a Department of Cardiopneumology, Coimbra Polytechnic Institute, ESTEC and b Institute of Cardiovascular Research, Coimbra, Portugal

Transcription

1 7 Devices and technology Reproducibility of aortic pulse wave velocity as assessed with the new Analyse Telmo Pereira a, João Maldonado b, Ivo Andrade a, Eva Cardoso a, Mafalda Laranjeiro a, Rita Coutinho a and Jorge Conde a Introduction The aim of this study was to assess the interobserver and intraobserver reproducibility, as well as the temporal variability of the new Analyse assessing aortic pulse wave velocity (PWV). Methods Eighty-seven participants (6% men), mean age 3.26±6.58 years, were enrolled in a cross-sectional study. All patients were subjected to sequential measures of aortic PWV by two experienced operators. In a group of 27 participants, PWV was also determined month after the first moment evaluation to assess the temporal stability of the PWV estimations with the device. Results The analysis of concordance showed a very good agreement for paired PWV values in terms of both the intraobserver and the interobserver variability and also the temporal variability. The intraclass correlation coefficients were above.98 for the three conditions (P <.), indicating an excellent strength of agreement. Further evidences in favor of a good overall performance of the device were determined from the Bland Altman analysis, with small mean differences for intrareproducibility, interreproducibility, and temporal reproducibility (respectively,.2±.38,.±.5, and.7±.5 m/s), and with differences mainly between 2 SDs of the mean difference. The correlations observed were independent of sex, age, arterial pressure, heart rate, and BMI. Conclusion The data showed an excellent reproducibility of the Analyse for the assessment of aortic PWV when used in ideal conditions and by experienced observers. The technical profile presented shows that this device meets the requirements of quality for its inclusion in integrated clinical follow-up programs. Blood Press Monit 9:7 75 c 2 Wolters Kluwer Health Lippincott Williams & Wilkins. Blood Pressure Monitoring 2, 9:7 75 Keywords: arterial stiffness,, pulse wave velocity a Department of Cardiopneumology, Coimbra Polytechnic Institute, ESTEC and b Institute of Cardiovascular Research, Coimbra, Portugal Correspondence to Telmo Pereira, MSc, PhD, Department of Cardiopneumology, Coimbra Polytechnic Institute, ESTEC, General Humberto Delgado, 2, Lousã, Portugal Tel: ; fax: ; telmo@estescoimbra.pt Received 26 September 23 Revised 26 December 23 Accepted 3 January 2 Introduction Cardiovascular diseases are responsible for a high proportion of global morbidity and mortality, constituting a public health problem of the highest magnitude worldwide []. Given the fact that arteries are the place, the target, and the common denominator of cardiovascular diseases, the study of vascular function has gained considerable support and reached a level of high importance. The role of the arterial system is also crucial in the physiopathology of one of the major cardiovascular risk factors, hypertension, considered a major problem and public health challenge [2]. Bearing in mind that the involvement of the arterial system expresses the early stages of the continuum of cardiovascular diseases, the development of methods for the study of arterial function is a critical clinical requirement. In this sense, several methodological options have been developed, including the device, which has been considered as the reference method for evaluation of aortic stiffness by measurement of aortic pulse wave velocity (PWV) [3]. Several cohort studies have reported the usefulness of PWV for the prediction of cardiovascular events in different clinical settings [ 2]. Furthermore, a recent meta-analysis [3] unequivocally reinforced this notion, showing the clinical utility of PWV for risk stratification, whereas others have shown that adding PWV to conventional risk factors improves cardiovascular risk stratification achieved with conventional risk scores [9 2]. However, several studies have evaluated the reproducibility of this method, indicating that the evaluation of PWV with the is reliable and reproducible, adding evidence to the suitability of the method to clinical practice as the standard method for PWV measurement [22 28]. However, all of these studies were based on the previous generations of the device, so that no data exist on the more recent version of the device the Analyse (Alam Medical, Paris, France). This new version of the device represents a step toward a simultaneous multisignal/parameter platform as it allows for the traditional PWV measurement simultaneously recording the carotid arterial pressure curve and estimating central pressure wave parameters. To achieve this important feature, considerable hardware and c 2 Wolters Kluwer Health Lippincott Williams & Wilkins DOI:.97/MBP.38

