Journal of Clinical Ophthalmology and Research

: 2016  |  Volume : 4  |  Issue : 3  |  Page : 137--141

Magnetic resonance imaging (MRI) in Duane retraction syndrome

Siddharth Agrawal1, Vinita Singh1, Anit Parihar2, Vishal Katiyar1, Rajat M Srivastava1, Vikas Chahal1,  
1 Department of Ophthalmology, King Georges' Medical University, Lucknow, Uttar Pradesh, India
2 Department of Radiodiagnosis, King Georges' Medical University, Lucknow, Uttar Pradesh, India

Correspondence Address:
Siddharth Agrawal
Department of Ophthalmology, King Georges' Medical University, Lucknow - 226 003, Uttar Pradesh


Purpose: To study the magnetic resonance imaging (MRI) findings in Duane retraction syndrome (DRS). Materials and Methods: In this case–control study, 16 consecutive cases of Duane syndrome underwent MRI of the brain and orbit, after informed consent. MRI (fast imaging enhancing state acquisition) was done with special focus on pontomedullary junction to look for the status of abducens nerve and associated abnormalities along with quasi-coronal sections of orbit to estimate thickness and cross sectional area of horizontal extra-ocular muscles (EOMs). Thickness and cross-sectional area of EOMs were compared to the contralateral side and to age-matched controls. Results: Of 16 cases, 14 had DRS Type I, and 2 had DRS Type II. MRI revealed absent abducens nerve on the ipsilesional side in 12 out of 14 in DRS I, whereas two DRS II patients revealed intact abducens nerve on the affected side. Thickness and cross-sectional area of the horizontal recti was statistically comparable to the contralateral side and age-matched controls. Conclusions: All the DRS patients in study had structurally normal horizontal recti muscles, whereas 12 out of 14 of DRS-I patients had an absent abducens nerve on MRI.

How to cite this article:
Agrawal S, Singh V, Parihar A, Katiyar V, Srivastava RM, Chahal V. Magnetic resonance imaging (MRI) in Duane retraction syndrome.J Clin Ophthalmol Res 2016;4:137-141

How to cite this URL:
Agrawal S, Singh V, Parihar A, Katiyar V, Srivastava RM, Chahal V. Magnetic resonance imaging (MRI) in Duane retraction syndrome. J Clin Ophthalmol Res [serial online] 2016 [cited 2022 Aug 12 ];4:137-141
Available from:

Full Text

Duane retraction syndrome (DRS) is a group of ocular motility disorders characterized by deficient horizontal eye movements, narrowing of palpebral aperture, eyelid retraction, and abnormal vertical movements of the eye ball.[1] Huber classified it into three clinical subtypes: Type I DRS with limited abduction, Type II with limited adduction, and Type III with variable limitation in both abduction and adduction.[2]

Various mechanisms have been suggested for this syndrome ranging from abnormal medial rectus (MR) insertion, presence of a nonelastic band, structurally abnormal lateral rectus (LR), paradoxical innervations, and anomaly in the brain stem.[3],[4] Enhanced genetic understanding of nonprogressive restrictive ocular motility disorders such as DRS has revealed of a dysgenetic nuclei of the affected cranial nerves in the brainstem and pons and such related disorders are now grouped as “congenital cranial dysinnervation disorders.”[5]

Imaging modalities such as magnetic resonance imaging (MRI) play a crucial role into diagnosing diseases and planning management. The aim of the study was to identify the anatomical defect in our patients of DRS by detailed imaging of the abducens nerve, orbit and the horizontal extra-ocular muscles (EOMs). We have performed high resolution T1-weighted MRI (including fast imaging enhancing state acquisition [FIESTA]) at the level of brain stem for abducens nerve and have done pulse sequencing for the orbit to study EOMs.

 Materials and Methods

After clearance from the Institutional Ethical Committee, 16 consecutive patients of DRS underwent MRI on SIGNA EXCITE 1.5T GEMSOW (GE) MR scanner. Informed consent was taken and any contraindication for the scan ruled out. Age-matched controls were imaged for comparison.

Technical aspects of the MRI were confirmed during the study, and all scans were taken by the same individual using the same model. Two scans were given utmost importance. FIESTA (TE-1.7, TR-10, Flip angle-45, Bandwidth-62.50, Matrix size 256 × 256, NEX-3) and T2-weighted coronal sequence with fat suppression (TE-85, TR-4300, inversion time-150, Bandwidth-27.78, matrix size-352 × 256, NEX-2). The MRI protocol is summarized in [Table 1].{Table 1}

FIESTA is an steady state free precession sequence which uses gradient echo sequences with a small flip angle and short relaxation times in which a steady state develops between the pulse repetitions. It gives strong signal from fluid tissues while suppressing background tissue signals.[6]

On MRI, the entire paths of both the abducens nerves were traced from the level of the upper medulla oblongata to the level of the upper pons. If the entire cisternal segment of the nerve was not identified, it was considered to be absent.[7] If the nerve was identified within a limited length from the cisternal segment, it was considered to be hypoplastic.[8]

To image the horizontal recti quasi-coronal, images of contiguous 2 mm thick quasi-coronal image planes were obtained perpendicular to the long axis of each orbit. Patients who were unable to cooperate were examined under general anesthesia.

Digital MRI images were obtained and were transferred to personal computer and were decoded using a locally available digital imaging and communications in medicine viewer software. In each MRI, the two sections were studied in detail. One, the section at pontomedullary junction to look for the exit of abducens nerve and the other, the quasi-coronal view to look for the thickness of horizontal recti. For each patient the width of the horizontal recti in the center of the entire visualized muscle. The borders of the horizontal recti were outlined with a digital cursor in the MRI software and their cross sectional area was measured by the “area” function in the scanner. Acquisitions were done in primary gaze and the other eye was not shielded. Patient was directed to fixate at a target placed inside the MRI gantry. The analysis was done on the console of the MR scanner.

