Professional Documents
Culture Documents
605–616
Review
Fifteen years experience with finger arterial pressure monitoring:
assessment of the technology
a,)
Ben P.M. Imholz , Wouter Wieling a , Gert A. van Montfrans a , Karel H. Wesseling b
a
Department of Internal Medicine, F4-222, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, Netherlands
b
TNO-Biomedical Instrumentation, Academic Medical Center, Amsterdam, Netherlands
Abstract
We review the Finapres technology, embodied in several TNO-prototypes and in the Ohmeda 2300 and 2300e Finapres NIBP.
Finapres is an acronym for FINger Arterial PRESsure, the device delivers a continuous finger arterial pressure waveform. Many papers
report on the accuracy of the device in comparison with intra-arterial or with noninvasive but intermittent blood pressure measurements.
We compiled the results of 43 such papers and found systolic, diastolic and mean accuracies, in this order, ranging from y48 to 30
mmHg, from y20 to 18 mmHg, and from y13 to 25 mmHg. Weighted for the number of subjects included pooled accuracies were
y0.8 ŽSD 11.9., y1.6 Ž8.3. and y1.6 Ž7.6. mmHg respectively. Subdividing the pooled group according to criteria such as reference
blood pressure, place of application, and prototype or commercial device we found no significant differences in mean differences or SD.
Measurement at the finger allows uninterrupted recordings of long duration. The transmission of the pressure pulse along the arm arteries,
however, causes distortion of the pulse waveform and depression of the mean blood pressure level. These effects can be reduced by
appropriate filtering, and upper arm ‘return-to-flow’ calibration to bring accuracy and precision within AAMI limits. For the assessment
of beat-to-beat changes in blood pressure and assessment of blood pressure variability Finapres proved a reliable alternative for invasive
measurements when mean and diastolic pressures are concerned. Differences in systolic pressure are larger and reach statistical
significance but are not of clinical relevance. Finger arteries are affected by contraction and dilatation in relation to psychological and
physical Žheat, cold, blood loss, orthostasis. stress. Effects of these phenomena are reduced by the built-in Physiocal algorithm. However,
full smooth muscle contraction should be avoided in the awake patient by comforting the patient, and covering the hand. Arterial state can
be monitored by observing the behaviour of the Physiocal algorithm. We conclude that Finapres accuracy and precision usually suffice for
reliable tracking of changes in blood pressure. Diagnostic accuracy may be achieved with future application of corrective measures.
q 1998 Elsevier Science B.V. All rights reserved.
of blood pressure are discussed, and finally we consider ‘Finapres’, enabling the measurement of continuous finger
the future place of finger arterial pressure measurement. arterial pressure in clinical settings w8–10x. They con-
tributed numerous technical adjustments and improve-
ments with respect to the finger cuff, the use of a stabilised
infrared light-emitting diode ŽLED., of a fast bi-directional
2. Finapres principles of operation proportional valve and a high speed servo system of at
least 30 Hz bandwidth w4,7,9x. Most important was the
The method is based on the development of the dy- development of the dynamic servo setpoint adjuster, i.e.
namic Žpulsatile. unloading of the finger arterial walls the search procedure and criterion for the automated deter-
using an inflatable finger cuff with built-in photo-electric mination and periodic adjustment of the arterial unloaded
plethysmograph. A fast pneumatic servo system and a volume ŽPhysiocal procedure. w10x. The start-up procedure
dynamic servo setpoint adjuster assure arterial unloading at is shown in Fig. 1. With standardised increments in cuff
zero transmural pressure and consequent full transmission pressure ŽFig. 1, trace I. pulsations in the plethysmogram
of arterial blood pressure to cuff air pressure w2–4x. The ŽDC. increase to reach a maximum and subsequently
principle of unloading has been described in detail w5–7x, decrease again ŽFig. 1, trace II.. The cuff pressure at
Fig. 1. Finapres start-up and physiocal adjustments. The three simultaneous traces show from above, finger cuff pressure, infrared plethysmogram Žmore
transmitted light plotted upwards., and total amount of infrared light passing through the finger. The initial square wave has a 1 volt or 100 mmHg
calibration. Next, the cuff pressure is increased stepwise and held constant until one beat is detected per step level. The plethysmographic amplitude and
the total infrared increase with increasing cuff pressure since blood is pressed out from under the cuff with consequent reduced absorption of incident
infrared and greater tissue transmission. At Ža. the largest plethysmogram is detected and its corresponding pressure level is set upon completion of the up
staircase. At this pressure, several further plethysmograms are analyzed for wave shape and the servo setpoint adjusted accordingly at Žb.. The servo loop is
then closed and finger pressure is measured. After ten beats the loop is opened and cuff pressure is set at a level half way between systolic and diastolic
and further plethysmograms examined for waveform. At Žc. the loop is closed at a slightly higher setpoint, corresponding to a further reduction of blood
under the cuff and increased transmission. Intervals between Physiocal interruptions are gradually increased until 70 beats elapse, or about 1 min.
B.P.M. Imholz et al.r CardioÕascular Research 38 (1998) 605–616 607
˜´
The practical design of the Penaz-method as imple- finger and more central blood pressures are by and large
mented in the Finaprese and the ambulant version the similar in appearance as shown in Fig. 2.
Portaprese devices, includes an inflatable finger cuff which We review all 43 selected evaluation studies published
comprises an infrared plethysmograph and a small box until March 1997. For each study average accuracies be-
attached to the wrist enclosing a fast servo-controlled tween finger and the reference techniques wFinger-Refer-
pressurising system for the continuous adjustment of cuff encex are taken. When averages are not reported we calcu-
pressure according to changes in the plethysmographic lated them from the individual data. From these data the
output w10–12x. The cuff and wrist-box are connected to a accuracy of finger blood pressure among studies was
main unit which holds the air pump, electronics and a calculated as weighted mean with the total number of
computer. patients as weights w55x. Possible influences of confound-
The performance of Finapres by comparing finger arte- ing factors on the accuracy were identified.
rial pressures with intra-radial, intra-brachial and indirect The weighted accuracy of finger arterial pressure mea-
oscillometric or Riva–Rocci–Korotkoff as reference tech- surement among these studies comprising a total of 1031
niques has been reported in a large number of studies subjects was y0.8 " 11.7 mmHg wrange y48 to 30 mmHgx
w8,11–53x. We address two questions in the following for systolic pressure, y1.6 " 8.5 mmHg wrange y20.1 to
3. Finapres for measurement of blood pressure levels Fig. 3A shows average values per study in the operating
theatre ŽO.Th; middle panels. versus the cardiovascular
laboratory ŽCVL; right panels.. Overall similar discrepan-
3.1. Physiological background cies are seen ŽTable 1.. If from all data only resting
—steady state— values are selected similar scatters are
Arterial waveforms in the finger and more central arter- seen in the operating room and in laboratory conditions
ies differ, since the shape of the pressure wave on its way ŽFig. 3B, Table 1..
to the periphery is changed by reflection and by a pressure The effect of age as an independent variable for the
gradient along the arterial tree w54x. Peripheral reflection difference between finger and invasive pressure was inves-
results in amplification of the pressure wave and increases tigated in an earlier report w56x. The main finding was that
systolic pressure; the amplification is frequency- and thus correlations between differences in finger and invasive
heart rate-dependent w20x. The pressure gradient causes pressures versus age were not significant Ž r s y0.24,
mean and diastolic pressure to be lower w12x. Nevertheless, y0.27 and y0.14 for systolic, mean and diastolic pressure
Fig. 2. Typical intra-arterial and finger arterial pressure registrations during the Valsalva manoeuvre Žleft. and standing-up Žright.. Pressure changes in
intraarterial pressure are followed closely but finger pulse pressures are systematically greater in this subject. ŽFigure is a composition of original tracings
published elsewhere w25,27x..
608 B.P.M. Imholz et al.r CardioÕascular Research 38 (1998) 605–616
differences, respectively. ŽFig. 4, left panel.. In the present 3.3. Methodological factors
review, the 43 studies cover an age range of 5 to 76 years.
In this group, correlations of the average finger-to-invasive Choosing a particular reference certainly bears upon
pressure differences with average age also were not signifi- these comparative studies that are based on intra-radial,
cant with values of y0.15, y0.24 and 0.09 for systolic, intra-brachial and indirect oscillometric or Riva–Rocci–
mean and diastolic pressure differences, respectively ŽFig. Korotkoff measurements. These different groups are shown
4, right panels.. in Fig. 5A. On average, Finapres underestimates mean and
Table 1
Average finger–reference blood pressures ‘accuracy’ and standard deviation ‘precision’ ŽmmHg.
Condition Systolic Mean Diastolic
Accuracy Precision Accuracy Precision Accuracy Precision
All w8,11,23–29,31–41,44,45,47,49–51,53x y0.8 11.9 y1.6 8.3 y1.6 7.6
All-CVL w11,12,14–28,34,35,41,44,45,47,49–51,53x 0.6 12.5 y3.8 8.3 y2.2 7.4
All-O.Th w8,13,29,31–33,36–40,42,43,46,48,52x y5.9 8.4 1.3 7.2 0.7 7.9
Rest w8,11,13–29,31–41,44,45,47,49–51,53x y0.5 8.8 y1.6 7.7 y2.5 6.0
Rest-CVL w11,12,14–28,34,35,41,44,45,47,49,50,53 x 0.6 9.3 y5.6 7.2 y4.1 5.0
Rest-O.Th w8,13,29,31–33,36–40,42,43,46,48,52x y3.0 7.8 2.0 7.3 1.7 6.8
Rest-I.Brach w11,12,14,16,17,20,23–25,27,28,35,44,45x y0.0 8.9 6.2 7.6 y3.9 5.6
Rest-I. Rad w13,15,26,29,31–34,34,36,37,40,42,43,46–49,52x y5.3 6.8 0.5 6.4 1.3 5.4
Rest-RRK w8,18,19,21,22,38,39,41,50–52x 0.8 9.4 y1.6 5.6 y4.9 5.0
Rest-F.Proto w8,37,38x 7.0 3.3 y3.7 5.2 y9.5 4.5
Rest-F.P45 w8,11,16,20,21,24–28,40,44x 1.3 8.5 y7.0 10.7 y2.8 7.1
Rest-F.Ohm w17–20,22,23,29,31–34,39,41–43,45–49,52,53x y1.3 9.0 y0.1 4.1 y2.0 5.1
Values are given as subgroup of the database containing 43 studies, and calculated from average accuracies among studies.
Accuracies are given as weighted means for the number of subjects participating in the study.
CVL s cardiovascular laboratory; O.Th s operating theatre; I.Brach s intra-brachial artery pressure; I.Rad s intra-radial artery pressure; RRK s
auscultatory pressure; F.Proto s use of all kinds of TNO-prototypes; F.P45s use of TNO-prototypes 4 and 5 only; F.Ohm s use of Ohmeda Finapres
device
B.P.M. Imholz et al.r CardioÕascular Research 38 (1998) 605–616 609
diastolic pressures as recorded in the brachial artery, For systolic pressures average differences were not differ-
whereas such differences tend to be positive in the radial ent; no relation between finger-to-reference difference and
artery ŽTable 1; Rest-I.Brach vs. Rest-I.Rad; p s 0.05.. measurement technique was found.
Fig. 5. Panel A: Average differences between finger and reference blood pressure selected per reference technique: Rad refers to intra-radial artery pressure
Ž N s 357.; bra refers to intra-brachial artery pressure Ž N s 202. and RRK refers to indirect auscultatoryroscillometric techniques Ž N s 404.. Panel B:
Average differences between finger and reference blood pressure selected per Finapres apparatus: P4.5 refers to TNO-Finapres prototypes 4 and 5
Ž N s 211., Ohm refers to the use of the commercial Ohmeda 2300 NIBP monitor Ž N s 787..
610 B.P.M. Imholz et al.r CardioÕascular Research 38 (1998) 605–616
Jones et al. investigated the first Ohmeda 2300, a w25,26x.. These differences were usually small as compared
commercially manufactured Finapres device, and the later to the magnitude of the responses ŽTable 2. and therefore
updated version Ž2300 e . and detected a significant im- not of clinical relevance w12,25,27x. As such, the specific
provement of its accuracy w48x. At comparison of the information that is contained within these blood pressure
performance of TNO prototypes and commercial devices responses w65x is always secured w12x. Second, minute-to-
the TNO prototypes show an underestimation for mean minute changes in some conditions ŽTable 2, head-up
and diastolic pressures ŽFig. 5B.. This may be largely tilting ŽHUT. w14–16,28x. show increases in differences for
attributed to the fact that in 8 of the 12 studies intra-brachial all pressures, i.e. systolic, mean and diastolic. Although
artery pressures were used which caused the expected the true origin of these differences is unknown, it has been
underestimation of brachial mean and diastolic pressures attributed to changes in local tissue turgidity w66,67x. Third,
ŽTable 1.. in addition to changes of pulse wave amplification —as a
result of modification of heart rate and left-ventricular
3.3.1. Conclusions ejection time w20x — different responses in systolic refer-
Although the accuracy of finger blood pressure in our ence and finger measurement can occur, such as during the
review showed considerable scatter, the weighted accuracy continuous infusion of phenylephrine and exercise. In such
611
612 B.P.M. Imholz et al.r CardioÕascular Research 38 (1998) 605–616
total of 1031 it was impossible to obtain finger pressure differences in 16 subjects overall differences between fin-
w18,35,40,41,43,51,52x. These were associated with techni- gers were 1.1 mmHg for systolic, 2.4 mmHg for mean and
cal problems in early prototype devices Ž N s 7. w35x, with 3.3 mmHg for diastolic pressure w11x. In contrast in studies
the inability to apply appropriate cuff sizes in the very investigating two-finger differences from simultaneous
young Ž N s 5. w18,51x, with perioperative cold temperature measurements large differences were unusual w17,22,51x.
Ž N s 11. w40,41x, or were unexplained Ž N s 10. w43,52x. Whether or not differences in cuff application are involved
Sometimes the impossibility to measure the plethysmo- may need further clarification.
gram in the finger was due to the combination of vascular
disease and ambient cold temperature Ž N s 3. w41x.
Arteriolar vasoconstriction can interfere with measure- 5.3. Warming of the hand
ment of the plethysmogram w8,17,20,50x. Vasoconstriction
was initiated experimentally by vasoactive agents w17,20x The beneficial effect of heating on finger pressure
or ambient cold temperature in patients known with spastic measurement was investigated systematically by Hilde-
problems of the arteries w8,41,51x. If graded vasoconstric- brandt et al. w22x and Tanaka and Thulesius w50x. Hilde-
tion with phenylephrine is performed measurement contin- brandt showed that during isometric exercise the systolic
According to the user manual the correct selection of 5.4. Assessment of the Finapres signal output
cuff size and application of cuffs is crucial. Just as with the
auscultatory technique, the proper application of finger Barras in 1973 used a micropuncture technique to actu-
cuffs can affect outcome; Jones et al. found a decrease in ally measure intra-arterial digital pressure. He found a
accuracy when cuffs were applied too loose w49x. His study waveform similar to the radial waveform; having a steep
does not mention specific alterations of the finger wave- systolic pressure increase and deep dicrotic notch w70x.
form, but in our experience oscillations are more frequent Another study compared finger artery aplanation tonome-
in case of loose cuffs. try with Finapres at adjacent fingers w4x. The waveforms
appeared similar, suggesting an adequate pneumatic servo
system bandwidth w4,10x.
5.2. Selection of fingers Low gave practical advice for obtaining adequate finger
blood pressures in the AAN newsletter w71x. If the cuff is
According to the Ohmeda users manual middle or annu- snugly applied, Finapres signals were considered accept-
lar fingers are preferred. The basis for this is experimental able when the signal was Ža. large in size, Žb. sharp in
rather than anatomical or physiological. From the time shape having a distinct dicrotic notch, and Žc. within
early prototypes were developed studies in the operation reasonable agreement with brachial ŽRRK. recordings. In
theatre described the use of the thumb using prototype addition the patient had to feel warm, and a temperature
thumb cuffs; however this proposal was never fully devel- controlled muff is used in order to warm actively the
oped to reach success w29,39,40x. fingers at 428C. These suggestions largely agree with our
Finapres users sometimes switch the site of finger mea- own experiences w72x. However, the major criteria for
surement until finger pressure values correspond close accepting the Finapres output should be that it is consid-
enough to auscultatory reference values w51,69x. This sug- ered stable. This is the case when the periods between
gests that finger pressures differ between fingers of one servo self adjusts ŽPhysiocals. are above 30 beats magni-
hand. With Portapres during 24-h recordings using middle tude ŽTable 3.. As a second criterion the fall in steepness
and annular fingers in alternation at 30 min interval, nearly of the pressure wave in case of increasing vasomotor tone
similar finger pressures were seen in 8 subjects but due to can be used w72x.
B.P.M. Imholz et al.r CardioÕascular Research 38 (1998) 605–616 613
Table 3
Optimal measurement conditions for finger arterial blood pressure measurement
Surrounding:
Clinic ambient temperature ) 228C
Outclinicrambulant hand covered with clothing
Subject:
Children no specific guidelines
Elderly
Vascular disease
Measurement:
Site: Temperature hands warm hands if cold
Position of hand always at heart level
Cuff: Type finger digit III en IV, do not use thumb
Signal: Stable measurement beyond 5 min, interval between physiocal ) 30 beats
Physiocal onroff minute-to-minute measurement: ON
Reproducibility:
journals w78–80x. Diagnostic accuracy is achieved only the underlying blood pressure changes w86x. In addition,
after application of corrective measures. the above mentioned measures of cardiac output applied
on the non-invasive finger pressure waveform supports its
6.1. Adjustments of finger blood pressure leÕels potential as a key-hole to the circulation.
w18x Tanaka H, Thulesius O, Yamaguchi H, Mino M, Konishi K. Contin- the finger, measured continuously by a new method Žthe Finapres..
uous non-invasive finger blood pressure monitoring in children. Acta Anesthesiology 1985;62:342–345.
Paediatr 1994;83:57–73. w39x Epstein RH, Kaplan S, Leighton BL, Norris MC, DeSimone CA.
w19x De Mey C, Schroeter VM, Butzer R, Roll S, Belz GG. Method Evaluation of a continuous noninvasive blood pressure monitor in
specificity of non-invasive blood pressure measurement: oscillome- obstetric patients undergoing spinal anesthesia. J Clin Monit
try and finger pulse pressure vs. acoustic methods. Br J Clin 1989;5Ž3.:157–163.
Pharmacol 1995;40:291–297. w40x Kurki TS, Smith NT, Sanford TJ Jr., Head N. Pulse oximetry and
w20x Bos WJW, van den Meiracker AH, Wesseling KH, Schalekamp finger blood pressure measurement during open-heart surgery. J Clin
MADH. Effect of regional and systemic changes in vasomotor tone Monit 1989;5:221–228.
on finger pressure amplification. Hypertension 1995;26:315–320. w41x Kurki TS, Piirainen HI, Kurki PT. Non-invasive monitoring of
w21x Nijboer JA, Dorlas JC, Lubbers J. The difference in blood pressure finger arterial pressure in patients with Raynaud phenomenon: ef-
between upper arm and finger during physical exercise. Clin Physiol fects of exposure to cold. Br J Anaesth 1990;65:558–563.
1988;8Ž5.:501–510. w42x Farquhar IK. Continuous direct and indirect blood pressure measure-
w22x Hildebrandt W, Schutze H, Stegemann J. On the reliability of the ment ŽFinapres. in the critically ill. Anaesthesia 1991;46Ž12.:1050–
Penaz cuff during systemic and local fingertip vasodilatation at rest 1055.
and in exercise. Eur J Appl Physiol 1991;62Ž3.:175–179. w43x Pace NL, East TD. Simultaneous comparison of intra-arterial, oscil-
w23x Silke B, Spiers JP, Boyd S, Graham G, McParland G, Scott ME. lometric and Finapres monitoring during anesthesia. Anesth Analg
Evaluation of non-invasive blood pressure measurement by the 1991;73:213–220.
Ross Russell RW. Value of non-invasive continuous blood pressure w73x Gizdulich P, Aschero G, Guerrisi M, Wesseling KH. Effect of
monitoring in the detection of carotid sinus hypersensitivity. Clin hydrostatic pressure of finger pressure measured with Finapres.
Auton Res 1991;1:157–159. Homeostasis 1995;36Ž3.:120–129.
w59x Kenny RA, Introduction. In: Kenny RA, editor. Syncope in the w74x Voogel AJ, van Montfrans GA. Reproducibility of 24-hour finger
Older Patient, -Causes, Investigations and Consequences of Syncope arterial blood pressure, variability and systemic hemodynamics. J
and Falls. Chap. 1. Chapman and Hall Medical, London, 1996;8–10. Hypertens 1997; in press.
w60x Wieling W, Karemaker JM. Measurement of heart rate and blood w75x Trazzi R, Mutti E, Frattola A, Imholz B, Parati G, Mancia G.
pressure to evaluate disturbances in neurocardiovascular control. In: Reproducibility of non-invasive and intra-arterial blood pressure
Bannister R, Mathias CJ, editors. Autonomic Failure. A Textbook of monitoring. Implications for studies on antihypertensive treatment. J
Clinical Disorders of the Autonomic Nervous System, 5th ed. Ox- Hypertens 1991;9:115–119.
ford University Press, Oxford, 1997, in press. w76x Imholz BPM, Dambrink JHA, Karemaker JM, Wieling W. Ortho-
w61x Partridge BL, Sanford TJ Jr.. Finger plethysmography in Anesthesia. static circulatory control in the elderly evaluated by non-invasive
Semin Anesth 1989;8Ž2.:102–111. continuous blood pressure measurement. Clin Sci 1990;79:73–79.
w62x Benditt DG, Ferguson DW, Grubb BP, Kapoor WN, Kugler J, w77x de Jong-de Vos van Steenwijk CCE, Wieling W, Johannes JM,
Lerman BB, Maloney JD, Raviele A, Ross B, Sutton R, Wolk MJ, Harms MP, Kuis W, Wesseling KH. Incidence and hemodynamic
Wood DL. Tilt table testing for assessing syncope — ACC expert characteristics of near-fainting in healthy 6- to 16-year old subjects.
consensus document. JACC 1996;28Ž1.:263–275. JACC 1995;25:1615–1621.
w63x Low PA. Assessment: clinical autonomic testing report of the thera- w78x Grassi G, Seravalle G, Cattaneo BM, et al. Sympathetic activation