Abstract / Summary
Abstract Background: Plethysmometry is a cuff based method that records photoplethysmographic pulse amplitudes at multiple cuff pressure levels and estimates ranges of systolic and diastolic blood pressure together with mean arterial pressure. Objective: To compare plethysmometric BP estimates with carefully obtained auscultatory sphygmomanometric measurements and to determine classification thresholds of plethysmometric pressures for hypertension diagnosis by comparison with sphygmomanometric reference. Methods: Ninety six adults underwent four video and audio recorded mercury sphygmomanometric measurements (M0 to M4), two before and two after a single plethysmometric recording. Two observers independently interpreted the recordings, with adjudication when their readings differed by more than 2 mmHg. The mean of measurements M1 to M3 was used as the reference sphygmomanometric pressure. Plethysmometry recorded cuff pressure and finger photoplethysmograms from the cuffed and uncuffed arms during 16 cuff pressure plateaus lasting for ten seconds each. From the relationship of pulse amplitude (in the cuffed arm) to cuff pressure during the plateaus, lower and upper limits of systolic and diastolic pressures were derived. Average systolic and diastolic pressures were calculated as midpoints of their ranges. Average pulse pressure was estimated. Mean arterial pressure (MAP) was estimated independently from cuff pressure oscillation amplitude; when the maximum cuff oscillation amplitude extended across adjacent pressure levels, lower and upper MAP estimates were reported and their midpoint used as the average MAP. Agreement of plethysmometry and sphygmomanometry for average systolic and diastolic pressures was evaluated using mean differences, Bland Altman analysis, the numerical requirements of AAMI/ESH/ISO Criterion 1, and the historical BHS grading system. ROC analysis was used to explore plethysmometric thresholds for classifying hypertension, as compared to sphygmomanometric thresholds of 140 mmHg for systolic and 90 mmHg for diastolic pressures. Results: The mean difference between plethysmometric and sphygmomanometric estimates was 0.97 [6.58] mmHg (mean[SD]) for systolic pressure and -3.02 [8.74] mmHg for diastolic pressure, meeting the numerical limits specified in AAMI/ESH/ISO Criterion 1 for systolic pressure, but not for diastolic pressure. Historical BHS grading was Grade A for systolic and Grade C for diastolic pressures. Thirteen participants were hypertensive by the sphygmomanometric method, using the conventional thresholds of 140 mmHg and 90 mmHg for systolic and diastolic respectively. In exploratory ROC analysis, a classification criterion of plethysmometric diastolic pressure more than or equal to 83 mmHg OR mean arterial pressure more than or equal to 98 mmHg provided the highest Youden index, with a sensitivity of 100% and specificity of 90.36% as compared to sphygmomanometric classification. However, in order not to miss isolated systolic hypertension, a criterion of (systolic more than or equal to 124 mmHg OR diastolic more than or equal to 83 mmHg) AND MAP more than or equal to 98 mmHg was employed in the study. Twelve of the thirteen hypertensives were identified by this criterion, with one discordant negative. Five additional participants were classified as hypertensive by plethysmometry. The five discordant positives and one discordant negative by plethysmometry would require further investigations to confirm the classification. Conclusions: A single plethysmometric recording showed small mean differences from the average of three reference sphygmomanometric measurements and promising discrimination of sphygmomanometrically defined hypertension. Because the complete AAMI/ESH/ISO validation protocol was not performed and the diagnostic threshold was derived and evaluated in the same small cohort, these findings require confirmation using repeated paired measurements and independent validation datasets.