Synkinesis Following Diabetic Third Nerve Palsy | Diabetic Retinopathy | JAMA Ophthalmology | JAMA Network
[Skip to Navigation]
Sign In
Case Reports and Small Case Series
January 2000

Synkinesis Following Diabetic Third Nerve Palsy

Arch Ophthalmol. 2000;118(1):132-134. doi:

A 59-year-old man developed a left pupil-sparing third nerve palsy related to diabetes that recovered gradually over 5 months; the only deficit remaining at 8 months was a trace underaction of adduction. Oculomotor synkinesis was evident at this time, manifest as left upper lid retraction on adduction and on downgaze. There were no signs of oculomotor or pupillary synkinesis. High-resolution magnetic resonance imaging of the third nerve pathway revealed no abnormality. Oculomotor synkinesis following diabetic third nerve palsy is extremely rare, with only one report in the literature to date.

Synkinesis may occur during the recovery phase of a traumatic third nerve palsy or after long-standing compression. The abnormal movements that constitute synkinesis may involve the upper lid, oculomotor system, and the pupil. Synkinesis is rare when the cause is ischemia1 and we have found only one previous report of synkinesis following diabetic third nerve palsy.2 Our report adds a second to the literature.

Report of a Case

A 59-year-old man with type 2 diabetes mellitus was seen with a 2-week history of left ocular and retro-orbital pain, progressive left upper lid ptosis, and diplopia in all directions of gaze. Diabetes was well controlled and there was no history of hypertension or previous neurologic disturbance. Medication consisted of oral hypoglycemic agents only.

The examination revealed corrected visual acuity of 20/30 OU, normal color vision, and full visual fields. The pupils were 4 mm in diameter and reacted briskly to light. There was no proptosis. The upper lid and ductions in the right eye were normal. On the left side, there was complete ptosis with no levator function. The left eye was exotropic, adduction and elevation were absent, and depression was markedly limited. Intorsion on attempted downgaze was present. The remainder of the cranial nerves were normal. A diagnosis of diabetic left pupil-sparing third nerve palsy was made and the patient was observed.

At 2 months, his pain had resolved completely, and left upper lid ptosis was only 3 mm with 5 mm of levator function. The left eye could be adducted just beyond the midline and there was improved but still limited upgaze and inferior rectus function. There was 2 mm of left upper lid retraction on attempted adduction, with the right upper lid remaining stationary on right gaze. There was complete descent of the left upper lid on downgaze at this time. Because of the signs of oculomotor synkinesis, a magnetic resonance scan of the third nerve pathway was obtained, but no intracranial abnormality from the midbrain to the orbit was found.

At 5 months, the patient was asymptomatic and had almost completely recovered. The patient had no ptosis or pupillary abnormality, normal levator function, and full motility except for a trace underaction of adduction on the left. Retraction of the left upper lid on adduction was unchanged and there was a new incomplete descent of the left upper lid by 2 mm on downgaze (Figure 1).

A, Right gaze demonstrates slight underaction of left medial rectus and elevation of left upper lid. B, Left gaze demonstrates the normal lower position of the left upper lid. C, On downgaze, the left upper lid does not descend fully.

A, Right gaze demonstrates slight underaction of left medial rectus and elevation of left upper lid. B, Left gaze demonstrates the normal lower position of the left upper lid. C, On downgaze, the left upper lid does not descend fully.

Although the patient's third nerve palsy had recovered almost completely, definite signs of oculomotor synkinesis were evident. These synkinetic movements were unchanged at the last examination, 8 months after the onset of symptoms.


Third nerve palsy is the commonest cranial mononeuropathy in patients with diabetes.3 Although the precise etiological role of diabetes-related third nerve palsy is uncertain,4,5 diabetes-related ischemia of the peripheral nerve may be important.6 And, with the use of magnetic resonance imaging, it is now apparent that isolated third nerve palsy (with or without pain and without pupillary involvement) in a diabetic individual may uncommonly be due to midbrain infarction or hemorrhage.7 It should be noted that if the third nerve palsy in a diabetic patient fails to recover, it is likely to be due to a coexistent compressive lesion such as a tumor or an aneurysm.

Recovery of third nerve palsy in diabetes usually occurs within 3 months.8-10 The mechanism of recovery is not known, but almost certainly involves remyelination, present at 6 months11 and complete by 3 years.12 Synkinesis may occur during recovery of any third nerve palsy, particularly following traumatic and compressive lesions,13-15 but only rarely when the nerve palsy appears to be due to microvascular disease1 or ischemia related to giant cell arteritis.16 Oculomotor synkinesis is manifest as abnormal movements of the upper lid, globe, or pupil, occurring in isolation or in a variety of combinations that may be subtle and easily missed by the examiner.

There are currently 3 contesting theories on the pathogenesis of oculomotor synkinesis (see review by Sibony et al17): (1) Aberrant regeneration of injured nerve fibers in which regenerating axons that grow into the wrong nerve sheaths are directed to incorrect targets. (2) "Release" phenomenon or synaptic reorganization of the third nerve prenuclear or nuclear complex.18,19 (3) Abnormal signal conduction (ephaptic transmission) between peripheral axons where there is a secondary partial or complete loss of myelin.

