dynamic reactions produced deflating pressurecirc.ahajournals.org/content/38/6/1085.full.pdfdynamic...

Dynamic Reactions Produced by

Deflating a Blood Pressure CuffBy EDWARD D. FREIs, M.D., AND R. FRANCIS SAPPINGTON, JR., M.D.

SUMMARYIntra-arterial pressure was recorded at various points in the compressed segment

of brachial artery during deflation of a standard blood pressure cuff in human subjectswith arms of normal girth. Cuff pressure and the Korotkoff sounds also were recordedsimultaneously. The data were analyzed in terms of the various dynamic reactionsproduced by cuff deflation, which may influence the auscultatory indications of systolicand diastolic blood pressure. Cuff pressure was incompletely transmitted to the com-

pressed arterial segment. As a result, muffling, the auscultatory indication of diastolicpressure, occurred at a cuff pressure higher than the directly recorded intra-arterialdiastolic blood pressure. The first Korotkoff sound, on the other hand, provided a closeapproximation of intra-arterial systolic pressure. This may be due to a delay in pene-tration of the diminutive pulse waves into the distal part of the compressed arterialsegment at systolic levels of cuff pressure. This effect appears to compensate for otherinfluences tending to raise the indirect reading of systolic blood pressure. The extentof the delay in penetration should be dependent on the length of the collapsed segment,which in turn is a function of cuff width.

Additional Indexing Words:Intra-arterial pressureBlood pressure measurement

THE auscultatory technique for estimat-ing blood pressure is one of the most

important and frequently used clinical mea-surements. Yet, the physical basis of themethod is poorly understood. The origin ofthe Korotkoff sounds still remains an un-settled question.1-5 In addition, there are sev-eral other aspects of the technique that havenot been clarified.4 For example, it is notknown whether the intra-arterial pressure inthe occluded segment is at the same or adifferent level than the cuff pressure. Neitherhas it been determined in man whether sig-nificant augmentation of systolic pressurecaused by summation of reflected waves oc-curs proximal to the occluded segment. Al-though it has been postulated that the com-pression is greater in the center than at the

Muffling Korotkoff sounds

ends the length of the occluded segmentremains unknown. Finally, the pressure gra-dient through the compressed segment duringthe various stages of cuff deflation has notbeen determined in man.The arterial pressure probably cannot be

measured with great precision by the indirectmethod. Nevertheless, clarification of thehemodynamic changes induced by the com-pressing cuff would provide a sound physicalbasis for establishing the criteria to be usedin estimating systolic and diastolic bloodpressure by the auscultatory method. In thepresent report some of the dynamic reactionsproduced by the compressing cuff are eluci-dated and their implications with regard tothe auscultatory criteria for estimating bloodpressure are discussed.

From the Veterans Administration Hospital andthe Department of Medicine, Georgetown Univer-sity School of Medicine, Washington, District ofColumbia.

Circulation, Volume XXXVIII, December 1968

Methods

Complete and technically satisfactory experi-ments were obtained in six male patients be-tween the ages of 26 and 38 years. Patients

1085

by guest on June 9, 2018http://circ.ahajournals.org/

Dow

nloaded from

http://circ.ahajournals.org/

FREIS, SAPPINGTON

with obesity or cardiovascular abnormalities were

excluded. The purpose and nature of the pro-cedure were explained to each patient prior toobtaining his consent.

After anesthetizing the skin and subcutaneoustissue with lidocaine a 16-gauge thin-walled hy-podermic needle was inserted into the brachialartery in the region of the antecubital fossa. A30-cm length of vinyl tubing (0.028 inch ID)was inserted through the needle and threadedup the brachial artery approximately 20 cm. Theneedle was then removed and compression ap-

plied immediately over the puncture site.The blood pressure cuff was the standard

wrap-around type held in place by a "Velcro"backing. The inflatable bladder measured 12.5by 23 cm and the nylon cloth covering, 14 by 58cm. It was applied to the upper arm in the usualmanner. To obtain fluoroscopic visualization ofthe location of the catheter tip thin lead stripswere applied along the width of the cuff with ad-hesive tape spaced at 2-cm intervals from theproximal to the distal edge. The catheter was

filled with 50% diatrizoate solution (Hypaque),and under direct observation, using an imageintensifying fluoroscope, the tip of the catheterwas positioned with reference to the lead strips.The diatrizoate solution was then withdrawn andthe catheter filled with heparin-saline.The proximal end of the catheter was connect-

ed to a P23db Statham strain-gauge pressuretransducer fixed at heart level. The catheter-transducer system was thoroughly and repeatedlyflushed with heparin-saline solution to excludeair bubbles and prevent clotting. Intra-arterialpressure was recorded continuously before andduring inflation and deflation with the cathetertip placed just above the proximal edge of thecuff. The catheter tip was then withdrawn to thelead marker 2 cm distal to the proximal edgeand the pressure recordings were repeated dur-ing cuff inflation and deflation. In this way a

sequence of eight recordings of intra-arterialpressure were taken during cuff inflation anddeflation spaced at intervals of 2 cm from justabove the proximal edge to just below the distaledge of the blood pressure cuff.The Korotkoff sounds were recorded simul-

taneously with the pressure recordings bymeans of a Sanborn dynamic microphone ap-

plied lightly to the arm with a rubber strap.The microphone was positioned over the brachialartery in the antecubital fossa. It was connectedto a Sanborn amplifier with a high-pass filter of12 db per octave and a nominal frequency cutoffof 100 cycles/second. The cuff was inflated rap-

idly by a hand bulb. An adjustable bleeder valvewas used-to control deflation, which occurred ata rate of approximately 3 mm Hg per second.

