Background/aims To investigate the effect of eyelid manipulation on the measurement of intraocular pressure (IOP) using two different tonometries (rebound tonometry (RT) vs Goldmann applanation tonometry (GAT)).

Methods 103 patients with primary open-angle glaucoma were prospectively enrolled. For all of the patients, IOP measurements were performed in three different ways: (1) RT with lid manipulation (LM), (2) RT without LM and (3) GAT. The order of the three measurements was randomly selected. Additionally, the palpebral fissure height (PFH; elliptical space between upper and lower eyelids) was measured.

Results The mean value of IOP measured by GAT was 13.97±2.80 mm Hg, which was not significantly different from that by RT without LM (13.75±2.44 mm Hg; P=0.096), but which was significantly lower than that by RT with LM (15.21±2.91 mm Hg; P<0.001). On a Bland-Altman plot, RT with LM was overestimated relative to GAT (mean: −1.5) and RT without LM (mean: −1.2). Among the high IOPs (>20 mm Hg), interestingly, those measured by RT without LM were significantly lower than those measured by GAT (P<0.001). In the subgroup analysis of PFH, the smaller the PFH, the more exaggerated the IOP difference between GAT (P=0.014) and RT with LM (P<0.001).

Conclusion RT-measured IOP was significantly exaggerated when manipulation was applied to the eyelid. This overall trend was more pronounced when PFH was small. GAT-measured IOP, meanwhile, showed a good correlation with IOP measured using RT without LM.