Journal of Applied Physiology Ad Instruments
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 101: 1207-1214, 2006. First published June 1, 2006; doi:10.1152/japplphysiol.00876.2005
8750-7587/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
101/4/1207    most recent
00876.2005v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stok, W. J.
Right arrow Articles by Karemaker, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stok, W. J.
Right arrow Articles by Karemaker, J. M.

Changes in finger-aorta pressure transfer function during and after exercise

Wim J. Stok,1 Berend E. Westerhof,2 and John M. Karemaker1

1Department of Physiology, Academic Medical Center, University of Amsterdam; and 2BMEYE, Academic Medical Center, Amsterdam, The Netherlands

Submitted 20 July 2005 ; accepted in final form 18 May 2006


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Noninvasive finger blood pressure has become a surrogate for central blood pressure under widely varying circumstances. We tested the validity of finger-aorta transfer functions (TF) to reconstruct aortic pressure in seven cardiac patients before, during, and after incremental bicycle exercise. The autoregressive exogenous model method was used for calculating finger-aorta TFs. Finger pressure was measured noninvasively using Finapres and aortic pressure using a catheter-tip manometer. When applying the individual TFs found during rest for reconstruction of aortic pressure during all workloads, systolic pressure was increasingly underestimated, with large variation between subjects: +4.0 to –18.1 mmHg. In most subjects, diastolic pressure was overestimated: –3.9 to +5.5 mmHg. Pulse pressure estimation varied between +4.5 and –21.9 mmHg. In all cases, wave distortion was present. Postexercise, error in reconstructed aortic systolic pressure slowly declined, and diastolic pressure was overestimated. During rest, the TF gain had a minimum between 3.65 and 4.85 Hz (Fmin). During exercise, Fmin shifted to frequencies between 4.95 and 7.15 Hz at the maximum workload, with no change in gain. Postexercise, gain in most subjects shifted to values closer to unity, whereas Fmin did not return to resting values. Within each subject, aorta-Finapres travel time was linearly related to mean pressure. During exercise, Fmin was linearly related to both delay and heart rate. We conclude that, during increasing exercise, rest TFs give an increasingly unreliable reconstruction of aortic pressure, especially at higher heart rates.

autoregressive exogenous

IN CARDIOVASCULAR FUNCTION testing, blood pressure (BP) and heart rate (HR) can be measured noninvasively at a peripheral site, for instance at the finger (Finapres), using the volume clamp technique of Penáz (32) and the Physiocal criteria of Wesseling et al. (44). In should be noted, however, that, during exercise, important changes in central-to-peripheral pressure differences have been described (2, 7, 35). The usual peripheral systolic pressure augmentation increases further during exercise, by an amount that may differ between subjects. Therefore, peripheral pressure may become an unreliable measure for accurate estimation of cardiac load. Many attempts have been made to reconstruct brachial or aortic pressures (AORs) from these peripheral measurements by the use of transfer functions (TFs).

The use of a generalized TF (gTF) for AOR reconstruction from the radial site has been proposed and tested in several studies (4, 8, 10, 31, 39), but its use is still criticized by others (5, 6, 14, 16, 17, 23, 24). Part of this dispute is, however, related to calibration errors, rather than to the use of a gTF itself (5, 15, 23, 29). For group averages, a gTF may perform well but less so in individual cases. This is even more of an issue when more parameters from the pressure wave are to be derived than just systolic and diastolic pressures (17, 26), and results might be disturbed by distortion of the reconstructed waves.

Therefore, several (model-based) attempts have been made to individualize TFs (20, 36, 41, 45). Even if a TF for an individual subject is known that works well under resting conditions, it is still unsettled how this TF will perform under challenging cardiovascular conditions like exercise. In a preliminary report, Sharman et al. (37) mentioned good results using a gTF in a study in 15 patients using radial tonometry. However, HR increased by only 13 beats/min during the exercise (37). In cardiovascular function testing, one will be typically interested in the responses in the individual only.

Most studies have involved responses to changes in BP by pharmacological interventions, but HR changes in those studies were minor. Chen et al. (4) studied aorta-to-radial TF during hemodynamic transients (Valsalva's maneuver, abdominal compression, nitroglycerin, or inferior caval vein obstruction). Averaged TFs changed relatively little during pressure transients, but intersubject variability increased. In 4 of 14 subjects, the variation in peak amplitude of the TF was >20%. The use of a gTF resulted in distortions in reconstructed systolic pressures, even though the error in reconstructed systolic peak pressure estimates was found to be moderate.