The cuff was connected via a T tube to both amercury manometer and a P23db strain-gaugepressure transducer.

Simultaneous tracings of intra-arterial pressure,cuff pressure, and Korotkoff sounds were record-ed on magnetic tape at a speed of 7X2 inches/second.* This permitted playback on a Sanbornmultichannel recorder at various paper speeds be-tween 5 and 50 mm/second as desired. In oneexperiment the intra-arterial pressure in the op-posite arm was recorded simultaneously througha 20-gauge hypodermic needle connected direct-ly to a P23db Statham transducer.

Calibration of pressure transducer was per-formed at the end of each experiment by apply-ing static pressures from 0 to 200 mm Hg. Noevidence of alinearity or hysteresis was found.The frequency response of the catheter-trans-ducer system was tested for us by Dr. PeterLuchsinger, using a mechanical pulse-wave gen-erator. When carefully filled to exclude air bub-bles the pressure recording system was flat andfree from phase shift to 30 cps.The blood pressure gradient through the com-

pressed segment was reconstructed from the se-quence of recordings as follows: first, the seriesof intra-arterial pressure recordings taken at 2-cm intervals from just above to just below thecuff was inspected to determine whether theblood pressure remained stable during the se-quence. This was done by measuring the averageintra-arterial systolic and diastolic blood pressure,which was recorded during the 1-minute intervalpreceding cuff inflation at each of the eight cathe-ter positions in the sequence of recordings. Ifthe average systolic or diastolic blood pressureduring these pre-inflation control periods showeddeviations in any of the eight recordings greaterthan 5 mm Hg as compared to the other controlreadings the results for that sequence were dis-carded. In six of the subjects tested the sequenceof control pre-inflation blood pressures showeddeviations within the acceptable range, andthese were used in plotting the blood pressuregradients.Each recording contained three channels of in-

formation, the Korotkoff sounds, the intra-arterialblood pressure taken at a given catheter-tip posi-tion, and the cuff pressure. The location of thefirst Korotkoff sound (Ki) was noted on therecord, and the cuff pressure was measured atthis point. Intra-arterial systolic and diastolicblood pressure also was measured. Next, intra-arterial blood pressure was measured at the pointat which cuff pressure was 50 mm Hg above the

*Model PR-3300, Consolidated ElectrodynamicsCorp. Pasadena, California.


1086


Dow

nloaded from


DYNAMIC REACTIONS

PRESSURE RECORDING 8 CM. DISTAL TO PROXIMAL EDGE

tx ~~~~~~~~~~MKOROTKOFF * eSOUNDS

ARTERIAL sPRESSURE LI

PRESSURE RECORDING X CK DISTAL TO PROXIMAL EDGE

KOROTKOFF HSOUNDS t

PRESSURE ¶ KK K K HM \mm.g 4a _

4 8 12 16 24SECONDS

Figure 1Simultaneous recordings in a 28-year-old man of Korotkoff sounds and intra-arterial pressurerecorded in the compressed arterial segment at 8 cm distal to the proximal edge of the 14-cmwide nylon cuff (above) and at 10 cm distal (below). The Korotkoff sounds have been retracedfor purposes of illustration. See text for further details.

level of the cuff pressure present at the time of KIas well as at various cuff pressures below this level.Then muffling was identified on the record asthe point at which the initial high-frequencycomponent of the Korotkoff sounds suddenlydiminished in amplitude. Cuff pressure as wellas systolic and diastolic intra-arterial blood pres-sure was measured at this point. Intra-arterialblood pressure also was measured at cuff pres-sures 30, 20, and 10 mm Hg above muffling.The above-described measurements were made

on each of the recordings representing theeight catheter-tip positions spaced 2 cm apart inthe sequence of recordings taken through thecompressed arterial segment. These measure-ments were then used to plot the spatial intra-arterial pressure gradients shown in figures 3and 4.

ResultsPenetration of the Pressure Pulse into theCompressed Arterial SegmentWhen the cuff pressure was 50 mm Hg

Circulation, Volume XXXVHII, December 1968

above systolic blood pressure, pulsatile pres-sure fluctuations were not present in thecompressed arterial segment. As cuff pressure

was lowered, pulsatile pressures penetrateddistally under the cuff. The initial pulsationswere very small but grew progressively inamplitude as cuff pressure was loweredtoward the systolic level.

Figure 1 shows Korotkoff sounds andintra-arterial blood pressure recordings fromtwo of the eight catheter positions as re-

corded in one of the subjects. The simul-taneously recorded cuff pressures have beenomitted. The upper record was taken 8 cm

distal to the proximal edge of the 14-cmwide cuff ( 1 cm distal to the midpoint ofthe compressed segment), whereas the lowerwas taken 2 cm further distally at a point10 cm from the proximal edge of the cuff.

