M. Rovina

Short-Term Repeatability of Noninvasive Aortic Pulse Wave Velocity Assessment: Comparison between Methods and Devices

BACKGROUND Aortic pulse wave velocity (PWV) is an indirect index of arterial stiffness and an independent cardiovascular risk factor. Consistency of PWV assessment over time is thus an essential feature for its clinical application. However, studies providing a comparative estimate of the reproducibility of PWV across different noninvasive devices are lacking, especially in the elderly and in individuals at high cardiovascular risk. METHODS Aimed at filling this gap, short-term repeatability of PWV, estimated with 6 different devices (Complior Analyse, PulsePen-ETT, PulsePen-ET, SphygmoCor Px/Vx, BPLab, and Mobil-O-Graph), was evaluated in 102 high cardiovascular risk patients hospitalized for suspected coronary artery disease (72 males, 65 ± 13 years). PWV was measured in a single session twice, at 15-minute interval, and its reproducibility was assessed though coefficient of variation (CV), coefficient of repeatability, and intraclass correlation coefficient. RESULTS The CV of PWV, measured with any of these devices, was <10%. Repeatability was higher with cuff-based methods (BPLab: CV = 5.5% and Mobil-O-Graph: CV = 3.4%) than with devices measuring carotid-femoral PWV (Complior: CV = 8.2%; PulsePen-TT: CV = 8.0%; PulsePen-ETT: CV = 5.8%; and SphygmoCor: CV = 9.5%). In the latter group, PWV repeatability was lower in subjects with higher carotid-femoral PWV. The differences in PWV between repeated measurements, except for the Mobil-O-Graph, did not depend on short-term variations of mean blood pressure or heart rate. CONCLUSIONS Our study shows that the short-term repeatability of PWV measures is good but not homogenous across different devices and at different PWV values. These findings, obtained in patients at high cardiovascular risk, may be relevant when evaluating the prognostic importance of PWV.

Influence of carotid atherosclerotic plaques on pulse wave assessment with arterial tonometry

Objective: Aortic stiffness and central pressure measurements have become increasingly important for the overall estimation of cardiovascular risk. The aim of this study is to verify whether the presence of stenosis in the carotid arteries due to atherosclerotic plaques may induce a bias in the measurement of carotid–femoral pulse wave velocity (PWV) and in the analysis of central pulse waveform variables assessed by carotid tonometry. Methods: Eighty-four patients (age: 67.1 ± 12.4 years) undergoing screening for carotid atherosclerosis were enrolled, divided into three groups according to carotid ultrasound findings (NASCET criteria): 28 patients without significant stenosis, 30 patients with bilateral plaques, and 26 patients with right or left monolateral stenosis. PWV and other variables derived from the central pulse waveform analysis (central blood pressure, augmentation index and forward and backward waves) were measured at both right and left carotid arteries by a validated PulsePen tonometer. A repeatability study was performed in 28 young healthy patients (age: 25.4 ± 2.9 years). Results: A high degree of correlation was found between bilateral measurements in all groups, and particularly in groups with monolateral carotid stenosis, with no significant difference attributable to lateralized stenosis. Right–left differences in asymmetric groups were 0.35 ± 5.12 mmHg (R2 = 0.960) for central blood pressure, −2.12 ± 7.39% (R2 = 0.743) for augmentation index, 0.64 ± 1.56 m/s (R2 = 0.947) for PWV, 0.08 ± 8.48 mmHg for forward wave (R2 = 0.742) and 0.35 ± 2.35 mmHg for backward wave (R2 = 0.907). Conclusion: Measurement of PWV and of variables derived from the central pulse waveform analysis by carotid tonometry is not biased by the presence of local atherosclerotic plaques.

