Journal Logo


Evaluation of the Vicorder, a novel cuff-based device for the noninvasive estimation of central blood pressure

Pucci, Giacomoa,b,*; Cheriyan, Josepha,*; Hubsch, Annettea; Hickson, Stacey S.a; Gajendragadkar, Parag R.a; Watson, Timothyc; O'Sullivan, Michaelc; Woodcock-Smith, Jeana; Schillaci, Giuseppeb; Wilkinson, Ian B.a; McEniery, Carmel M.a

Author Information
doi: 10.1097/HJH.0b013e32835a8eb1
  • Free



High blood pressure (BP) is a major risk factor for cardiovascular disease and a powerful predictor of cardiovascular events [1]. Central (aortic) BP (cBP) is the ‘true’ pressure exerted on the heart, kidney and brain, and may differ substantially from peripheral BP, which is the result of amplification of the pressure waveform along the arterial tree. Whereas mean arterial pressure (MAP) and DBP remain largely unchanged throughout the large conduit arteries [2], SBP and pulse pressure (PP) tend to increase from the aorta to the peripheral arteries, under the influence of a number of factors that affect the timing and amplitude of incident and reflected waves [3–5].

Direct cBP measurement during invasive catheterization of the ascending aorta remains the gold standard, but the interventional nature of the technique limits its clinical use. The analysis of peripheral waveforms with dedicated devices, most of which are based on applanation tonometry integrated with an invasively validated radial-to-aortic generalized transfer function, first enabled the noninvasive assessment of cBP [6–9] and provided the opportunity to assess central hemodynamics noninvasively. Central SBP (cSBP) and PP (cPP), and the differences with the corresponding brachial values, named SBP and PP amplification, may have prognostic relevance and refine individual cardiovascular risk stratification [10–13]. Moreover, the effects of BP-lowering drugs on cBP may differ from those observed on brachial BP (bBP) [14].

The SphygmoCor device (AtCor Medical, Sydney, Australia) typically calibrates the radial waveform to brachial SBP (bSBP) and DBP (bDBP), assuming that no significant amplification of the waveform occurs between the brachial and the radial artery. However, there is increasing evidence that the brachial-to-radial SBP amplification could significantly affect estimations of cBP [15–18].

The Vicorder system is a commercially available, operator-independent device (Skidmore Medical, Bristol, UK), which determines brachial oscillometric BP using a cuff placed around the upper arm. Brachial pressure waveforms are recorded with the same cuff using a volume displacement technique. cBP parameters are then derived from bBP waveforms self-calibrated to bSBP and bDBP by applying a previously described brachial-to-aortic transfer function [19].

The aim of this study was to evaluate the performance of the Vicorder device in deriving aortic waveforms and related cBP values from brachial waveforms, against direct aortic BP measurement during invasive coronary angiography. Vicorder cBP values were also compared with cBP values obtained by the SphygmoCor, an established method for noninvasive cBP assessment. As biases introduced by brachial-to-radial SBP amplification may theoretically be offset by calibrating the SphygmoCor radial waveform to brachial mean arterial pressure (bMAP)/bDBP, we re-calibrated radial waveforms to bMAP/bDBP in order to quantify the effects of different methods of radial waveform calibration on SphygmoCor-based cBP estimation.



Fifty patients (mean age 61 ± 11 years, 62% men) undergoing routine diagnostic coronary angiography at the Department of Cardiology, Cambridge University Hospitals NHS Trust, were approached for participation in the invasive validation study between September 2009 and May 2011.

In the noninvasive study, 90 patients from the general population (mean age 42 ± 20 years, 53% men) were enrolled among those referred to the Clinical Pharmacology Unit, University of Cambridge between May and October 2009. Thirty patients were selected in three age ranges: less than 30 years, 30–60 years and more than 60 years.

In both studies, patients were excluded if they had a history of peripheral arterial disease, aortic aneurysm, absent brachial or radial pulses or known obstructive large artery atherosclerotic disease, active malignancy, hypotension (SBP <90 mmHg), valvular heart disease, known left ventricular dysfunction (ejection fraction <50%) or arrhythmias (including frequent ventricular and supraventricular premature beats). Smoking, caffeine or alcohol consumption was not permitted for at least 4 h before the measurement. Patients were required to abstain from strenous physical activity for at least 24 h before the procedure. The study had favorable review by the local Research Ethics Committee and written informed consent was obtained from all patients.


