Macular hole formation after vitrectomy for retinal detachment

Editor, Earlier theories of macular hole (MH) formation include the suggestions that its development is caused by tangential vitreomacular traction (Gass 1995) or cystoid degeneration of foveal inner retinal layers. Further studies using optical coherence tomography (OCT) suggest that vitreofoveal traction plays a major role in the pathogenesis of MH (Haouchine et al. 2001). The primary purpose of vitrectomy is to relieve vitreous traction at the macula; hence, MH formation in an eye that has previously undergone vitrectomy remains a conundrum. We describe five cases of full-thickness MH formation in eyes that had previously undergone vitrectomy performed by five different surgeons (Table 1). The mean age of the patients was 59.6 years and all were male. Four patients initially presented with a macula-off rhegmatogenous retinal detachment (RD) and one with a macula-on tractional RD. They underwent RD repair with uncomplicated three-port pars plana vitrectomy (PPV), cryotherapy to the breaks and gas tamponade (20% SF6 or 14% C3F8). Macular hole was diagnosed 12 days to 4 months postoperatively (Fig. 1A) and confirmed by Stratus OCT (Carl Zeiss Meditec AG, Jena, Germany) (Fig. 1B). Four patients opted for further three-port PPV, internal limiting membrane (ILM) peel and 14% C3F8 gas tamponade with postoperative facedown posturing. In all cases, the retina remained flat after RD repair and the MH remained closed after MH surgery. There was no evidence of a MH in the fellow eye. The mean follow-up period was 12.4 months. Visual acuity improved after MH surgery in three patients and remained stable in the fourth. (A) Fundus photograph of the left eye of patient 1 showing a small macular hole. (B) Optical coherence tomography of the same eye showing a full-thickness macular hole but no clear evidence of residual cortical vitreous or an epiretinal membrane. Macular hole formation after vitrectomy has been reported infrequently. The mechanism is not fully understood. Theories of pathogenesis related to various pathologies include tangential traction by remnant vitreous, the ILM or an epiretinal membrane, foveal cystic degeneration causing umbo dehiscence with breakdown of a macrocyst and iatrogenic hole formation caused by mechanical stress on the fovea or trauma to the ILM (Li et al. 2008). It is plausible that residual undetected preretinal cortical vitreous contracts to induce tangential traction, resulting in hole formation. This was difficult to detect clinically or with Stratus OCT, but it may be below the resolution of OCT. A complete posterior vitreous detachment (PVD) may not occur in all eyes with RD. Peripheral localized vitreous separation may occur, but there may still be a persistent focal vitreomacular attachment. Alternatively, vitreoschisis may occur at the macula or across the retina. This would give the appearance of a vitreous detachment but would leave vitreous cortex attached at the macula. Persistent prefoveal vitreous has been demonstrated despite spontaneous PVD in 44% of normal autopsy eyes using scanning electron microscopy (Kishi et al. 1986). Kimura et al. (2004) showed remnant vitreous at the macula using triamcinolone acetonide during vitrectomy in eyes with RD. Whatever the mechanism, the strongly adherent vitreous may be difficult to remove from the macula during vitrectomy, particularly with a bullous detachment. It is possible that the holes developed (perhaps initially as undetected microholes) as a result of vitreous traction during spontaneous PVD. Alternatively, during intraoperative manipulation and induction of PVD, the transmission of antero–posterior mechanical forces may have caused significant vitreous traction at the fovea with subsequent MH formation. This hypothesis is difficult to substantiate. In our cases, the MHs were closed with surgery. Peeling of the ILM during MH repair may have removed the remaining contractile vitreous, thereby allowing successful hole closure. It would have been interesting to use triamcinolone acetonide or trypan blue dye intraoperatively or to send the ILM specimen for histopathology to identify residual adherent vitreous. Macular hole can occur after vitrectomy for RD repair and can compromise the results of successful retinal reattachment surgery. Although it is uncommon, surgeons should be aware of this complication because it may be prevented with careful examination and intraoperative manipulation, and the removal of any preretinal vitreous during vitrectomy.