Editor, There is accumulating evidence that the vitreomacular interface, vitreomacular traction and mechanical stress might be risk factors for the development of age-related macular degeneration (AMD): vitreomacular adhesion may promote exudative AMD, total posterior vitreous detachment (PVD) as well as vitrectomy have been found to be protective against the development of AMD (Krebs et al. 2011; Schulze et al. 2012). The vitreous cortex touches the retinal surface, and developmental and age-related remodelling of the vitreous as well as the presence of a bursa premacularis has been described before (Worst & Los 1995). Los (1997) described in her PhD thesis that the human vitreous shows perimacular concentric circles of longitudinal cisterns as well as a bursa premacularis. Moreover, the perpendicular pillar-like cisterns undergo changes in time. More than 100 autopsy eyes, studied by Worst, were derived from patients above the age of 50 (mean age: 70); the findings of Los, described in her thesis, were based upon the examination of more than one hundred donor eyes from younger patients, with varying ages. All autopsy eyes were studied with courtesy of the Corneal Eye Bank in Amsterdam, and the anatomical studies were performed by the Jan Worst Research Group in Haren, the Netherlands. However, as it is technically very difficult to reveal and to photograph intact intravitreal structures in their three-dimensional organization after injection dye into fresh, unfixed vitreous bodies suspended in a balanced salt solution (BBS) – several specimens got damaged during preparation or were even lost!– it is impossible to give the exact number of eyes in which the typical perimacular elongated cistern-like structures and the bursa premacularis could be visualized (see Fig. 1A). With informed consent, clinical data were collected from ten patients above the age of 50 during triamcinolone-assisted vitrectomy in the Alkmaar Medical Center in Alkmaar, the Netherlands. The human vitreous shows in vivo identical perimacular concentric circles of longitudinal cisterns as well as a bursa premacularis (see Fig. 1B and C). Based upon these identical anatomical aspects of the vitreous and the location of perimacular and macular lesions in AMD, we hypothesize that there is a relationship: as such retinal and choroidal lesions geographically correspond with the location of the longitudinal vitreal cisterns and the bursa premacularis, they can be interpreted as mechanical effects that may progress to AMD, depending on the magnitude and direction of vector forces during eye rotations. The usual chronic course of AMD with exacerbations and remissions and often progressive, bilateral appearance are suggestive of intermittent traumatic effects. Physiological vitreous changes and eye rotations occur in everyone, but vitreous structures will only touch the retina in the presence of elongated cisterns and/or significant vitreous movements, which process might be facilitated by vitreomacular adhesions. Our hypothesis that mechanical stress might contribute to the pathogenesis of AMD is supported by the clinical findings of others, who used spectral domain optical coherence tomography to evaluate the vitreoretinal interface: a positive relationship between a persistent attachment of the posterior vitreous and early signs of AMD was found (Schulze et al. 2012) as well as a complete correspondence between the adhesion and subretinal neovascularization (Krebs et al. 2011). We realize that observational and anatomical, experimental studies have their limitations, and further histopathological as well as case–control studies are required to confirm our hypothesis. Bursa premacularis filled with ink in an autopsy eye (A) and clinically after injection of triamcinolone during pars plana vitrectomy (B and C). Area of Martegiani on the left of the bursa in the autopsy eye (A). The bursa premacularis in the center is surrounded by distended cisterns of the circulus petaliformis: these structures are thought to make contact with the retinal surface periodically. We thank Peter J. de Lint, vitreoretinal surgeon in the Medical Center Alkmaar; the Corneal Eye Bank, Amsterdam; the Jan Worst Research Group, Haren; the ophthalmic photography units of the VU University Medical Center, Amsterdam; and the Medical Center Alkmaar, the Netherlands.