Reduced vitreal concentration of periostin after vitrectomy in patients with proliferative diabetic retinopathy.

Abstract

In an earlier study, we identified periostin, a 90-kDa-secreted matricellular protein involved in wound repair (Kudo et al. 2007), as a molecule whose expression is enhanced in proliferating epiretinal fibro(vascular) membranes (FVMs) and in the vitreous of patients with both proliferative diabetic retinopathy (PDR) and proliferative vitreoretinopathy (PVR; Ishikawa et al. 2013). These results identified periostin as an important molecule for FVM formation. However, to the best of our knowledge, there is little direct evidence on how vitrectomy affects the periostin-mediated fibro(vascular) proliferation. Thus, the purpose of our study was to determine whether vitrectomy alters the levels of periostin in the vitreous of eyes with PDR. In our University Hospital, we have been using a two-step surgical strategy for the treatment of patients with severe PDR (Yoshida et al. 2012). It was decided that patients with severe PDR would undergo vitrectomy without the insertion of an IOL at the initial surgery, and an IOL would be implanted only after confirming that the activity of retinopathy had calmed down after the initial vitrectomy. We collected vitreous samples from 54 eyes of 54 patients with macular hole as control and 36 eyes of 33 patients with PDR before the vitrectomy without IOL implantation. We obtained 36 samples from the same patients at the time of the IOL implantation during a second vitrectomy. The interval between the initial vitrectomy and IOL implantation was 3.1 to 25.7 (mean 6.7) months. The mean concentration of periostin in the vitreous was significantly higher in the 36 vitreous samples collected from 33 patients with PDR (11.44 ± 2.62 ng/ml) than in the vitreous of the control patients (0.12 ± 0.03 ng/ml, p < 0.001, Fig. 1A). At the time of the IOL implantation, the periostin level was significantly higher (3.39 ± 1.33 ng/ml, p = 0.007) than that in the control patients, but the periostin level was significantly lower than the level in the vitreous collected at the initial vitrectomy (11.4 ± 2.56 ng/ml, p < 0.001, Fig. 1A). Finally, the periostin level was significantly and inversely correlated with the interval between the first vitrectomy and the second vitrectomy for IOL implantation (r = −0.38, p = 0.02, Fig. 1B). The reason why vitrectomy is effective in reducing periostin has not been determined. Because the level of periostin protein was reduced in a time-dependent manner (Fig. 1B), one possible explanation is that there is an increase in the diffusion of periostin away from the retina by the replacement of the vitreous gel with less viscous saline. Another reason for the quieting effects of vitrectomy may be the removal of FVMs because we have shown that the smooth muscle cells and M2 macrophages in FVMs actively produce periostin as well as a variety of adhesion-molecules (Yoshida et al. 2011). Anti-VEGF therapy is being used to treat neovascular diseases of the eye, and its use has led to significant advances in the management of PDR. However, this treatment turns on an ‘angiofibrotic switch’ to favour a fibrotic phase, and tractional retinal detachment have been reported in PDR patients following the administration of anti-VEGF agents in spite of the neovascular inhibition (Sohn et al. 2012). This indicates that anti-VEGF therapy is not effective in inhibiting fibrotic proliferation. Our findings that successful vitrectomy can reduce both periostin and VEGF for a long time (Yoshida et al. 2012; Fig. 1B) suggest that a successful vitrectomy may inhibit both angiogenesis and subsequent fibrosis by reducing the level of these molecules.