Invasive study

Invasive cBP was obtained at the end of routine coronary angiography for 50 patients, with individuals in the supine position, using a fluid-filled Cordis 6F catheter inserted via the femoral artery. The tip of the catheter was advanced to the aortic root and placed 2 cm proximal to the aortic valve. The correct position was confirmed for all patients under direct radiographic guidance. The catheter was connected to a pressure transducer (Kimal, Uxbridge, UK) using an infusion system, and care was taken to remove air bubbles and blood from the catheter. Static calibration of the system was tested and confirmed before each study by connecting the tip of the catheter to a mercury sphygmomanometer with a t-tube connector. The zero-reference point was obtained by placing the catheter at the level of the heart of the patient before each femoral insertion. Invasive BP waveforms obtained by the pressure transducer were digitally converted and analyzed by a centralized data acquisition system (GE Healthcare, Hatfield, UK) with a flat frequency response up to 180 Hz.

During continuous invasive aortic pressure wave recording, bBP was measured by a digital oscillometric technique using the Vicorder system (Skidmore Medical, Bristol, United Kingdom) and a brachial cuff placed around the nondominant arm. Cuffs were chosen for each patient out of three standard-sized cuffs based on the patient's arm circumference. Values were obtained in duplicate, according to the British Hypertension Society guidelines [20], and the average values were used in subsequent analyses.

Immediately afterwards, brachial wave tracings were digitally computed by the Vicorder with a cuff statically inflated at the level of 70 mmHg using a volume–displacement technique. Waveforms were digitally inspected and poorly reproducible traces were excluded by the in-built Vicorder software algorithm. A previously described brachial-to-aortic transfer function [19] was then applied by the Vicorder software to the brachial waveform in order to obtain the cBP waveform. Radial waveforms were recorded at the wrist of the same arm with a high-fidelity applanation tonometer (SPC-301, Millar Instruments, Inc., Houston, Texas, USA). A previously validated generalized radial-to-aortic transfer function [7] was then automatically applied to the waveform by the SphygmoCor device in order to derive cBP waveform. Overall, for each patient, two brachial waveforms by Vicorder and two radial waveforms were recorded.

During each Vicorder and SphygmoCor waveform recording, a single value of invasive cBP was obtained; since the coefficient of variation among the four invasive cBP values (two during Vicorder and two during SphygmoCor waveform recording) was less than 3% for each patient, the average value was used for the analyses. Invasive central mean arterial pressure (cMAP) was obtained by integrating the invasive cBP curve with the area method.

Brachial and radial waveforms were initially calibrated to the invasively determined MAP and DBP. For each individual, two cSBP values were then calculated from each device and averaged. With this cSBP value, the central form factor was also calculated as (invasive cMAP − invasive cDBP)/(cSBP − invasive cDBP). This nondimensional value expresses the ratio between the area under the waveform curve and the area of a rectangle with the same base and height, and is directly proportional to how ‘peaked’ the waveform is.

Brachial waveforms obtained by Vicorder were further calibrated to Vicorder oscillometric bSBP and bDBP. For each waveform, cSBP and cPP were estimated and individually averaged for the analysis. In the Vicorder device, bDBP is taken as a measure of cDBP. Radial waveforms recorded by tonometry using the SphygmoCor system were also calibrated in two different ways: to oscillometric bSBP and bDBP and to bMAP and bDBP, where bMAP was calculated using the Vicorder device by integrating the brachial waveform calibrated to bSBP/bDBP.

bBP obtained with Vicorder was also compared with a clinically validated oscillometric device (Omron HEM 705CP; Omron Healthcare, Kyoto, Japan) [21] in a subgroup of 29 patients from the invasive study, immediately after bBP measurement by Vicorder and before peripheral waveform assessment.

Noninvasive study

All patients were required to lie supine and rest for at least 15 min. Patients were categorized as hypertensive if their bSBP was more than 140 mmHg or bDBP more than 90 mmHg, or if they were receiving BP-lowering drug treatment; PP was defined as SBP − DBP. Amplification of SBP or PP was calculated as bSBP/cSBP or bPP/cPP.

As in the invasive study, two bBP values were obtained with the Vicorder cuff and averaged for calibrating peripheral waveforms. Immediately afterwards, two brachial waveforms (Vicorder) and two radial waveforms (SphygmoCor) were obtained. Brachial and radial waveforms were calibrated as previously described in the invasive study and the resulting cBP estimates were compared between devices.

Statistical analysis

Statistical analyses were performed using SPSS 17.0 for Windows (SPSS Inc., Chicago, Illinois, USA). We predicted a sample size of 40 patients in the invasive study would provide 95% power to detect a 5 mmHg absolute difference between estimated and invasively measured cSBP, assuming a SD of the difference of 8 mmHg and using a two-side α error of 0.05.

