. Patients with less than 2 months of follow-up and patients who received botulinum toxin injections to the medial rectus muscle were excluded from the study.
The following data were extracted from the medical record: VRT type, whether a medial rectus recession (MRR) was performed at the time of the VRT, preoperative deviation, postoperative day 2 deviation, and postoperative month 2 deviation. The two surgical procedures that were included were: full-tendon transposition with Foster augmentation by posterior fixation suture to the sclera approximately 8 mm posterior to the lateral rectus insertion (FTT+FA)12 and partial-tendon transposition with a 4-mm resection of the transposed segment of the vertical rectus muscle and simultaneous medial rectus recession (PTT+R+MRR).14 The degree of medial rectus contracture was determined by forced duction testing. For the patients with a tight medial rectus muscle, the PTT+R+MRR was the preferred procedure. Patients with adjustable sutures were adjusted in the clinic either the same day of surgery or the morning after. Ocular alignment in the primary position was measured at distance by the alternate prism cover test. Preoperative and postoperative month 2 deviations (in prism diopters [PD]) were compared to determine the total correction. Postoperative day 2 and postoperative month 2 deviations were compared to determine the total postoperative drift.
One- and two-way frequency distributions were created for categorical variables. The continuous outcome variables (drift and correction) were tested for normal distributions. Based on the results of these normal distribution tests, appropriate statistics (parametric and/or non-parametric) were used to compare the difference in drift and correction between the procedures.
The main analysis dataset included only one eye per person because, with such a small data set and the need to use non-parametric statistics, no model could be built to account for the correlation between two eyes of a single person. As a sensitivity analysis, the entire analysis was re-run using the same statistical tests, including the “second eyes” and the “first eyes.” All analyses were done using SAS software (version 9.4; SAS, Inc., Cary, NC). A P value of less than .05 was considered statistically significant.
Twenty-seven patients who underwent VRT for total sixth cranial nerve palsy were included. Sixteen underwent FTT+FA and 11 had PTT+R+MRR. Of the 4 patients with bilateral sixth cranial nerve palsy, 3 underwent FTT+FA and 1 underwent PTT+R+MRR. Only 1 patient with a prior MRR had a reoperation. The mean preoperative deviation was 53 PD (range: 35 to 90 PD) in the FTT+FA group and 55 PD (range: 35 to 95 PD) in the PTT+R+MRR group. Table 1 demonstrates the distance preoperative deviation, postoperative day 2 and month 2 deviations, and postoperative drift and total net correction at month 2 for each patient.
Preoperative and Postoperative Deviations, Drift, and Net Correction for Each Surgical Procedure (N = 27)
The mean preoperative alignment in patients who had FTT+FA was 53 PD of esotropia (range: 35 to 90 PD of esotropia). The mean postoperative day 2 deviation was 9 PD of esotropia (range: 12 PD of exotropia to 45 PD of esotropia), whereas the mean postoperative month 2 deviation was 17 PD of esotropia (range: 12 PD of exotropia to 75 PD of esotropia). Three patients were orthotropic at the 2-month postoperative follow-up. There was a mean total esotropic drift of 6 ± 9 PD, but it ranged from 10 PD of exotropic drift to 25 PD of esotropic drift. The total correction at 2 months was 40 ± 13 PD (range: 15 to 68 PD). This procedure exhibited the smaller postoperative drift and total correction of the two procedures.
The mean preoperative alignment in patients who underwent PTT+R+MRR was 55 PD of esotropia (range: 35 to 95 PD of esotropia). The mean MRR was 5.2 mm after adjustment (range: net correction 1 to 11 mm). One patient with prior recession of the medial rectus muscle underwent a re-recession at the time of the PTT+R+MRR, which accounted for an 11-mm recession. The mean deviation was 13 PD of exotropia (range: 35 PD of exotropia to 4 PD of esotropia) on postoperative day 2 and 4 PD of esotropia (range: 15 PD of exotropia to 20 PD of esotropia) on postoperative month 2. At the 2-month follow-up, 2 patients were orthotropic (25%). There was a mean total esotropic drift of 16 ± 11 PD (range: 5 PD of exotropic drift to 33 PD of esotropic drift). The mean total correction at 2 months was 52 ± 19 PD (range: 27 to 87 PD). This procedure demonstrated the larger total correction and postoperative drift.
