Abstract
Aims
Continuous noninvasive blood pressure measurement by photoplethysmography has been regularly used in the experimental paradigm of angiotensin challenges, applied to the phase I clinical testing of angiotensin-converting enzyme inhibitors and angiotensin receptor antagonists. This work aims to evaluate the performance of this measurement method, in terms of reliability, reproducibility and dependence on technical settings.
Methods
Data have been gathered from 13 clinical studies on antihypertensive drugs, using the Finapres® device for measuring the response to exogenous angiotensin challenges. The agreement between simultaneous recordings at different fingers and the influence of the reading method are assessed. A literature review addresses the question of the concordance between results obtained noninvasively and through arterial cannulation.
Results
The relative precision of blood pressure monitoring by photoplethysmography allows reproducible determination of angiotensin-induced blood pressure peaks (agreement limits for systolic and diastolic peaks: 12 and 6 mmHg respectively). The reading method influences the results to a similar extent. As compared with blood pressure measured intra-arterially, the difference is usually within limits of clinical acceptability.
Conclusions
In the context of phase I studies using the angiotensin challenges methodology, the reliability and reproducibility of noninvasive blood pressure measurement appear satisfactory, despite the technical limitations of this method. The impact of selected changes in the settings and reading methods is limited.
Keywords: angiotensin, blood pressure, pharmacometrics, photoplethysmography
Introduction
During the past two decades, the early clinical development of cardiovascular drugs acting on the renin-angiotensin system (RAS) has increasingly relied on effect measurements in healthy volunteers [1, 2]. For such drugs, phase I studies are expected to provide accurate information about dose–response relationships, effect duration and pharmacokinetic-pharmacodynamic profile, in addition to safety and tolerability. The quality of early clinical testing is of utmost importance for a sound choice of doses and schedules to be applied in phase II-III trials.
The study paradigm of repeated challenges with exogenous angiotensin (Ang) has gained wide acceptance for determining a drug’s ability to block the pressor effect of this vasoactive peptide in vivo [3, 4]. Ang I challenges have been used to quantify the inhibition of angiotensin-converting enzyme (ACE), and Ang II challenges to assess the AT1 angiotensin receptor blockade induced by specific antagonists. Such studies require an accurate measurement method for continuous blood pressure recording over the minutes preceding and following the challenges. In most instances, and in particular with healthy subjects, there is little need or justification to use invasive methods for continuous blood pressure recording [5]. Arterial cannulation is associated with significant discomfort for the subjects and risks of complications [6, 7]. With the emergence of the first prototype photoplethysmographic arterial pressure monitor in 1982 (Finapres®, Ohmeda, Englewood CO), beat-to-beat blood pressure could be measured noninvasively and continuously at the finger [4]. The measures obtained noninvasively with the Finapres® device have been claimed to reflect closely those determined simultaneously through the use of an intra-arterial catheter [8]. For these reasons, during recent years, the pharmacodynamic profile of ACE inhibitors and Ang II receptor antagonists have been mainly assessed with this device. A similar methodology will probably continue to be used for the development of future drugs targeted towards the RAS. Therefore, the precision, reproducibility and predictive value to be expected with this experimental paradigm deserves critical evaluation. It is to be mentioned that the Finapres® apparatus is no longer commercially available, but another noninvasive continuous blood pressure recorder, based on applanation tonometry, has been recently developed and seems to fulfil the same promises (Pilot®, Colin Medical Instruments, San Antonio TX).
This work aims to assess the validity of non-invasive blood pressure recording in measuring the effect of angiotensin boluses, as used in phase I trials of drugs acting on the RAS. The companion paper [9] addresses the pharmacodynamics of angiotensin and the relevance of this study paradigm for clinical drug development. The objectives of this work are:
to review the published evidence on the reliability of noninvasive beat-to-beat blood pressure monitoring at the finger with the Finapres® (by reference to intra-arterial recordings)
to establish the reproducibility of noninvasive blood pressure monitoring by examining the agreement between simultaneous recordings at two different fingers
to evaluate the influence of the reading methods applied to the recordings to derive a summary measure of the peak response to the angiotensin challenge.