SphygmoCor-derived and Vicorder-derived cBP parameters were compared using Student's paired t-test with further analyses using Bland–Altman plots [22]. Pearson's correlation coefficients were used to assess the strength of correlation between cBP values, and their difference was examined using the z-test. The overall strength of agreement between estimated and measured cBP values was also quantified according to Lin's concordance correlation coefficient [23], whereas the Pitman–Morgan test was used to test the difference between correlated variances [24]. The differences between invasive cBP and noninvasively estimated cBP (SphygmoCor and Vicorder) were assessed according to the American Association for the Advancement of Medical Instrumentation (AAMI) [25]. British Hypertension Society [26] criteria for the evaluation of BP measuring devices were also applied to test the interdevice agreement between the two devices for the noninvasive estimate of cSBP. Patient characteristics and results are expressed as mean (SD). All analyses were two sided and a probability of less than 0.05 was considered significant.


Invasive study

Eighty patients undergoing elective coronary angiography were studied. Thirty patients were excluded, including one for bigeminy during the procedure, six for low-quality brachial Vicorder waveforms, seven for low-quality invasive cBP traces, eight for low-quality SphygmoCor radial waveforms and eight for periprocedural hemodynamic instability. Clinical characteristics of the remaining 50 patients are reported in Table 1 and cBP values are reported in Tables 2 and 3.

Demographics and clinical characteristics of invasive and noninvasive study populations
Values of invasively measured and estimated central SBP and form factor by Vicorder and SphygmoCor, both calibrated to invasive mean and DBP
Values of invasively measured and estimated central blood pressure by Vicorder, SphygmoCor brachial SBP/DBP-calibrated and SphygmoCor brachial MAP/DBP-calibrated

When peripheral waveforms were calibrated to invasive cMAP and cDBP, cSBP estimated by Vicorder was strongly correlated with invasive cSBP (r = 0.92, P < 0.001), although slightly lower (137 ± 18 versus 141 ± 19 mmHg, Δ −4.0 ± 7.4, P < 0.001, Table 2). Compared with invasive values, cSBP noninvasively estimated by Vicorder met the standards proposed by the AAMI [25] (Δ versus invasive ≤5 ± 8 mmHg) for the validation of automatic BP measurement devices. The relevant Bland–Altman plot did not reveal any relationship between the observed differences and average cBP (Fig. 1). SphygmoCor calibrated to invasive cMAP/cDBP also provided a good estimate of invasive cSBP (140 ± 18 versus 141 ± 19 mmHg, Δ −1.4 ± 7.9, P = 0.21) and a high degree of correlation (r = 0.93, P < 0.001). Vicorder-estimated cSBP did not differ from SphygmoCor-estimated cSBP in terms of absolute mean difference (Δ −2.6 ± 9.2, P = 0.18), variance (t for correlated variances = 0.95 versus Vicorder, P = 0.35) and strength of the correlation with invasive cSBP (z-test, P = 0.73). Overall, Vicorder-based and SphygmoCor-based estimates of cSBP both showed a moderate strength-of-agreement with invasive cSBP according to Lin's concordance correlation coefficients (0.91 for Vicorder and 0.92 for SphygmoCor). The central invasive form factor was overestimated by the Vicorder device (0.45 versus 0.41) and slightly by SphygmoCor (0.42 versus 0.41).

(a) Scatterplot of invasive cSBP and Vicorder cSBP calibrated to invasive MAP/DBP, and linear regression equation. Solid line: line of equality; dotted line: linear regression line. (b) Bland–Altman plot of the differences between invasive cSBP and Vicorder cSBP calibrated to invasive MAP/DBP. Solid line: mean value; dotted lines: mean ± 2 SDs. cSBP, central systolic blood pressure; MAP, mean arterial pressure.

When brachial (Vicorder) and radial (SphygmoCor) waveforms were calibrated to the same oscillometric bSBP/bDBP values obtained by Vicorder, despite high correlation with invasive cSBP (both r = 0.92, P < 0.001), both cSBP estimates were lower than invasive cSBP (Vicorder versus invasive: −6.4 ± 7.4 mmHg, P < 0.001, Table 3, SphygmoCor versus invasive: −11.9 ± 7.2 mmHg, P < 0.001). When SphygmoCor radial waveforms were re-calibrated to bMAP/bDBP (as opposed to bSBP/bDBP), underestimation of cSBP was significantly attenuated (Δ versus invasive −2.8 ± 9.4 mmHg, P = 0.04), but a larger SD of the differences was observed (t for correlated variances = 2.07 versus Vicorder, P = 0.04; t = 2.32 versus SphygmoCor bSBP/bDBP calibrated, P = 0.02) together with a nonsignificantly lower degree of correlation with invasive cSBP (r = 0.87, z-test P = 0.12). The relevant Bland–Altman plots did not reveal any relationship between the observed differences and average cBP (Fig. 2). Compared with invasive cSBP values, none of the three methods for the noninvasive estimate of cSBP met the standards proposed by the AAMI [25] for the validation of automatic BP measurement devices.