Systolic time intervals assessed from analysis of the carotid pressure waveform

OBJECTIVE The timing of mechanical cardiac events is usually evaluated by conventional echocardiography as an index of cardiac systolic function and predictor of cardiovascular outcomes. We aimed to measure the systolic time intervals, namely the isovolumetric contraction time (ICT) and pre-ejection period (PEP), by arterial tonometry. APPROACH Sixty-two healthy volunteers (age 47  ±  17 years) and 42 patients with heart failure and reduced ejection fraction were enrolled (age 66  ±  14 years). Pulse waves were recorded at the carotid artery by arterial tonometry together with simultaneous aortic transvalvular flow by Doppler-echocardiography, synchronized by electrocardiographic gating. The ICT was determined from the time delay between the electrical R wave and the carotid pressure waveform, after adjustment for the pulse transit time from the aortic valve to the carotid artery site, estimated by an algorithm based on the carotid-femoral pulse wave velocity. The PEP was evaluated by adding the electrical QR duration to the ICT. MAIN RESULTS The ICT derived from carotid pulse wave analysis was closely related to that measured by echocardiography (r  =  0.90, p  <  0.0001), with homogeneous distribution in Bland-Altman analysis (mean difference and 95% confidence interval  =  0.2 from  -14.2 to 14.5 ms). ICT and PEP were higher in cardiac patients than in healthy volunteers (p  <  0.0001). The ratio between PEP and left ventricular ejection time was related to the ejection fraction measured with echocardiography (r  =  0.555, p  <  0.0001). SIGNIFICANCE The timing of electro-mechanical cardiac events can be reliably obtained from the carotid pulse waveform and carotid-femoral PWV, evaluated using arterial tonometry. Systolic time intervals assessed with this approach showed good agreement with measurements performed with conventional echocardiography and may represent a promising additional application of arterial tonometry.

NON-INVASIVE MEASUREMENT OF AORTIC PULSE WAVE VELOCITY: A COMPARATIVE EVALUATION OF EIGHT DEVICES

Objective: Several non-invasive devices purport to measure aortic pulse wave velocity (PWV), by applying different approaches and sensors, with the aim of evaluating cardiovascular risk. Purpose of this study is to compare the PWV measured by eight commercially available devices in patients with cardiovascular disease. Design and method: In this study, 102 patients (70% males, mean age 65 ± 13 years) were enrolled among those who were going to undertake an elective cardiac catheterization study. For each patient, the following device was used to non-invasively evaluate aortic PWV, in a random order: BPLab, Complior Analyse, Mobil-O-Graph, pOpmètre, PulsePen-ET, PulsePen-ETT and SphygmoCor. Data were analyzed by computing the coefficient of the correlation (r) and determination (r2) between measured values and with age of patients. Results: The mean blood pressure, heart rate and PWV measured in the population were: 102 ± 16 mmHg, 65 ± 12 s-1 and 11.2 ± 3.6 m/s. Comparative data are shown in Table 1. Devices evaluating carotid-femoral PWV (Complior Analyse, PulsePen-ET, PulsePen-ETT, SphygmoCor) presented a very strong agreement between each other (r > 0.80) and moderate correlation with the PWV measured by the Mobil-O-Graph (r 0.45 to 0.65), while a weak correlation was found between carotid-femoral PWV measurements and the BPLab or the pOpmètre (r < 0.30). A moderate-strong relationship was found between age and cf-PWV (r2 0.20 to 0.38), whereas PWV measured by pOpmètre and BPLab showed a weak correlation with age (r2 0.05 and 0.06 respectively). On the contrary, a very strong relationship was found between Mobil-O-Graph and age (r2 = 0.90). Conclusions: Devices measuring carotid-femoral PWV, considered the gold-standard measure for aortic PWV, present a very good agreement between each other, in our population of patients with cardiovascular disease. The Mobil-O-Graph, which estimates aortic PWV from age and blood pressure values, also present a good correlation with measures of carotid-femoral PWV. The two other measuring devices (BPLab, pOpmètre) does not provide a PWV measure in agreement with carotid-femoral PWV. Our results support the use of devices measuring carotid-femoral PWV for a proper and consistent evaluation of aortic PWV.

