Globe subluxation is a rare orbital event. It is characterized by anterior displacement of the eye, usually when the equator of the globe protrudes beyond the retracted eyelids. Globe subluxation can fall into 3 categories: spontaneous, voluntary, or traumatic. Spontaneous globe subluxation occurs without conscious effort or without a precipitating factor. Voluntary globe subluxation is defined as the ability to express one's globe without any precipitating trigger. Traumatic globe subluxation results without conscious effort after direct orbital trauma. Fewer than 50 cases of spontaneous subluxation of the globe have been reported, with the first reported case in 1907.1 The most common presentation is spontaneous subluxation after a precipitating trigger, usually involving eyelid manipulation. Patients typically present with varying underlying risk factors such as exophthalmos, severe lid retraction, floppy eyelid syndrome, thyroid orbitopathy, or conditions that predispose patients to shallow orbits such as Crouzon syndrome.1 There are many potentially severe complications of cataract surgery; however, there have been no reported cases of clinically significant globe subluxation intraoperatively. We present an unusual case of intraoperative globe subluxation during phacoemulsification in a patient with complex ocular comorbidities and previous surgery.
A 38-year-old woman with a history of congenital glaucoma and previous bilateral 350 mm2 Baerveldt tube (Johnson & Johnson Vision Care, Inc.) implantation presented with worsening vision. On examination, corrected distance visual acuities were 20/125 and 20/100 with intraocular pressures of 27 mm Hg and 22 mm Hg in the right and left eyes, respectively. There were moderate nuclear sclerosis changes bilaterally. The Baerveldt tubes were well positioned supratemporally in the anterior chamber, and the plates were 10 mm from the limbus bilaterally. Visual field testing showed a right superior arcuate scotoma and left superior hemifield defect and inferior arcuate scotoma. On optical coherence tomography (Cirrus HD-OCT, Carl Zeiss Meditec AG), the mean peripapillary retinal nerve fiber layer thicknesses were 70 μm in the right eye and 41 μm in the left eye. The axial lengths, anterior chamber depths, and horizontal white-to-white corneal diameters on biometry were 30.45 mm, 3.58 mm, and 12.60 mm, respectively, in the right eye and 27.24 mm, 3.57 mm, and 12.70 mm, respectively, in the left eye. Her body mass index was 31 kg/m2.
The patient was scheduled for a left irrigation of the Baerveldt tube, phacoemulsification, and intraocular lens insertion. The procedure was performed under general anesthesia to reduce the risk for globe perforation from a periocular block and at the patient's request to alleviate anxiety. The Baerveldt tube was flushed with 1 mL of balanced salt solution using a 27-gauge Rycroft cannula. A temporal corneal wound was created, followed by insertion of a Malyugin ring (MicroSurgical Technology) to dilate the small pupil. Routine continuous curvilinear capsulorhexis, hydrodissection, and phacoemulsification of the lens were completed. During cortical aspiration, posterior positive vitreous pressure was noted with anterior chamber shallowing. This persisted despite injection of opthalmic viscosurgical device into the anterior chamber. A 12.5 D SN60WF (Alcon) intraocular lens was placed in the capsular bag, the corneal wounds sutured with 10/0 nylon, and the speculum loosened in an attempt to reduce the posterior positive vitreous pressure. Intravenous mannitol 20% (40 g in 200 mL over 30 minutes) was administered. Aqueous misdirection syndrome was considered, and a 2-port pars plana vitrectomy was performed. The anterior chamber deepened to normal depth following vitrectomy, allowing for completion of the soft lens matter removal and suture closure of the paracenteses.
On removal of the speculum, extreme unilateral globe subluxation of the operative eye was evident, with the upper and lower lids retracted behind the equator (Figure 1A). After 40 minutes of gentle manual decompression of the orbit, with a Desmarres retractor to pull the lower lid anteriorly and using cotton tip applicators to massage the globe either side of the horizontal recti (Figure 1B), the globe was repositioned into the orbit and the eyelids able to be closed (Figure 1C). On day 1 postoperatively, the uncorrected distance visual acuity was 20/200 and with pinhole to 20/100, the intraocular pressure 6 mm Hg, and the eye in the normal position (Figure 1D). The anterior chamber was deep with the posterior chamber intraocular lens in the capsular bag and the retina flat with no choroidal effusion or suprachoroidal hemorrhage.
Phacoemulsification and insertion of a posterior chamber intraocular lens in the fellow right eye proceeded 3 weeks later. Precautions were used to minimize the possibility of a further episode of globe subluxation. The patient was administered intravenous mannitol 20% (100 g in 500 mL over 30 minutes) before the surgery with insertion of an indwelling urinary catheter. General anesthetic and a van Lint facial nerve block was performed. A Barrett adjustable speculum (Duckworth & Kent) was suspended to the drapes with 2/0 silk sutures, aimed at keeping the blades of the speculum anterior to the globe (Figures 2, A and B). The Baerveldt tube for this eye was not flushed to reduce egress of fluid into the sub-Tenon space; however, there was no attempt to stent the tube intraoperatively to prevent egress of the irrigation fluid during phacoemulsification. Equipment to perform a pars plana vitrectomy was readily available, but not used. The operation was completed with no complications of posterior positive vitreous pressure, anterior chamber shallowing, or globe subluxation. At 2 months postoperatively, the corrected distance visual acuities were 20/120 in the right eye and 20/80 in the left eye, similar to her vision before development of cataract, limited by amblyopia.