Correction and Drift
Figure 1 demonstrates the overall correction from the preoperative visit to the postoperative month 2 visit for the two VRT procedures. Figure 2 demonstrates the net drift between the postoperative day 2 and month 2 visits. There was a trend toward greater correction with PTT+R+MRR than FTT+ FA, but the difference was not statistically significant (mean difference: 11, 95% confidence interval: −1 to 24, P = .071). The difference in the amount of drift was statistically significant (mean difference: −11, 95% confidence interval: −19 to −3, P = .009). If surgical success is defined as less than 10 PD of horizontal deviation, the success rate was 73% (8 of 11) for PTT+R+MRR and 44% (7 of 16) for FTT+FA.
Box plot of net correction of the deviation in prism diopters between the preoperative and postoperative month 2 visits of the two vertical rectus transposition procedures. FTT = full-tendon transposition; PTT+R and simultaneous MRR = partial-tendon transposition with resection and simultaneous medial rectus recession
Box plot of drift in prism diopters between the postoperative day 2 and postoperative month 2 visits. Positive drifts are esotropic and negative drifts are exotropic. FTT = full-tendon transposition; PTT+R and simultaneous MRR = partial-tendon transposition with resection and simultaneous medial rectus recession
The surgical treatment of total sixth cranial nerve palsy will ideally include a transposition procedure because resection of a paralytic muscle has limited long-term effect. Methods to balance the adducting force generated by the medial rectus muscle are necessary and can be achieved by VRT. The degree of function of the lateral rectus muscle, degree of contracture of the ipsilateral medial rectus muscle, and preoperative deviation and anticipated postoperative correction and drift are crucial in selecting a procedure for total sixth cranial nerve palsy. Thus, we sought to evaluate the magnitude of change and amount of postoperative drift for two commonly used VRT procedures.
Depending on the type of VRT, the amount of correction varies greatly. Prior studies have shown that FTT procedures correct 30 to 50 PD of esotropia.1,3,13 When combined with resection of the vertical rectus muscles, VRT can correct 33 to 64 PD of esotropia.2 The addition of a medial rectus recession or posterior fixation suture further increases the total correction up to 70 PD (range: 25 to 70 PD).1–3,6,10–12 The Foster augmentation increases the total correction for a full tendon transposition due to the increased tonic abducting force of the transposed muscle.1,2,10–12 The addition of a medial rectus recession decreases the antagonistic tonic force of the medial rectus muscle, allowing for increased abduction of the eye.
When comparing the two groups, the preoperative alignment was similar (53 PD of esotropia in the FTT+FA group vs 55 PD of esotropia in the PTT+R+MRR group). Despite the similar preoperative magnitude of esotropia, the net correction was larger in the PTT+R+MRR group (52 PD) than in the FTT+FA group (40 PD). However, this difference was not statistically significant (P = .071).
The net correction of the PTT+R+MRR procedure ranged from 27 to 87 PD. This is similar but slightly larger than the range reported by Couser et al.14 (range: 29 to 53 PD). In addition, the mean correction in our patients (52 PD) tended to be larger when compared to Couser et al., who found a mean correction of 39 PD. This may be explained by the fact that their cohort had a smaller preoperative deviation (43 vs 55 PD) and a slightly smaller average medial rectus recession (4.9 vs 5.2 mm). The net correction of the FTT+FA procedure averaged 40 PD (range: 15 to 68 PD). The mean correction in our series is similar to that of other authors, who reported a mean of 41.2 PD (range: 37 to 72 PD),10 41.3 PD (range: 25 to 70 PD),2 and 47.5 PD (range: 30 to 80 PD).12
An advantage of PTT+R+MRR is that the medial rectus muscle can be adjusted postoperatively. However, to determine the optimal alignment at the time of adjustment, the postoperative drift should be taken into account. Few studies have investigated or compared the postoperative drift that occurs after VRT procedures used to surgically treat total sixth cranial nerve palsies. Our results show that both procedures had a mean esotropic drift at postoperative month 2, but the mean drift was greater for PTT+R+MRR (16 PD) than FTT+FA (6 PD). This difference was statistically significant (P = .009). We postulate that this may be due to the increase in the abducting tone of transposed muscles that results from the posterior fixation augmentation, which can possibly minimize the esotropic drift.