(a) Scatterplot of invasive cSBP, Vicorder cSBP, SphygmoCor cSBP calibrated to brachial SBP/DBP and SphygmoCor cSBP calibrated to brachial MAP/DBP, and linear regression equations (invasive study). Solid lines: lines of equality; dotted lines: linear regression lines. (b) Bland–Altman plots of the differences between Vicorder cSBP, SphygmoCor brachial SBP/DBP-calibrated cSBP and SphygmoCor brachial MAP/DBP-calibrated cSBP. Solid lines: mean value; dotted lines: mean ± 2 SDs. cSBP, central systolic blood pressure; MAP, mean arterial pressure.

Invasive cDBP was lower than oscillometric bDBP (74 versus 78 mmHg, P < 0.001) and, as a consequence, all the three noninvasive estimates of cPP were lower than invasive cPP (Δ Vicorder versus invasive cPP: −9.6 ± 8.2 mmHg, Δ SphygmoCor bSBP/bDBP-calibrated cPP versus invasive: −16.4 ± 8.2 mmHg, Δ SphygmoCor bMAP/bDBP-calibrated cPP versus invasive: −7.5 ± 10.4 mmHg).

In the subgroup of 29 patients in whom bBP was also measured with Omron HEM 705CP, Vicorder bSBP values were not dissimilar from those provided by Omron bSBP (Δ Vicorder versus Omron: −0.5 ± 7.2 mmHg, P = 0.72), whereas Vicorder bDBP was lower than Omron bDBP (Δ Vicorder versus Omron bSDP −2.9 ± 4.3 mmHg, P < 0.01).

Noninvasive study

Good-quality brachial and radial waveforms were available for all 90 patients. Twenty-one of the patients were hypertensive, including 11 on BP-lowering drug treatment. Clinical characteristics are reported in Table 1.

When peripheral waveforms were calibrated to the same bSBP and bDBP, the estimated cSBP was higher with Vicorder than with SphygmoCor (Δ versus Vicorder: −6.2 ± 4.6 mmHg, P < 0.001, Table 4). cPP was also higher with Vicorder than with SphygmoCor (Δ versus Vicorder: −7.1 ± 4.6 mmHg, P < 0.001). Consequently, central-to-peripheral PP amplification was lower with Vicorder than with SphygmoCor (1.17 versus 1.42, P < 0.001).

Values of central SBP, PP and central-to-peripheral PP amplification estimated by Vicorder, SphygmoCor brachial SBP/DBP-calibrated and SphygmoCor brachial MAP/DBP-calibrated

When radial waveforms were re-calibrated with bMAP and bDBP (bMAP calculated by integrating the Vicorder brachial waveform), no significant difference between Vicorder and SphygmoCor was found in the estimate of cSBP (Δ −0.5 ± 3.3 mmHg, P = 0.24) and cPP (Δ −0.7 ± 3.2 mmHg, P = 0.08), whereas a small difference was found in estimated central-to-peripheral PP amplification (1.17 versus 1.20, P = 0.02).

According to the British Hypertension Society criteria [26], there was an excellent interdevice agreement between Vicorder and SphygmoCor for the noninvasive estimate of cSBP when the radial waveform was calibrated to bMAP/bDBP (90% of the differences ≤5 mmHg, 100% ≤10 mmHg, 100% ≤15 mmHg, grade A), but not when the radial waveform was calibrated to bSBP/bDBP (41% of the differences ≤5 mmHg, 77% ≤10 mmHg, 98% ≤15 mmHg, grade D).

The Bland–Altman plots in Fig. 3 show that the Vicorder–SphygmoCor difference in estimated cSBP had a significant relationship with the average of the two estimates of cSBP. The difference between SphygmoCor and Vicorder increased at higher values of cSBP (β = −0.51 and −0.26, both P < 0.001) and cPP (β = −0.44 and −0.20, both P < 0.001).

Vicorder cSBP estimates were strongly and equally correlated with SphygmoCor cSBP estimates, irrespective of the calibration method (r = 0.93 for Vicorder and SphygmoCor bSBP/bDBP-calibrated, r = 0.93 for Vicorder and SphygmoCor bMAP/bDBP calibrated, z-test, P = 0.80).