Methods
The data were gathered from 13 phase I clinical studies performed in our centre, involving nine Ang II receptor antagonists (losartan, tasosartan, candesartan cilexadil, TAK-536, SC-52458, L-159,282, LRB-081, UP-269–6, CS-866), one ACE inhibitor (CS-622) and one dual ACE-NEP inhibitor (MDL-100 240). Preliminary angiotensin dose-findings were performed in each subject at their entry in the study, to determine the dose of Ang I or Ang II needed to increase blood pressure by approximately 30 mmHg. The same dose was to be used throughout the trial in that subject. The blocking effect profile of a single dose of the Ang II antagonist or the ACE inhibitor studied was then evaluated from the response to serial boluses of angiotensin. Blood pressure was recorded continuously and noninvasively with the Finapres® device, from 7 min before up to 7 min after each angiotensin injection. This apparatus consists of a servo-photoplethysmomanometer enabling blood pressure to be recorded beat-to-beat at the finger by transdigital photometry. The technique uses the principle of the unloaded artery, the cuff pressure around the finger being dynamically adjusted by a servo-system to equal intra-arterial pressure, leading to annulation of the blood flow pulses [4]. Data acquisition of blood pressure values at the finger is transmitted either as an analogic signal to a paper recorder or as a digital signal to a computer. The reading of blood pressure is performed manually or through a software specifically designed for that purpose (Figure 1). Both reading methods aim at determining the height of the blood pressure peak induced by angiotensin, which occurs about 3 min after injection. While manual reading relies on eyeballing with pencil and ruler, computer reading formally calculates the difference between the minimum of a 3 min window moving average (included in a larger interval of 0.5–6.5 min) determined on baseline blood pressure and the maximum of a 30 s window moving average determined in a standard time interval (usually 0.25–3.25 min) following the angiotensin bolus. Beat-to-beat pressure data are generated by the electronics of the Finapres® and transferred to a personal computer through a serial interface to perform the calculations, using a home-made software (available from the author on request). The response to the angiotensin challenge is thus expressed as the height of the blood pressure peak induced:
 |
Figure 1.
Example of beat-to-beat systolic (SBP), mean and diastolic (DBP) blood pressure (shadowed area) and heart rate (HR) curves recorded by photoplethysmography (Finapres® with computer data acquisition) during an angiotensin bolus challenge (arrow). The graph indicates how the preinjection and postinjection levels are compared with assess the effect of angiotensin in terms of SBP, DBP and HR peaks.
This applies to systolic, diastolic and mean blood pressure (BP) values as well as to heart rate (for which the postinjection nadir is compared with the preinjection level). The blocking action of an ACE inhibitor or Ang II antagonist is reflected in the decrease of the peak height observed at various times after a given dose of Ang I, respectively, Ang II. The results can be expressed as a percentage of inhibition:
 |
With this approach, even in the absence of effective blockade, some variation in the response to angiotensin boluses provides both moderately positive and negative inhibition values (usually less than±30%). On the other hand, even in the presence of complete blockade, small pressure peaks may be identified after an angiotensin challenge as a result of physiological fluctuations, leading to inhibition values infrequently reaching 100%.
Data from several preliminary angiotensin dose-findings were used to assess the performance of this method. A literature review completes the validation.
Reliability of non-invasive blood pressure recording
This is assessed by reviewing the published evidence validating different Finapres®prototypes (Finapres 2300 and Finapres 2300e) by comparison with simultaneous intra-arterial measurements. Intra-arterial blood pressure recordings are considered to have very small errors and can be therefore treated in most instances as the method of reference [7]. The criteria to assess the reliability of the noninvasive measurement by comparison with the reference method are the accuracy and the precision for systolic and diastolic blood pressure over a series of simultaneous measurements with both techniques. The accuracy is quantified as the bias, that is the average difference (Δ¯) between both recordings. The precision is assessed by the spread of differences between photoplethysmographic and intra-arterial values, expressed as standard deviation (s.d.Δ); the limits of agreement between the two methods can be evaluated as the 95% confidence interval for the expected differences (CI95%Δ=Δ¯±2 s.d.Δ) [10].