This degree of extreme intraoperative globe subluxation during phacoemulsification has not been previously reported in the literature, to our knowledge. We hypothesize several reasons why this complication occurred. Although high myopia has not been shown to be an independent risk factor for globe subluxation, a long axial length in the presence of a shallow orbit could lead to anterior displacement of the globe.1 Our patient did not have any of the other risk factors such as floppy eyelids or thyroid orbitopathy.2
The mechanism behind globe subluxation usually involves forced opening of the eyelids, causing posterior pressure against the globe, in turn causing the globe to subluxate anteriorly. The patient also underwent a vitrectomy, creating a softer globe that is more inclined to herniate under pressure. It is therefore possible that the Lieberman speculum that causes lateral and posterior pressure may have been contributory. In replacing the globe postoperatively, it was found that the upper lid was retracted, the globe not able to be retropulsed, and the lower eyelid rolled inward suggesting raised intraorbital pressure as the main mechanism. The patient was under a general anesthetic with muscle paralysis, so no eyelid spasms or tone should have occurred throughout the procedure. However, in preparation for the second eye, a general anesthesia with vecuronium and a van Lint facial nerve block were undertaken to eliminate any possibility of eyelid or extraocular muscle contraction.
Intraoperative flushing of the Baerveldt tube may have disrupted the capsule over the plate creating a communication for fluid to traverse into the sub-Tenon space from the anterior chamber via the Baerveldt tube. Further egress of irrigation fluid during phacoemulsification and irrigation/aspiration of the cortex into the sub-Tenon space and subsequently into the retrobulbar space could have contributed to globe subluxation by increased retrobulbar volume. The manual decompression of globe contents back into the bony orbit required prolonged but gentle pressure. Fortunately, there was no retrobulbar mass such as a spontaneous orbital hemorrhage to prevent the decompression. An orbital computed tomography scan would have been performed in the case that decompression was not possible and retrobulbar hemorrhage suspected. Return of the globe to the bony orbit was critical to avoid exposure keratopathy, corneal abrasions, and traumatic optic neuropathy, particularly as this was the patient's better seeing eye.3,4 Lateral cantholysis was considered in view of the extreme subluxation, however, it was decided that it would result in destabilization of the eyelid complex and could result in further difficulty maintaining the globe within the bony orbit.
In conclusion, the complications during this patient's phacoemulsification occurred due to the concomitance of factors. This included a longer myopic eyeball secondary to congenital glaucoma and increased orbital volume from the egress of irrigation fluid via the Baerveldt tube into the retrobulbar space. The speculum exerted lateral and posterior pressure on a soft vitrectomized eye, allowing the globe to subluxate. By reducing the egress of fluid, suspending the speculum to the drapes, and avoiding the need for vitrectomy, this eye-threatening scenario was prevented in the other eye.
WHAT WAS KNOWN
- Phacoemulsification in patients with a history of congenital glaucoma and previous glaucoma drainage devices can present intraoperative challenges.
- Globe subluxation is a rare occurrence, typically in patients with risk factors, including exophthalmos, severe lid retraction, floppy eyelid syndrome, and thyroid orbitopathy, or in those with shallow orbits.
WHAT THIS PAPER ADDS
- Extreme globe subluxation can occur during phacoemulsification, resulting in a sudden rise in posterior vitreous pressure, shallowing of the anterior chamber, and the risk for vitreous loss and expulsive hemorrhage. A combination of intravenous mannitol, pars plana vitrectomy, and manual decompression was required to successfully complete the surgery.
- The proposed pathophysiology is egress of irrigation fluid into the sub-Tenon and retrobulbar space via the glaucoma drainage device with resulting mechanical pressure from the eyelids behind the equator of the globe.
- Prevention of this complication can be achieved using pretreatment with intravenous mannitol, regional and general anesthesia, anterior suspension of the speculum to the drapes, and reduction of irrigation fluids.
1. Jones ST, Ho ST, Maleki B, Valenzuela AA. Eye popping disease: common characteristics and management of spontaneous globe subluxation. Asia Pac J Ophthalmol (Phila) 2012;1:198–201
2. Ezra DG, Derriman L, Mellingtion FE, Jayaram H, Badia L. Spontaneous globe luxation associated with shallow orbits and floppy eyelid syndrome. Orbit 2008;27:55–58
3. Zeller J, Murray SB, Fisher J. Spontaneous globe subluxation in a patient with hyperemesis gravidarum: a case report and review of the literature. J Emerg Med 2007;32:285–287
4. Apostolopoulos M, Papaspirou A, Damanakis A, Theodossiadis G, Moschos M. Bilateral optic neuropathy associated with voluntary globe subluxation and floppy eyelid syndrome. Arch Ophthalmol 2004;122:1555–1556