2 Aortic pulse wave velocity measurement Pereira et al. 7 software changes had to be made in the, whose impact in terms of overall performance is not known, although expected. Thus, the aim of this study was to evaluate the interobserver and intraobserver reproducibility of the Analyse in measuring aortic PWV and relating it to previous generations of the same equipment. Methods Population The study included 87 participants (6% men) whose fundamental characteristics are summarized in Table. The mean age of the sample was 3.26±6.58 years, ranging from 8 to 9 years. In clinical terms, about 39% of the patients (n = 3) were smokers, 27.8% (n = 25) had hypertension, 7.2% (n = 5) were diabetic, and 23% (n = 2) had dyslipidemia. Coronary heart disease was found in 5 patients (7.2% of the sample). The overall mean BMI was 23.83±3.37 kg/m 2, ranging from 8.29 to kg/m 2. Twenty-six patients were overweight (3%) and seven patients (8%) were obese. Systolic blood pressure (SBP) ranged from 9 to 95 mmhg, with a mean of 25.98±6.89 mmhg. The mean diastolic blood pressure (DBP) was 7.99±8.37 mmhg, ranging from 56 to 95 mmhg. The mean PWV was 7.55±2.9 m/s (range: 5 5 m/s). In comparison, women were younger and had significantly lower BP, PWV, and BMI. All clinical evaluations were performed at rest without previous consumption of coffee or tobacco in a period of 2 h before the procedure. All participants provided their informed consent to participate in the study. Procedure All evaluations were performed in the morning in a laboratory under appropriate conditions and controlled humidity, temperature, light, and sound. Relevant demographic and clinical data were collected. We used a clinically validated Class A automatic blood pressure measurement device (Colson MAM BP 3AA-2; Colson, Paris) [29] with the cuff adjusted to the Table Demographic characteristics of the participants studied (N =87) participant s arm circumference. Three consecutive measurements were taken in the right arm, with an interval of 2 min between measurements. The average SBP, DBP, and heart rate was considered for analysis. Pulse pressure (PP) was calculated as PP = SBP DBP and the mean arterial pressure as MAP = DBP + /3PP. Aortic stiffness was assessed in all participants by measuring the carotid-femoral PWV according to a previously described technique [22]. PWV was measured with the patient in the supine position after a -min resting period. The operator located the carotid and femoral arteries and marked the point for capturing the corresponding pressure curves with two specific pressuresensitive transducers. The distance between the carotid and femoral arteries was than measured directly and entered into the Analyse software. Brachial blood pressure was measured and entered into the Analyse software, and then signal acquisition was launched. The operator positioned the carotid sensor with the help of its specific holder and manually held the femoral sensor on the femoral artery. When the operator observed pulse waveforms of sufficient quality, simultaneous carotid and femoral pressure curves were recorded for 5 s. The time delay (aortic transit time) between the two pulse waveforms was calculated automatically according to the intersecting tangent algorithm (as recommended previously in Reference Values for Arterial Stiffness Collaboration [3]). The distance traveled by the pulse waveforms was measured between the two recording sites directly on the body surface and was corrected automatically according to the equation.8 direct distance, subtracting the manubrium-to-carotid distance as recommended previously in Van Bortel et al. [3]. PWV was then calculated using measurements of transit time and distance traveled by the pulse wave between the two recording sites. PWV evaluations were performed sequentially by two independent and properly trained operators. Thus, three PWV measurements were obtained per participant, in a factorial design comprising one device ( Analyse) and two independent observers (Fig. ), to enable the evaluation of intraobserver and interobserver reproducibility. In a group of 27 participants, the PWV was again determined by one observer (always the same operator) about month after the first sampling time to evaluate the stability of results over time. Mean±SD Minimum Maximum Age (years) 3.26± Height (cm) 68.38± Weight (kg) 67.56± BMI (kg/m 2 ) 23.83± SBP (mmhg) 25.98± DBP (mmhg) 7.99± PP (mmhg) 5.99± MBP (mmhg) 89.2±9. 