In our patient, it is assumed that the third nerve palsy was due to a lesion of the peripheral nerve since magnetic resonance imaging revealed no evidence of a brainstem, cavernous sinus, or subarachnoid lesion along the third nerve pathway. Magnetic resonance imaging is unhelpful in confirming the peripheral location of the lesion in diabetic third nerve palsy since, as occurred in our patient and in other studies,20,21 the third nerve did not enhance.

Histological studies of diabetic third nerve palsies may provide insight about the pathogenesis of synkinesis in our patient. These studies11,12,22 found that the dominant feature was central focal demyelination of the subarachnoid, intracavernous third nerve with relative axonal sparing. The fascicles and nuclei were normal in two of the examinations. These findings mitigate against the misdirection and central theories but favor the ephaptic theory of electrotonic conduction between bare axons in our case.

In conclusion, this case report is the second of synkinesis following diabetic third nerve palsy. It is emphasized that signs of oculomotor synkinesis may be subtle and thus easily missed unless specifically looked for, but the diagnosis of synkinesis following diabetic third nerve palsy remains a diagnosis of exclusion.

Corresponding author: Bryan S. Sires, MD, PhD, University of Washington, Ophthalmology, Box 356485, Seattle, WA 98195-6485 (e-mail:

This study was supported in part by an unrestricted departmental award from Research to Prevent Blindness Inc, New York, NY.

Zorrilla  EKozak  GP Ophthalmoplegia in diabetes mellitus.  Ann Intern Med. 1967;67968- 976Google ScholarCrossref
Weinstein  EADolger  H External ocular muscle palsies occurring in diabetes mellitus.  Arch Neurol Psychiatry. 1948;60597- 603Google ScholarCrossref
Rush  JAYounge  BR Paralysis of cranial nerves III, IV, and VI.  Arch Ophthalmol. 1981;9976- 79Google ScholarCrossref
Teuscher  AUMeienberg  O Ischaemic oculomotor nerve palsy.  J Neurol. 1985;232144- 149Google ScholarCrossref
Watanabe  KHagura  RAkanuma  Y  et al.  Characteristics of cranial nerve palsies in diabetic patients.  Diabetes Res Clin Pract. 1990;1019- 27Google ScholarCrossref
Smith  BEDyck  PJ Subclinical histopathological changes in the oculomotor nerve in diabetes mellitus.  Ann Neurol. 1992;32376- 385Google ScholarCrossref
Hopf  HCGutmann  L Diabetic 3rd nerve palsy: evidence for a mesencephalic lesion.  Neurology. 1990;401041- 1045Google ScholarCrossref
Goldstein  JECogan  DG Diabetic ophthalmoplegia with special reference to the pupil.  Arch Ophthalmol. 1960;64144- 152Google ScholarCrossref
Green  WRHackett  ERSchlezinger  NS Neuro-ophthalmic evaluation of oculomotor nerve paralysis.  Arch Ophthalmol. 1964;72154- 167Google ScholarCrossref
Capo  HWarren  FKupersmith  MJ Evolution of oculomotor nerve palsies.  J Clin Neuroophthalmol. 1992;1221- 25Google Scholar
Weber  RBDaroff  RBMackey  EA Pathology of oculomotor palsy in diabetics.  Neurology. 1970;20835- 838Google ScholarCrossref
Asbury  AKAldredge  HHerschberg  RFisher  CM Oculomotor palsy in diabetes mellitus: a clinico-pathological study.  Brain. 1970;93555- 566Google ScholarCrossref
Krohel  GB Blepharoptosis after traumatic third-nerve palsies.  Am J Ophthalmol. 1979;88598- 601Google Scholar
Hamer  J Prognosis of oculomotor palsy in patients with aneurysms of the posterior communicating artery.  Acta Neurochir (Wien). 1982;66173- 185Google ScholarCrossref
Elston  JS Traumatic third nerve palsy.  Br J Ophthalmol. 1984;68538- 543Google ScholarCrossref
Sibony  PALessell  S Transient oculomotor synkinesis in temporal arteritis.  Arch Neurol. 1984;4187- 88Google ScholarCrossref
Sibony  PALessell  SGittinger  JW Acquired oculomotor synkinesis.  Surv Ophthalmol. 1984;28382- 390Google ScholarCrossref
Wartenberg  R Associated movements in the oculomotor and facial muscles.  Arch Neurol Psychiatry. 1946;55439- 488Google ScholarCrossref
Lyle  DG Experimental oculomotor nerve regeneration.  Am J Ophthalmol. 1966;611239- 1243Google Scholar
Mark  ASBlake  PAtlas  SWRoss  MBrown  DKolsky  M Gd-DTPA enhancement of the cisternal portion of the oculomotor nerve on MR imaging.  AJNR Am J Neuroradiol. 1992;131463- 1470Google Scholar
Blake  PYMark  ASKattah  JKolsky  M MR of oculomotor nerve palsy.  AJNR Am J Neuroradiol. 1995;161665- 1672Google Scholar
Dreyfus  PMHakim  SAdams  RD Diabetic ophthalmoplegia: report of a case, with postmortem study and comments on vascular supply of human oculomotor nerve.  Arch Neurol Psychiatry. 1957;77337- 349Google ScholarCrossref