Dissociated vertical deviation (DVD) is a poorly understood ocular motor phenomenon characterized by a slow disconjugate vertical drifting of the non-fixing eye.1 It is usually bilateral, variable, asymmetric, and associated with latent or manifest latent nystagmus.2 Along with the upward drift there is also an exodeviation and excyclotorsional movement of the eye, together referred as “dissociated strabismus complex”.3 Although it is more commonly noted in association with early-onset strabismus,4-6 it is not infrequently seen with late-onset acquired and sensory deviations.7-9
Due to complex motor disturbances present in DVD, the success rate of surgical treatment is lower than that in other strabismus forms. Over the years, a variety of different surgical procedures have been described for the treatment of DVD such as weakening of the inferior oblique (IO),10,11 anteriorization of the IO muscle alone12-15 or combined with resection,16 recession of the superior rectus (SR) muscle alone17-20 or combined with a posterior fixation suture,21,22 SR posterior fixation suture alone,22 resection or plication of the inferior rectus,23,24 and all 4 oblique muscle-weakening procedures.25,26
The magnitude of DVD in different gazes and asymmetry between the 2 eyes are important factors when planning the type and amount of surgery.27 In cases of DVD without significant oblique muscle dysfunction, unilateral or bilateral SR recession is the preferred option for treatment. Several authors have studied the results of unilateral SR recession for asymmetric DVD or DVD manifesting predominantly in 1 eye.17,18,21 However, there is an increased risk of decompensation of DVD in the non-operated eye with such an approach.23,28 There are few studies that mentioned results of bilateral SR recession, albeit with small sample sizes.19,20,22 Thus, in the present study we evaluated the results of bilateral symmetric and asymmetric SR recessions in patients with bilateral DVD without oblique dysfunction and the factors that might influence the outcome of the surgical treatment of this disease.
A retrospective review of medical records of patients who had undergone bilateral SR recession for bilateral DVD from January 2012 to December 2016 at Dr Shroff's charity eye hospital in New Delhi, India was performed. The study adhered to the tenets of the Declaration of Helsinki and was approved by the institutional review board of the hospital. Informed consent was also obtained from the patients. Only those patients who had cooperated for a prism under cover test to measure the DVD were included in the study. Patients with significant oblique muscle dysfunction (grade 2 or more) were excluded. Inferior oblique muscle overaction was graded based on the upward displacement of the eye in adduction during lateral versions: grade 1 - no hypertropia in adduction but slight upward displacement in the field of action of IO; grade 2 - small hypertropia in adduction; grade 3 - obvious hypertropia in adduction; and grade 4 - large hypertropia in adduction along with an abduction movement when the eye is moved into the field of action of the IO.29 Similarly superior oblique overaction was graded based on the downward displacement of the eye in adduction. Patients with inadequate measurement of DVD, incomplete preoperative and postoperative orthoptic examination, prior SR muscle surgery, and with less than 6 months of follow-up were also excluded. Simultaneous or prior horizontal muscle surgery was not considered to be exclusion criteria.
Horizontal deviation was measured by prism alternate cover test. True hypertropia (non-DVD) was differentiated from DVD by observing the fixing eye for corresponding hypotropia during alternate cover test. Once this hypotropia was neutralized, DVD was measured using the prism under cover test. All measurements were made for both distance (patient fixating at 6 m vision chart) and near (patient fixating at 30 cm at near vision chart) with the patient's refractive error corrected. For the prism under cover test, a base down prism and cover was placed in front of the affected eye. The eye was occluded approximately for 10 seconds, then the cover was switched to the fixing eye and the downward movement of only the uncovered eye was observed. This movement was gradually neutralized by increasing the dioptric power of prisms to obtain the maximum amount of DVD. For the purpose of analysis only the distance measurements were considered. Care was taken to distinguish non-DVD especially IO overaction from DVD by measuring in lateral gazes. Incomitance was described as a difference of 7 prism diopters (PD) or more between the lateral gaze and primary position DVD.27 Patients with significant incomitant DVD were excluded from the study. The asymmetry of DVD was determined by the magnitude of difference in the deviation in primary position between the 2 eyes. A difference of 10 PD or more between the 2 eyes was considered an asymmetric DVD.
The additional data collected included the following: age at surgery, sex, best-corrected visual acuity (BCVA) for distance as recorded by logMAR chart, presence or absence of fixation preference, amount and type of horizontal deviation, presence and amount of vertical deviation, amount of SR recession performed, and details of any other previous surgery. Amblyopia was defined as a difference of 2 or more lines in BCVA between the 2 eyes.