We have shown, for the first time, that the Vicorder device is adequate for the estimation of cSBP when brachial pressure waveforms are calibrated to invasively recorded aortic mean and diastolic pressures. When the waveforms were calibrated to brachial cuff pressures, as commonly used in clinical practice, both devices significantly underestimated cSBP, although Vicorder estimates of cSBP were still reasonably close and highly correlated to invasive cSBP. Vicorder and SphygmoCor provided estimates of cSBP values that were in better agreement with each other when SphygmoCor-derived radial waveforms were calibrated to brachial MAP/DBP rather than to brachial SBP/DBP, although calibration to brachial MAP/DBP appeared to be associated to a slightly higher imprecision of the estimate. Taken together, these results confirm the validity of a brachial cuff oscillometry-based approach for the noninvasive determination of aortic pressure, as an easy-to-use and operator-independent alternative to conventional radial tonometry [27].

In clinical practice, there is a need for devices that are suitable for high-throughput settings that enable the noninvasive estimation of cSBP and of PP amplification, as indicators of cardiovascular risk beyond brachial BP levels [10–13]. We showed that the brachial cuff-based approach used by the Vicorder device provided results generally in line with those obtained with the SphygmoCor device, which has been demonstrated to estimate cSBP with high accuracy compared with invasively measured cSBP [6,8,9] when radial waveforms are calibrated to invasive pressures. Although SphygmoCor cSBP appeared to be slightly more accurate than Vicorder cSBP, in terms of both absolute BP values and form factor, both devices estimated cSBP with acceptable error, when compared with the standards proposed by AAMI for validation of electronic devices for BP measurement. These results support the concept that the Vicorder device, which uses a built-in brachial-to-aortic transfer function, is able to reproduce the relative differences in the amplitude of pulse waveforms between the aorta and the periphery. A similar method, based on the application of a transfer function to an oscillometric brachial waveform and recently published by Weber et al.[28], has provided results in keeping with those obtained by Vicorder in our study.

When oscillometric brachial SBP/DBP was used for calibrating peripheral waveforms, the true aortic SBP was underestimated by Vicorder and, to a higher degree, by SphygmoCor. Moreover, oscillometric bDBP was higher than invasive cDBP. We hypothesize that these discrepancies are due in large part to differences between intrabrachial BP (the ‘true’ BP in the brachial artery) and the cuff pressure, as measured by the oscillometric device, which is widely acknowledged to underestimate SBP and overestimate DBP [29–31] irrespective of the approach used (oscillometric or auscultatory). As reported in other studies, inaccuracies of cuff pressure measurements may be the source of the underestimation of central SBP and PP [32–34]. Moreover, this is also likely to explain the apparent lack of central-to-peripheral SBP amplification in our invasive study (mean invasive cSBP 141 mmHg versus oscillometric brachial SBP 143 mmHg). Another potential reason for the lack of BP amplification is that we included patients of advanced age, with a high prevalence of cardiovascular risk factors and referred for coronary angiography for suspected or known coronary artery disease, all conditions associated with low BP amplification.

Despite the significant differences of estimated cSBP values with the corresponding invasive values, their strong correlation with invasive cSBP suggests the utility of both SphygmoCor and Vicorder for clinical use. The inherent inaccuracy of noninvasive brachial SBP measurement devices increases the imprecision of any device, which estimates cSBP from peripheral waveform. Nevertheless, brachial BP has proven to be a strong cardiovascular risk factor and the inaccuracy of BP measurement may not jeopardize its ability to predict cardiovascular risk. Similarly, noninvasive devices should be evaluated not only in terms of their ability to reproduce absolute invasive cSBP values, but on the basis of their power to predict cardiovascular risk.

Calibration of the SphygmoCor to brachial SBP/DBP underestimated the true aortic pressure by approximately 12 mmHg, whereas its re-calibration to MAP/DBP derived from brachial oscillometry resulted in a much closer agreement between the two devices and a closer estimate of the true invasive pressure by SphygmoCor. The underestimation of cSBP by the SphygmoCor might be due to the presence of a brachial-to-radial pressure amplification [15,16,18,35,36], which the Vicorder is not sensitive to. However, in our study, when bMAP was used for radial calibration, the SD of the difference in cSBP between invasive and SphygmoCor measurements was larger than that for either the Vicorder cSBP or SphygmoCor cSBP calibrated to SBP/DBP. Thus, any potential imprecision in the assessment of brachial MAP may affect the accuracy of cSBP estimate. In this regard, the Vicorder has an advantage in that bMAP is obtained by calculating the integral of the area under the curve of the brachial waveform, which may theoretically be considered more rigorous than bMAP determined from oscillometry. Further studies are required to clarify the impact of inaccuracies in the determination of MAP on the calibration of a given pressure waveform.