During leg exercise, apart from increases in BP and HR, peripheral vasoconstriction by sympathetic activation in the arm can be expected (before vasodilatation to allow heat dissipation sets in) (43). Any change in pulse wave velocity or reflection coefficient is most likely to change the TF (21, 40, 41). In addition, during higher HRs, the lower harmonic components of the pressure wave, with the highest energy, may come closer to the point where the gain in the TF deviates most from unity, thus making the wave form much more sensitive to even minor changes in this TF.

In the present study, we, therefore, investigated the finger-to-central (aorta) wave reconstruction during exercise using individualized inverse TFs. Data were used from experiments performed earlier by Blum et al. (2) in seven subjects using Finapres and AOR during static and dynamic exercise.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Subjects.   Measurements were performed in seven male subjects after coronary angiography for diagnostic evaluation of ischemic heart disease. Age, weight, and height of the subjects are presented in Table 1. To allow comparison, we will use the initials for the subjects as Blum et al. (2) did. beta-Blockers, nitrates, and calcium channel blockers were taken by most of the subjects. One subject (JB) was not under beta-blockade. Four subjects were moderately hypertensive (JD, AK, WS, and HG). The subjects gave their informed consent before the experiments. The study was approved by the Ethics Committees of the Academic Hospital of the University of Utrecht (UMCU) and the Eemland Hospital in Amersfoort, the Netherlands, where the measurements were performed.


View this table:
[in this window]
[in a new window]
  		Table 1. Subject data, maximal workload, heart rate, and blood pressures during rest period and at the maximum workload

 
Measurements.   AOR was measured after coronary angiography using a calibrated Millar type VPC 663 catheter (frequency response, including amplifier, >5 kHz), diameter 2 mm, inserted in the right brachial artery. The sensor was located ~8 cm from the aortic valve. The sensor measured side pressure, which minimizes the kinetic energy factor that may induce erroneous pressure readings during larger cardiac output values in exercise. Finger arterial pressure was measured on the midphalanx of the middle finger of the left hand using a Finapres model 5 (BMI, TNO, The Netherlands) (formerly commercially available as the Ohmeda Finapres 2300 E NIBP monitor). The finger was held on the examination table, some 10–15 cm below heart level. During data processing, all finger pressure (FINAP) results were compensated for this height difference with a fixed –10 mmHg offset.

Exercise protocol.   We used the data from the dynamic exercise part of the protocol. The test was performed using an electromagnetically braked cycle ergometer with the subject in the supine position. The protocol consisted of bicycling at 70 rpm, incrementing the workload by 25 W every 2 min. The first step was 25 or 50 W, depending on the expected maximal capacity of the subject, which had been tested recently before. Only one subject reached the maximal workload of 175 W (Table 1). After the maximal attained exercise level, the workload was reduced to 25 W for 2 min (the first minute indicated as R25a, the second as R25b) and thereafter to 0 W (also two measurements, R0a and R0b), after which the catheter was retracted. The total duration of the postexercise period was 4.5 min or less.

Data analysis.   The Finapres and AOR signals were sampled at 100 Hz, with 0.25-mmHg resolution. Beat-to-beat data of systolic and diastolic pressure and heart interval were calculated using custom-made software. Mean pressures (MAPs) were calculated from the integral over one beat and HR from the interbeat interval. For analysis of the TF, the data at the end of an exercise step were selected. In most cases, more than 1 min of artifact-free continuous data was available. The minimum accepted length was 10 s. The maximum length used for the filter calculations was 60 s.

TF estimates were calculated using an autoregressive exogenous (ARX) model method (25) with 10 poles and zeros, as proposed by Chen et al. (4, 8).

For the implementation, Matlab (The Mathworks) version 5.21 with System Identification Toolbox was used. Before calculation of the TFs with ARX, both signals were detrended and low-pass filtered at a cutoff frequency just above the 10th harmonic. At the maximal observed FINAP pulse pressure of 126 mmHg during exercise, the amplitude of the 10th harmonic was less than ~1.9 mmHg. The TF is calculated in the finger-aorta direction; therefore, for the ARX model, the FINAP signal is defined as the input signal and the aorta as output. The input must occur in time before the output, which was ensured by shifting the FINAP signal back in time using the largest delay between the first 10 harmonics of both pressure waves, as calculated by an fast Fourier transform-based TF estimate. A recursive filter of higher order (15 poles and zeros) was fitted to the TF's gain and phase data. Gain at zero frequency was incorporated in this filter by forcing it to be equal to MAPAOR/MAPFINAP. Care was taken to leave the filter's gain and phase characteristics equal to the original TF at frequencies above one-half of the first harmonic of the pressure signal. The filter was then used to transform the sampled Finapres signal in the time domain.