Indication for surgery was a decompensated DVD in 1 or both eyes. Decompensated DVD was considered when a spontaneous upward drift of the eye was noted while examining the patient in the clinic. All patients underwent bilateral SR recession regardless of fixation preference. The techniques used were hangback, hemi-hangback, or fixed scleral suture. The same technique was used in both eyes of each patient. Irrespective of the technique, meticulous dissection of the SR muscle was performed to avoid postoperative lid or superior oblique muscle dysfunction. In general, large recessions of more than 8 mm were performed using the hemi-hangback or hangback techniques. The amount of SR recession performed was guided by the established nomogram provided in Table 1. Asymmetric surgery was performed if the patient had a difference of 10 PD or more in primary position DVD of both eyes or in the presence of an additional non-DVD of more than 5 PD. The amount of SR recession in asymmetric cases was also guided by the amount of DVD in individual eye (Table 1). In cases where asymmetric surgery was done due to the presence of non-DVD, 1 mm of SR recession was added per 3 PD of hypertropia. Asymmetric surgery was considered if the difference in the amount of SR recession performed was 2 mm or more between the 2 eyes.
Postoperative measurements were recorded at 1 month and at the last follow-up. Success was defined as absence of manifest DVD in both eyes at final postoperative visit. Dissociated vertical deviation was considered latent when the eye drifted upwards only under the cover and returned to primary position as soon as the cover was removed. Additionally, categories of success were defined according to the magnitude of residual DVD for distance as excellent (0-4 PD), good (5-9 PD), fair (10-14 PD), and poor (≥15 PD).17,18
Statistical analysis was performed using the Statistical Package for the Social Sciences version 21.0. Normality of data was tested by Kolmogorov-Smirnov test. Quantitative variables were compared using unpaired t-test or Mann-Whitney test between the 2 groups. Wilcoxon signed rank test was used for comparison of preoperative and postoperative DVD. Qualitative variables were correlated using Chi-square test/Fisher exact test. Kaplan-Meier survival curve was plotted to assess the actuarial success rate. For all statistical tests, a P value less than 0.05 was considered statistically significant.
A total of 27 patients (12 male and 15 female) who underwent bilateral SR recessions for bilateral DVD during the study period met the inclusion criteria. The median age at surgery was 8 years [interquartile range (IQR), 6-10 years] and mean follow-up was 16.3 ± 10.6 months (range, 6-48 months). The median BCVA was 0.1 logMAR (range, 0-0.5) in the right eye and 0.1 logMAR (range, 0-0.8) in the left eye. Amblyopia was seen in 8 patients (29.6%) at the time of surgery. Strabismus was the cause of amblyopia in 4 patients and the other 4 patients had both strabismus and anisometropia. The mean lines of vision loss in the amblyopic eye was 3.12 logMAR. All the patients with amblyopia were undergoing 2 to 3 hours of occlusion of the better eye at the time of surgery. Fixation preference for 1 eye was seen in 24 patients before surgery. Horizontal deviation in the form of either exotropia (n = 16) or esotropia (n = 10) was observed in 26 patients along with the DVD. Among these patients, 48% had large horizontal deviation at presentation, ie, more than 30 PD whereas 19.2% had less than 15 PD deviation and 32.8% had deviation between 15 and 30 PD. Overall, 23 patients underwent simultaneous horizontal rectus muscle surgery (either medial rectus recession for esotropia or lateral rectus recession for exotropia). None of the patients underwent more than 2 rectus muscle surgeries per eye. A pattern and V pattern was seen in 5 patients and 1 patient, respectively before surgery. None of the patients had significant IO or superior oblique muscle dysfunction. V pattern was corrected by half muscle-width vertical upshifting of both lateral rectus muscles. A pattern of less than 15 PD was corrected by the SR recession alone. A pattern of more than 15 PD in 3 patients was corrected by additional half muscle-width vertical downshifting of the lateral rectus muscles. Abnormal head posture was recorded in 12 patients at the time of presentation. Latent nystagmus was seen in 3 patients.
