RESUMO
BACKGROUND: Glaucomatous eyes often show strong intraocular pressure (IOP) fluctuations and individual measurements at different time points are necessary for personalized therapy. To survey IOP variations 48-hours diurnal and nocturnal IOP measurements were performed on two consecutive days. Aims of this study were to investigate the short-term repeatability of 48-hours measurements within one patient's IOP profile and long-term repeatability between two separate IOP profiles of the same patient. METHODS: A retrospective cohort study was performed evaluating data of 90 glaucoma patients in a German university medical center between 2006 and 2013. All patients underwent two separate diurnal IOP profiles of 48 h. IOP was measured at 8 am, 2 pm, 6 pm, 9 pm using Goldmann applanation tonometry and at 12 midnight using Perkins tonometry in supine position on two consecutive days. Intraclass correlation coefficients (ICC) were calculated to evaluate agreement for the same time points (each time point agreement) and for consecutive measurements within the IOP profiles (between time point agreement). ICC ≤ 0.4 was defined as poor agreement, 0.4-0.75 as moderate and ≥ 0.75 as excellent. Differences between time points were investigated by Bland Altman plots. RESULTS: Each time point measurements of profile 1 showed moderate to excellent agreement (ICCs 0.62-0.93). There was a moderate to excellent agreement for measurements between time points of profile 1 (ICCs day one 0.57-0.86, day two 0.71-0.90). Profile 2 was performed at a median interval of 12.0 months (quartiles 11.0 to 21.0). Each time point agreements within profile 2 showed ICCs from 0.23 to 0.81. It showed moderate to excellent agreement for changes between time points (ICCs 0.53-0.94). Day two demonstrated ICCs from 0.74 to 0.88. Long term IOP repeatability (over both pressure profiles) showed moderate to good agreement (ICCs 0.39-0.67). CONCLUSIONS: Short and long-term agreement of IOP measurements evaluated by diurnal IOP profiles is moderate to good. Due to mostly moderate agreements, which we believe represent IOP fluctuations, we conclude that it is necessary to perform 48-hours IOP profiles to gain a better overview of the individual IOP fluctuations.