Delay between the AOR and FINAP pressure signals used in the parameter dependencies section was calculated using the classical upstroke foot-to-foot method.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Individual data for HR, MAP, and systolic pressure (both aorta and finger) during the resting period and the maximal exercise load are presented in Table 1. The Finapres level correction of 10 mmHg (see METHODS) was appropriate, as in most cases, aortic and finger MAPs were in close agreement or FINAP was lower (subject AK), which could be physiological.

TF during rest.   First, FINAP-AOR inverse TFs for each of the subjects were estimated during the resting period (TFrest). The waveform reconstruction using the ARX method was excellent. Only the incisura at the moment of aortic valve closure is less well defined. Fig. 1 shows two examples of the reconstructed AOR (rAOR) of subjects CT and JB. Differences between rAOR and AOR systolic pressure, diastolic pressure, and MAP were calculated beat to beat and averaged. Systolic differences did not exceed 1 mmHg, whereas pulse pressure differences were 2 mmHg or less (Table 2). A first minimum in the TF (Fmin) was found between 3.65 and 4.85 Hz, and in five of the seven subjects a second minimum between 6.8 and 8.2 Hz, which in most cases was shallower. The TF modulus at the first minimum was rather different between subjects, ranging from 0.08 to 0.31. DC gain was close to one in most subjects (1.08 in subject AK), indicating only a small difference in MAP between finger and aorta (Table 1). A gain >1 implies that MAP in the finger is lower.


Figure 1
View larger version (8K):
[in this window]
[in a new window]
  		Fig. 1. Aortic pressure (AOR) wave reconstruction with individualized transfer functions (TFs) from the Finapres signal in 2 subjects [CT (A) and JB (B)] during the resting period. Top traces, right axis: upward shifted Finapres signal. Bottom traces, left axis: AOR (thin line) and reconstructed AOR (rAOR; thick line).

 

View this table:
[in this window]
[in a new window]
  		Table 2. Measured and estimated pulse pressure differences

 
TFrest during exercise and postexercise.   We applied TFrest to estimate AOR from FINAP, both during the periods of exercise with increasing workload and during postexercise. During exercise, small changes in aorta-finger MAP differences developed. In the TFrest filter, only the MAP relation between AOR and FINAP during rest is incorporated, and therefore the rest filter does not correct these differences. For calculation of the systolic and diastolic reconstructed pressures, we corrected, therefore, for the differences in MAP. This correction does not influence pulse pressure estimation.

Results for the systolic and diastolic estimation are presented in Fig. 2, C and D, together with the differences between the raw Finapres signal and AORs (Fig. 2, A and B). During exercise, TFrest underestimated systolic rAOR increasingly in six out of seven subjects but with a large variation between subjects. Diastolic rAOR was moderately overestimated between 0 and 6 mmHg in six out of seven subjects. The reconstructed pulse pressure was underestimated by >5 mmHg in four of seven subjects and at the higher workloads by >15 mmHg in three of seven subjects (Table 2). Postexercise, this increased initially to six of seven subjects, while diastolic rAOR was overestimated in all subjects. Average differences for all subjects for all workloads and postexercise, rest values excluded, and corrected for MAP were –6.3 ± 6.4, 2.2 ± 2.6, and –8.5 ± 8.7 mmHg for systolic, diastolic, and pulse pressure, respectively.


Figure 2
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 2. Differences between Finapres and AORs (Fin-AOR) (A and B) and between rAOR and measured AORs (rAOR-AOR) (C and D) when using the TF calculated for each of the subjects during the resting period for both the exercise and the recovery period. Both Finapres and reconstructed pressures are corrected for mean pressure. A and C: systolic pressures; B and D: diastolic pressures.

 
Although in some subjects (CT, HG, WS) the error in pulse pressure was smaller, also in those cases the reconstructed pressure waveform is distorted (Fig. 3A). When the pulse pressure error was large, only part of the pressure wave appeared to be reconstructed, and the peak systolic part was cut off (Fig. 3B).


Figure 3
View larger version (9K):
[in this window]
[in a new window]
  		Fig. 3. Pressure reconstruction with individual rest TFs (TFrest) during rest, increasing levels of exercise (steps of 25 W each) and postexercise in subjects CT (A) and JB (B). From top to bottom: AOR, rAOR, and finger pressure (FINAP). In subject CT, systolic pressure is almost correct but with large-wave distortion; in subject JB, systolic pressure is largely underestimated.