We have shown that Vicorder provides lower cSBP values than SphygmoCor at higher BP levels (see the Bland–Altman plots in Figs 1 and 3). In fact, when using bMAP/bDBP for calibration, we noted a difference of approximately 4 mmHg in SphygmoCor cSBP compared with Vicorder cSBP in the invasive study, whereas no systematic difference between the two devices was found in the noninvasive study. One potential explanation for this divergence may be the difference between radial high-fidelity tonometry and the volume–displacement technique adopted by the Vicorder device for brachial waveform sampling, or on the difference between the two in-built transfer functions.

(a) Scatterplot of Vicorder cSBP, SphygmoCor cSBP calibrated to brachial SBP/DBP and SphygmoCor cSBP calibrated to brachial MAP/DBP, and linear regression equations (noninvasive study). Solid lines: lines of equality; dotted lines: linear regression lines. (b) Bland–Altman plots of the differences between invasive cSBP, Vicorder cSBP, SphygmoCor cSBP calibrated to brachial SBP/DBP and SphygmoCor cSBP calibrated to brachial MAP/DBP. Solid lines: mean values; dotted lines: mean ± 2 SD. cSBP, central systolic blood pressure; MAP, mean arterial pressure.

The findings of the present study should be considered in the light of its limitations. The algorithm provided by Vicorder for the brachial BP oscillometric measurement has not been validated according to standard protocols. However, the good agreement between Vicorder and a validated device (Omron HEM 705CP) [21] in a subgroup of 29 patients supports its use as a reliable tool to measure bBP. Concerns have been raised about the use of fluid-filled catheters for invasive cBP measurement. Even with the highest technical skill, a dampening effect due to microscopic air bubbles or blood clots in the pressure transducer system cannot be ruled out. Also, fluid-filled systems show a relatively slow response to BP changes. However, whereas this may distort the waveform profile, there should be little effect on the absolute values of cSBP and cDBP absolute values and on the waveform area. On the contrary, micro-tip catheters are not immune from bias and dynamic pressure issues. Another limitation of our study is the lack of intrabrachial BP data that did not permit us to validate, invasively, the estimation of SBP and PP amplification.

Some concerns have been raised about the use of the same mathematical generalized transfer functions to generate cBP waveforms in different clinical settings [37]. As we were not aware of the built-in algorithms used for mathematical transformations, we could not compare the performances of the two transfer functions themselves. Furthermore, it should be mentioned that results provided by our invasive study have been obtained in a selected population at high cardiovascular risk undergoing coronary angiography, and the findings may not be extended to patients without vascular disease.

In our invasive study, a sizeable minority of the patients were excluded due to low-quality waveforms. This was largely related to technical difficulties in making recordings during an interventional procedure. Nevertheless, when the two devices were tested in the more routine clinical setting of our noninvasive study, all waveforms were of high quality. Finally, similarly to Hope et al. [38], we chose to test the agreement of cSBP estimates with invasive measures according to the AAMI criteria. We acknowledge that such criteria were not developed to validate devices for the noninvasive assessment of cBP against invasive pressures and may not be appropriate in this setting, also considering the relatively wide BP limits allowed. However, neither the specific criteria for validation of noninvasive devices have yet been developed nor the cut-off values for cBP related to different prognosis have been established.

In conclusion, there is increasing interest in brachial cuff-based devices for the evaluation of cBP, primarily because they are easy to use and avoid the need for skilled technicians. Therefore, such devices may facilitate the widespread introduction and use of cBP parameters for risk stratification and follow-up [39]. We have shown that both Vicorder and SphygmoCor provide accurate estimates of cSBP when calibrated to invasively recorded MAP/DBP. However, calibrating the devices to brachial cuff pressures, derived using oscillometry, resulted in an underestimation of the true cSBP. These data indicate that the accuracy of noninvasive estimation of cSBP depends on the method of waveform calibration and is subject to the widely acknowledged inaccuracy of cuff-based assessments of brachial pressure. Nevertheless, the Vicorder is adequate for the noninvasive estimation of central BP and may be considered as a simple alternative to tonometry-based methods. Further improvements to brachial cuff-based techniques for estimating cBP may eventually allow better cardiovascular risk stratification at the population level beyond the research domain.


J.C., I.B.W. and C.M.M. acknowledge funding support from the NIHR Cambridge Biomedical Research Centre, Cambridge University Hospitals and the Comprehensive Local Research Network. I.B.W. also acknowledges British Heart Foundation funding for his academic posts.

Conflicts of interest

There are no conflicts of interest. No funds have been obtained for preparing this article.