HOW TO IMPROVE THE CALCULATION OF MEAN ARTERIAL PRESSURE AT THE BRACHIAL ARTERY LEVEL

Objective: Mean arterial pressure (MAP) is the time-averaged pressure through the cardiac cycle and may be calculated from brachial pressure values. Previous studies proposed thumb-rules, as adding 40% of pulse pressure to diastolic BP, to calculate MAP, but this approach is not unanimously accepted. We aimed to find the best way of calculating MAP by analyzing the brachial pressure wave. Design and method: We examined the pressure waveform obtained with brachial arterial tonometry (PulsePen, DiaTecne) in 1526 subjects from 3 cohorts (age 64.4 ± 18.2 years, 44.1% males), one from general population (n = 490, age 49.6 ± 12.7 years, 39.4% males), one of elderly patients (n = 284, age 87.6 ± 4.7 years, 25.4% males) and one of hypertensive patients (n = 752, age 59.2 ± 14.4 years, 54.3% males). Brachial pressure wave was calibrated with oscillometric systolic and diastolic brachial blood pressure measurement. The “real” MAP and the percentage of pulse pressure that needs to be added to diastolic blood pressure (PP%) to obtain the MAP were calculated from the time-averaged brachial pressure waveform. Results: The mean PP% in the pooled population was 42.2 ± 5.5% and was lower in the elderly cohort (40.8 ± 5.4%, p < 0.0001) than in the general population cohort (42.8 ± 6.0%) and in the hypertensives (42.2 ± 5.0%). PP% was higher in women (42.9 ± 5.6%) than in men (41.2 ± 5.1%, p < 0.0001), and was significantly correlated in multiple regression analysis with diastolic pressure (&bgr;=0.337, p < 0.0001), heart rate (&bgr;=0.091, p < 0.0001), while it was weakly related with age (&bgr;=-0.053, p = 0.05) and not related to systolic pressure. An equation to obtain an improved calculation of MAP in a single subject was derived from our data: PP% = 25.361 + 0.047*heart rate + 0.163*Diastolic pressure (+2.137 if female). Conclusions: Our data provide an estimate of the PP% required to be added to diastolic pressure to obtain the “real” MAP, which is 42.2% (with and SD of 5.5%). PP% presents a marked inter-individual variability, which discourages the use of a unique PP% for everyone. Our results offer the possibility to improve the calculation of MAP in the single subject by applying a formula derived from the analysis of the brachial waveforms in a large population.

COMPARISON BETWEEN AORTIC PULSE WAVE VELOCITY MEASURED INVASIVELY AND NON-INVASIVELY BY EIGHT DIFFERENT DEVICES

Objective: Aortic pulse wave velocity (PWV) is the best indicator of aortic viscoelastic properties. Aim of this study is to investigate if invasively measured aortic PWV is accurately estimated by non-invasive methods which purport to assess it. Design and method: One-hundred and two patients (30% female, mean age 65 ± 13 years) planned to undertake a cardiac catheterization were enrolled in the study. Different non-invasive methods were evaluated for each subject by randomly alternating the following devices: BPLab, Complior Analyse, Mobil-O-Graph, pOpmètre, PulsePen-ET, PulsePen-ETT and SphygmoCor. Immediately after, aortic PWV was evaluated by aortic catheterization and simultaneous measurement of pressure wave above the aortic valve and at the aortic bifurcation (FS-Stiffcath). Invasive data were analyzed by proprietary software and compared with non-invasive PWV values by Bland-Altman analysis and paired parametric tests (for the whole population) and non-parametric tests (for quartiles of population according to PWV). Results: Devices evaluating carotid-femoral PWV (Complior Analyse, PulsePen-ET, PulsePen-ETT, SphygmoCor) and the Mobil-O-Graph presented a strong agreement with aortic invasive PWV (respectively, Pearson R = 0.64, 0.78, 0.71, 0.70, 0.66), while a moderate agreement was present for the BPLab and the pOpmètre (R = 0.23, 0.23). In the whole population, a significant underestimation of invasive PWV was present for Complior Analyse (−0.73 m/s, p = 0.016), SphygmoCor (−0.61 m/s, p = 0.024), Mobil-O-Graph (−1.01 m/s, p < 0.001) and pOpmètre (−1.55 m/s, p = 0.003). A tendency toward the overestimation of aortic PWV for lower PWV values and the underestimation of PWV for higher values was present for all devices, and was significant for the PulsePen-ET and the BPLab in the lowest quartile (PWV < 8.5 m/s, p < 0.05) and for Complior Analyse, SphygmoCor, BPLab and Mobil-O-Graph for the highest quartile (PWV > 13 m/s, p < 0.05). Conclusions: Devices measuring carotid-femoral PWV and the Mobil-O-Graph, which estimates aortic PWV from age and blood pressure values, present a good correlation with invasive aortic PWV in a large population with cardiovascular disease, while a less good agreement was found for other measuring devices (BPLab, pOpmètre). The underestimation of high PWV values may lead to erroneous estimation of cardiovascular risk by non-invasive devices.

