Transient regression of choroidal neovascularization membrane after vitrectomy in age-related macular degeneration with vitreomacular traction syndrome


Chung-May Yang, MD
Department of Ophthalmology
National Taiwan University Hospital
College of Medicine
National Taiwan University
No. 7, Chung-Shan South Road
Tel: + 886 2 23123456, ext 5187
Fax: + 886 2 2393 4420


Age-related macular degeneration (AMD) is one of the most common causes of severe visual loss in the developed countries. Besides genetic factors, inflammation and ischaemia, several recent studies have found vitreomacular adhesion may contribute to the formation of choroidal neovascularization (Krebs et al. 2007; Francesca et al. 2008; Lee et al. 2009), and surgical release of the vitreomacular traction may be beneficial in promoting the regression of choroidal neovascularization membrane (CNVM) (Ikeda et al. 2000; Francesca et al. 2008). The definite relationship between vitreoretinal traction and CNVM as well as the long-term therapeutic effect of induction of vitreomacular separation remains not well established. We report a case of exudative AMD with vitreomacular traction and traction retinal detachment. Vitrectomy resulted in only temporal quiescence of CNVM.

A 84-year-old man with macular disciform scar in the right eye (OD) had cataract operation in the left eye (OS) in 2005. Best-corrected visual acuity (BCVA) was 20/50 after surgery. However, progressive blurred vision developed in July 2006. Fluorescein angiography revealed CNVM (Fig. 1). Partial regression of CNVM was noted after photodynamic therapy (PDT) with Visudyne (verteporfin); further PDT showed minimal effect. Optical coherence tomography (OCT) showed obvious vitreomacular adhesion (Fig. 2A). Further decrease of visual acuity to finger counting vision in OS was noted in the following 4 months. Repeated OCT showed significant vitreomacular traction and macular sensory retina detachment (Fig. 2B). Pars plana vitrectomy and separation of adherent posterior vitreous cortex were performed in OS in February 2007. After surgery, retina was reattached, and CNVM regressed significantly (Fig. 2C). Visual acuity in OS improved to 20/150. Twelve months after the surgery, recurrent CNVM with decreased vision was noted (Fig. 2D). Repeated courses of intravitreal injection of bevacizumab were necessary to keep a dry macula. BCVA was 20/500 at the most recent follow-up.

Figure 1.

 Fundus photograph and late-stage fluorescein angiography showed active choroidal neovascularization in OS before treatments.

Figure 2.

 Serial OCT pictures in our case. (A) OCT in October 2005 showed thin posterior hyaloid adhesion and the presence of undulation and thickening of retinal pigment epithelium. (B) OCT in February 2006 before the surgery showed obvious vitreomacular traction with sensory retinal detachment at the central macular area. (C) postsurgery OCT showed reattached sensory retina and smooth contour of the retinal pigment epithelium after the surgery, indicating regression of choroidal neovascularization membrane (CNVM). (D) OCT showed the recurrence of CNVM with cystoid macular oedema 12 months after the surgery.

Recent studies raise the possibility that vitreomacular adhesion may be a risk factor of CNVM formation in AMD. The proposed pathogenesis includes the following: (i) interference of the metabolism of the cells of macula caused by traction-induced local ischaemia and disruption of choroidal supply, and subsequently leading to increased secretion of vascular endothelium growth factors; (ii) chronic low-grade inflammation induced by persistent traction; (iii) mechanical force on the macula leading to pigment epithelial detachment or tear; (iv) macular oedema induced by chronic stimuli of persistent traction (Krebs et al. 2007; Francesca et al. 2008; Schulze et al. 2008).

In our case, surgical release of vitreomacular traction induced prompt regression of CNV. However, the stable postsurgical condition could not be maintained. Recurrence of CNV was observed 12 months after the surgery. The recurrence pattern of CNVM in our case has not been reported in those case series.

From our case study, we postulate that vitreoretinal traction may not be one of the initial events leading to AMD. Rather, it is possible that chronic inflammation and scarring process of the macula in AMD make posterior vitreous more adherent to the retina, and the vitreomacular adhesion further aggravates the pathologic process of AMD, thus forming a vicious cycle. Surgical release of vitreomacular adhesion, or more recently, chemical vitreolysis with intravitreal injection of microplasmin, may block this pathogenic process at preretina level but cannot eliminate the existent pathologic changes of RPE cells occurred at subretina level. The experience of our case suggested that surgical intervention may have beneficial effect on the disease process; however, it may not inhibit disease recurrence. Further, long-term longitudinal studies are necessary to determine the therapeutic effect of these interventions on the treatment of AMD.