72 HR (bpm) 69.5±6.2 5 PWV 7.55± DBP, diastolic blood pressure; HR, heart rate; MBP, mean blood pressure; PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure. Fig. Observer A st evaluation Observer B st evaluation Observer A 2nd evaluation Flowchart of the evaluation of aortic pulse wave velocity. Observer A Late evaluation

3 72 Blood Pressure Monitoring 2, Vol 9 No 3 The quality of the PWV records was subsequently evaluated by two independent observers with considerable experience in this methodology. Statistical analysis The data of participants in the sample were computerized and processed using STATA software, version. (StataCorp. 29, STATA Statistical Software: Release ; StataCorp LP, College Station, Texas, USA). The distribution of variables was tested for normality by Kolmogorov Smirnov s test and the homogeneity of variances was addressed using the Levene test. A simple descriptive statistic was applied for demographic and clinical characterization. Data are presented as mean±sd for continuous variables and as frequency (%) for categorical variables. Comparisons between independent groups were performed using Student s t-test and for repeated measures by Student s paired t-test or repeatedmeasures analysis of variance (as suitable). Reproducibility analysis was based on the procedures recommended by Bland Altman [32]. The correlation between measurements was determined (Pearson s r and regression analysis) and Bland Altman plots (a chart with relative differences between each pair of measurements in relation to its average) were obtained and analyzed. Differences between measurements were expressed as mean difference±sd, with corresponding confidence intervals as recommended in Bland Altman analysis [32]. Diagnostic agreement analysis was based on the determination of the K coefficient. The intraclass correlation coefficients were determined to assess the overall strength of agreement. The criterion for statistical significance used was a P-value of r.5 to a 95% confidence interval. Results The strength of agreement for intraobserver, interobserver, and temporal measurements are shown in Table 2, documenting a very strong overall agreement in all the situations considered (lowest limit of intraclass correlation s 95% confidence interval greater than.9). The mean PWV values obtained for each pair of measurement conditions provided added evidence on this strong Table 2 Intraobserver, interobserver, and temporal strength of agreement of PWV measurements with the analyse device PWV PWV 2 ICC CI P Intraobserver (n = 87) 7.6± ± <. Interobserver (n = 87) 7.6± ± <. Temporal (n = 26) 9.6± ± <. CI, confidence interval of the ICC; ICC, intraclass correlation; PWV, pulse wave velocity; PWV, first measured PWV; PWV2, second measured PWV. agreement as the mean differences were quite small and nonsignificant. The Bland Altman analysis for the three conditions considered is shown in Fig. 2 and showed very good concordance between measurements, with differences mainly located between 2 SDs, substantiating the abovementioned correlations. The reproducibility was maintained along the continuum of PWV values, indicating clearly that the performance of the methods was independent of the estimated value. A regression analysis was also carried out to evaluate the influence of some relevant clinical variables (blood pressure, heart rate, BMI, age, and sex) on the reproducibility of the PWV measurements, showing no relation with intraobserver, interobserver, and temporal variability under any of the conditions considered. The mean differences for intraobserver and interobserver measurements are shown in Fig. 3, and paralleled the results of our previous study for the preceding generations of the device [28]. For the intraobserver variation, the mean difference was.2±.38 m/s and for the interobserver variation, a difference of.