[PP.09.25] DEVICES FOR THE NON-INVASIVE ASSESSMENT OF AORTIC PULSE WAVE VELOCITY: EVALUATION OF SHORT-TERM REPEATABILITY

Objective: Aortic pulse wave velocity (PWV) is a validated indicator of central arterial stiffness and cardiovascular risk. We aimed to compare the repeatability of PWV measures obtained with non-invasive devices. Design and method: We evaluated the repeatibility of non-invasive measures of PWV, obtained with 4 devices measuring two-points carotid-femoral PWV (Complior, PulsePen ETT, PulsePen ET, SphygmoCor), and with 2 devices estimating PWV from the oscillometric cuff-derived brachial pulsewave (BPLab, Mobil-O-Graph). 102 patients planned to undertake a cardiac catheterization (age 65 ± 13 years, 70.6% males) were enrolled. Repeated measures of PWV were obtained with all devices in a single session, 15 minutes apart. Duplicate PWV and carotid-femoral PTT measurements were compared using different indices. Coefficients of variation (CV%) and their confidence intervals (CI) are reported. Results: Devices evaluating carotid-femoral PWV showed a good repeatability (CV%[CI] for Complior: 8.8[7.3–10.1]; PulsePen ETT: 8.0[6.2–9.5]; PulsePen ET: 5.8[4.9–6.6]; SphygmoCor: 9.5[7.7–11.0]), whereas the repeatability of PWV estimated by cuff-based devices was slightly higher (BPLab: 5.5[4.2–6.6], Mobil-O-Graph: 3.4[2.9–3.8]). A lower repeatability of carotid-femoral PWV was present for greater arterial stiffness values (CV%[CI] for PWV<10 m/s vs PWV>=10 m/s: Complior 7.0[5.4–8.3] vs 10.5[8.0–12.5], PulsePen ETT 6.3[3.6–8.1] vs 9.2[6.5–11.3], PulsePen ET 4.9[3.5–6.0] vs 6.5[5.3–7.6], Sphygmocor 8.5[5.7–10.6] vs 10.3[7.7–12.3]. No such difference was observed with cuff-based devices (BPLab 6.0[3.6–7.7] vs 5.1[3.5–6.4], Mobil-O-Graph 3.5[2.8–4.1] vs 3.2[2.6–3.7]). Differences between repeated PWV measurements were not correlated with concomitant blood pressure (R2: 0.005) or heart rate differences (R2: 0.013). Conclusions: Short-term repeatability of PWV measures was good but not homogenous among different devices. A greater repeatability was observed with cuff-based devices, compared to devices measuring carotid-femoral PWV. This is probably due for Mobil-O-Graph to the algorithm for PWV assessment, which considers age and mean blood pressure, and for BPLab to the automated editing procedure which eliminates highly variable PWV values. Repeatability of PWV is not influenced by blood pressure or heart rate concomitant changes. For carotid-femoral PWV, the repeatability of measures is lower for higher PWV values. These results could be usefully considered when assessing PWV in a clinical setting.

AB0753 Arterial stiffness increase in the early phase of arthropathy related with psoriatic arthritis is not further modified by stable prolonged retention of a minimal disease activity condition

Background Increased large arteries wall stiffness (AS) is a well known independent morbidity and mortality cardiovascular risk factor, not only related to hypertension and diabetets, but also induced by long standing systemic inflammation, as observed in inflammatory rheumatic or gut chronic diseases. Psoriatic arthritis (PsA) was also shown to increase cadiovascular morbidity, but this observation was commonly related with the associated occurrence of long term arthritis, disease activity and CV traditional risk factors Objectives To assess whether in a selected group of PsA patients, affected by recent onset (4–12 months) arthritis, not suffering from CV disease risk factors, evidence could be obtained of an early alteration of arterial wall function as expressed by an increased arterial pulse wave velocity (aPWV) and secondary reduced tonometric subendocardial viability ratio (tSEVR). Further, we evaluated if a 18 months (Mth) treatment with synthetic disease modifying antirheumatic drugs (sDMARD) other than cyclosporine, leading to a demonstrable target of clinical “minimal disease activity”(MDA) could modify such possible modifications Methods Conclusive data were obtained in a selected goup of 12 PsA patients (Pts) (CASPAR criteria classified) (M/F, 6/5; mean age, 50.8; range, 40–65) not suffering from axial spondylitis who firsly underwent (without any previous treatment) aPWV measurement (PulsePen, Diatecne Srl, Milan) and tSEVR calculation, and then were re-evaluated for the same calculations after a 18 Mth time of sDMARD treatment. Each of the PsA Pts had a 3–4 Mth follow up and showed to reach the target of stable MDA (with also ultrasound confirmed remission of active synovitis) after a 4–8 Mth treatment. Before treatment, the group of PsA Pts was compared with an age, body weight, CV parameters and risk factors–matched control group of voluntary 22 healthy subjects (M/F, 11/11; mean age, 51.3; range, 41–67) Results Before any treatment, aPWV was higher in the group of patients with PsA than in control subjects (median, 8.667 m/s vs 6.963 m/s, p<0.02) while tSEVR was decreased (median, 1.44 vs 1.50, p<0.05). Aortic PWV was not modified after the sDMARD treatment (median m/s, before=8.691, after=8.048, p=0.55), despite statistically significant improvements of the disease activity scores (DAS28; modified CPDAI; DAPSA) as well as cutaneous PASI, and stable retention of a MDA condition. A direct correlation (Spearman rank) between aPWV and DAS28 (rho=0,70; p-value=0,04), BASDAI (rho=0,77; p-value=0,01), and HAQ (rho=0,66; p-value=0,05) was found, not instead between aPWV and ESR or CRP. Pts with PsA, at the end of the follow up, had increased levels of systolic (134,1±14,3 vs 122±11,2, p<0,05) and diastolic (82±7,4 vs 73,5±6,6, p<0,05) blood pressure, with unmodified heart rate Conclusions Early onset of PsA seems to be associated with already established and stable increase of AS. Subclinical previous phase of inflammation and length of psoriatic disease could be addressed as possible causes Disclosure of Interest None declared

[OP.7C.05] AORTIC STIFFNESS AND PULSE WAVE ANALYSIS IN CHILDREN AND ADOLESCENTS WITH MARFAN SYNDROME.

Objective: Marfan syndrome (MFS) is an autosomal dominant genetic disorder characterized by aortic root dilation beginning in childhood. Aortic stiffness is increased in patients with MFS but data in pediatric age are lacking. Aim of this study was to evaluate aortic stiffness and pulse wave analysis in children and adolescents with MFS, compared to general pediatric population, and its association with aortic root diameters. Design and method: Fifty-two children with MFS (age: 12.0 ± 3.3, 5.6–17.5 years), identified according to 2010 Revised Ghent Criteria and without history of cardiovascular surgery were enrolled. Patients underwent a clinical evaluation and an echocardiography. Viscoelastic aortic properties were studied assessing carotid-femoral pulse wave velocity (PWV) and analysing central blood pressure waveform with carotid tonometry. Hemodynamic parameters of MFS patients were compared with those of 73 age, sex, mean blood pressure and heart rate matched controls. Results: Central pulse pressure (CPP) was significantly higher (38.2 ± 12.2 vs 33.6 ± 8.0 mmHg, p = 0.01) and pulse pressure amplification (PPA) was significantly reduced in MFS patients than controls (18.0 ± 15.1% vs 31.8 ± 20.3%, p < 0.0001). PWV was not significantly different between MFS and controls (4.98 ± 1.00 vs 4.79 ± 0.68 m/s). In MFS cohort, after correction for age, sex, mean arterial pressure and heart rate both CPP and PPA were significantly associated with aortic root diameter at the sinuses of Valsalva (CPP R = 0.393 p = 0.008; PPA R = −0.306 p = 0.041). Conclusions: CPP and PPA are increased in children with MFS and are related to aortic root diameter. PWV is similar to general pediatric population. Pulse wave analysis variables are able to predict aortic abnormalities better than PWV in children and adolescents with MFS.