Reproducibility of non-invasive blood pressure monitoring
The agreement between blood pressure values recorded simultaneously at two distant sites (left and right hands) by Finapres® were assessed using data collected during a study on a dual ACE-NEP inhibitor (MDL-100 240). Two cuffs were placed on the middle finger of the right and left hands of 13 healthy subjects and blood pressure was recorded at both sites during 7 min before and 7 min after an injection of Ang I. No preliminary standardization was made before the recordings. Both raw beat-to-beat blood pressure values and peak effect of Ang I (determined by computer reading) were considered for systolic, diastolic, and mean blood pressure. The left-right comparison of beat-to-beat blood pressure recording was performed within each subject by considering the bias (accuracy, Δ¯) and standard deviation of the differences (precision, s.d.Δ) [10, 11]. The results are summarized as the median and range of absolute values observed for those indicators over the 13 volunteers. The reproducibility of the determination of angiotensin-induced blood pressure peak height, as measured on both hands, was evaluated by plotting the pairs of corresponding values along the identity line and their absolute differences along their averages, and by calculating the within-subject standard deviation (s.d.r), from which a 95% confidence interval for the difference between two determinations was derived (limit of reproducibility, CI95% =2·√2·s.d.r) [12]. All computations were done with the Microsoft Excel® software (version 7.0a).
Influence of the reading method
The concordance between the manual and the computer reading of angiotensin-induced peaks was assessed from 544 pairs of data obtained during a study on an ACE-NEP inhibitor (MDL-100 240). The peaks of systolic and diastolic blood pressure and heart rate responses obtained with both reading methods were graphed and compared by evaluating the bias, the standard deviation of the differences and the agreement limits [10].
Results
Reliability of non-invasive blood pressure recording
The agreement observed in several published studies between beat-to-beat photoplethysmographic measurement of blood pressure and intra-arterial catheterization is summarized in Table 1. Most of the 15 identified studies report a fairly good correlation between beat-to-beat changes in systolic and diastolic blood pressure recorded at the finger by different Finapres® devices and simultaneous intra-arterial monitoring, with correlation coefficients higher than 0.93 for diastolic and systolic blood pressure [8, 13, 14]. However, the average group discrepancies at rest and during anaesthesia indicate an over-or under-reading by 5–8 mmHg diastolic and mean blood pressure [4, 6–8, 13–18], exceeding 10 mmHg in one study [19]; the discrepancies are in general greater for systolic blood pressure values. The corresponding standard deviations are variable, ranging from 3 mmHg up to 20 mmHg. Similar findings are reported during various tests (Valsalva manoeuvre, cold pressor, leg raising, injection of phenylephrine or nitroglycerine, exercise or orthostatic stress). Several investigators [7, 20–25] mention the fact that Finapres® does not always reflect arterial pressure faithfully. Some authors are also concerned by the unpredictability and duration of progressive drifts in blood pressure measurement [22]. They even report differences greater than ±20 mmHg, which they consider unacceptable for the clinical monitoring of patients.
Table 1.
Agreement between noninvasive blood pressure recording by photoplethysmography and intra-arterial blood pressure monitoring. SBP: systolic blood pressure; DBP: diastolic blood pressure; Δ¯: average difference (bias); s.d.Δ: standard deviation of differences (precision).
The large interindividual variability may partly explain the wide range of biases (underestimation of intra-arterial pressure in some patients and overestimation in others). It may reflect both technical, physiological and pathological influences (disease states affecting peripheral blood flow, presence of a pressure gradient) [6, 15, 26]. A study [6] compared the accuracy between different cuff sizes, cuff positions and choice of finger and showed that the thumb gives the most accurate results (correlation coefficients of 0.95 for systolic and 0.88 for diastolic blood pressure vs values obtained invasively). The correlation coefficients with the other fingers ranges from 0.50 to 0.92 for systolic and 0.76–0.93 for diastolic blood pressure.
