Ophthalmoplegia as a complication of cosmetic facial filler injection

Interest in rejuvenation has increased the number of aesthetic soft tissue augmentation of the face using the filler injections (Park et al. 2014). However, complications of filler injection include blindness, ophthalmoplegia or strokes (Coleman 2002; Peter Park et al. 2012). Herein we describe the patterns and prognosis of ophthalmoplegia as a complication of cosmetic facial filler in four patients. All four patients developed unilateral ophthalmoplegia with sudden visual loss during or just after filler injection into the glabella and perinasal areas including the nasolabial folds. They all showed a total paralysis of infraduction and adduction (Fig. 1), and three of them had a palsy of the superior oblique. Supraduction was limited in two patients and abduction only in one patient. Based on the non-dilated pupil fixed at midposition without reflexes, associated oculo-sympathetic injury was suspected in one. Two patients showed necrotic changes of the skin, and one reported hypesthesia around the injection sites. Severe ocular pain was reported in the involved side in all patients. The visual loss was so profound that three patients showed no light perception and one was able to perceive only hand motions. Three patients had normal brain MRIs, while one patient showed a high signal intensity on diffusion-weighted images (DWI) in the right retrobulbar optic nerve, which is suggestive of posterior ischaemic optic neuropathy (PION). Cerebral angiography showed an occlusion of the ophthalmic artery without a choroidal flush in a patient. All patients had fluorescein angiography that disclosed patchy choroidal filling delays along with diffuse retinal artery perfusion defects, diminished retinal artery perfusion with preservation of the superior branch, or normal perfusion in the patient with PION. In our patients, the hyaluronic acid appeared to have flowed retrograde into the ophthalmic artery via its distal branches supplying the superficial face, which acted as a conduit for the retrograde migration of the filler particles during high-pressure injections (Coleman 2002; Peter Park et al. 2012). Subsequently, the ophthalmic artery or its branches including the central retinal artery appeared to be occluded, resulting in blindness and ophthalmoplegia. Our patients showed a relative sparing of the lateral rectus muscle, while previous studies reported total ophthalmoplegia after filler injections (Kim Park et al. 2014). Occlusion of the medial/inferior muscular trunk arising from the ophthalmic artery paralyses the medial and inferior recti and inferior oblique muscles, while obstruction of the lateral/superior trunk causes palsies of the levator palpebrae, superior rectus and superior oblique muscles. Independent branches also supply the extraocular muscles separately, and the branches to the lateral rectus can arise from the lacrimal or recurrent meningeal artery instead of the ophthalmic artery. The relative sparing of the lateral rectus in our patient may be explained by this anatomical variation, or mechanical effect of swollen tissues around the injection sites, which were mostly located in the nasal side of the orbits, inducing a further obstruction of the arteriovenous flow. During the follow-up, the extraocular movements improved in three patients, and one with a complete ophthalmoplegia initially developed phthisis bulbi later. Despite recovery of the ophthalmoplegia, ocular misalignment persisted in all patients, and two of them were subjected to strabismus surgery. This indicates an unfavourable prognosis of ophthalmoplegia from a facial filler injection. Whereas previous studies reported a worse visual prognosis and more frequent cerebral infarctions with the use of autologous fat as a filler material than with hyaluronic acid (Park et al. 2014), further progression of the emboli beyond the ophthalmic artery and resultant cerebral infarctions did not occur in our patients with hyaluronic acid injection. However, all the patients including the one who had intra-arterial thrombolysis using hyaluronidase remained without light perception in the involved eye during the follow-up of more than 8 months.