These values in patients of DRS were compared with the muscles of the contralateral side and 9 age-matched controls. For statistical analysis, Statistical Package for the Social Sciences (SPSS) (IBM, SPSS Statistics for Windows Version 21.0, Armonk, NY) was used and interpretation of the resultant data done. Unpaired t-test was applied and P < 0.05, 95% confidence interval was taken as statistically significant.


Of 16 cases, 14 had DRS Type I and 2 had DRS Type II. MRI revealed absent abducens nerve on the ipsilesional side in 12 out of 14 in DRS I, whereas two DRS II patients revealed intact abducens nerve on the affected side [Figure 1], [Figure 2], [Figure 3], [Figure 4]. The mean age of the patients was 19.25 years (range 6–27 years). The mean age of the controls was 18.44 years (range 12–26 years).{Figure 1}{Figure 2}{Figure 3}{Figure 4}

The results are summarized in [Table 2],[Table 3],[Table 4],[Table 5]. The area and thickness of horizontal recti muscles (MR, LR) of the involved side were compared to the same of the age matched controls. On unpaired t-test, there was no statistically significant difference in the thickness and area [Table 4].{Table 2}{Table 3}{Table 4}{Table 5}

Statistically no significant difference was found when the area and thickness of the involved EOMs was compared with the unaffected side [Table 5].

Other incidental findings included absence of 6th nerve on uninvolved (left) side in patient number 8 and a neurocysticercosis lesion in left frontal lobe in patient number 6.


The precise nature of pathogenesis of DRS has been a puzzle, which still requires a lucid explanation. MRI provides an excellent noninvasive modality to study the anatomy of EOMs, cranial nerves, and their nuclei.[6] The present study on MRI in DRS is the largest to the best of our knowledge with 16 patients (PubMed search).

Various MRI protocols have been evaluated to image cranial nerves and their nuclei and FIESTA has shown to increase the visualization to almost 100%.[8] We employed FIESTA to study the pontomedullary junction and VI th (Abducens) cranial nerve. About 12 out of 14 of our Type I DRS patients were found to have absent abducens nerve where as it was found present in both of our Type II DRS patients which was similar to the results obtained by Xia et al.[9] Lack of abducens nerve innervation to the affected LR explains in part the pathogenesis in Type I DRS patients. A study by Kang and Demer [10] highlighted a marked atrophy of LR muscle in cases of abducens nerve palsy unlike in Type I DRS patients. In our study, we did not find any statistically significant difference in the area and thickness of the affected horizontal recti muscles when compared to age-matched controls and uninvolved eyes suggesting a lack of influence of the age and duration of disease on the anatomical normalcy of EOMs. Anatomical normalcy of LR in Type I DRS may be due to anomalous innervations by the 3rd cranial nerve, which has been demonstrated in the past studies.[9] This could have been established by a dynamic MRI which was difficult to perform as majority of the patients in our study were children with limited capacity to cooperate. Strachan and Brown concluded with a possibility of an structurally abnormal LR to explain the electromyographic results.[4] Structural normalcy of EOMs on MRI may suggests anomalous/paradoxical innervation of EOMs to explain such results in DRS patients.

MRI studies by Xia et al.[9] revealed double innervation of LR muscle in all patients with Type II DRS. In our study, both our patients had normally innervated LR muscle. This could be indicative of different entities with similar clinical features being wrongly grouped together. A limited number of cases in our study prevent any further deductions. Our observations could be extended to differentiate DRS from clinically similar presentation such as infantile esotropia and abducens nerve palsy.[11]


We thus conclude that in DRS, horizontal EOMs are structurally normal and in Type I DRS, abducens nerve is frequently absent.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Alexandrakis G, Saunders RA. Duane retraction syndrome. Ophthalmol Clin North Am 2001;14:407-17.
2Huber A. Electrophysiology of the retraction syndromes. Br J Ophthalmol 1974;58:293-300.
3Apple C. Congenital abducens paralysis. Am J Ophthalmol 1939;22:169-73.
4Strachan IM, Brown BH. Electromyography of extraocular muscles in Duane's syndrome. Br J Ophthalmol 1972;56:594-9.
5Nentwich MM, Nentwich MF, Maertz J, Brandlhuber U, Rudolph G. Congenital cranial dysinnervation disorders (CCDD). Klin Monbl Augenheilkd 2015;232:275-80.
6Sheth S, Branstetter BF 4th, Escott EJ. Appearance of normal cranial nerves on steady-state free precession MR images. Radiographics 2009;29:1045-55.
7Parsa CF, Grant PE, Dillon WP Jr., du Lac S, Hoyt WF. Absence of the abducens nerve in Duane syndrome verified by magnetic resonance imaging. Am J Ophthalmol 1998;125:399-401.
8Hatipoglu HG, Durakoglugil T, Ciliz D, Yüksel E. Comparison of FSE T2W and 3D FIESTA sequences in the evaluation of posterior fossa cranial nerves with MR cisternography. Diagn Interv Radiol 2007;13:56-60.
9Xia S, Li RL, Li YP, Qian XH, Chong V, Qi J. MRI findings in Duane's ocular retraction syndrome. Clin Radiol 2014;69:e191-8.
10Kang NY, Demer JL. Comparison of orbital magnetic resonance imaging in Duane syndrome and abducens palsy. Am J Ophthalmol 2006;142:827-34.
11Agrawal S, Singh V, Agrawal S. Congenital sixth nerve palsy or type I Duane syndrome? Oman J Ophthalmol 2011;4:92-4.