±.5 m/s was found. In either case, the differences were smaller than the ones reported previously for the third generation of the device, although the general profile of this technology remains excellent across the gamut of the method. The differences for the temporal variability were also very small (.7±.5 m/s), indicating an excellent performance of the methodology, an aspect that was reinforced in a diagnostic agreement analysis, in which PWV was dichotomized into normal/ abnormal, on the basis of previously published normalcy criteria for the Portuguese population [32]. The K coefficient was above.98 (P <.), indicating that PWV classification was not affected by the intraobserver or interobserver differences, thus reinforcing the good diagnostic profile of this methodology for the measurement of aortic PWV. Discussion The large arteries are known to be affected significantly by clinical conditions that determine cardiovascular diseases that, in turn, are the major cause of mortality and morbidity in modern societies [,2]. This finding emphasizes the importance of an early assessment of arterial function, which is considered a common bond of organ damage because of cardiovascular risk factors. This has led to the development of new methodologies aiming to evaluate hemodynamic features that are dependent on the normal arterial functioning. Particularly, arterial stiffness has received increasing support as an important factor for cardiovascular risk stratification, given the accumulated evidence showing its usefulness in several clinical settings, such as diabetes, arterial hypertension, dyslipidemia, impaired renal failure,

4 Aortic pulse wave velocity measurement Pereira et al. 73 Fig. 2 (a) Intra-observer reproducibility 6 m/s R 2 = (b) Inter-observer reproducibility 6 m/s R 2 = (c) Temporal reproducibility 6 m/s R 2 = Difference between two measurements Difference between two measurements Difference between two measurements Mean + 2SD Mean 2SD 3 5 Mean pulse wave velocity Mean + 2SD Mean 2SD 3 5 Mean pulse wave velocity Mean + 2SD Mean 2SD 3 5 Mean pulse wave velocity Intraobserver, interobserver, and temporal reproducibility of pulse wave velocity (PWV) estimation with the Analyse. Intraobserver (a), interobserver (b), and temporal (c) relationship between two independent measurements of PWV (left panels) and PWV reproducibility with Bland Altman plots showing observer difference in measurements (right panels). and aging, among others [ 7]. Other studies have also shown the importance of assessing central arterial pressures, as well as other parameters of ventricular vascular interaction, such as the augmentation index [27 3]. Currently, the method is considered the gold standard for aortic PWV evaluation [22], combining a good overall technical profile in terms of simplicity and low cost, which adds to the vast number of clinical studies supporting this technology [ 2]. Naturally, one fundamental requirement for the clinical implementation of any technology refers to its reproducibility, an aspect that has been demonstrated previously for the different generations of the device [28]. However, a new generation of the device Analyse has been made available recently, offering the possibility for simultaneous PWV and central pulse wave analysis. The extent to which the technological modifications

5 7 Blood Pressure Monitoring 2, Vol 9 No 3 Fig. 3 (a) (b) m/s m/s.5 Previously published data.6 3rd generation 3rd generation st generation Previously published data st generation Manual..6 Manual analyse introduced in this new version of the device will affect its reproducibility profile remains uncertain, an aspect that motivated the present work. Our results showed an excellent reproducibility of the method, both in terms of immediate intraobserver and interobserver variability and even when considering the stability of the PWV measured at different times (-month interval) by the same operator. The high reproducibility profile documented in our results, with very small mean differences for paired measures, and strong correlations, even when adjusting for important clinical features, such as age, BMI, sex, heart rate, and BP, confers additional support to the excellent technical features of the method for the evaluation of aortic PWV. In fact, the discrepancies between measurements were so small that they did not produce significant diagnostic changes when PWV was classified as normal/ abnormal in accordance with previously published criteria (Cohen s K >.98, P <.) [3]. However, and comparing the mean differences obtained for the Analyse with the ones documented previously by our group for previous devices [28], we can see that the performance of this new version is in line with that of the immediately preceding generation ( SP), the contemporary version having the huge advantage of. analyse Mean intraobserver and interobserver differences in pulse wave velocity estimation with the Analyse. Intraobserver (a) and interobserver (b) differences and its relation to previous generations of the device. Adapted from Pereira and Maldonado with permission [28]. simultaneously assessing central BP and other indexes of ventricular vascular interaction. The reproducibility profile of the Analyse for central pulse wave analysis remains uncertain as the results presented here refer only to PWV reproducibility; hence, data should not be extrapolated for other purposes. It is important to stress that the reproducibility profile of the method relies on a strict compliance with the technical requirements of this method [22]. It is essential to include a resting period of B min before arterial pulse acquisition to ensure that PWV is acquired under a baseline and hemodynamically stable condition. Also, and notwithstanding its technical simplicity, this methodology demands an adequate amount of training; thus, it should be performed by an experienced, adequately trained professional with a good theoretical knowledge of arterial physiology and hemodynamics. An important clinical challenge that remains to be answered refers to the classification of PWV and its integration into riskscorealgorithms[9 2].Someproposalsarealready available, although no consensual approach for the definition of normality still exists, with some groups arguing in favor of an operational approach, whereas others propose a more conservative statistical definition [,5,22,3,33]. Given the current scientific evidences, it is becoming increasingly clear that PWV must be included in daily practice in all clinical settings in which cardiovascular impact is clear and expected. This method combines a remarkable technical simplicity with a minor cost, in addition to an excellent reproducibility (as shown in this study), and to a large body of scientific evidence supporting its use, thus making it a good choice for clinical and epidemiological purposes. The next challenge will be to validate PWV as a therapeutic target as its usefulness for risk stratification is clearly beyond doubt. Acknowledgements Conflicts of interest There are no conflicts of interest. References World Health Organization. The World Health Report 22: reducing risks, promoting life. Geneva: World Health Organization; World Health Organization. A global brief on hypertension Silent killer, global public health crisis. Geneva: World Health Organization; Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol 2; 55: Blacher J, Guérin AP, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness on survival in end-stage renal disease. Circulation 999; 99: Boutouyrie P, Tropeano AI, Asmar R, Gautier I, Benetos A, Lacolley P, Laurent S. Aortic stiffness is an independent predictor of primary coronary events in hypertensive patients: a longitudinal study. Hypertension 22; 39: 5. 6 Cruickshank K, Riste L, Anderson SG, Wright JS, Dunn G, Gosling RG. Aortic pulse-wave velocity and its relationship to mortality in diabetes and glucose intolerance: an integrated index of vascular function? Circulation 22; 6:

6 Aortic pulse wave velocity measurement Pereira et al Guerin AP, Blacher J, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness attenuation on survival of patients in end-stage renal failure. Circulation 2; 2: Mattase-Raso FU, Van der Cammen TJ, Hofman A, Van Popele NM, Bos ML, Shalekamp MA, et al. Arterial stiffness and risk of coronary heart disease and stroke: the Rotterdam Study. Circulation 26; 3: Meaume S, Benetos A, Henry OF, Rudnichi A, Safar ME. Aortic pulse wave velocity predicts cardiovascular mortality in subjects > 7 years of age. Arterioscler Thromb Vasc Biol 2; 2: Pannier B, Guerin AP, Marchais SJ, Safar ME, London GM. Stiffness of capacitive and conduit arteries: prognostic significance for end-stage renal disease patients. Hypertension 25; 5: Shoji T, Emoto M, Shinohara K, Kakiya R, Tsujimoto Y, Kishimoto H, et al. Diabetes mellitus, aortic stiffness, and cardiovascular mortality in end-stage renal disease. J Am Soc Nephrol 2; 2: Shokawa T, Imazu M, Yamamoto H, Toyofuku M, Tasaki N, Okimoto T, et al. Pulse wave velocity predicts cardiovascular mortality: findings from the Hawaii Los Angeles