Purpose: To study the concordance of diurnal intraocular pressure (IOP) between fellow eyes in primary open-angle glaucoma (POAG).
Design: Retrospective chart review.
Participants: Ninety-three POAG patients.
Methods: Patients who met the definition of POAG and underwent diurnal curve measurement were included. Subjects were excluded if there was a history of surgery, trauma, ocular vascular disease, incomplete diurnal curve, or asymmetric ocular medication use. Patients on symmetric ocular medication were included and analyzed separately. Goldmann applanation tonometry was performed at 10 am, 1 pm, 4 pm, 7 pm, 10 pm, and 7 am (the next day). The following statistical analyses were performed: (1) average Pearson correlation coefficient (r) from individual correlations of right and left eye IOP over the 6 time points for each subject; (2) linear mixed model analysis for repeated measures, with eye (right and left) and time as the within-subject fixed effects, and (3) absolute difference in change in IOP between fellow eyes over each time interval and probability that the difference was within 2 or 3 mmHg.
Main outcome measure: The concordance of the IOP between fellow eyes as measured by absolute difference in change in IOP between fellow eyes and probability of the difference being within 2 or 3 mmHg.
Results: Thirty-seven patients were untreated and 56 were treated on symmetric IOP-lowering medications. The diurnal curves of fellow eyes exhibited parallel profiles according to the linear mixed model. The average difference in the change of IOP between fellow eyes over given time intervals ranged from 1.6 to 2.0 mmHg. The estimated probability that the absolute change in IOP between fellow eyes was within 2 mmHg was 68% to 90%, and within 3 mmHg was 78% to 95% for all time intervals.
Conclusion: The diurnal variation of IOP in POAG is largely concordant between fellow eyes. For any given time interval, the fellow eye IOPs may fluctuate asymmetrically a minority of the time. Clinicians who utilize the uniocular trial should be aware of the limit of the IOP concordance.