1087


Dow

nloaded from


FREIS, SAPPINGTON

In each record the Korotkoff sounds areshown above the intra-arterial pressure re-cordings. The two tracings represent the areaof transition in the pressure recordings be-tween the proximal and distal regions of thecompressed segment.

In the upper recording taken at the 8-cmposition small pressure fluctuations pene-trated through to this level prior to the onsetof the first Korotkoff sound. In more proximallocations of the catheter these pressurefluctuations appeared progressively earlier inthe deflation cycle. Note that the pulsepressure increased gradually to the point ofmuffling (indicated by M in the sound trac-ing). The increase of pulse pressure was dueto a fall in the diastolic level as the cuffpressure was gradually reduced, thus permit-ting a greater pulse amplitude to penetrateinto the compressed segment.The recording taken 2 cm distal to this

point (lower half of figure 1) shows markedchanges in the intra-arterial pressure tracing.No pressure fluctuations occurred until theonset of the first Korotkoff sound, indicatingcomplete arterial occlusion at or just proximalto this point prior to KI. The intra-arterialpressure preceding KI was much lower thanat the 8-cm location, being somewhat lessthan 40 mm Hg as compared to approximate-ly 100 mm Hg at the point 2 cm proximally(upper half of figure 1). Thus, just prior toKI a pressure gradient in excess of 60 mmHg was present over this 2-cm length ofcompressed arterial segment. The initialpressure pulse that penetrated to the 10-cmlevel occurred coincident with KI and wasgreatly altered in contour as compared to thepressure pulses recorded proximal to this lev-el. The wave front was exceedingly steep,ending in a spike of very short duration.The diastolic pressure rose during the firsteight pulse cycles following KI, which wasopposite to its behavior in the segment 2 cmproximal. Although the absolute blood pres-sure was lower than in the proximal segmentat Ki, the pulse pressure was greater.The rate of penetration of the arterial pulse

into the proximal region of the compressed

segment at suprasystolic cuff pressures variedsomewhat in the different subjects, the twoextreme examples being shown in figure 2.In all subjects penetration was negligible ata cuff pressure 50 mm Hg above the level ofKI. At 20 mm Hg pulsations had progressedinto the proximal third of the compressedsegment in three subjects and to the mid-segment region in the three others. At a cuffpressure 5 mm Hg above level of Ki thepresence of pulsations still was limited to theproximal half of the compressed segment inthe former subjects but had moved a fewcentimeters beyond the midpoint in the latter.The initial pulsation appearing in the distal2 cm of the compressed segment and beyondwas simultaneous with the appearance of thefirst Korotkoff sound in all subjects.Pressure Gradients at Suprasystolic CuffPressuresThe pressure gradients in the compressed

arterial segment at the level of Kl as well as

PENETRATION OF PULSATILE PRESSURE UNDER THE CUFF+50

\\ ' '~~~~~~~~~~~T.J.

-*- W.vV.+40

+30

o +20

+10

PROXIMAL * D I STAL

K I , I10 2 4 6 8 10 12 14

POSITION UNDER CUFF - CK

Figure 2Graph showing relationship between the supra-systoliclevel of cuff pressure shown on the ordinate andthe farthest penetration of the pressure pulse into thecompressed arterial segment shown on the abscissa.Two examples are shown. See text for further details.


1088


Dow

nloaded from


DYNAMIC REACTIONS

0 2 4 6 8 10

POSITION UNDER CUFF - CM.

12 14

Figure 3

Graph showing systolic intra-arterial pressure on theordinate, and position with relation to the cuff fromwhich the pressure recordings were taken on theabscissa, in a 31-year-old normal man. Intra-arterialsystolic pressure gradients are shown at cuff pressures

of 50, 30, and 0 mm Hg, respectively, above thelevel of the onset of the first Korotkoff sound. Intra-arterial pressures at 50 and 30 mm Hg above Kl were

nonpulsatile except near the proximal edge of the cuff.

at 50 and 30 mm Hg cuff pressure above thislevel are shown in figure 3. The abscissaindicates the location under the cuff, and -the

ordinate the level of intra-arterial systolic

pressure. The gradients were reconstructedfrom pressure recordings taken at eightsuccessive catheter locations spanning theregion from the proximal to the distal edgesof the cuff as described in the methods sec-

tion.At suprasystolic cuff pressures the pressure

gradient increased through the proximal halfof the compressed arterial segment; the great-er the cuff pressure, the greater was the slopeof the pressure gradient (fig. 3). Distal tothe midsegment intra-arterial pressure fell atfirst gradually and then steeply. The regionof steepest fall varied somewhat in the differ-ent subjects but usually was within theboundaries encompassed by the segment 8 to12 cm distal to the proximal edge of the 14cm cuff. The pressure drop occurring in thisdistal segment varied between 80 and 90 mmHg in the different subjects.Thus, at suprasystolic cuff pressures the

compression was not uniform, but rather thetransmitted pressure was greatest in themiddle of the compressed arterial segmentand least near the ends where the pressures

tended to equilibrate with the intra-arterialpressures existing beyond the respectiveedges of the cuff.At the time of Kl there was a positive