In contrast to our results, Agarwal et al.15 reported a postoperative exotropic drift following superior rectus transpositions in their 10 patients with sixth cranial nerve palsy. This difference may be attributed to the different transposition procedure, longer follow-up period, and larger mean MRR (5.6 mm in Agarwal et al. vs 5.2 mm in our study). Another contributing factor may be that 7 of the 10 patients also had posterior fixation augmentation sutures, which may have prevented the esotropic drift by enhancing the effect of the VRT.
In our study, 4 patients had an exotropic drift. The only patient in the PTT+R+MRR group who had an exotropic draft was the patient who underwent a re-recession of the medial rectus muscle to a position 15 mm from the limbus. In addition, 3 of 16 patients in the FTT+FA had an exotropic drift at the 2-month postoperative visit. These findings, along with Agarwal et al.'s results, suggest that larger recessions of the medial rectus muscle and posterior fixation augmentation sutures may be associated with less esotropic drift after VRT.
Our study had several limitations, including the retrospective nature of chart review, a relatively small patient population, and a different number of patients in each surgical group. We did not have complete preoperative and postoperative data regarding the limitation to abduction or ocular torticollis. A wide range of correction was obtained for each group, which could be due to multiple surgeons with differing techniques. This might also have resulted from patients with a −4 duction deficit being compared to those with −5 or worse deficits. It is also possible that some patients may have had only a partial palsy and a tight medial rectus muscle that masked some of the abducting capability. A final discrepancy could lie in the fact that there were different examiners assessing and measuring alignment, thus introducing another source of variability. Additionally, there was a wide range in the postoperative drift and it is possible that additional changes in alignment could occur at postoperative time intervals longer than 2 months. Despite these limitations, our study provides valuable information because it directly compares two VRT procedures by evaluating not only the postoperative correction, but also the immediate postoperative drift, which has not been reported to this extent in prior studies.
If surgical success is defined as less than 10 PD of horizontal deviation, the success rate of PTT+R+MRR was 73% and that of FTT+FA was 44%. We identified that PTT+R+MRR had a larger correction despite a greater postoperative esotropic drift when compared to FTT+FA. With this in mind, targeting a small immediate postoperative overcorrection may be desirable in PTT procedures.
- Gunton KB. Vertical rectus transpositions in sixth nerve palsies. Curr Opin Ophthalmol. 2015;26:366–370. doi:10.1097/ICU.0000000000000178 [CrossRef]
- del Pilar Gonzalez M, Kraft SP. Outcomes of three different vertical rectus muscle transposition procedures for complete abducens nerve palsy. J AAPOS. 2015;19:150–156. doi:10.1016/j.jaapos.2015.01.007 [CrossRef]
- Lee YH, Lambert SR. Outcomes after superior rectus transposition and medial rectus recession versus vertical recti transposition for sixth nerve palsy. Am J Ophthalmol. 2017;177:100–105. doi:10.1016/j.ajo.2017.02.019 [CrossRef]
- Yazdian Z, Rajabi MT, Ali Yazdian M, Rajabi MB, Akbari MR. Vertical rectus muscle transposition for correcting abduction deficiency in Duane's syndrome type 1 and sixth nerve palsy. J Pediatr Ophthalmol Strabismus. 2010;47:96–100. doi:10.3928/01913913-20100308-07 [CrossRef]
