The Age-Related Impact on Applanation vs. Schiotz IOP Discrepancy: Implications for Community Glaucoma Screening

Abstract

Glaucoma is a major cause of irreversible blindness, and although it is not defined solely by intraocular pressure (IOP), IOP remains the most important modifiable risk factor, making accurate tonometry critical for community screening in aging populations. This prospective, observational, comparative study assessed age-related agreement between Goldmann applanation tonometry (GAT) and portable Schiøtz indentation tonometry, which is widely used in outreach settings but is more sensitive to biomechanical changes that accompany aging. Adults aged 18–75 years with clear corneas, open angles, and no prior glaucoma were recruited from an eye OPD and community screening programs and stratified into 18–39 years (n=48), 40–59 years (n=314), and 60–75 years (n=14). A single optometrist followed a standardized protocol (GAT first, then Schiøtz; average of two consistent readings), and refractive error was categorized by spherical equivalent to evaluate its influence. Across all age groups, Schiøtz measured higher mean IOP than GAT, with GAT versus Schiøtz values of 14.66±2.38 vs 15.27±2.64 mmHg (18–39), 14.96±2.50 vs 15.63±2.77 mmHg (40–59), and 15.36±3.34 vs 16.12±3.87 mmHg (60–75), indicating a widening discrepancy with age. In older adults, Bland–Altman analysis demonstrated a mean bias of +0.76 mmHg with 95% limits of agreement from −0.47 to +2.00 mmHg, greater scatter at higher mean IOP, and an upper limit approaching the prespecified clinically relevant threshold (>2 mmHg). Discrepancies were largest in myopic eyes—especially among those aged 60–75 years—moderate in hyperopes, and smallest in emmetropes, identifying elderly myopes as the subgroup most prone to clinically meaningful IOP overestimation by indentation tonometry. Overall, Schiøtz tonometry remains a pragmatic low-cost screening tool, but borderline or elevated Schiøtz readings in elderly individuals, particularly those with refractive errors, should be confirmed with applanation-based methods to reduce misclassification and inappropriate referral.