Sir,

The Ozurdex implant (Allergan, Inc., Irvine, CA, USA) is licensed in the UK for treatment of adult patients with macular oedema following retinal vein occlusion (RVO). We present a case of inadvertent injection of dexamethasone implant into the crystalline lens, its effects and management.

Case report

A 56-year-old Asian woman presented with a 2-week history of blurred vision in her right eye. Her BCVA was recorded at: OD 6/60; OS 6/6. Angiography confirmed hemiretinal vein occlusion with diffuse macular oedema. Treatment with intravitreal dexamethasone implant (Ozurdex, Allergan, Inc.) was planned. The patient was extremely nervous and apprehensive about the treatment. As she could not communicate in English, an appropriate interpreter was arranged and efforts were made to allay her anxiety by explaining to her about the procedure and what was expected of her during the process.

The eye was anaesthetised with topical benoxinate 1% and subconjunctival injection of 2% lignocaine in the superotemporal quadrant. She was still anxious and despite further attempts to reassure her and control the eye position, the implant was accidentally injected into the crystalline lens (Figure 1).

Figure 1
figure 1

(a) Anterior segment photograph showing dexamethasone implant within the patient’s crystalline lens. (b) OCT scan demonstrating diffuse macular oedema with the presence of intraretinal and subretinal fluid. (c) OCT map overlays on the colour fundus photograph confirming increase in central retinal thickness. Disc and retinal haemorrhages are also evident in the background.

Full size image

The patient was reviewed on days 1, 2 and 7. No inflammatory reaction was seen and intraocular pressure (IOP) remained normal. Hence, a conservative management plan was put in place and she was monitored closely for complications related to Ozurdex such as progression of pre-existing cataract, phacolytic glaucoma, steroid-induced glaucoma, and retinal detachment.

After 2 weeks the IOP rose to 28 mm Hg, which was managed well with topical anti-glaucoma medications. At this point, the macular oedema had already started to resolve and central retinal thickness (CRT) on OCT scan had reduced from 560 μm to 375 μm. Conservative management was continued to see whether the implant would biodegrade within the lens.

After 3 months, the implant did not show any notable signs of degenerating. At this point, despite the implant being within the natural lens, there was a remarkable improvement in macular oedema, resolving almost completely with CRT at 256 μm.

Besides, the cataract had progressed significantly; hence, the decision was taken to proceed with surgery. Cataract extraction was performed through a 2.8-mm clear-corneal incision. The cataract was posterior subcapsular with a soft nucleus so was easily aspirated along with the Ozurdex implant using a phacoemulsification probe. A small capsular defect was noted, which enlarged during soft lens matter removal and vitreous humour entered the anterior chamber. After cortical clean-up and straightforward anterior vitrectomy, an AcrySof MA60 (Alcon Laboratories, Inc., Fort Worth, TX, USA) intraocular lens was placed in the sulcus with optic captured within the intact anterior capsule. The patient had an uncomplicated post-operative recovery. At 1 month BCVA was 6/15 and the macula was dry (Figure 2) and this remains stable at 12 months follow-up.

Figure 2
figure 2

(a) Anterior segment photograph showing posterior chamber intraocular lens implant placed in the sulcus. (b) OCT scan demonstrating dry macula with almost complete resolution of fluid and presence of good foveal contour. (c) OCT map overlays on the colour fundus photograph confirming significant reduction in central retinal thickness. Note disc and retinal haemorrhages have also resolved in just over 3 months.

Full size image

Comment

We present a case of accidental intralenticular injection of Ozurdex, its effects and management. Various reports1, 2, 3 in the literature describe migration of intraocular implants but none relate to the intralenticular position. Despite the implant being within the lens, our patient had good response to the treatment. Furthermore, the side effects of treatment were no worse than that reported in the Geneva study.4, 5 Although the implant landed in the crystalline lens and potentially contributed to earlier manifestations of known side effects, watchful and conservative management was prudent and outcomes, both in terms of VA and anatomical improvement of macular oedema, were good.

This case also highlights the need to ensure patient education, particularly keeping in mind the language barriers. Extra time with the patient to explain the procedure, with the help of an interpreter if required, may calm and acclimatise the patient.