Intravitreal corticosteroids are frequently used to treat diabetic macular edema (DME), particularly when use of anti-vascular endothelial growth factor (VEGF) agents has not yielded sufficient improvement. Long-acting dexamethasone intravitreal implants (Ozurdex; Allergan, Dublin, Ireland) and fluocinolone acetonide implants (Retisert; Bausch + Lomb, Rochester, NY, and Iluvien; Alimera Sciences, Alpharetta, GA) have been used to enhance the action of intravitreal steroids for longer duration of effect. The most common drug-related complications of increased cataract formation and raised intraocular pressure are well-known and frequently managed. However, reports of retinal impact defects, macular hole formation, injection into the crystalline lens or into the subretinal space represent serious, though rare, previously reported adverse events (AEs) for the dexamethasone implant.1–3 This case represents a novel complication showing degradation of nearby retina from an abutting dexamethasone implant.
A 73-year-old male presented to clinic for routine follow-up. The patient had extensive previous treatment for bilateral DME including previous pars plana vitrectomy (PPV) with internal limiting membrane peeling in both eyes and multiple intravitreal injections of both anti-VEGF and intravitreal corticosteroid medications. During the patient's exam three new adjacent inferior retinal holes with a cuff of subretinal fluid were noted in the left eye (Figure 1). A posterior vitreous detachment had been created during PPV of the left eye 4 years prior. There were no operculums or retinal flaps to denote recent traction. Juxtaposed to the holes was the remainder of an Ozurdex dexamethasone intravitreal implant, placed 3 months prior to this visit. The patient had been seen 4 weeks earlier, and no holes were noted at that time. The formation of retinal holes occurred well after insertion of the drug, reducing the possibility of impact damage and without flap or operculum formation decreasing the likelihood of a tractional event. Options were discussed with the patient including laser barricade and observation. Without evidence of traction and a history of PPV with induction of a posterior vitreous detachment, the risk of progression was low. After counseling, the patient opted for laser barricade (Figure 2). The holes were surrounded with laser extending to the ora serrata to reduce the risk of a retinal detachment.
Fundus photograph of the left eye showing three adjacent inferior retinal holes with resting dexamethasone implant in the hole closest to 6-o'clock and a cuff of subretinal fluid temporally. Photo also shows evidence of severe nonproliferative diabetic retinopathy.
Three juxtaposed retinal holes and subretinal fluid surrounded by laser barricade. Degradation of the implant is noted.
The dexamethasone 0.7 mg sustained release intravitreal implant has been safely used to decrease macular edema in the setting of DME and retinal vein occlusion as well as treat noninfectious uveitis.4 This case exhibits a previously unreported adverse event with the development of full-thickness retinal defects in the far periphery adjacent to a resting implant. The injection had been performed in standard fashion without complication.
In eyes with previous PPV, the most common adverse events were found to be conjunctival hemorrhage and conjunctival hyperemia (72.7%).5 In the study by Boyer et al., of 56 patients, no adverse events of retinal hole or tear formation were noted. Also discussed in this study, the implant dissolves into inert substances (carbon dioxide and water) over time.5 The NOVADUR solid polymer degrades into lactic acid and glycolic acid. Prior to dissolution, the implant may theoretically cause mechanical irritation and surface disruption.
In an online video interview, Michael Singer, MD, of Medical Center Ophthalmology Associates in San Antonio, Texas, discussed the mechanism of injection of the Ozurdex implant. According to the video, the implant is not a spring-released mechanism but instead accordion released; essentially, the ejection of the implant from the applicator is directly proportional to the force applied to the button rather than the potential energy of a spring.6 This application force can therefore be adjusted by the physician performing the injection: the more gradual the button is pushed, the less force the implant exits with from the applicator.6
Christenson et al. continued this line of thinking, reporting the force of administration of an Ozurdex implant was the possible cause of an eccentric MH.2 The patient's posterior hyaloid was still intact and the hole developed only after receiving a dexamethasone implant from an outside ophthalmologist. This study further elucidates the force of the applicator through both air and basic saline solution, finding a reduction from 0.77 Newtons to 0.024 Newtons respectively.2 That study patient was not vitrectomized, and though the average generated force through basic saline solution was not thought to be sufficient for retinal impact damage, ample force is generated through air. Given the physician variability of force proposed by Dr. Singer, retinal damage could be attributed to the Ozurdex implant. However, our patient's retinal holes were not seen during an interval office visit between insertion of the implant and their discovery. There are also three holes and it would be unlikely that a single implant caused all three defects through impact.