systolic pressure gradient through the proxi-mal 8 cm of the compressed segment (fig. 3),which averaged 6.7, SD 2.5 mm Hg in thedifferent subjects (table 1). At 10 cm distalto the proximal edge the intra-arterial sys-

tolic pressure fell moderately, averaging 10.6,

able 1

Systolic Pressure Gradient and Cuff Pressure at Time of First Korotkoff Sound

Systolic pressure (mm Hg) Cuff pressureProximal At mid Distal to

Subject to cuff cuff Diff. cuff Diff.* mm Hg Diff.*

J.W.J.P.T.J.L.C.W.W.C.B.MeanSD

99122

89120112123

10813097126114130

+9+8+8+6+2+7+6.7

2.5

405115

554153

-68-79-82-71-73-77-74.3

5.3

10212387122111

125

-6-7-10-4-3-5

-5.82.4

*Difference between measurement and midsegment systolic pressure.


140r

IN

E 100

"I

a:CL.

< 80

- 60

VIf

40

201

____ AT CUFF P. 50 mm. Hg ABOVE KlCUFF P. 30 mm. Hg ABOVE KI

AT FIRST KOROTKOFF SOUND K1

Pr a DsI

1089

,. DistalProxi ma


Dow

nloaded from


FREIS, SAPPINGTON

SYSTOLIC PRESSURE GRADIENTS

l\

-----AT FIRST KOROTKOFF SOUND

-- AT 20 mm. Hg ABOVE MUFFLING L.

-AT MUFFLING

0 2 4 6 8 10


12 14

DIASTOLIC PRESSURE GRADIENTS

100 I

cn

EfE

80:r

V)

-.

Vr 40

0 2 4 6 8 10


Figure 4

Graphs of systolic (left) and diastolic (right) intra-arterial pressure gradients through thecompressed arterial segment in the same subject as shown in figure 2. Intra-arterial pressuregradients are plotted at Kl, 20 mm Hg above mnuffling, and at muffling, respectively. Othernotations as in figure 3.

SD 5.1 mm Hg, lower than the midsegmentpressure. Distal to this point the intra-arterialsystolic pressure fell steeply, averaging 74.3,SD 5.3 mm Hg, from the midsegment to thedistal edge (table 1).

Pressure Gradients from KI to Muffling

As cuff pressure was lowered below thelevel at Kl the systolic pressure gradientsgradually diminished (fig. 4). In most sub-jects systolic pressure gradients disappearedat cuff pressures approximately 10 mm Hgabove muffling, although in two a smallnegative gradient in the distal segment per-

sisted to the level of muffling, when it alsodisappeared.The diastolic pressure gradient rose con-

siderably through the proximal half of thecompressed segment at the time of Ki (fig. 4).This was due to the fact that cuff pressure

closed the midsegment region when the intra-arterial pressure fell below the transmitted

cuff pressure. Thus, diastolic pressure in themidsegment at the time of K1 was at the levelof effective transmitted cuff pressure. Be-tween 8 and 12 cm distal to the proximaledge of the cuff (in most subjects in theshorter distance of 10 and 12 cm) diastolicpressure fell steeply, averaging 57.2, SD 9.3mm Hg, in the different subjects. Althoughdiminishing gradually, a steep negative-pres-sure gradient continued in this segmentthroughout the period of loud sounds. At a

cuff pressure 20 mm Hg above muffling thediastolic pressure gradient through the proxi-mal half of the cuff was positive, averaging17.0, SD 4.7 mm Hg, and was negative throughthe distal half, averaging 31.2, SD 7.2 mm Hg(fig. 4 and table 2).The onset of muffling was accompanied

by a distinct change in the pressure gradient,characterized by the disappearance of boththe positive gradient in the proximal segmentand the negative gradient in the distal.

Citrclation, Volme XXXVIII, December 1968

100

CL'

a-

-i,c

UN

80

60

40

/ \

/ ..--- ' l,

-.-.- --AT FIRST KOROTKOFF SOUND

-*AT 20 mm. Hg ABOVE MUFFLING

AT MIUFFLING I

Proximal * DistalDistalI I I A I

12 1420 , . . .-.A I 4U - -~~U0

1090

ProximalI


Dow

nloaded from


DYNAMIC REACTIONS

Table 2Diastolic Pressure Gradients at Cuff Pressures of 20 mm Hg above Muffling and atMuffling

At cuff pressure 20 mm Hg above mufflingdiastolic pressure (mm Hg)

Proximal At mid Distal toto cuff cuff Diff. cuff Diff.

63 79 +16 45 -3469 92 +23 68 -2445 65 +20 15 -4073 82 +9 44 -3864 80 +16 57 -2365 83 +18 55 -28

+17.0 -31.24.7 7.2

Proximal

Proximalto cuff

687350706566

At mufflingdiastolic pressure (mm Hgj

At mid Distal tocuff Diff. cuff

68 0 6879 +6 7853 +3 5277 +7 7568 +3 6268 +2 66

+3.52.6

Table 3Difference between Midsegment Nonpulsatile Pressure and Cuff Pressure at Supra-systolic Cuff Pressures

Level of cuff pressure above Kl (mm Hg)50 mm Hg 30 mm Hg 20 mm Hg 10 mm Hg

Subject MSP* CPt Diff MSP CP Diff. MSP CP Diff. MSP CP Diff.

J.W. 140 152 -12 120 132 -12 114 122 -8 101 112 -11J.P. 152 173 -21 131 153 -22 122 143 -21 116 133 -17T.J. 118 140 -22 107 120 -13 95 110 -15 98 100 -2L.C. 155 172 -17 136 152 -16 126 142 -16 121 132 -11W.W. 145 160 -15 125 140 -15 115 130 -15 103 120 -17C.B. 159 175 -16 141 155 -14 132 145 -13 121 135 -14Mean -17.2 -15.3 -14.7 -12.0SD 3.8 3.6 4.3 5.6

*MSP = midsegment nonpulsatile pressure.tCP = cuff pressure.

Diastolic pressure henceforth remained es-

sentially unchanged through the compressedsegment (table 2 and fig. 4).

Effective Pressure Transmitted from the Cuffto the Underlying Artery

As noted above, the pressure exerted by thecuff on the compressed arterial segment isgreatest in the middle of the segment andleast at the ends. The difference between in-tra-arterial pressure in the maximally com-

pressed midsegment and the cuff pressure

should provide a measure of the transmittedpressure loss. However, since the occludedsegment functions as a closed end-reflectingpoint, the systolic pressure recorded just proxi-mal to this region may be raised above thelevel of transmitted pressure because of theconversion of kinetic energy of flow to pres-

sure energy. Nevertheless, the transmittedpressure loss can be estimated during completeCirculation, Volume XXXVIII, December 1968

occlusion, when pulsatile pressures are rnotpresent in the midsegment. At lower cuffpressures it may be estimated from the differ-ence between cuff pressure and midsegmentdiastolic pressure, since the diastolic level isnot significantly influenced by reflections.4The beginning of the systolic upstroke in thiscase represents the level of pressure in theartery capable of opening the occluded seg-

ment. Therefore, at supra-diastolic cuff pres-

sures the foot of the pressure pulse recordedin the midsegment region may be used as a

measure of the effective transmural pressure

being exerted on the artery by the cuff at thatmoment.

At a cuff pressure 50 mm Hg above Kithe nonpulsatile intra-arterial pressure in themidsegment region averaged 17.2, SD 3.8 mmHg, less than cuff pressure. The magnitudeof the difference decreased as cuff pressure

Subj ect

J.W.J.P.T.J.L.C.W.W.C.B.MeanSD

Diff.

0-1-1-2-6-2-2.0

2.1

1091


Dow

nloaded from


FREIS, SAPPINGTON

P t 1-p I II

C') In

,v00) r-: c'1

m

P 00 0 Om1 1>V c 0oom - oOC

00 CID 0 C to m110 t- 10 C0 co 1-

tin - Co C')coqi

q0/2

I

C) 00 0 10 q- 0) c0 000 O

000 c:0- co coco1- 101- - 00

0t0 't co

Q I HI

CO-

C.Co m001tu 000C 1- 0) 0) 0'X)o b O

0 01n N- 0) t- 0) c0 O0 1- 0)

t t o(:: 1- -l -1II-

c1)~0 rq0 10 "t

u- -iw00r-

cl 0 coC 0 C'0)0)0

1-m 00 0 o C 00 .-

p I I~- 11-

.,-cq C'.) j> C-I - 10P, 00100 Cot --I ClU r- r- q ?- -i

00 Cot .0o

.- =

m r-i L--c--im'-0 --i

P-4 ri ?-q r-

approached KI and at 10 mm Hg above KImidsegment intra-arterial pressure averaged12.0, SD 5.6 mm Hg, less than cuff pressure(table 3).At the time of KI midsegment intra-arterial

diastolic pressure averaged 8.1, SD 1.9 mmHg, less than cuff pressure. Between 30 and10 mm Hg cuff pressures above muffling themidsegment diastolic pressure was approxi-mately 10, SD 3.5 mm Hg, below cuff pressure,and at muffling the average difference was7.3, SD 2.5 mm Hg (table 4). Since themidsegment region represented the area ofmaximum compression exerted by the cuffon the brachial artery, it was apparent thatthe cuff pressure was incompletely transmit-ted to the artery since in all stages of thedeflation cycle prior to muffling, cuff pressurewas higher than midsegment intra-arterialdiastolic pressure.

Difference between Auscultatory and DirectIndication of Blood Pressure

In each of the control recordings of intra-arterial pressure preceding each cuff inflationthe mean of the respiratory fluctuation insystolic and diastolic pressures were mea-sured. The eight control intra-arterial pres-sure recordings of systolic and diastolic pres-sure taken prior to cuff inflation in eachsubject were averaged and are shown intable 5. Also, the recorded cuff pressurees atthe time of the microphone recordings of Kiand of muffling also were measured and theeight readings for each subject were aver-aged (table 5).The results indicated that there was close

agreement between the average cuff pressureat the time of Kl and the average controllevel of systolic pressure, the deviation be-tween the two estimates being -0.1, SD 2.2mm Hg, in the six subjects. On the otherhand, the average cuff pressure at the time ofmuffling provided an estimate of diastolicpressure that was significantly higher thanthe average control direct recording ofdiastolic pressure. The cuff pressure estimatedthe diastolic level was 8.7, SD 2.1 mm Hg,higher than the intra-arterial diastolic pres-sure.


606

00)

C) 9

OC0)CA

'e6

C) m

£ vcc: Cd-E w

H

11 =4Q

* -I-

1092

_/

beD

8

¢beD0

08

0

EO-

08-1

be

0

v

8

0

0

0U)

.6

6

be

bebe

;be

C

co

-be

beC)

be

*0n6

0

bev)


Dow

nloaded from


DYNAMIC REACTIONS

Table 5Difference between Auscultatory Indication of Blood Pressure and Actual Values

Subject

J.W.J.P.,r.j.L.C.W.W.C.B.MeanSD

Average*systolicpressure(mm Hg)

103.6122.893.3122.3110.9124.8113.0

Averagetcuff pressure

at Ki(mm Hg)

102.3124.690.1

125.4110.1124.4112.8

Diff.(mm Hg)

-1.3+1.8-3.2+3.1-0.8-0.4-0.1

2.2

Average*diastolicpressure(mm Hg)

56.271.550.074.363.773.564.9

Averagetcuff pressureat muffling(mm Hg)

62.883.057.481.173.683.773.6

Diff.(mm Hg)

+6.6+11.5+7.4+6.8+9.9

+10.2+8.7

2.1

*Average of eight recordings of systolic and diastolic pressure, respectively, taken prior tovarious cuff inflations.

tAverage of eight recordings of cuff pressures at time of first Korotkoff sound taken duringvarious cuff deflations.

DiscussionThe possibility was considered that the

insertion of the catheter may have producedlocal spasm of the arterial wall, which inter-fered with the production of normal Korotkoffsounds. Care was used in inserting the needleand catheter to avoid trauma to the artery.The needle was removed immediately afterinsertion of the catheter, and pressure wasapplied to avoid hematoma. After waiting 10minutes the Korotkoff sounds recorded withthe catheter in place appeared no differentfrom those recorded prior to the arterialpuncture.There also was no evidence to indicate

that the presence of the catheter interferedwith normal closure of the compressed arteri-al segment. If such had occurred, KI wouldhave been recorded earlier, that is, at highercuff pressures when the catheter protrudedthrough the midsegment than when it wasbelow it. However, there was no significantvariation in cuff pressures at Ki when thecatheter tip lay in the proximal or in thedistal end of the cuff. As a further checkintra-arterial pressures were recorded simul-taneously from the contralateral arm in onesubject before as well as after insertion of thecatheter. The latter record failed to revealany indication that the catheter altered theKorotkoff sound indications of systolic ordiastolic pressure.Circulation, Volume XXXVIII, December 1968

To our knowledge a study of the dynamicpressure changes in the compressed arterialsegment has not been previously carried out.Bazett and associates6 recorded pressuresproximal and distal to the compressed seg-ment of the femoral artery of the dog butthe artery was severed in order to place itwithin the compression chamber. They notedby inspection that the occlusion was conicaland opened out both proximally and distallyfrom the maximally compressed midsegment.This observation is consistent with the pres-ently recorded pressure gradients duringocclusion in the compressed segment of man.The central closure of the compressed

segment divides the proximal and distal partsinto separate systems that equilibrate withthe pressures existing in their respectivevasculatures. The first recorded Korotoffsound always occurs simultaneously with thefirst pulse that traverses the distal segment.When the first pulse traverses the midseg-ment at the time of KI it passes into an arterialsegment, which has a pressure drop of ap-proximately 75 mm Hg over a length of only4 or 5 cm. The steepness of this gradient mustinduce a marked acceleration of the blood atthe moment that the occluded segment opens.

Compression of the brachial artery by anair-filled cuff produces a variety of dynamicreactions, some of which elevate and othersreduce the apparent readings. Some of these

1093


Dow

nloaded from


FREIS, SAPPINGTON

influences were recognized as early as 1901by von Recklinghausen7 and were furtheramplified by Erlanger8 and Wiggers4 in lateryears. However, instrumentation was not suf-ficently developed to measure their effectsexperimentally in man.The present results indicate that the pres-

sure in the cuff was incompletely transmittedto the underlying arterial segment. At supra-systolic cuff pressures the pressure within thecompressed arterial segment was lower thanthe cuff pressure. This was the case even inthe completely occluded midsegment region.Between Ki and muffling the beginning ofthe systolic upstroke in the arterial midseg-ment occurred at an intra-arterial pressuresomewhat lower than the cuff pressure. Thisindicated that the occluded segment wasopening at an intra-arterial pressure belowthat exerted by the cuff. The inefficient trans-mission of pressure energy probably was aconsequence of the resistance imposed bythe viscoelastic tissues of the arm.4A pressure loss occurring between cuff and

artery would produce a falsely high indica-tion of blood pressure by the indirect method.This appeared to be the case with regard tothe muffling criterion for estimating diastolicblood pressure. At muffling the beginning ofthe arterial upstroke in the midsegmentregion occurred at an intra-arterial pressureaveraging 7.3 mm Hg lower than cuff pres-sure. This approximated the discrepancyfound between directly measured intra-arteri-al diastolic pressures taken prior to cuff in-flation and that taken by the auscultatorymethod using muffling as the diastolic crite-rion.

Since transmission of cuff pressure also wasincomplete at Kl, why was there good agree-ment between the direct and indirect mea-surements of systolic blood pressure? Webelieve that an additional phenomenon wasoperative at Kl (but not at muffling), whichacted in the direction of lowering the indirectreading. At the time of Kl the pressure ex-erted on the artery was sufficient to produceocclusion throughout most of the pulse-wave

cycle. Only the peaks of the pressure pulseswere able to penetrate into the midsegment.Because they were transient and of smallmagnitude (top half of figure 1) they con-tained too little pressure energy to open theentire length of the collapsed segment untilthe effective pressure exerted by the cuff fellsignificantly below the level of the intra-arterial systolic pressure. Kl always occurredsimultaneously with the passage of the firstpulse wave which passed completely throughthe compressed segment. In order to causesuch complete penetration a pulse pressureof sufficient amplitude to drive the pulsewave through the occluded segment was re-quired. The extent of the delay in penetrationwill be dependent on the length of theoccluded segment, which in turn is a functionof cuff width. This compensatory phenome-non cannot occur at muffling when the cuffpressure is near the level of diastolic pressureand pulse waves of normal amplitude arepassing through the compressed segment.Another factor that possibly could influence

the indirect readings of blood pressure iswave reflections, since the arterial occlusionacts as a closed end-reflecting point. Reflec-tions would elevate the systolic peaks proxi-mal to the occlusion but would have noeffect on the diastolic portions of the pulsewave. The initial snapping portion of theKorotkoff sounds is associated with openingof the occluded segment.3' As muffling isapproached the occluded segment opens nearthe foot of the pulse wave which is unaffectedby wave reflections. However, at KI reflec-tions could influence the level at which thefirst pulse passes through the occluded seg-ment since reflections elevate the systolicpeaks and could lead to a falsely highestimate of systolic blood pressure.Bazett and associates" have indicated that

the conical shape of the arterial occlusionprobably would prevent major reflectionsfrom developing. At the time of Ki there wasa slight positive systolic pressure gradient inthe proximal half of the cuff (table 1 andfig. 3), which may have been caused by

Circulation, Volame XXXVIII, December 1968

1094


Dow

nloaded from


DYNAMIC REACTIONS

wave reflections. However, if significant re-flections were present, they were not re-flected back for any considerable distancesince the systolic pressure recorded justproximal to the cuff at the time of Kl was nohigher than that recorded prior to cuff in-flation. Whatever influence reflections mighthave had in elevating the systolic bloodpressure at the time of Ki their effect on theindirect reading appeared to be compensatedfor by the delay in penetration of thediminutive pulsations through the occludedmidsegment.The present results indicate that when

muffling is used as the criterion the diastolicpressure reads higher by the indirect methodthan the intra-arterial diastolic blood pres-sure measured directly. Simultaneous record-ings of intra-arterial and auscultatory bloodpressures have been carried out by numerousother investigators. A minority9'-' observedmuffling at cuff pressures 3 to 5 mm Hg lowerthan intra-arterial diastolic, whereas the ma-jority12-22 found that muffling occurred atcuff pressures usually averaging 8 to 10 mmHg higher than the intra-arterial diastolicpressure. For this reason some authors be-lieve that disappearance of sound provides abetter estimate of diastolic pressure.Disappearance of sounds could not be

assessed as a criterion of diastolic pressurein the present study because in some record-ings sounds of low frequency and intensitypersisted to levels far below diastolic pres-sure. It was apparent that the point selectedfor disappearance will vary with the acuityof hearing of the observer and possibly alsowith the type of stethoscope head used. Inaddition, it is known that following exerciseor reactive hyperemia and in patients withaortic insufficiency clearly audible, post-muf-fling sounds may extend far below diastoliclevels of intra-arterial pressure. Thus, mufflingrather than disappearance seems to provide amore dependable end point for estimatingdiastolic pressure.Although further investigation is needed to

determine the point in the sequence of Korot-koff sounds that most accurately representsCirculation, Volume XXXVIII, Decemrber 1968

the diastolic pressure, the present study indi-cates that muffling is the acoustic representa-tion of the disappearance of a diastolicpressure gradient through the compressedsegment. However, because of the loss intransmission of pressure from the cuff to theartery, the indirect reading will be too high.The amount of correction needed requiresfurther study. The present results as well asthe majority of reports in the literature, com-paring the auscultatory and direct methods,suggest that approximately 9 mm Hg sub-tracted from the manometric reading atmuffling may provide the closest estimate ofintra-arterial diastolic blood pressure using astandard cuff in nonobese adults.

References1. RODBARD, S.: Significance of the intermediate

Korotkoff sounds. Circulation 8: 600, 1953.2. MEISNER, J. E., AND RUSHAMER, R. F.: Produc-

tion of sounds in distensible tubes. Circula-tion Research 12: 651, 1963.

3. CHUNGCHAROEN, D.: Genesis of Korotkoffsounds. Amer J Physiol 207: 190, 1964.

4. WIGGERS, C. J.: Dynamic reactions induced bycompression of an artery. Circulation Re-search 4: 4, 1956.

5. FRANK, O., AND WEZLER, K.: Die Bestimmungdes Blutdrucks beim Menschen. Z Biol 91:439, 1931.

6. BAZETT, H. C., LAPLACE, L. B., AND SCOTT,J. C.: Pressure changes induced in the vas-cular system as the result of compression ofa limb, and their effect on the indirect mea-surement of lateral pressures. Amer J Physiol112: 182, 1935.

7. VON RECKLINGHAUSEN, H.: Ueber Blutdruck-messung beim Menschen. Arch Exp PathPharm 46: 78, 1901.

8. ERLANGER, J.: Movements in the artery undercompression during blood pressure determina-tions. Amer J Physiol 55: 84, 1921.

9. ROBERTS, L. N., SMILEY, J. R., AND MANNING,G. W.: Comparison of direct and indirectblood-pressure determinations. Circulation 8:232, 1953.

10. VAN BERGEN, F. H., WEATHERHEAD, D. S.,TREOLAR, A. E., DOBKIN, A. B., AND BUCK-LEY, J. J.: Comparison of indirect and directmethods of measuring arterial blood pressure.Circulation 10: 481, 1954.

11. HOLLAND, W. W., AND HUMERFELT, S.: Mea-surement of blood pressure: Comparison ofintra-arterial and cuff values. Brit Med J 2:1241, 1964.

1095


Dow

nloaded from


FREIS, SAPPINGTON

12. BERLINER, K., FUJiY, H., LEE, D. H., YILDIZ,M., AND GARNIER, B.: Blood pressure mea-surements in obese persons: Comparison ofintra-arterial and auscultatory measurements.Amer J Cardiol 8: 10, 1961.

13. RAGAN, C., AND BORDLEY, J.: Accuracy of clin-ical measurement of arterial blood pressure.Bull Hopkins Hosp 69: 504, 1941.

14. STEELE, J. M.: Comparison of simultaneous in-direct (auscultatory) and direct (intra-ar-terial) measurements of arterial pressure inman. J Mount Sinai Hosp (NY) 8: 1042,1942.

15. HAMILTON, W. F., WOODBURY, R. A., ANDHARPER, H. T.: Physiologic relations be-tween intrathoracic, intraspinal, and arterialpressures. JAMA 107: 853, 1936.

16. KOTTE, J. K., IGLAUER, A., AND McGuIRE, J.:Measurement of arterial blood pressure inthe arm and leg: Comparison of sphygmo-manometric and direct intra-arterial pressures.Amer Heart J 28: 476, 1944.

17. HENSCHEL, A., DE LA VEGA, F., AND TAYLOR,

H. L.: Simultaneous direct and indirectblood pressure measurements in man at restand work. J Appl Physiol 6: 506, 1954.

18. TROUT, K. W., BERTRAND, C. A., AND WILLIAMS,M. H.: Measurement of blood pressure inobese persons. JAMA 162: 970, 1956.

19. HARRISON, E. G., JR., ROTH, C. M., ANDHINES, E. A.: Bilateral indirect and directarterial pressures. Circulation 22: 419, 1960.

20. BERLINER, K., FujiY, H., Ho LEE, D., YILDIZ,M., AND GARNIER, B.: Accuracy of bloodpressure determinations: Comparison of directand indirect measurements. Cardiologia 37:118, 1960.

21. KORVONEN, M. J., TELIVHO, L. J., AND JAR-VINEN, E. J. K.: Sphygmomanometer cuffsize and the accuracy of indirect measure-ment of blood pressure. Amer J Cardiol 13:688, 1964.

22. LONDON, S. B., AND LONDON, R. E.: Critique ofindirect diastolic end point: "Muffling" vs."last" sound. Arch Intern Med (Chicago) 119:39, 1967.

100 Years AgoRoutine Temperature of Patients

For the last sixteen years my attention has been uninterruptedly directed to thecourse pursued by the temperature in diseases of various kinds. The thermometer hasbeen regularly employed at least twice daily, and in febrile patients from four to eighttimes a day, and even oftener, in special circumstances, for all the patients in my wards.I have also experienced the applicability of this method of investigation in very nu-merous cases in private practice. In this way I have gradually got together a materialwhich comprises many thousand complete cases of thermometric observations of disease,and millions of separate readings of the temperature. The more my observations weremultiplied the more firmly rooted did my conviction become of the unparalleled valueof this method of investigation, as giving an accurate and reliable insight into thecondition of the sick.-C. A. WUNDERLICH: Preface of First Edition. On the Temperaturein Diseases: A Manual of Medical Thermometry. London, The New Syndenham Society,1871 [ p. v].


1096


Dow

nloaded from


EDWARD D. FREIS and R. FRANCIS SAPPINGTON, JR.Dynamic Reactions Produced by Deflating a Blood Pressure Cuff

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1968 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.38.6.1085

1968;38:1085-1096Circulation.

http://circ.ahajournals.org/content/38/6/1085located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. and Rights Question and Answer

Permissionsthe Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle column ofClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:


Dow

nloaded from

http://circ.ahajournals.org/content/38/6/1085

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


dynamic reactions produced deflating pressurecirc.ahajournals.org/content/38/6/1085.full.pdfdynamic...

Documents