Enhancing central blood pressure accuracy through statistical modeling: A proof-of-concept study

Figures

• 

  FIGURE 1. Scatter plots of brachial SBP/estimated central SBP with invasive aortic SBP. Each gray dot represents the relationship between invasive aortic SBP and brachial SBP/estimated central SBP for a given individual. Correlation coefficients, regression slopes and mean accuracy for aortic SBP are displayed with their 95th confidence interval. (A) Brachial SBP; (B) C1SBP; (C) C2SBP; and (D) adjusted central SBP (full model). SBP, systolic blood pressure.

• 

  FIGURE 2. Differences between brachial SBP/estimated central SBP and invasive aortic SBP by age before and after adjustment. The distribution of differences by age strata are displayed by boxplots before and adjustment (full model). The horizontal black line in the middle of each box indicates the median, the square extremities indicate the interquartile range, and the vertical black lines end at either the minimum (first quartile minus 1.5 times the interquartile range) or the maximum (third quartile plus 1.5 times the interquartile range). Gray dots indicate outliers. The full models include the age, sex, heart rate, estimated SBP amplification, pulse wave velocity, augmentation index at 75 bpm, reflection magnitude, height, weight, eGFR, active smoking, diabetes, and antihypertensive treatment. C1SBP, Type I calibrated central SBP; C2SBP, Type II calibrated central SBP; eGFR, estimated glomerular filtration rate; SBP, systolic blood pressure.

• 

  FIGURE 3. Differences between brachial SBP/estimated central SBP and invasive aortic SBP by sex before and after adjustment. The distribution of differences by sex strata are displayed by boxplots before and adjustment (full model). The horizontal black line in the middle of each box indicates the median, the square extremities indicate the interquartile range, and the vertical black lines end at either the minimum (first quartile minus 1.5 times the interquartile range) or the maximum (third quartile plus 1.5 times the interquartile range). Gray dots indicate outliers. The full models include the age, sex, heart rate, estimated SBP amplification, pulse wave velocity, augmentation index at 75 bpm, reflection magnitude, height, weight, eGFR, active smoking, diabetes, antihypertensive treatment. C1SBP, Type I calibrated central SBP; C2SBP, Type II calibrated central SBP; eGFR, estimated glomerular filtration rate; SBP, Systolic blood pressure.

• 

  FIGURE 4. Bland–Altman plots for aortic SBP before and after adjustment. Modified Bland–Altman plots show the reference intra-arterial measurements (aortic SBP) on the X-axis instead of the mean of both measurements. Left to right panels show in order the agreement of brachial cuff SBP, C1SBP, and C2SBP with aortic SBP. All values are in mmHg. Red dashed lines and red solid lines represent mean differences (bias) and limits of agreement (bias ± 2 SDs) respectively. C1SBP, Type I central systolic blood pressure; C2SBP, Type II systolic blood pressure; SBP, systolic blood pressure.