Preoperatively the median DVD was 9 PD (IQR, 6-14 PD) in the right eye and 9 PD (IQR, 7-12 PD) in the left eye. Asymmetric DVD (>10 PD) was seen in 6 patients (22.2%). Additionally, non-dissociated hypertropia (range, 2-10 PD) was seen in 6 patients preoperatively. This hypertropia was seen in the eye with the larger DVD. The amount of SR recession performed ranged from 5 to 10 mm with a median of 7 mm (IQR, 6-9 mm) in the right eye and 7 mm (IQR, 6-9 mm) in the left eye. The frequencies of hangback, hemi-hangback, and fixed scleral-suture surgery were 13 (48.1%), 7 (25.9%), and 7 (25.9%), respectively. Postoperatively the median DVD reduced to 5 PD (IQR, 3-8 PD) in the right eye (P = 0.015) and 6 PD (IQR, 3-10 PD) in the left eye (P = 0.016) (Fig. 1). Figure 2 shows the graphical representation of outcome of DVD after surgery in each patient. At the last follow-up, DVD remained latent in both eyes in 63.0% of patients. The amount of residual DVD was found to be <5 PD (excellent outcome, n = 23) in 42.6% of the eyes, 5-9 PD (good outcome, n = 17) in 31.5%, 10-14 PD (fair outcome, n = 6) in 11.1%, and ≥15 PD (poor outcome, n = 8) in 14.8% eyes after surgery. Residual DVD of 10 PD or more resulted in manifestation of DVD in 57% of these eyes as compared with 7.5% eyes with residual DVD of less than 10 PD (P = 0.0003). The Kaplan-Meier survival curve is shown in Figure 3. The cumulative success rate was 81.5% after 6 months as well as 1 year following surgery. The success rate reduced to 60.4% and 41.4% at 2 years and 4 years of follow-up, respectively.
Overall 8 patients underwent bilateral asymmetric SR recession. The median amount of recession asymmetry performed was 3.5 mm (range, 2-5 mm). Indication for asymmetric recession in 6 patients was asymmetric DVD of 10 PD or more and in 2 patients was the presence of non-DVD of more than 5 PD preoperatively. Patients with asymmetric surgery showed a slightly higher success rate (75%) as compared with those with symmetric surgery (57.9%) at the last follow-up. However, this difference was not statistically significant (P = 0.67).
The probability of a successful outcome did not depend on age at surgery, sex, preoperative type or magnitude of horizontal deviation, preoperative vertical deviation, amblyopia, and postoperative residual horizontal deviation. Also the outcome did not differ in moderate (<15 PD) and large DVD (≥15 PD). The only factor that favored a successful outcome was the presence of fixation preference (P = 0.041). Table 2 shows the distribution of various preoperative and postoperative factors between the successful and failure groups.
Three eyes developed elevation limitation (grade -1 or -2) in the immediate postoperative period. However it did not persist till the last follow-up visit. Additionally other surgical complications like upper lid retraction, ptosis, or superior oblique dysfunction were not seen even in those patients of whom 10-mm recession was performed. Two patients developed small hypotropia (<7 PD) after symmetrical recession of SR muscle. One patient who underwent asymmetric surgery developed pseudo-IO overaction in the eye with lesser amount of SR recession with residual DVD in the better eye and a switch in fixation preference. Three additional patients switched fixation preference after surgery; all of whom underwent symmetric surgery.
Several publications reported their results of unilateral SR recession for DVD.17,18,21 The reason for unilateral intervention was that the DVD was mostly manifested in the non-preferred eye and majority of their patients displayed a fixation preference. This approach may, however, lead to hypotropia in the operated eye and/or manifestation of DVD in the nonoperated eye.17,23,28 Shwartz and Scott17 showed that after unilateral SR recession, 17% patients developed DVD in the contralateral eye and 21% patients developed hypotropia in the operated eye postoperatively. By statistical analysis, they demonstrated that in patients with any DVD present in the contralateral eye preoperatively, there was 15 times greater risk of developing a significant DVD (≥15 PD) in that eye postoperatively. Therefore, bilateral SR recession is preferred by most strabismologists and was also the technique used in the current study. We performed bilateral SR recessions in spite of 89% of our patients displaying fixation preference before surgery. None of the patients included in the study had profound amblyopia or poor vision in 1 eye. In such cases we performed unilateral surgery as the worse eye is unlikely to resume fixation any time in future.