 
TFs during exercise and postexercise.   To further investigate the obvious changes in the TFs, for each workload, individual TFs were calculated (exercise TF). In all subjects, the first Fmin shifted to higher frequencies at increasing workload (Fig. 4A), but the gain (TF modulus) at those frequencies remained almost constant in six out of seven subjects (Fig. 4B). Postexercise, Fmin only slowly shifted back and did not reach the initial resting values, whereas gain at Fmin increased in most subjects. In fact, the TF modulus curve was shifted slightly upwards toward unity at all frequencies, quite distinct in one subject (JB). DC gain was also increased in four out of seven subjects, implying a larger MAP decrease between aorta and finger than during rest and exercise. Two examples of changes in TFs during exercise and postexercise are presented in Fig. 5, A and B, respectively, together with the frequencies of the first four harmonics of the pressure waves. In comparison (Fig. 2), in subject CT, the error in rAOR introduced by the use of TFrest during exercise remained small at first, while in subject JB the error increased rapidly.


Figure 4
View larger version (8K):
[in this window]
[in a new window]
  		Fig. 4. Change in frequency (A) of the first minimum and gain at that minimum (B) in the finger-aortic TF in 7 subjects during and after incremental bicycle exercise. Note that not all subjects reached the same maximum workload. R25a and R25b, first and second minute of recovery period at 25 W workload; R0a and R0b, first and second minute after workload reduced to zero.

 

Figure 5
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 5. Finapres-aortic TFs (moduli) calculated on the first 10 harmonics of the pressure waves during rest, increasing workloads, and the recovery period in subject CT (A) and JB (B). For convenience, curves are shifted upward relative to the TFrest. The left axis ticks indicate the 1x gain level for each TF. On each TF, the dots indicate the frequencies of the first 4 harmonics of the pressure wave at that workload. The vertical line is at the frequency of the minimum in the TFrest.

 
Parameter dependencies.   We investigated parameter relations expected to be of influence on the TF, in particular on the frequency of the first minimum (21, 40, 41, 45): BP level, wave front delay, and (possibly) HR. These relations are, however, complicated by the fact that, during exercise, both BP values and HR change in parallel.

Within each subject, AOR-FINAP pulse-wave delay changed almost linearly with aortic MAP during exercise and postexercise. Both returned to or below the resting value during postexercise. During exercise, Fmin increased almost linearly with the decrease in AOR-FINAP delay (Fig. 6A) and with the increase in HR (Fig. 6B, HR presented in hertz). However, during postexercise, Fmin remained unexpectedly high in four of seven subjects compared with the decrease in both delay and HR.


Figure 6
View larger version (7K):
[in this window]
[in a new window]
  		Fig. 6. A: relation between averaged aorta-Finapres delay and the first minimum in the TF (Fmin) during rest and exercise. B: relation between heart rate (HR) (presented in Hz) and Fmin during rest and exercise.

 
The quotient of Fmin and HR at the corresponding load changes from an average of 3.6 at rest to 2.9 at maximum load, indicating that Fmin is shifting from between the third and fourth harmonic to a position below the third (see also Fig. 5).

Use of a general TF.   For testing the possible use of a general TF, we averaged the TFs from the resting periods of the seven subjects and used this as a single TF on the rest and exercise data (postexercise was excluded). Results for pulse pressure are presented in Table 2. During rest, the difference between AORs and rAORs was equally small as using individual TFs, except in one subject (JD: +5.6 mmHg). In this subject, the large underestimation during exercise with the individual rest filter disappeared with this gTF. During exercise, the differences in the other six subjects remained almost the same or shifted slightly to more positive values compared with the use of the individual rest TFs.


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
We calculated individualized finger (Finapres)-aortic TFs in seven subjects during rest and exercise in the supine position. We showed that the TFs change during exercise, in most subjects invalidating the reconstruction of AOR from FINAP by a single TF that performed well under resting conditions. In all subjects, wave distortions developed using the rest TFs during exercise, even in the present study, where we did not have to resort to a gTF, but had the individualized TFs available from pressure measurements at both central and peripheral sites. Using individualized TFs has the advantage that calibration errors are avoided. Calibration errors have in part confused the discussion about the validity of the use of general TFs (23, 29). Although Finapres gives an absolute pressure measurement, any additional calibration that might be necessary is automatically incorporated in the individual filter as a constant factor that is valid for all frequencies.

In their recent study, Gallagher et al. (10) examined central-to-peripheral TFs in a large population. They stressed that a TF can only be useful in clinical testing if it remains stable over the various interventions. They conclude that, in their cohort, a gTF may be used for the central-to-radial transfer. However, their comparisons for various BP levels are restricted to interindividual data. The intraindividual interventions are restricted to moderate decreases in BP at about normal HR. The natural course of events during spontaneous exercise differs in many aspects from the physiology observed during pharmacological interventions (4, 12, 26, 31, 39). During leg exercise, both BP and HR will increase considerably, and, in the arm, initial vasoconstriction, followed by a later vasodilatation, can be expected (33, 43). Due to vessel wall properties, where the wall is becoming increasingly stiffer at higher BP values, it can be expected that the TF is much more affected by an increase in BP than by a moderate decrease in pressure.

The shift of the first Fmin to higher frequencies can thus be explained by a less compliant behavior of the vascular bed at higher pressures, as supported by the shorter delays between aortic and FINAP waves. An alternative explanation, a higher reflection coefficient and a lower compliance of the small arteries due to vasoconstriction, is less likely, since both will result in a lower gain at the first minimum (21). Our findings are probably not specific for FINAP. Rowell et al. (35) published radial and AOR waves during rest and exercise. We digitized one of his figures of one subject and calculated the inverse TFs as described above. This revealed a shift of the TF minimum to higher frequencies, from 5.1 Hz at rest to 7.45 Hz at the second exercise step (at 47% maximal oxygen uptake).

The shift of the lower harmonics of the pressure wave, with the highest power, to higher frequencies with increasing HR, makes the shape of the reconstructed pressure wave clearly more sensitive to small changes in the TF, since these harmonics are approaching its first minimum. This is more true when TFrest is used during exercise. The vertical lines at the frequency of Fmin in Fig. 5 show why TFrest fails at increasing HRs: at rest Fmin relates to the fourth (A) or third (B) harmonic with relative little power. At the higher HRs, this shifts to the second harmonic and even below that, explaining the observed distortion in Fig. 3. Small changes in TF at normal resting HRs will have a much lesser effect on the reconstructed waveform, because only the higher harmonics, with less power, are affected. This explains the reasonable success in the use of general TFs applied to pressure recordings from pharmacological intervention studies that were the basis for previous gTF-validations (4, 10, 31, 39).

During exercise, the Fmin was linearly related to HR within subjects and in the low range remarkably similar between subjects. HR has been mentioned as a determinant of pulse wave velocity (11, 22), although this is questioned by others (30, 46). In the present study, we cannot distinguish the possible effects of HR from those of pressure. However, an effect of HR alone seems unlikely, since the viscoelastic properties of arteries are almost frequency independent in the range of prevailing HRs and harmonics of the pressure wave (1, 27). Segers et al. (36), in their model study of the upper limb, found no direct relation between TF model parameters and intersubject resting BP, HR, or age. Sugimachi et al. (41), however, found a strong improvement in estimations of central pressure by use of transmission delay to individualize the TF, which was confirmed in a study by Westerhof (45).

We measured peripheral pressure at the finger. Reliability of Finapres has been validated during exercise and peripheral vasoconstriction (2, 9, 12, 13, 18, 19, 38). During vasodilatation in the recovery period, a diminished peripheral systolic pressure amplification has been observed (13, 28). The same has been observed during vasodilatation by heat stress (42). This is due to a lower forearm vascular resistance with less pulse wave reflections (34), as confirmed in model studies (21, 36). Vasodilatation may, thereby, lead to erroneous conclusions about changes in central systolic pressure during this period. Translated to TFs, in the present study, we found the overall gain of the TFs to be increased toward unity, which indeed implies diminished pressure amplification in the aorta-finger direction. This combination of factors, including small-vessel capacitance, is probably also responsible for the high TF minimum postexercise in some of the subjects, as both MAP and the delay returned to near resting values, and we consider the moderately increased postexercise HR insufficient as a sole explanation.

Use of a gTF.   During standard clinical exercise testing, it would be preferable if a general TF could be used.

To our knowledge, a gTF for the finger-aortic pathway, based on measurements in a larger subject population, has not been published yet. An estimate can be made from a study of Karamanoglu and Feneley (20). The frequency of the Fmin of that TF was higher than in our subjects' TFrest, and, therefore, this TF did not perform well during rest. We have, therefore, averaged the seven rest TFs from our study and used it as a gTF for the rest and exercise period. The individual scalings in the TFs are lost by the averaging. Although this is not verified for Finapres, we corrected this in the individual using the equal central and peripheral diastolic and MAP method that is also used to calibrate tonometry data. Using no correction introduced differences in estimated pulse pressure up to 8 mmHg already during rest. It should be realized that, in this study, invasively measured data are used for the correction.

In some subjects, the differences shifted to more positive values, but the range remained equally large (–17.9 to +6.0 compared with –21.9 to +4.5 mmHg). The mean absolute difference was smaller (4.8 compared with 7.3 mmHg). Differences were largest in the subjects with the highest HR during exercise (Table 1). The better performance of the gTF during exercise in subject JD (Table 2) can be explained by the low frequency (3.65 Hz) of Fmin in the individual TFrest of this subject. The higher Fmin of the gTF will, therefore, be closer to the optimum during exercise in this subject, at the same time worsening the performance during rest.

Limitations.   We incorporated the central-to-peripheral drop in MAP in the filters as gain at zero frequency. As long as the central-to-peripheral pressure drop is small and does not change, realistic central pressures can be derived from the FINAP alone. However, in our study, small central-to-peripheral MAP differences developed during exercise. We could not discriminate between physiological changes and changes in the height difference between the measurement sites. We, therefore, corrected for these changes, although this may not be possible using only noninvasive data. It will affect the systolic and diastolic, but not the pulse pressure, estimates.

It may be questioned whether a linear approach for the reconstruction is still feasible at high pressures and whether the much larger kinetic energy component during systole can be neglected (3). Because of ethical reasons, this kind of study may almost exclusively be performed in patients with cardiovascular disease, who may expect to benefit from the outcome. For patient history, we had to rely on the original publication (2): apart from resting BP and the use of beta-blockers, no individual data on cardiovascular medication were available.

The age range is limited, and exercise performance and maximal HRs are different. However, we feel that this does not detract from the general conclusions.

In conclusion, during increasing exercise, changes in the aorta-finger TF develop. The frequency of the minimum in the inverse TF is shifting to higher frequencies at increasing workload. This makes central pressure reconstruction making use of only rest aorta-finger TFs unreliable, especially at higher HRs.


   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
The authors thank Dr. Viktor Blum for providing the original data set (2), thereby making this study possible.


   FOOTNOTES
 

Address for reprint requests and other correspondence: Wim J. Stok, Dept. of Physiology, Academic Medical Center, Univ. of Amsterdam, Rm. M01–214, Meibergdreef 9, NL-1105 AZ Amsterdam, The Netherlands (e-mail: w.stok{at}amc.uva.nl)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Bergel DH. The dynamic elastic properties of the arterial wall. J Physiol 156: 458–469, 1961.
  2. Blum V, Carriere EG, Kolsters W, Mosterd WL, Schiereck P, and Wesseling KH. Aortic and peripheral blood pressure during isometric and dynamic exercise. Int J Sports Med 18: 30–4, 1997.[Web of Science][Medline]
  3. Burton AC. Physiology and Biophysics of the Circulation. Chicago, IL: Year Book Medical, 1972, p. 105.
  4. Chen CH, Nevo E, Fetics B, Pak PH, Yin FC, Maughan WL, and Kass DA. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure. Validation of generalized transfer function. Circulation 95: 1827–1836, 1997.[Abstract/Free Full Text]
  5. Cloud GC, Rajkumar C, Kooner J, Cooke J, and Bulpitt CJ. Estimated central aortic pressure by SphygmoCor requires intra-arterial peripheral pressures. Clin Sci (Lond) 105: 219–225, 2003.[Medline]
  6. Davies JI, Band MM, Pringle S, Ogston S, and Struthers AD. Peripheral blood pressure measurement is as good as applanation tonometry at predicting ascending aortic blood pressure. J Hypertens 21: 571–576, 2003.[CrossRef][Web of Science][Medline]
  7. Eckert S and Horstkotte D. Comparison of Portapres non-invasive blood pressure measurement in the finger with intra-aortic pressure measurement during incremental bicycle exercise. Blood Press Monit 7: 179–183, 2002.[CrossRef][Web of Science][Medline]
  8. Fetics B, Nevo E, Chen CH, and Kass DA. Parametric model derivation of transfer function for noninvasive estimation of aortic pressure by radial tonometry. IEEE Trans Biomed Eng 46: 698–706, 1999.[CrossRef][Web of Science][Medline]
  9. Gabriel A, Lindblad LE, and Angleryd C. Non-invasive vs. invasive beat-to-beat monitoring of blood pressure. Clin Physiol 12: 229–235, 1992.[Web of Science][Medline]
  10. Gallagher D, Adji A, and O'Rourke MF. Validation of the transfer function technique for generating central from peripheral upper limb pressure waveform. Am J Hypertens 17: 1059–1067, 2004.[CrossRef][Web of Science][Medline]
  11. Giannattasio C, Vincenti A, Failla M, Capra A, Ciro A, De Ceglia S, Gentile G, Brambilla R, and Mancia G. Effects of heart rate changes on arterial distensibility in humans. Hypertension 42: 253–256, 2003.[Abstract/Free Full Text]
  12. Gizdulich P, Imholz BP, van den Meiracker AH, Parati G, and Wesseling KH. Finapres tracking of systolic pressure and baroreflex sensitivity improved by waveform filtering. J Hypertens 14: 243–250, 1996.[Web of Science][Medline]
  13. Hildebrandt W, Schutze H, and Stegemann J. On the reliability of the Penaz cuff during systemic and local fingertip vasodilatation at rest and in exercise. Eur J Appl Physiol Occup Physiol 62: 175–179, 1991.[CrossRef][Web of Science][Medline]
  14. Hoeks AP, Meinders JM, and Dammers R. Applicability and benefit of arterial transfer functions. J Hypertens 21: 1241–1243, 2003.[CrossRef][Web of Science][Medline]
  15. Hope SA, Meredith IT, and Cameron JD. Effect of non-invasive calibration of radial waveforms on error in transfer-function-derived central aortic waveform characteristics. Clin Sci (Lond) 107: 205–211, 2004.[Medline]
  16. Hope SA, Tay DB, Meredith IT, and Cameron JD. Comparison of generalized and gender-specific transfer functions for the derivation of aortic waveforms. Am J Physiol Heart Circ Physiol 283: H1150–H1156, 2002.[Abstract/Free Full Text]
  17. Hope SA, Tay DB, Meredith IT, and Cameron JD. Use of arterial transfer functions for the derivation of aortic waveform characteristics. J Hypertens 21: 1299–1305, 2003.[CrossRef][Web of Science][Medline]
  18. Idema RN, van den Meiracker AH, Imholz BP, Man in't Veld AJ, Settels JJ, Ritsema van Eck HJ, and Schalekamp MA. Comparison of Finapres non-invasive beat-to-beat finger blood pressure with intrabrachial artery pressure during and after bicycle ergometry. J Hypertens Suppl 7: S58–S59, 1989.[Medline]
  19. Imholz BP, Parati G, Mancia G, and Wesseling KH. Effects of graded vasoconstriction upon the measurement of finger arterial pressure. J Hypertens 10: 979–984, 1992.[Web of Science][Medline]
  20. Karamanoglu M and Feneley MP. On-line synthesis of the human ascending aortic pressure pulse from the finger pulse. Hypertension 30: 1416–1424, 1997.[Abstract/Free Full Text]
  21. Karamanoglu M, Gallagher DE, Avolio AP, and O'Rourke MF. Pressure wave propagation in a multibranched model of the human upper limb. Am J Physiol Heart Circ Physiol 269: H1363–H1369, 1995.[Abstract/Free Full Text]
  22. Lantelme P, Mestre C, Lievre M, Gressard A, and Milon H. Heart rate: an important confounder of pulse wave velocity assessment. Hypertension 39: 1083–1087, 2002.[Abstract/Free Full Text]
  23. Lehmann ED. Regarding the accuracy of generalized transfer functions for estimating central aortic blood pressure when calibrated non-invasively. J Hypertens 18: 347–350, 2000.[CrossRef][Web of Science][Medline]
  24. Lehmann ED. Where is the evidence that radial artery tonometry can be used with a generalised transfer function to accurately and non-invasively predict central aortic blood pressure? J Hum Hypertens 15: 145–146, 2001.[CrossRef][Web of Science][Medline]
  25. Ljung L. System Identification: Theory for the User. Upper Saddle River, NJ: Prentice Hall, 1987.
  26. Millasseau SC, Patel SJ, Redwood SR, Ritter JM, and Chowienczyk PJ. Pressure wave reflection assessed from the peripheral pulse: is a transfer function necessary? Hypertension 41: 1016–1020, 2003.[Abstract/Free Full Text]
  27. Nichols WW and O'Rourke MF. McDonalds Blood Flow in Arteries, Theoretic, Experimental and Clinical Principles. London: Arnold, 2005.
  28. Nijboer JA, Dorlas JC, and Lubbers J. The difference in blood pressure between upper arm and finger during physical exercise. Clin Physiol 8: 501–510, 1988.[Web of Science][Medline]
  29. O'Rourke MF. Estimation of central aortic pressure by SphygmoCor requires accurate peripheral pressure measurement. Clin Sci (Lond) 106: 434–437, 2004.[Medline]
  30. O'Rourke MF and Hayward CS. Arterial stiffness, gender and heart rate. J Hypertens 21: 487–490, 2003.[CrossRef][Web of Science][Medline]
  31. Pauca AL, O'Rourke MF, and Kon ND. Prospective evaluation of a method for estimating ascending aortic pressure from the radial artery pressure waveform. Hypertension 38: 932–937, 2001.[Abstract/Free Full Text]
  32. Penáz J. Photoelectric measurement of blood pressure, volume and flow in the finger (Abstract). Digest of the International Conference on Medicine and Biological Engineering 10: 104, 1973.
  33. Piepoli M, Coats AJ, Adamopoulos S, Bernardi L, Feng YH, Conway J, and Sleight P. Persistent peripheral vasodilation and sympathetic activity in hypotension after maximal exercise. J Appl Physiol 75: 1807–1814, 1993.[Abstract/Free Full Text]
  34. Piepoli M, Isea JE, Pannarale G, Adamopoulos S, Sleight P, and Coats AJ. Load dependence of changes in forearm and peripheral vascular resistance after acute leg exercise in man. J Physiol 478: 357–362, 1994.[Abstract/Free Full Text]
  35. Rowell LB, Brengelmann GL, Blackmon JR, Bruce RA, and Murray JA. Disparities between aortic and peripheral pulse pressures induced by upright exercise and vasomotor changes in man. Circulation 37: 954–964, 1968.[Abstract/Free Full Text]
  36. Segers P, Carlier S, Pasquet A, Rabben SI, Hellevik LR, Remme E, de Backer T, De Sutter J, Thomas JD, and Verdonck P. Individualizing the aorto-radial pressure transfer function: feasibility of a model-based approach. Am J Physiol Heart Circ Physiol 279: H542–H549, 2000.[Abstract/Free Full Text]
  37. Sharman JE, Lim R, Franco J, Philips KP, Padgett MJ, Garrahy PJ, and Marwick TH. Validation of the Shygmocor generalised transfer function during exercise (Abstract). Cardiac Society of Australia and New Zealand Annual Meeting 14S: S46, 2005.
  38. Silke B, Spiers JP, Boyd S, Graham E, McParland G, and Scott ME. Evaluation of non-invasive blood pressure measurement by the Finapres method at rest and during dynamic exercise in subjects with cardiovascular insufficiency. Clin Auton Res 4: 49–56, 1994.[CrossRef][Medline]
  39. Soderstrom S, Nyberg G, O'Rourke MF, Sellgren J, and Ponten J. Can a clinically useful aortic pressure wave be derived from a radial pressure wave? Br J Anaesth 88: 481–488, 2002.[Abstract/Free Full Text]
  40. Stergiopulos N, Westerhof BE, and Westerhof N. Physical basis of pressure transfer from periphery to aorta: a model-based study. Am J Physiol Heart Circ Physiol 274: H1386–H1392, 1998.[Abstract/Free Full Text]
  41. Sugimachi M, Shishido T, Miyatake K, and Sunagawa K. A new model-based method of reconstructing central aortic pressure from peripheral arterial pressure. Jpn J Physiol 51: 217–222, 2001.[CrossRef][Web of Science][Medline]
  42. Tanaka H and Thulesius O. Effect of temperature on finger artery pressure evaluated by volume clamp technique. Clin Physiol 13: 535–545, 1993.[Web of Science][Medline]
  43. Toska K and Eriksen M. Peripheral vasoconstriction shortly after onset of moderate exercise in humans. J Appl Physiol 77: 1519–1525, 1994.[Abstract/Free Full Text]
  44. Wesseling KH, de Wit B, van der Hoeven GMA, van Goudoever J, and Settels JJ. Physiocal, calibrating finger vascular physiology for Finapres. Homeostasis 36: 76–82, 1995.
  45. Westerhof BE. Blood Pressure Analysis on Time Scales From Seconds to Days (Dissertation). Amsterdam: University of Amsterdam, 2005.
  46. Wilkinson IB, Mohammad NH, Tyrrell S, Hall IR, Webb DJ, Paul VE, Levy T, and Cockcroft JR. Heart rate dependency of pulse pressure amplification and arterial stiffness. Am J Hypertens 15: 24–30, 2002.[CrossRef][Web of Science][Medline]



This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
E. A. Dawson, M. A. Black, J. Pybis, N. T. Cable, and D. J. Green
The impact of exercise on derived measures of central pressure and augmentation index obtained from the SphygmoCor device
J Appl Physiol, June 1, 2009; 106(6): 1896 - 1901.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
B. E. Westerhof, I. Guelen, W. J. Stok, H. A. J. Lasance, C. A. P. L. Ascoop, K. H. Wesseling, N. Westerhof, W. J. W. Bos, N. Stergiopulos, and J. A. E. Spaan
Individualization of transfer function in estimation of central aortic pressure from the peripheral pulse is not required in patients at rest
J Appl Physiol, December 1, 2008; 105(6): 1858 - 1863.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
S. Munir, B. Jiang, A. Guilcher, S. Brett, S. Redwood, M. Marber, and P. Chowienczyk
Exercise reduces arterial pressure augmentation through vasodilation of muscular arteries in humans
Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1645 - H1650.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
101/4/1207    most recent
00876.2005v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stok, W. J.
Right arrow Articles by Karemaker, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stok, W. J.
Right arrow Articles by Karemaker, J. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.