Background: Following the growing clinical interest towards blood pressure (BP) variability (V), various indices for its quantification have been proposed. Aim of this study was to explore the relation between different BPV indices derived from 24h ambulatory BP monitoring (ABPM) and some potential determinants in a large group of untreated hypertensive patients. Methods: 6688 out of 11291 patients included in the Dublin study were eligible for BPV analysis, after strict ABPM quality check according to the 2013 ESH ABPM position paper criteria. They were divided into 7 age subgroups (18-29, 30-39, 40-49, 50-59, 60-69, 70-79, ≥80years) across which we explored changes in several 24h BPV indices, including indices of overall BPV [standard deviation (SD), Coefficient of variation (CV) and ratio of systolic (S) and diastolic (D)SD]; indices of short term 24h BPV after excluding circadian BP changes [Average real variability (ARV=1/(∑▒w_i ) ∑_(i=2)^N▒〖w_i×|〖BP〗_i-〖BP〗_(i-1) |〗), Weighted 24hSD (wSD=(〖SD〗_day∙T_day+〖SD〗_night∙ T_night)/(T_day+T_night )), Successive BP Variation (SV=√(1/(N-1) ∑_(i=1)^N▒〖(〖〖BP〗_(i+1)-〖BP〗_i)〗^2 〗)), Time Rate of BP changes (TR=(〖∑_(i=1)^(N-1)▒| r〗_i |)/(N-1), where r_i=〖(BP〗_(i+1)-〖BP〗_i)/〖(t〗_(i+1)-t_i)], and C) indices of extreme BP fluctuations such as range of BP values, maximum trough [mean(BPi)-min(BPi)] and maximum peak [max(BPi)-mean(BPi)]. The impact of mean BP, age, gender, smoking, diabetes and history of CV diseases on these BPV indices was assessed through multiple regression analysis. Results: Subjects were unevenly distributed across age subgroups (n=173, 336, 662, 957, 746, 366 and 85, respectively). All 24h SBPV indices showed constant increase with age and BP levels (regression coefficients <0.29, p <0.001 always) and were higher in smokers (p<0.001), while changes in DBPV indices with age were weaker and less univocal (Figure 1). Changes of BPV indices with categorical factors (gender, diabetes and history of CV diseases) were all very small, with regression coefficients<1 except for changes in SBPV with diabetes and previous CV diseases (regr.coef.=1.87, 2.19 respectively). Conclusions: Indices estimating different BPV components are related to age, smoking and mean BP levels, while they seem not to differ systematically as a function of gender, diabetes and previous CV diseases. These findings may be relevant when exploring the clinical relevance of different BPV indices
Liu, X., Faini, A., Bilo, G., Dolan, E., O’Brien, E., Parati, G. (2016). Determinants of different blood pressure variability indices in untreated hypertensive patients. Data from the Dublin study. Intervento presentato a: 26th European meeting on Hypertension and Cardiovascular pretection, Paris, France [10.1097/01.hjh.0000492151.09088.ff].
Determinants of different blood pressure variability indices in untreated hypertensive patients. Data from the Dublin study
LIU, XIAOQIUPrimo
;FAINI, ANDREASecondo
;BILO, GRZEGORZ;PARATI, GIANFRANCO
2016
Abstract
Background: Following the growing clinical interest towards blood pressure (BP) variability (V), various indices for its quantification have been proposed. Aim of this study was to explore the relation between different BPV indices derived from 24h ambulatory BP monitoring (ABPM) and some potential determinants in a large group of untreated hypertensive patients. Methods: 6688 out of 11291 patients included in the Dublin study were eligible for BPV analysis, after strict ABPM quality check according to the 2013 ESH ABPM position paper criteria. They were divided into 7 age subgroups (18-29, 30-39, 40-49, 50-59, 60-69, 70-79, ≥80years) across which we explored changes in several 24h BPV indices, including indices of overall BPV [standard deviation (SD), Coefficient of variation (CV) and ratio of systolic (S) and diastolic (D)SD]; indices of short term 24h BPV after excluding circadian BP changes [Average real variability (ARV=1/(∑▒w_i ) ∑_(i=2)^N▒〖w_i×|〖BP〗_i-〖BP〗_(i-1) |〗), Weighted 24hSD (wSD=(〖SD〗_day∙T_day+〖SD〗_night∙ T_night)/(T_day+T_night )), Successive BP Variation (SV=√(1/(N-1) ∑_(i=1)^N▒〖(〖〖BP〗_(i+1)-〖BP〗_i)〗^2 〗)), Time Rate of BP changes (TR=(〖∑_(i=1)^(N-1)▒| r〗_i |)/(N-1), where r_i=〖(BP〗_(i+1)-〖BP〗_i)/〖(t〗_(i+1)-t_i)], and C) indices of extreme BP fluctuations such as range of BP values, maximum trough [mean(BPi)-min(BPi)] and maximum peak [max(BPi)-mean(BPi)]. The impact of mean BP, age, gender, smoking, diabetes and history of CV diseases on these BPV indices was assessed through multiple regression analysis. Results: Subjects were unevenly distributed across age subgroups (n=173, 336, 662, 957, 746, 366 and 85, respectively). All 24h SBPV indices showed constant increase with age and BP levels (regression coefficients <0.29, p <0.001 always) and were higher in smokers (p<0.001), while changes in DBPV indices with age were weaker and less univocal (Figure 1). Changes of BPV indices with categorical factors (gender, diabetes and history of CV diseases) were all very small, with regression coefficients<1 except for changes in SBPV with diabetes and previous CV diseases (regr.coef.=1.87, 2.19 respectively). Conclusions: Indices estimating different BPV components are related to age, smoking and mean BP levels, while they seem not to differ systematically as a function of gender, diabetes and previous CV diseases. These findings may be relevant when exploring the clinical relevance of different BPV indices
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