Further evidence of mechanical irritation is found in a report by Uwaydat et al. A patient with noninfectious uveitis was treated with serial dexamethasone intravitreal implants. When the patient developed an epiretinal membrane (ERM), a PPV with ERM and internal membrane peeling was conducted. An implant was placed at the time of surgery, and 30% sulfur hexafluoride was added due to concern for a macular hole. At the postoperative 1-week visit, the implant was adhered to the superior macula causing distortion of the retinal layers on ocular coherence tomography (OCT). A second surgery was performed to move the implant and follow-up OCT confirmed retinal layer compression and hyperautofluorescence possibly indicating retinal pigment epithelial dysfunction. These findings normalized 10 days on both OCT and fundus autofluorescence after the second surgery with inferior displacement of the Ozurdex implant.1 The aforementioned products of lactic acid and glycolic acid, as well as high-dose concentrated dexamethasone, which would exist at the site of adherence in this case, have been shown to cause corneal edema, apoptosis, and necrosis.7 These same forces could be the cause of retinal damage. In our case, it is possible the implant localized to one area with possible adhesion and produced the causative damage shown.
Retinal defects are a potential AE with intravitreal dexamethasone 0.7-mg implant injection. The likely mechanism is similar to that of corneal decompensation. This case underscores the importance of thorough retinal exam during a treatment course, to not be solely focused on the macular edema. Also, care must be taken with each injection, particularly in vitrectomized eyes and those with gas or air as the dynamics of force have changed. Our patient opted for laser barricade prior to any further complications and has not experienced further AEs.
- Uwaydat SH, Sallam AB, Wang H, Goyal S. Retinal indentation by a dexamethasone implant in a gas-filled eye: Report of an unusual complication. JAMA Ophthal. 2017;135(10):1125–1127. doi:10.1001/jamaophthalmol.2017.3276 [CrossRef]
- Christensen L, Sanders R, Olson J. “Magic Bullet”: Eccentric macular hole as a complication from dexamethasone implant insertion. Case Rep Ophthalmol Med. 2016;2016:1706234. Epub 2016 Oct 5.
- Karandikar SS, Manayath GJ, Saravanan V, Narendran S, Narendran V. Iatrogenic subretinal injection of Ozurdex implant and its effect on macular edema. Oman Journal of Ophthalmology. 2017;10(2):100–102.
- Haller JA, Dugel P, Weinberg DV, Chou C, Whitcup SM. Evaluation of the safety and performance of an applicator for a novel intravitreal dexamethasone drug delivery system for the treatment of macular edema. Retina. 2009;29(1):46–51. doi:10.1097/IAE.0b013e318188c814 [CrossRef]
- Boyer D, Faber D, Gupta S, et al. Dexamethasone intravitreal implant for treatment of diabetic macular edema in vitrectomized patients. Retina. 2011;31(5):915–923. doi:10.1097/IAE.0b013e318206d18c [CrossRef]
- Prenner JL, Halperin L, Singer M. The Three-Click Technique. [video] https://eyetube.net/series/retina-today-journal-club/fpbej. Published December 11, 2017. Accessed February 17, 2018.
- Khurana RN, Appa SN, McCannel CA, et al. Dexamethasone implant anterior chamber migration. Ophthalmology. 2014;121(1):67–71. doi:10.1016/j.ophtha.2013.06.033 [CrossRef]