Reviewers’ Summary Evaluations Reviewer 1

The increasing importance of estimating central blood pressure (CBP) as a diagnostic and prognostic tool in the management of arterial hypertension is well known. The device tested by the authors seems to be as accurate as traditional applanation tonometry. It could have important advantages over the later. Firstly, this methodology is not operator-dependent, which can contribute to increase its use. Secondly, the calibration of the brachial waveform to the pressure also obtained from the brachial artery eliminates possible differences (amplification) between brachial and radial artery. This advantage could, at least theoretically, favor the accuracy of Vicorder system.

Reviewer 2

Pucci and co-authors report a thorough study of the effect of the different techniques and calibration methods utilized in two commercially available devices for the noninvasive assessment of central blood pressure. They report relatively large differences between devices which emphasizes the potential requirement for device specific ‘normal’ values - an important issue if estimation of central BP is to be applied clinically, since individuals may visit multiple practitioners using different devices. An important finding in the invasive arm of their study is that (possibly due to inherent inaccuracies in brachial measurement) no significant peripheral pressure amplification was apparent, questioning the usefulness of estimation of central BP in some cohorts.


1. Franklin SS, Lopez VA, Wong ND, Mitchell GF, Larson MG, Vasan RS, Levy D. Single versus combined blood pressure components and risk for cardiovascular disease: the Framingham Heart Study. Circulation 2009; 119:243–250.
2. Pauca AL, Wallenhaupt SL, Kon ND, Tucker WY. Does radial artery pressure accurately reflect aortic pressure? Chest 1992; 102:1193–1198.
3. Nichols WW, O’Rourke MF. McDonald's blood flow in arteries. Theoretical, experimental and clinical principles. V ed. London: Arnold; 2005.
4. Wilkinson IB, Franklin SS, Hall IR, Tyrrell S, Cockcroft JR. Pressure amplification explains why pulse pressure is unrelated to risk in young subjects. Hypertension 2001; 38:1461–1466.
5. McEniery CM, Yasmin, McDonnell B, Munnery M, Wallace SM, Rowe CV, et al., Anglo-Cardiff Collaborative Trial Investigators. Central pressure: variability and impact of cardiovascular risk factors – the Anglo-Cardiff Collaborative Trial II. Hypertension 2008; 51:1476–1482.
6. Karamanoglu M, O’Rourke MF, Avolio AP, Kelly RP. An analysis of the relationship between central aortic and peripheral upper limb pressure waves in man. Eur Heart J 1993; 14:160–167.
7. Pauca AL, O’Rourke MF, Kon ND. Prospective evaluation of a method for estimating ascending aortic pressure from the radial artery pressure waveform. Hypertension 2001; 38:932–937.
8. Chen CH, Nevo E, Fetics B, Pak PH, Yin FC, Maughan WL, Kass DA. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure. Validation of generalized transfer function. Circulation 1997; 95:1827–1836.
9. Fetics B, Nevo E, Chen CH, Kass DA. Parametric model derivation of transfer function for noninvasive estimation of aortic pressure by radial tonometry. IEEE Trans Biomed Eng 1999; 46:698–706.
10. Roman MJ, Devereux RB, Kizer JR, Okin PM, Lee ET, Wang W, et al. High central pulse pressure is independently associated with adverse cardiovascular outcome the strong heart study. J Am Coll Cardiol 2009; 54:1730–1734.
11. Vlachopoulos C, Aznaouridis K, O’Rourke MF, Safar ME, Baou K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with central haemodynamics: a systematic review and meta-analysis. Eur Heart J 2010; 31:1865–1871.
12. Benetos A, Thomas F, Joly L, Blacher J, Pannier B, Labat C, et al. Pulse pressure amplification a mechanical biomarker of cardiovascular risk. J Am Coll Cardiol 2010; 55:1032–1037.
13. Safar ME, Blacher J, Pannier B, Guerin AP, Marchais SJ, Guyonvarc’h PM, et al. Central pulse pressure and mortality in end-stage renal disease. Hypertension 2002; 39:735–738.
14. Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D, et al. CAFE InvestigatorsDifferential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes. Circulation 2006; 113:1213–1225.
15. Mahieu D, Kips J, Rietzschel ER, De Buyzere ML, Verbeke F, Gillebert TC, et al. Noninvasive assessment of central and peripheral arterial pressure (waveforms): implications of calibration methods. J Hypertens 2010; 28:300–305.
