Purchase this article with an account.
J. D. Peterson, S. D. Peterson, C. J. Roberts, A. M. Mahmoud, P. A. Weber; Longitudinal Goldmann Applanation Tonometry Measurement and Age: A New Explanation. Invest. Ophthalmol. Vis. Sci. 2008;49(13):689. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To evaluate the longitudinal relationship of age with Goldmann Applanation Tonometry (GAT) measurement.
GAT measurements from left eyes of 171 subjects were collected retrospectively from patient charts of an independent optometrist. Subjects with a history of glaucoma, LASIK, or other corneal surgery besides uncomplicated cataract surgery were excluded. Subjects had at least 6 years of data with a minimum of 3 measurement points. All GAT measurements were obtained by the same optometrist using the same Goldmann tonometer that was calibrated by the same technician over a period of 23 years. Group and individual regression analyses for each subject were performed and subjects were divided into 3 groups: significant (p<.05), marginally significant (p=0.05-0.1), and not significant (p>0.1).
Age of subjects at time of first measurement ranged from 13 to 82 years with an average starting age of 43.8 ± 15.1 years. Number of measurements recorded per subject ranged from 3 to 17 with an average of 7.44 ± 2.71 measurements. The time span for individual subject measurements was 6 to 23 years, with an average time span of 13.6 ± 4.22 years. When all subject data were combined in a single regression, as would be done in a cross-sectional study, IOP showed a weak, but significant positive relationship with age (p<.0001, R2=0.0291). However, analysis of the regressions for individual subjects revealed the following relationships: of the 171 eyes analyzed, 116 eyes (67.8%) were not statistically significant, 14 eyes (8.2%) were marginally significant, 1 eye (0.6%) was significantly negative, and 40 eyes (23.4%) were significantly positive. The range of R2 for those eyes that had a significant positive relationship was 0.378-0.996, with a mean R2 of 0.702 ± 0.147.
There is a very strong positive relationship between GAT IOP and age in a small percentage of the population (23% in this study), while the majority of the population (76%) shows no statistically significant relationship with age. The correlation between age and GAT IOP in the small subset of the population is so strong that it drives the overall relationship to be weakly positive. These results may explain why the relationship between GAT and age has been debated for decades, with some studies reporting a significantly positive relationship while others report no relationship or even a negative relationship. Further research into what factors cause the small subset of the population to have such a strong GAT relationship with age is indicated.
This PDF is available to Subscribers Only