BACKGROUND AND OBJECTIVE: Effective control of intraocular pressure (IOP) is essential for the successful management of glaucoma. IOP exhibits diurnal variation, yet continuous monitoring is impractical. To date, no clear evidence exists on the number of sampling timepoints required to characterize diurnal IOP and when those measurements should be collected. The objective of this study was to develop an optimized sampling scheme to estimate diurnal IOP and to provide sampling windows for practicality. METHODS: Baseline IOP values for glaucoma patients were collected from the published literature. A population model-based meta-analysis was performed to develop a model for diurnal IOP that accounts for covariates and inter-study variability. Optimization was performed using the D-optimality criteria to determine optimal sampling times. In addition, various reduced sampling designs were tested to investigate the minimum number of sampling timepoints to precisely estimate diurnal IOP. Also, sampling windows were calculated around the final optimal sampling times to allow flexibility in data collection. The final reduced optimized model was validated by simulating and estimating 500 datasets with reduced optimal sampling times. RESULTS: The final baseline IOP model included type of glaucoma as a covariate. Bootstrap analysis and visual predictive check plots revealed the adequacy of the model to describe the observed IOP data. Optimization results indicated an increasing trend in bias with decreasing sampling timepoints. A reduced model with four sampling times resulted in acceptable precision (<40 %). Restricting the sampling time between 8 a.m. and 4 p.m. underestimates the fluctuation in diurnal IOP. Sampling windows with ≥95 % efficiency were calculated around the optimized sampling times. Validation results indicated acceptable precision and relative bias for model estimates in the reduced optimized model. CONCLUSION: A physiologically based mechanistic model was developed to describe the diurnal variation in baseline IOP and inter-study variability was estimated on key diurnal model parameters. Optimization of the final covariate model indicated a reduced sampling time of at least four samples should be collected at 5:45 a.m., 2:15 p.m., 8:00 p.m., and 12:00 a.m. for reliable estimation of diurnal IOP variation.
BACKGROUND AND OBJECTIVE: Effective control of intraocular pressure (IOP) is essential for the successful management of glaucoma. IOP exhibits diurnal variation, yet continuous monitoring is impractical. To date, no clear evidence exists on the number of sampling timepoints required to characterize diurnal IOP and when those measurements should be collected. The objective of this study was to develop an optimized sampling scheme to estimate diurnal IOP and to provide sampling windows for practicality. METHODS: Baseline IOP values for glaucomapatients were collected from the published literature. A population model-based meta-analysis was performed to develop a model for diurnal IOP that accounts for covariates and inter-study variability. Optimization was performed using the D-optimality criteria to determine optimal sampling times. In addition, various reduced sampling designs were tested to investigate the minimum number of sampling timepoints to precisely estimate diurnal IOP. Also, sampling windows were calculated around the final optimal sampling times to allow flexibility in data collection. The final reduced optimized model was validated by simulating and estimating 500 datasets with reduced optimal sampling times. RESULTS: The final baseline IOP model included type of glaucoma as a covariate. Bootstrap analysis and visual predictive check plots revealed the adequacy of the model to describe the observed IOP data. Optimization results indicated an increasing trend in bias with decreasing sampling timepoints. A reduced model with four sampling times resulted in acceptable precision (<40 %). Restricting the sampling time between 8 a.m. and 4 p.m. underestimates the fluctuation in diurnal IOP. Sampling windows with ≥95 % efficiency were calculated around the optimized sampling times. Validation results indicated acceptable precision and relative bias for model estimates in the reduced optimized model. CONCLUSION: A physiologically based mechanistic model was developed to describe the diurnal variation in baseline IOP and inter-study variability was estimated on key diurnal model parameters. Optimization of the final covariate model indicated a reduced sampling time of at least four samples should be collected at 5:45 a.m., 2:15 p.m., 8:00 p.m., and 12:00 a.m. for reliable estimation of diurnal IOP variation.
Authors: Anastasios G P Konstas; Kostantinos Boboridis; Despina Tzetzi; Kostantinos Kallinderis; Jessica N Jenkins; William C Stewart Journal: Arch Ophthalmol Date: 2005-07
Authors: Anastasios G P Konstas; Symeon Lake; Athanasios I Economou; Kostantinos Kaltsos; Jessica N Jenkins; William C Stewart Journal: Arch Ophthalmol Date: 2006-11
Authors: Tomas M Grippo; John H K Liu; Nazlee Zebardast; Taylor B Arnold; Grant H Moore; Robert N Weinreb Journal: Invest Ophthalmol Vis Sci Date: 2013-01-17 Impact factor: 4.799
Authors: F Topouzis; S Melamed; H Danesh-Meyer; A P Wells; V Kozobolis; H Wieland; R Andrew; D Wells Journal: Eur J Ophthalmol Date: 2007 Mar-Apr Impact factor: 2.597