16. Verbeke F, Segers P, Heireman S, Vanholder R, Verdonck P, Van Bortel LM. Noninvasive assessment of local pulse pressure: importance of brachial-to-radial pressure amplification. Hypertension 2005; 46:244–248.
17. Segers P, Mahieu D, Kips J, Rietzschel E, De Buyzere M, De Bacquer D, et al. Amplification of the pressure pulse in the upper limb in healthy, middle-aged men and women. Hypertension 2009; 54:414–420.
18. Davies JE, Shanmuganathan M, Francis DP, Mayet J, Hackett DR, Hughes AD. Caution using brachial systolic pressure to calibrate radial tonometric pressure waveforms: lessons from invasive study. Hypertension 2010; 55:e4.
19. O’Rourke MF. Influence of ventricular ejection on the relationship between central aortic and brachial pressure pulse in man. Cardiovasc Res 1970; 4:291–300.
20. Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JP, et al. The BHS Guidelines Working PartyBritish Hypertension Society Guidelines for Hypertension Management, 2004 – BHS IV: summary. BMJ 2004; 328:634–640.
21. O’Brien E, Mee F, Atkins N, Thomas M. Evaluation of three devices for self-measurement of blood pressure according to the revised British Hypertension Society Protocol: the Omron HEM-705CP, Philips HP5332, and Nissei DS-175. Blood Press Monit 1996; 1:55–61.
22. Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310.
23. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989; 45:255–268.
24. Gardner RC. Psychological statistics using SPSS for Windows. New Jersey: Prentice Hall; 2001.
25. White WB, Berson AS, Robbins C, Jamieson MJ, Prisant M, Roccella E, Sheps SG. National standard for measurement of resting and ambulatory blood pressures with automated sphygmomanometers. Hypertension 1993; 21:504–509.
26. O’Brien E, Petrie J, Littler W, de Swiet M, Padfield PL, Altman DG, et al. The British Hypertension Society protocol for the evaluation of blood pressure devices. J Hypertens 1993; 11 (Suppl 2):S43–S62.
27. Cheng HM, Wang KL, Chen YH, Lin SJ, Chen LC, Sung SH, et al. Estimation of central systolic blood pressure using an oscillometric blood pressure monitor. Hypertens Res 2010; 33:592–599.
28. Weber T, Wassertheurer S, Rammer M, Maurer E, Hametner B, Mayer CC, et al. Validation of a brachial cuff-based method for estimating central systolic blood pressure. Hypertension 2011; 58:825–832.
29. Van Berghen FH, Weatherheads DS, Treloar AE, Dobkin AB, Buckley JJ. Comparison of indirect and direct methods of measuring arterial blood pressure. Circulation 1954; 10:481–490.
30. Holland WW, Humerfelt S. Measurement of blood-pressure: comparison of intra-arterial and cuff values. BMJ 1964; 2:1241–1243.
31. Rasmussen PH, Staats BA, Driscoll DJ, Beck KC, Bonekat HW, Wilcox WD. Direct and indirect blood pressure during exercise. Chest 1985; 87:743–748.
32. Smulyan H, Siddiqui DS, Carlson RJ, London GM, Safar ME. Clinical utility of aortic pulses and pressures calculated from applanated radial-artery pulses. Hypertension 2003; 42:150–155.
33. Takazawa K, O’Rourke MF, Fujita M, Tanaka N, Takeda K, Kurosu F, Ibukiyama C. Estimation of ascending aortic pressure from radial arterial pressure using a generalised transfer function. Z Kardiol 1996; 85:137–139.
34. Cloud GC, Rajkumar C, Kooner J, Cooke J, Bulpitt CJ. Estimation of central aortic pressure by SphygmoCor requires intra-arterial peripheral pressures. Clin Sci 2003; 105:219–225.
35. Rezai MR, Goudot G, Winters C, Finn JD, Wu FC, Cruickshank JK. Calibration mode influences central blood pressure differences between SphygmoCor and two newer devices, the Arteriograph and Omron HEM-9000. Hypertens Res 2011; 34:1046–1051.
36. O’Rourke MF, Takazawa K. Measurement of central aortic pressure: an acceptable compromise? [letter]. J Hypertens 2011; 29:2038– reply 2039–2040, 2040–2041.
37. Hope SA, Meredith IT, Cameron JD. Arterial transfer functions and the reconstruction of central aortic waveforms: myths, controversies and misconceptions. J Hypertens 2008; 26:4–7.
38. Hope SA, Meredith IT, Cameron JD. Effect of noninvasive calibration of radial waveforms on error in transfer-function-derived central aortic waveform characteristics. Clin Sci (Lond) 2004; 107:205–211.
39. Schillaci G, Grassi G. Central blood pressure: getting to the heart of the matter. J Hypertens 2010; 28:237–239.

applanation tonometry; central blood pressure; hypertension; oscillometric signals; pulse wave analysis; wave reflection

© 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins