 |
|
Choroidal Neovascularization After Epiretinal Membrane Removal
Scott M. Warden, MD;
Sophia I. Pachydaki, MD;
John B. Christoforidis, MD;
Donald J. D’Amico, MD;
John I. Loewenstein, MD
Arch Ophthalmol. 2006;124:1652-1654.
Surgical removal of an epiretinal membrane (ERM) is a relatively common procedure, often resulting in significant visual improvement. Several complications, however, are well recognized, and include cataract formation, retinal breaks and detachments, cystoid macular edema, retinal phototoxicity, and endophthalmitis.1 Choroidal neovascularization (CNV) is a rare complication of ERM surgery, and, to our knowledge, only 3 cases have been reported in the literature.2-4 We report 3 additional cases of CNV after surgical removal of idiopathic macular puckers, and comment on the literature regarding this entity.
Report of Cases
Case 1. A 47-year-old man complained of increasing metamorphopsia in his left eye for 3 years. His ophthalmic history was notable for myopia of 7.75 diopters, cataract surgery in his left eye at the age of 40 years, and a small retinal tear in the nasal periphery in the left eye treated with a laser. His visual acuity (VA) was 20/60 OS, and the result of a fundus examination revealed a macular pucker (Figure 1A). Examination results were unremarkable in the right eye. The patient underwent pars plana vitrectomy and an uncomplicated ERM peel using fine forceps.
|
|
|
|
Figure 1. For case 1, fundus photographs of when the patient was initially seen, showing an epiretinal membrane (ERM) with macular striae (A), and 2 months after ERM removal, showing elevation of the macula by a new and extensive subretinal membrane (B).
|
|
|
Four weeks postoperatively, the patient reported a central scotoma and blurred vision. Eight weeks postoperatively, his VA decreased to 20/800 OS and juxtafoveal CNV was noted on fundus examination and fluorescein angiography (FA) (Figure 1B and Figure 2A). The patient underwent pars plana vitrectomy, and the subretinal membrane was removed. Two weeks later, his VA was 20/125–1 OS and the results of a fundus examination revealed significant improvement (Figure 2B). However, 8 months postoperatively, the patient developed a rhegmatogenous retinal detachment requiring vitrectomy and silicone oil placement; his VA 2 years later was 20/500 OS.
|
|
|
|
Figure 2. For case 1, a corresponding fluorescein angiogram 2 months after epiretinal membrane removal documents intense focal hyperfluorescence consistent with fibrovascular ingrowth in the nasal macula (A) and a fundus photograph 2 weeks after subretinal membrane removal shows an improved macular appearance (B).
|
|
|
Case 2. A 77-year-old man complained of decreased VA in his right eye for 2 years. His VA was 20/80 OD. The results of an ophthalmic examination revealed cataracts in both eyes, a mild ERM in his left eye, and an ERM associated with cystic retinal changes in his right eye. The patient underwent uncomplicated pars plana vitrectomy and an ERM peel in his right eye using Tano forceps (Synergetics Inc, Wilsonville, Ore). A trace intraretinal hemorrhage was noted inferotemporal to the fovea postoperatively.
Four weeks postoperatively, the patient's VA improved to 20/50 OD. Because of a residual intraretinal hemorrhage, FA was performed, revealing no evidence of CNV.
Five months postoperatively, the patient developed metamorphopsia and his VA declined to 20/125 OD. The results of a fundus examination revealed an elevated grayish lesion temporal to the fovea. The FA showed classic extrafoveal CNV that was treated with laser photocoagulation. Three weeks following treatment, his VA was 20/100 OD. The FA showed no evidence of CNV. Seven weeks later, his VA remained stable at 20/100 OD.
Case 3. An 80-year-old man complained of deteriorating vision in his left eye 1 month after uneventful cataract extraction and intraocular lens implantation. His ocular history was significant for cataract extraction in his left eye and medically managed glaucoma in both eyes. The result of a fundus examination was unremarkable in his right eye and revealed a macular ERM and cystoid macular edema in his left eye, which was confirmed by FA. Eighteen months after cataract surgery, the cystoid macular edema had resolved; however, the patient's VA remained poor, at 20/50–1OS, secondary to the ERM. A pars plana vitrectomy and membrane peel using Tano forceps were performed. Minimal surface bleeding was noted during the procedure; 1 week later, a focal intraretinal hemorrhage was seen nasal to the fovea and a preretinal hemorrhage was seen inferior to the fovea. One month postoperatively, the patient's VA was 20/200 OS. The FA showed classic extrafoveal CNV that was treated with laser photocoagulation.
Three weeks following laser photocoagulation, the patient's VA deteriorated further to 20/400 OS. The FA showed recurrence of subfoveal CNV, which was then treated with photodynamic therapy. The patient required 4 sessions of photodynamic therapy, 2, 5, 10, and 13 months after ERM removal.
The FA 17 months following surgery showed subretinal fibrosis with no evidence of leakage. At the 2-year follow-up, the patient's VA remained poor, at 20/800 OS.
Comment
In this series, we report 3 cases of CNV that developed after idiopathic ERM removal. Three additional cases have been reported in the literature.2-4 Preoperatively, there was no evidence of age-related macular degeneration in any of the patients.
The mean duration of the ERM before surgery in the 6 cases was 22.6 months (range, 5 months to 4 years). Choroidal neovascularization was diagnosed between 1 month and 2 years postoperatively. The mean preoperative VA was 20/65 (range, 20/50-20/200). The CNV complexes were predominantly classic and were initially extrafoveal or juxtafoveal. Management included surgical removal, photocoagulation, and photodynamic therapy. The mean follow-up after CNV treatment was 1 year (range, 4-24 months). The mean final VA was 20/270 (range, 20/40-20/800). The visual outcome was largely poor when the follow-up interval was longer, and variable at shorter follow-ups.
The cause of CNV and its relation to ERM surgery is not entirely clear. It is intriguing that 3 of the 6 patients developed intraretinal or preretinal hemorrhaging in the perioperative period. While peeling the membrane from the macula, tractional forces transferred to the underlying retinal pigment epithelium and Bruch membrane could have resulted in a direct break or localized trauma reflected by areas of hemorrhaging. Causes for variable times of CNV onset postoperatively are uncertain, but may include the amount of induced trauma and the patient's predisposition to developing CNV.
Another possible explanation is that subclinical CNV coexisted with the ERM when the patient was initially seen. Gass5 described 2 patients with idiopathic macular puckers who developed CNV years later and 1 patient who had an ERM and CNV on initial examination. However, the results of a fundus examination in our patients did not reveal any clinical signs of CNV before ERM surgery. In addition, 1 patient (case 2) underwent FA 1 month after ERM removal, which showed no hyperfluorescence in the macula.
We do not believe FA is necessary before performing ERM removal to rule out preexisting CNV. However, FA is valuable in examining patients postoperatively when they develop a decreased VA and metamorphopsia. Because of the limited number of reported cases, there is no consensus about the best treatment modality. Each patient should be examined and treated based on the CNV characteristics and available treatment options.
In summary, CNV is a rare complication of ERM surgery that is likely secondary to iatrogenic trauma that is transferred to the outer retina and Bruch membrane during ERM removal. Ophthalmologists who perform ERM surgery should consider CNV a possible cause of poor visual outcome in the early and late postoperative periods.
AUTHOR INFORMATION
Correspondence: Dr Loewenstein, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114-3096 (jloewenste{at}meei.harvard.edu).
Author Contributions: Drs Warden and Pachydaki contributed equally to this article.
Financial Disclosure: None reported.
REFERENCES
|
1. Graham K, D’Amico DJ. Postoperative complications of ERM surgery. Int Ophthalmol Clin. 2000;40:215-223.
WEB OF SCIENCE
| PUBMED
2. Korobelnik JF, Dureau P, Sarfati A, Chauvaud D, Huang-Xuan T. Choroidal neovascularization complicating epiretinal membrane removal. Acta Ophthalmol Scand. 1997;75:320-321.
WEB OF SCIENCE
| PUBMED
3. Ng EW, Bressler NM, Boyer DS, De Juan E. Iatrogenic choroidal neovascularization occurring in patients undergoing macular surgery. Retina. 2002;22:711-718.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
4. Ferrari TM, Dammacco R, Quaranta GM, Cardascia N, Boscia F, Sborgia C. Choroidal neovascularization secondary to vitrectomy for idiopathic epiretinal membrane: report of an unusual case. Eur J Ophthalmol. 2003;13:218-220.
WEB OF SCIENCE
| PUBMED
5. Gass JDM. Macular dysfunction caused by vitreal and vitreoretinal interface abnormalities. In: Stereoscopic Atlas of Macular Diseases: Diagnosis and Treatment. Vol 2. 4th ed. St Louis, Mo: Mosby–Year Book Inc; 1997:944-945.
SECTION EDITOR: W. RICHARD GREEN, MD
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
RELATED LETTERS
Choroidal Neovascularization Following Macular Surgery
Thomas J. Federici
Arch Ophthalmol. 2007;125(6):856.
EXTRACT
| FULL TEXT
Choroidal Neovascularization Following Macular Surgery—Reply
John I. Loewenstein, Scott M. Warden, Sophia I. Pachydaki, John B. Christoforidis, and Donald J. D’Amico
Arch Ophthalmol. 2007;125(6):857.
EXTRACT
| FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Choroidal Neovascularization Following Macular Surgery
Federici
Arch Ophthalmol 2007;125:856-856.
FULL TEXT
|