We observed a successful outcome (no manifest DVD) in 63% of patients after bilateral SR recession. Esswein et al21 studied outcome of large SR recessions (7-9 mm) in 86 patients and reported a success rate of 72% at 3 years of follow-up. They, however, included both unilateral and bilateral cases in their analysis. Also the successful outcome in their study was classified as either DVD not manifesting at all (corrected) or reduced to cosmetically acceptable levels (improved). In contrast, we classified failure as decompensated DVD in either of the eyes irrespective of the magnitude of the DVD. This could explain the slightly reduced success rate in our study. Magoon et al19 studied the outcome of large bilateral SR recession in 25 patients with bilateral DVD. They also performed bilateral surgery despite fixation preference and asymmetry of DVD. They found that this approach was highly successful with 19 (76%) of 25 patients having small DVD in both eyes postoperatively. The majority of their patients had large DVD (>20 PD) in both eyes prior to surgery and underwent 10 mm or more SR recession for DVD. They also found decompensated hypertropia of more than 10 PD in 10 (40%) patients that failed to get corrected in 5 patients postoperatively. Therefore, they suggested asymmetric surgery in such patients. Although we observed hypertropia in 22% of patients, the magnitude was small (≤10 PD).
We did not observe upper lid retraction, ptosis, or limitation of elevation in any of the eyes that underwent SR recession similar to other authors.17,19,21 One patient who underwent asymmetrical SR recession (10 mm in the right eye and 6 mm in the left eye) for asymmetrical DVD developed larger decompensated DVD in the lesser operated eye postoperatively. Additionally this patient showed pseudo-IO overaction in the left eye. This complication is similar to that reported by Freeman and Rosenbaum.30
We did not find overcorrection of DVD in any of our patients. The only major complication of the surgery was undercorrection as was found in many other studies.19,22,30,31 The rate of undercorrection, however, was not found to be affected by any of the parameters analyzed including age, amblyopia, type of surgical technique, residual postoperative horizontal deviation, asymmetry of DVD, asymmetry of surgery, and even the amount of preoperative DVD except fixation preference. None of the 3 patients who lacked fixation preference had a successful outcome. Among the 3 patients, 1 maintained alternate fixation, whereas the other 2 developed fixation preference for 1 eye following surgery. Varn et al31 suggested combining IO-weakening procedure along with large recession (≥10 mm) of SR muscle for large DVD (>20 PD) in order to prevent residual deviation. Since we did not find any difference in our results for DVD greater than 15 PD, we do not propose combined procedure as the first modality of treatment.
The results of this study should be interpreted within the limitations of its retrospective design. There were 3 surgeons involved in management of the patients. However, since the surgical dosages and techniques followed were similar, we do not consider this a major limitation. There were 3 types of surgical techniques performed: hangback, hemi-hangback, and fixed scleral sutures. Although there are some concerns of anterior creeping of the muscle with hemi-hangback and hangback technique,32,33 we did not find more undercorrection when using these techniques. Another limitation is the involvement of multiple examiners for measurement of DVD. Dissociated vertical deviation is a highly variable deviation and the measurements may change depending on the length of the time eye is dissociated and the attention span of the patient. We therefore considered the manifestation of DVD (which is perceived both by the patient and the ophthalmologist as residual deviation) to be the criteria for failure and not the absolute magnitude of the DVD.
In our series, 63% of our patients attained a successful outcome (DVD not manifesting spontaneously in either eye) after bilateral asymmetric or symmetric SR recession. All the failures were undercorrections of DVD and no overcorrections were found. Even in those patients with excellent results, DVD could not be completely abolished. Whether combination of other surgical procedures or a larger amount of SR recession will improve the success rate of this procedure warrants further study.
1. Christoff A, Raab EL, Guyton DL, et al. DVD—a conceptual, clinical, and surgical overview. J AAPOS
2. Helveston EM. Dissociated vertical deviation
: a clinical and laboratory study. Trans Am Ophthalmol Soc
3. Bielschowsky A. Lectures on Motor Anomalies
. Hanover (NH): Dartmouth College Publications; 1943: 11-20.
