The traditional WT equation was significantly biased in underpredicting true unbound phenytoin concentrations in neurointensive care unit patients and should not be used in this setting. Two modified equations were more accurate and precise and should be considered for use when unbound phenytoin concentrations are not readily available in an NICU population.

The ESRD WT equation was not accurate in predicting free phenytoin concentration in patients with ESRD on HD. A revised ESRD WT equation was found to be significantly more accurate. Given the small study sample, further studies are required to fully evaluate the clinical utility of the revised ESRD WT equation.

To individualize treatment, phenytoin doses are adjusted based on free concentrations, either measured or calculated from total concentrations. As a mechanistic protein binding model may more accurately reflect the protein binding of phenytoin than the empirical Winter-Tozer equation that is routinely used for calculation of free concentrations, we aimed to develop and validate a mechanistic phenytoin protein binding model. Methods: Data were extracted from routine clinical practice. A mechanistic drug protein binding model was developed using nonlinear mixed effects modelling in a development dataset. The predictive performance of the mechanistic model was then compared with the performance of the Winter-Tozer equation in 5 external datasets. Results: We found that in the clinically relevant concentration range, phenytoin protein binding is not only affected by serum albumin concentrations and presence of severe renal dysfunction, but is also concentration dependent. Furthermore, the developed mechanistic model outperformed the Winter-Tozer equation in 4 out of 5 datasets in predicting free concentrations in various populations. Conclusions: Clinicians should be aware that the free fraction changes when phenytoin exposure changes. A mechanistic binding model may facilitate prediction of free phenytoin concentrations from total concentrations, for example for dose individualization in the clinic.

Background: Data from the US National Health and Nutrition Examination Survey are freely available and can be analyzed to produce hypertension statistics for the noninstitutionalized US population. The analysis of these data requires statistical programming expertise and knowledge of National Health and Nutrition Examination Survey methodology. Methods: We developed a web-based application that provides hypertension statistics for US adults using 10 cycles of National Health and Nutrition Examination Survey data, 1999 to 2000 through 2017 to 2020. We validated the application by reproducing results from prior publications. The application’s interface allows users to estimate crude and age-adjusted means, quantiles, and proportions. Population counts can also be estimated. To demonstrate the application’s capabilities, we estimated hypertension statistics for noninstitutionalized US adults. Results: The estimated mean systolic blood pressure (BP) declined from 123 mm Hg in 1999 to 2000 to 120 mm Hg in 2009 to 2010 and increased to 123 mm Hg in 2017 to 2020. The age-adjusted prevalence of hypertension (ie, systolic BP≥130 mm Hg, diastolic BP≥80 mm Hg or self-reported antihypertensive medication use) was 47.9% in 1999 to 2000, 43.0% in 2009 to 2010, and 44.7% in 2017 to 2020. In 2017 to 2020, an estimated 115.3 million US adults had hypertension. The age-adjusted prevalence of controlled BP, defined by the 2017 American College of Cardiology/American Heart Association BP guideline, among nonpregnant US adults with hypertension was 9.7% in 1999 to 2000, 25.0% in 2013 to 2014, and 21.9% in 2017 to 2020. After age adjustment and among nonpregnant US adults who self-reported taking antihypertensive medication, 27.5%, 48.5%, and 43.0% had controlled BP in 1999 to 2000, 2013 to 2014, and 2017 to 2020, respectively. Conclusions: The application developed in the current study is publicly available and produced valid, transparent, and reproducible results.