Beat-to-beat continuous recording by photoplethysmography has also been compared with other non-invasive methods, such as office sphygmomanometry and 24 h ambulatory blood pressure monitoring [27]. A lack of agreement for systolic and diastolic values (±36 mmHg for office sphygmomanometry and ±40 mmHg for ambulatory monitoring) revealed that the two values, though related, were far from being equivalent.
Continuous non-invasive blood pressure measurement by radial artery aplanation tonometry has also been validated against the invasive reference method, and found to have non-negligible bias (average −5.8 mmHg for systolic, and 7.2 mmHg for diastolic BP) and imprecision (15.2 and 10.9 mmHg, respectively) [28], while good agreement was found with blood pressure measured by brachial sphygmomanometry [29]. We have not identified a study comparing these two non-invasive methods directly.
Reproducibility of non-invasive blood pressure monitoring
The agreement between raw beat-to-beat blood pressure data measured on the left and right fingers during 13 challenges are summarized in Table 2. The correlation between simultaneous measurements at two different fingers is good (median 0.91 for systolic, 0.95 for diastolic blood pressure). However, blood pressure values frequently deviate from the identity line by a constant value (Figure 2), which leads to low accuracy despite good precision. This systematic bias can be as large as 25–30 mmHg.
Table 2.
Concordance between left and right beat-to-beat measurements of blood pressure by photoplethysmography. Median (range) over 13 subjects. SBP: systolic blood pressure; DBP: diastolic blood pressure; HR: heart rate; |Δ¯|: absolute average difference (bias); s.d.Δ: standard deviation of differences (precision).
Figure 2.
Comparison between simultaneous beat-to-beat diastolic blood pressure (DBP) measured by finger photoplethysmography at the right and left hands of one representative subject. The raw data from both sides are plotted along the identity line (top), and the differences are plotted along the average values (bottom, average±2 s.d. indicated by the continuous and dashed lines respectively).
The left-right comparison of the measurements of peak blood pressure response to angiotensin challenges, evaluated over the 13 pairs of determinations, is satisfactory: the within-subject standard deviations are 4.3 mmHg for systolic, 2.3 mmHg for diastolic blood pressure, and 0.22 beats min−1 for heart rate (Figure 3). This means for the recording of angiotensin-induced peaks a reproducibility (CI95%) of 12 mmHg, 6.6 mmHg and 0.63 beats min−1 for systolic and diastolic blood pressure and heart rate, respectively. This indicates that the frequent occurrence of a systematic difference between measurements obtained at two distant sites has only a limited influence on the peak height determination, which is obtained after subtraction of the baseline.
Figure 3.
Comparison between peak diastolic blood pressure responses (DBP peak) evaluated simultaneously at the right and left hand by finger photoplethysmography in 13 subjects. The raw data from both sides are plotted along the identity line (top), and the absolute difference is plotted along the average (bottom, upper 2 s.d. indicated by the dashed line).
Influence of the reading method
The comparison of peak blood pressure values provided by the computer software vs the manual readings over 544 angiotensin-induced peaks is shown in Table 3. The agreement between both methods is good (Figure 4). Although statistically significant (P<0.001), the overestimation by computer reading is small (0.5 mmHg), and is not expected to have any clinical relevance on study results. The spread of individual differences (CI95%Δ=−5.5, 6.5 mmHg for systolic, −3, 7.4 mmHg for diastolic blood pressure) covers about 25% of the range of measured values.
Table 3.
Comparison of 544 manual and computer readings of angiotensin-induced blood pressure peaks measured by photoplethysmography. SBP: systolic blood pressure; DBP: diastolic blood pressure; HR: heart rate; Δ¯: average difference (bias); s.d.Δ: standard deviation of differences (precision).
Figure 4.
Comparison between the computer and the manual reading of the diastolic blood pressure increase (DBP peak) induced by angiotensin injections and recorded by photoplethysmography during 544 Ang I and II challenges during one study. The raw readings by both methods are plotted along the identity line (top), and the differences are plotted along the average values (bottom, average±2 s.d. indicated by the continuous and dashed lines respcetively).