4. Cherfan CG, Diehl NN, Mohney BG. Prevalence of dissociated strabismus
in children with ocular misalignment: a population-based study. J AAPOS
5. Hunter DG, Kelly JB, Buffenn AN, et al. Long-term outcome of uncomplicated infantile exotropia. J AAPOS
6. Olson RJ, Scott WE. Dissociative phenomena in congenital monocular elevation deficiency. J AAPOS
7. Lim HT, Smith DR, Kraft SP, et al. Dissociated vertical deviation
in patients with intermittent exotropia. J AAPOS
8. Kutluk S, Avilla CW, von Noorden GK. The prevalence of dissociated vertical deviation
in patients with sensory heterotropia. Am J Ophthalmol
9. Wilson ME, Saunders RA, Berland JE. Dissociated horizontal deviation and accommodative esotropia: treatment options when an eso- and an exodeviation co-exist. J Pediatr Ophthalmol Strabismus
10. Vodicková K, Autrata R, Rehůrek J. Anterior transposition or myectomy of the inferior oblique muscle in vertical deviation—long-term results [in Czech]. Cesk Slov Oftalmol
11. Uncovska E, Vancurova J. Anterior transposition versus myectomy of the inferior oblique muscle in the treatment of dissociated vertical deviation
. Scripta Medica (BRNO)
12. Bothun ED, Summers CG. Unilateral inferior oblique anterior transposition for dissociated vertical deviation
. J AAPOS
13. Pineles SL, Velez G, Velez FG. Asymmetric inferior oblique anterior transposition for incomitant asymmetric dissociated vertical deviation
. Graefes Arch Clin Exp Ophthalmol
14. Nabie R, Anvari F, Azadeh M, et al. Evaluation of the effectiveness of anterior transposition of the inferior oblique muscle in dissociated vertical deviation
with or without inferior oblique overaction. J Pediatr Ophthalmol Strabismus
15. Farid MF. Anterior transposition vs anterior and nasal transposition of inferior oblique muscle in treatment of dissociated vertical deviation
associated with inferior oblique overaction. Eye (Lond)
16. Farvardin M, Attarzadeh A. Combined resection and anterior transposition of the inferior oblique muscle for the treatment of moderate to large dissociated vertical deviation
associated with inferior oblique muscle overaction. J Pediatr Ophthalmol Strabismus
17. Schwartz T, Scott W. Unilateral superior rectus recession
for the treatment of dissociated vertical deviation
. J Pediatr Ophthalmol Strabismus
18. Scott WE, Sutton VJ, Thalacker JA. Superior rectus recessions for dissociated vertical deviation
19. Magoon E, Cruciger M, Jampolsky A. Dissociated vertical deviation
: an asymmetric condition treated with large bilateral superior rectus recession
. J Pediatr Ophthalmol Strabismus
20. Loba P, Nowakowska O, Broniarczyk-Loba A. Large hang-back recession of superior rectus muscles as effective treatment of patients with dissociated vertical deviation
[in Polish]. Klin Oczna
21. Esswein MB, von Noorden GK, Coburn A. Comparison of surgical methods in the treatment of dissociated vertical deviation
. Am J Ophthalmol
22. Can D, Ozkan SB, Kasim R, et al. Surgical results in highly asymmetric dissociated vertical deviations. Strabismus
23. Noel LP, Parks MM. Dissociated vertical deviation
: associated findings and results of surgical treatment. Can J Ophthalmol
24. Esswein Kapp MB, von Noorden GK. Treatment of residual dissociated vertical deviation
with inferior rectus resection. J Pediatr Ophthalmol Strabismus
25. Gamio S. A surgical alternative for dissociated vertical deviation
based on new pathologic concepts: weakening all four oblique eye muscles. Outcome and results in 9 cases. Binocul Vis Strabismus Q
26. Hatt SR, Wang X, Holmes JM. Interventions for dissociated vertical deviation
. Cochrane Database Syst Rev
27. Santiago AP, Rosenbaum AL. Dissociated vertical deviation
. In: Rosenbaum AL, Santiago AP, eds. Clinical Strabismus Management
. 1st ed. Philadelphia: WB Saunders Company; 1999: 237-247.
28. Sargent RA. Surgical correction of dissociated hyperdeviations. Am Orthopt J
29. Wright KW. Alphabet patterns and oblique muscle dysfunctions. In: Wright KW, Spiegel PH, Thompson LS, eds. Handbook of Pediatric Strabismus and Amblyopia
. 3nd ed. New York: Springer; 2006: 284-322.
30. Freeman RS, Rosenbaum AL. Residual incomitant DVD following large bilateral superior rectus recession
. J Pediatr Ophthalmol Strabismus
31. Varn MM, Saunders RA, Wilson ME. Combined bilateral superior rectus muscle recession and inferior oblique muscle weakening for dissociated vertical deviation
. J AAPOS
32. Repka MX, Fishman PJ, Guyton DL. The site of reattachment of the extraocular muscle following hang-back recession. J Pediatr Ophthalmol Strabismus
33. Wysenbeek Y, Wygnanski-Jaffe T, Rosner M, et al. Evaluation of superior rectus muscle attachment following hang-back recession in rabbit eyes. Eur J Ophthalmol