CHARACTERIZATION OF MEAN DAYTIME AND NIGHTTIME SYSTOLIC AND DIASTOLIC BLOOD PRESSURE: COMPARISON BETWEEN DIFFERENT METHODS.

PEREZ-LLORET S; RISK M; CARDINALI DP; SANCHEZ, R; RAMIREZ A

Milan, Italia

Congreso; Congreso de la Sociedad Europea de Hipertension; 2005

OBJECTIVES: To compare mean daytime and nighttime systolic and diastolic blood pressure (SBP and DBP) assessed by three commonly used methods. Concordance in classification of normotension or hypertension and dipping status is re-evaluated. METHODS: This is a transversal and observational study including 407 subjects. After a 24 h ABPM (Spacelab 90207), day-time (Dt) and nighttime (Nt) periods and their respective SBP and DBP were identified by: 1) the CUSUM method, as the 6 h-long periods with higher or lower slopes in the CUSUM plot; 2) fixed clock-time intervals (Nt: 23.00 to 07.00 h); 3) time to and getting out of bed, reported in subjects’ logbook. Subjects with mean Dt SBP and DBP greater than 135/85 mmHg or under antihypertensive treatment were classified as hypertensives. Subjects with a nocturnal dip lower than 10% of daytime SBP were classified as non-dippers. Bland Altman plots were constructed and concordance was tested by ANOVA for quantitative data and by kappa statistic (k) for categorical data. Data are expressed as Mean±2SD. RESULTS: CUSUM yielded higher Dt SBP values as compared to fixed clock-time intervals and to logbook results, by 5.2±7.4 and 4.6±7.5 mmHg, respectively. Conversely, it yielded lower Nt SBP by -3.8±6.2 and -2.3±5.8 mmHg when compared to fixed clock-time intervals and logbook, respectively. Similarly, CUSUM overestimated Dt DBP by 3.3±6.0 and 2.8±6.0mmHg, and underestimated Nt DBP by -2.5±5.4 and -1.2±5.0 mmHg, as compared to fixed clock-time intervals and logbook, respectively. In the classification as normotensives or hypertensives, higher concordance were observed between fixed time intervals and subjects’ logbook (k=0.9±0.01; p<0.01) than between CUSUM and fixed time intervals (k= 0.8±0.03, p<0.01) or subjects’ logbook (k= 0.8±0.02; p<0.01). As far as the diagnosis of non-dipper status, concordance between CUSUM and fixed time intervals (k=0.1±0.03) or subjects’ logbook (k=0.3±0.04) was lower than between fixed time intervals and subjects’ logbook (k=0.6±0.03; p<0.01). CONCLUSION: Overestimation of daytime BP means and the underestimation of nighttime BP mean values together with a low concordance to identify normotensive and hypertensive subjects restrain the use of CUSUM analysis method in clinical practice and research.