Discussion
Angiotensin challenges with continuous noninvasive blood pressure measurement by photoplethysmography have been regularly used in the early clinical evaluation of drugs acting on the RAS. For this type of study to provide consistent results requires that the noninvasive measurement method reliably reflects blood pressure changes, with satisfactory reproducibility, even if different settings and reading methods are applied. Furthermore, angiotensin boluses must induce reproducible, preferably dose-dependent peaks, without much variability over daytime. The evaluation of the internal validity of this study paradigm is addressed in this and in the companion paper, which also introduces some statements about the external validity of phase I studies based on this method [9].
First, as compared with blood pressure measured intra-arterially, several studies reviewed here indicate that the differences meet the requirements for blood pressure-measuring devices of the Association for the Advancement of Medical Instrumentation [30]. Some authors emphasize that both the physiological variability and the artefacts affecting photoplethysmographic measurements preclude the use of this method for the monitoring of patient in whom continuous accurate blood pressure measurements are essential. In our hands too, the blood pressure values reported by the Finapres® device not infrequently differ from the sphygmomanometer determination by 20 mmHg or more. But for investigational studies on cardiovascular agents, blood pressure values recorded over short periods of time by photoplethysmography provide an acceptable degree of precision and reproducibility to evaluate the acute effect of angiotensin boluses. A stable difference with a reference method, or a slow, progressive drift of a few millimetres of mercury are of little or no clinical significance in this context.
The peak heights measured after an angiotensin bolus appear relatively independent of the site of the cuff application and of the position of the hand, despite the marked influence of these settings on absolute blood pressure values. This is in accordance with the good relative accuracy of the beat-to-beat measurements (absolute differences cancel out in the determination of peak height). The reproducibility of peak height determination, using two separate devices, is better when considering the diastolic than the systolic blood pressure (6.6 mmHg vs 12 mmHg, respectively). These numbers must be interpreted in relation to the target peak effect of 30 mmHg set for angiotensin injections, and the physiological variability of angiotensin-induced peaks (see companion paper [9]). Blood pressure represents a highly fluctuating physiological variable, it is affected by numerous regulations, and investigators experienced in continuous recording know that a sudden noise in the room is sufficient to induce a 20–30 mmHg peak over several seconds. The reproducibility reported here for peak height measurements can thus be considered satisfactory.
As compared with the manual reading of the peak height, the computer reading with a dedicated software appears valuable, time-sparing, user-independent and convenient. The differences between both methods are more marked for diastolic blood pressure peaks. They result conceivably from differences in the strategies for peak identification and averaging between the investigator working with pencil and ruler and the computer algorithm. Some peculiarities in certain recorded curves indicate that a minority of readings display a difference larger than 10 mmHg. These few cases have been reviewed individually, and have revealed either a clearly erroneous manual reading, or a suspected artefact superimposed on the peak, which had been discarded by the investigator but taken into account by the computer. Thus, there is no definite argument for recommending either method, but the methods are not interchangeable during a study.
In conclusion, the reliability and reproducibility of the non-invasive measurement of blood pressure changes induced by angiotensin injections appear satisfactory for clinical studies based on such challenges, despite the technical limitations inherent to this method. The impact of selected changes in the settings and reading methods is limited.
Acknowledgments
This study was supported in part by the Swiss National Fund for Scientific Research (grant no. 3200–043264.95/1).
References
- 1.Csajka C, Buclin T, Brunner HR, Biollaz J. Pharmacokinetic-pharmacodynamic profile of angiotensin II receptor antagonists. Clin Pharmacokinet. 1997;32:1–29. doi: 10.2165/00003088-199732010-00001. [DOI] [PubMed] [Google Scholar]
- 2.Brunner HR, Waeber B, Nussberger J. Does pharmacological profiling of a new drug in normotensive volunteers provide a useful guideline to antihypertensive therapy? Hypertension. 1983;5:101–107. doi: 10.1161/01.hyp.5.5_pt_2.iii101. [DOI] [PubMed] [Google Scholar]
- 3.Brunner RH, Christen Y, Munafo A, Lee RJ, Waeber B, Nussberger J. Clinical Experience with Angiotensin II Receptor Antagonists. Am J Hypertension. 1995;5S:243–246. doi: 10.1093/ajh/5.12.243s. [DOI] [PubMed] [Google Scholar]
- 4.Christen Y, Waeber B, Nussberger J, Brunner HR. Noninvasive blood pressure monitoring at the finger for studying short lasting pressor responses in man. J Clin Pharmacol. 1990;30:711–714. doi: 10.1002/j.1552-4604.1990.tb03631.x. [DOI] [PubMed] [Google Scholar]
- 5.De Mey C, Schroeter V, Butzer R, Roll S, Belz GG. Method specificity of non-invasive blood pressure measurement: oscillometry and finger pulse pressure vs acoustic methods. Br J Clin Pharmacol. 1995;40:291–297. doi: 10.1111/j.1365-2125.1995.tb04549.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kurki T, Smith NT, Head N, December-Silver H, Quinn A. Noninvasive continuous blood pressure measurement from the finger: otpimal measurement conditions and factors affecting reliability. J Clin Monit. 1987;3:6–13. doi: 10.1007/BF00770876. [DOI] [PubMed] [Google Scholar]
- 7.Jones RDM, Brown AG, Roulson CJ, Smith ID, Chan SC. The upgraded Finapres 2300e: a clinical evaluation of a continuous noninvasive blood pressure monitor. Anaesthesia. 1992;47:701–705. doi: 10.1111/j.1365-2044.1992.tb02396.x. [DOI] [PubMed] [Google Scholar]
- 8.Gabriel A, Lindblad LE, Angleryd C. Non-invasive vs invasive beat-to-beat monitoring of blood pressure. Clin Physiol. 1992;12:229–235. doi: 10.1111/j.1475-097x.1992.tb00309.x. [DOI] [PubMed] [Google Scholar]
- 9.Buchwalder-Csajka C, Buclin T, Brunner HR, Biollaz J. Evaluation of the angiotensin challenge methodology for assessing the pharmacodynamic profile of antihypertensive drugs acting on the renin-angiotensin system. Br J Clin Pharmacol. 1999;48:594–604. doi: 10.1046/j.1365-2125.1999.00050.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;i:307–310. [PubMed] [Google Scholar]
- 11.Chinn S. The assessment of methods of measurement. Stat Med. 1990;9:351–362. doi: 10.1002/sim.4780090402. [DOI] [PubMed] [Google Scholar]
- 12.Altman DG, Bland JM. Measurement in medicine: the analysis of method comparison studies. Statistician. 1983;32:307–317. [Google Scholar]
- 13.Parati G, Casadei R, Groppelli A, Di Rienzo M, Mancia G. Comparison of finger and intra-arterial blood pressure monitoring a rest and during laboratory testing. Hypertension. 1989;13:647–655. doi: 10.1161/01.hyp.13.6.647. [DOI] [PubMed] [Google Scholar]
- 14.Smith NT, Wesseling KH, de Wit B. Evaluation of two prototype devices producing noninvasive, pulsatile, calibrated blood pressure measurement from a finger. J Clin Monit. 1985;1:17–29. doi: 10.1007/BF02832685. [DOI] [PubMed] [Google Scholar]
- 15.Bos WJB, van Goudoever J, van Montfrans GA, Van den Meiracker AH, Wesseling KH. Reconstruction of brachial artery pressure from noninvasive finger pressure measurements. Circulation. 1996;94:1870–1875. doi: 10.1161/01.cir.94.8.1870. [DOI] [PubMed] [Google Scholar]
- 16.Molhoek GP, Wesseling KH, Settels JM, et al. Evaluation of the Penàz servo-plethysmo-manometer for the continuous, non-invasive measurement of finger blood pressure. Basic Res Cardiol. 1984;79:598–609. doi: 10.1007/BF01910489. [DOI] [PubMed] [Google Scholar]
- 17.Imholz BPM, Settels JJ, van der Meiracker AH, Wesseling KH, Wieling W. Non-invasive continuous finger blood pressure measurement during orthostatic stress compared to intra-arterial pressure. Cardiovasc Res. 1990;24:214–221. doi: 10.1093/cvr/24.3.214. [DOI] [PubMed] [Google Scholar]
- 18.Pace NL, East TD. Simultaneous comparison of intraarterial, oscillometric, and Finapres monitoring during anesthesia. Anesth Analg. 1991;73:213–220. doi: 10.1213/00000539-199108000-00017. [DOI] [PubMed] [Google Scholar]
- 19.Rongen GA, Bos WJW, Lenders JWM, et al. Comparison of intrabrachial and finger blood pressure in healthy elderly volunteers. Am J Hypertens. 1995;8:237–248. doi: 10.1016/0895-7061(94)00000-2. [DOI] [PubMed] [Google Scholar]
- 20.Stokes DN, Clutton-brock T, Patil C, Thompson JM, Hutton P. Comparison of invasive and non-invasive measurement of continuous arterial pressure using the finapres. Br J Anaesth. 1991;67:26–35. doi: 10.1093/bja/67.1.26. [DOI] [PubMed] [Google Scholar]
- 21.Lindqvist A. Beat-to-beat agreement of non-invasive finger artery and invasive radial artery blood pressure in hypertensive patients taking cardiovascular medication. Clin Physiol. 1995;15:219–229. doi: 10.1111/j.1475-097x.1995.tb00513.x. [DOI] [PubMed] [Google Scholar]
- 22.Gibbs MN, Faracs BS, Larach DR, Derr JA. The accuracy of Finapres noninvasive mean arterial pressure measurements in anesthetized patients. Anesthesiology. 1991;74:647–652. doi: 10.1097/00000542-199104000-00004. [DOI] [PubMed] [Google Scholar]
- 23.Kermode JL, Davis NJ, Thompson WR. Comparison of the Finapres blood pressure monitor with intra-arterial manometry during induction of anaesthesia. Anaesth Intensive Care. 1989;17:470–486. doi: 10.1177/0310057X8901700413. [DOI] [PubMed] [Google Scholar]
- 24.Van Egmond J, Hasenbos M, Crul JF. Invasive vs non-invasive measurement of arterial pressure. Br J Anaesth. 1985;57:434–444. doi: 10.1093/bja/57.4.434. [DOI] [PubMed] [Google Scholar]
- 25.Aitken HA, Todd JG, Kenny GNC. Comparison of the Fianapres and direct arterial pressure monitoring during profound hypotensive anaesthesia. Br J Anaesth. 1991;67:36–40. doi: 10.1093/bja/67.1.36. [DOI] [PubMed] [Google Scholar]
- 26.Jones RDM, Kornberg JP, Roulson CJ, Visram AR, Irwin MG. The Finapres 2300e finger cuff. Anaesthesia. 1993;48:611–615. doi: 10.1111/j.1365-2044.1993.tb07129.x. [DOI] [PubMed] [Google Scholar]
- 27.Musso NR, Giacchè M, Galbariggi G, Vergassola C. Blood pressure evaluation by noninvasive and traditional methods. Am J Hypertens. 1996;9:293–299. doi: 10.1016/0895-7061(95)00354-1. [DOI] [PubMed] [Google Scholar]
- 28.Weiss BM, Spahn DR, Rahmig H, Rohling R, Pasch T. Radial artery tonometry: moderately accurate but unpredictable technique of continuous non-invasive arterial pressure measurement. Br J Anaesth. 1996;76:405–411. doi: 10.1093/bja/76.3.405. [DOI] [PubMed] [Google Scholar]
- 29.Cameron JD, McGrath BP, Dart AM. Use of radial artery applanation tonometry and a generalized transfer function to determine aortic pressure augmentation in subjects with treated hypertension. J Am Coll Cardiol. 1998;32:1214–1220. doi: 10.1016/s0735-1097(98)00411-2. [DOI] [PubMed] [Google Scholar]
- 30.Association for the Advancement of Medical Instrumentation. A.A.M.I. Standards. Arlington: 1980. American National Standard for electronic or automated sphygmomanometers; pp. 29–42. [Google Scholar]