- Nabie R, Andalib D. Augmented vertical recti transposition with intraoperative botulinum toxin for complete and chronic sixth nerve palsy. Eye (London). 2017;31:148–151. doi:10.1038/eye.2016.226 [CrossRef]
- Patil-Chhablani P, Kothamasu K, Kekunnaya R, Sachdeva V, Warkad V. Augmented superior rectus transposition with medial rectus recession in patients with abducens nerve palsy. J AAPOS. 2016;20:496–500. doi:10.1016/j.jaapos.2016.07.227 [CrossRef]
- Singh P, Vijayalakshmi P, Shetty S, Vora P, Kalwaniya S. Double augmented vertical rectus transposition for large-angle esotropia due to sixth nerve palsy. J Pediatr Ophthalmol Strabismus. 2016;53:369–374. doi:10.3928/01913913-20160810-01 [CrossRef]
- Nowakowska O, Loba P, Broniarczyk-Loba A. The efficacy of vertical rectus transposition and its modalities in patients with abducens nerve palsy. Eur J Ophthalmol. 2011;21:223–227. doi:10.5301/EJO.2010.3250 [CrossRef]
- Bansal S, Khan J, Marsh IB. Unaugmented vertical muscle transposition surgery for chronic sixth nerve paralysis. Strabismus. 2006;14:177–181. doi:10.1080/09273970601026201 [CrossRef]
- Simons BD, Siatkowski RM, Neff AG. Posterior fixation suture augmentation of full-tendon vertical rectus muscle transposition for abducens palsy. J Neuroophthalmol. 2000;20:119–122. doi:10.1097/00041327-200020020-00012 [CrossRef]
- Struck MC. Augmented vertical rectus transposition surgery with single posterior fixation suture: modification of Foster technique. J AAPOS. 2009;13:343–349. doi:10.1016/j.jaapos.2009.02.009 [CrossRef]
- Foster RS. Vertical muscle transposition augmented with lateral fixation. J AAPOS. 1997;1:20–30. doi:10.1016/S1091-8531(97)90019-7 [CrossRef]
- Park KA, Lyu I, Yoon J, et al. Muscle union procedure in patients with paralytic strabismus. PLoS One. 2015;10:e0129035. doi:10.1371/journal.pone.0129035 [CrossRef]
- Couser NL, Lenhart PD, Hutchinson AK. Augmented Hummelsheim procedure to treat complete abducens nerve palsy. J AAPOS. 2012;16:331–335. doi:10.1016/j.jaapos.2012.02.015 [CrossRef]
- Agarwal R, Sharma M, Saxena R, Sharma P. Surgical outcome of superior rectus transposition in esotropic Duane syndrome and abducens nerve palsy. J AAPOS. 2018;22:12–16. doi:10.1016/j.jaapos.2017.10.004 [CrossRef]
Preoperative and Postoperative Deviations, Drift, and Net Correction for Each Surgical Procedurea (N = 27)
|Preoperative Deviation (PD)||Postoperative Day 2 Deviation (PD)||Postoperative Month 2 Deviation (PD)||ET Drift (PD)||Correction (PD)|
|FTT+FA (n = 16)|
| 85 ET||20 ET||30 ET||10||55|
| 90 ET||45 ET||50 ET||5||40|
| 50 ET||12 ET||15 ET||3||35|
| 45 ET||15 ET||12 ET||3 XT||33|
| 50 ET||6 XT||2 ET||8||48|
| 64 ET||10 ET||30 ET||20||34|
| 54 ET||Ortho||5 ET||5||49|
| 45 ET||10 XT||Ortho||10||45|
| 35 ET||12 XT||12 XT||0||47|
| 50 ET||10 ET||Ortho||10 XT||50|
| 70 ET||6 ET||2 ET||4 XT||68|
| 35 ET||Ortho||Ortho||0||35|
| 60 ET||20 ET||25 ET||5||35|
| 50 ET||Ortho||25 ET||25||25|
| 35 ET||10 ET||20 ET||10||15|
| 35 ET||Ortho||4 ET||4||31|
|PTT+R+MRR (n = 11)|
| 50 ET||10 XT||15 XT||5 XT||65|
| 95 ET||(8 XT) Ortho||8 ET||8||87|
| 35 ET||14 XT||2 ET||16||33|
| 35 ET||6 XT||8 ET||14||27|
| 45 ET||(25 XT) 20 XT||2 ET||22||43|
| 80 ET||4 ET||20 ET||16||60|
| 58 ET||4 XT||Ortho||4||58|
| 35 ET||30 XT||Ortho||30||35|
| 50 ET||(6 XT) 8 XT||16 ET||24||34|
| 66 ET||35 XT||2 XT||33||68|
| 60 ET||(14 XT) 16 XT||2 ET||18||58|