Hiroshima study. Circ J 25; 69: Sutton-Tyrrel K, Najjar SS, Boudreau RM, Venkitachalam L, Kupelian V, Simonsick EM, et al. Health ABC Study. Elevated aortic pulse wave velocity, a marker of arterial stiffness, predicts cardiovascular events in well-functioning older adults. Circulation 25; : Terai M, Ohishi M, Ito N, Takagi T, Tatara Y, Kaibe M, et al. Comparison of arterial functional evaluations as a predictor of cardiovascular events in hypertensive patients: the Non-Invasive Atherosclerotic Evaluation in Hypertension (NOAH) study. Hypertens Res 28; 3: Wang KL, Cheng HM, Sung SH, Chuang SY, Li CH, Spurgeon HA, et al. Wave reflection and arterial stiffness in the prediction of 5-year all-cause and cardiovascular mortalities: a community-based study. Hypertension 2; 55: Willum-Hansen T, Staessen JA, Torp-Pedersen C, Rasmussen S, Thijs L, Ibsen H, Jeppesen J. Prognostic value of aortic pulse wave velocity as index of arterial stiffness in the general population. Circulation 26; 3: Maldonado J, Pereira T, Polónia J, Silva JA, Morais J, Marques J. participants in the EDIVA project. Arterial stiffness predicts cardiovascular outcome in a low-to-moderate cardiovascular risk population: the EDIVA (Estudo de DIstensibilidade VAscular) project. J Hypertens 2; 29: Livingstone KM, Lovegrove JA, Cockcroft JR, Elwood PC, Pickering JE, Givens DI. Does dairy food intake predict arterial stiffness and blood pressure in men?: evidence from the Caerphilly Prospective Study. Hypertension 23; 6: Mitchell G, Hwang SJ, Vasan RS, Larson MG, Pencina MJ, Hamburg NM, et al. Arterial stiffness and cardiovascular events: The Framingham Heart Study. Circulation 2; 2: Sehestedt T, Jeppesen J, Hansen TW, Wachtell K, Ibsen H, Torp-Peterson C, et al. Risk prediction is improved by adding markers of subclinical organ damage to SCORE. Eur Heart J 2; 3: Pereira T, Maldonado J, Polónia J, Silva JA, Morais J, Rodrigues T, Marques M. for the Participants in the Ediva Project. Aortic pulse wave velocity and HeartSCORE: Improving cardiovascular risk stratification. A sub-analysis of the EDIVA (Estudo de DIstensibilidade VAscular) project. Blood Press 23 [Epub ahead of print]. 22 Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, et al. European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 26; 27: Levy B, Targett RC, Bardou A, McIlroy MB. Quantitative ascending aortic Doppler blood velocity in normal human subjects. Cardiovasc Res 985; 9: Targett RC, Levy B, Bardou A, McIlroy MB. Simultaneous Doppler blood velocity measurements from aorta and radial artery in normal human subjects. Cardiovasc Res 985; 9: Asmar R, Benetos A, Topouchian J, Laurent P, Pannier B, Brisac AM, et al. Assessment of arterial distensibility by automatic pulse wave velocity measurement. Validation and clinical application studies. Hypertension 995; 26: Asmar R, Topouchian J, Pannier B, Benetos A, Safar M. Scientific, Quality Control, Coordination and Investigation Committees of the Study. Pulse wave velocity as endpoint in large-scale intervention trial. The study. J Hypertens 2; 9: Mancia G, Fagard R, Narkiewicz K, Redon J, Zanchetti A, Bohm M, et al. 23 Guidelines for the management of arterial hypertension. The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 23; 3: Pereira T, Maldonado J. Comparative study of two generations of the device for aortic pulse wave velocity measurements. Blood Press Monit 2; 5: Pereira T, Maldonado J. Performance of the Colson MAM BP 3AA-2 automatic blood pressure monitor according to the European Society of Hypertension validation protocol. Rev Port Cardiol 25; 2: Reference Values for Arterial Stiffness Collaboration. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: establishing normal and reference values. Eur Heart J 2; 3: Van Bortel LM, Laurent S, Boutouyrie P, Chowienczyk P, Cruickshank JK, De Backer T, et al. Artery Society; European Society of Hypertension Working Group on Vascular Structure and Function; European Network for Noninvasive Investigation of Large Arteries. Expert consensus document on the measurement of aortic stiffness in daily practice using carotid-femoral pulse wave velocity. J Hypertens 22; 3: Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 986; : Pereira T, Maldonado J, Polónia J, Alberto Silva J, Morais J, Marques M; participants of the EDIVA project. A statistical definition of aortic pulse wave velocity normality in a Portuguese population: a subanalysis of the EDIVA project. Rev Port Cardiol 2; 3: