PREVENTIVE MEDICINE 12, 715-719 (1983)

SHORT REPORT
Quality of Korotkoff Sounds: Bell vs Diaphragm, Cubital Fossa
vs Brachial Artery’
RONALD J. PRINEAS~ AND DAVID JACOBS

Laboratory of Physiological Hygiene, School of Public Health, Stadium Gate 27, 611 Beacon
Street SE, University of Minnesota, Minneapolis, Minnesota 55455
Comparison of clarity of Korotkoff sounds was made between those heard with the bell
or diaphragm of the stethoscope and between those heard over the cubital fossa or more
proximally and medially over the maximal brachial artery pulse-during indirect sphyg-
momanometry. Multiple measurements were made by two trained technicians on 48 supine
adults. The measurements were balanced as to stethoscope-head/arm-placement combina-
tions, stethoscope used, control of the pressure bulb, and use of right or left arm. The
Korotkoff sounds were heard better with the bell of the stethoscope placed over the brachial
artery pulse (BB) than with the diaphragm placed over the cubital fossa (DC). The former
combination (BB) gave significantly higher blood pressure estimates for systolic blood pres-
sure and fourth phase diastolic blood pressure than the latter combination (DC).

INTRODUCTION
The auscultatory estimation of systemic blood pressure is still the method of
choice for clinical purposes, population studies, and clinical trials. Based on
sounds heard with a stethoscope below a pressure cuff wrapped around the upper
arm, the method was first described by Korotkoff for establishing complete oc-
clusion of arteries in some surgical experiments (4). The four phases of Korotkoff
were later modified to five phases by Goodman and Howell in 1911 (1). Since
that time it has been noted that many factors affect these casual indirect auscul-
tatory measurements. Apart from the resting state (in relation to time, drugs,
food, physical activity, etc.) of the subject, and environment and observer effects
that have been summarized (3, S), the effect of technique on blood pressure
measurements has also been extensively studied (10). Most of these factors alter
the blood pressure measurement systematically, though only by a few millimeters
of mercury. For example, in a recent screening of school children, the morning
systolic blood pressures were on the average 1.5 mm Hg lower than those re-
corded in the afternoon (9).
Nonetheless, when each factor possibly affecting blood pressure measurement
is applied differently in different blood pressure measuring circumstances, sizable
differences in blood pressure levels may be noted for the same subjects. With the
recent results of the Hypertension Detection and Follow-up Program (2) indi-
cating the advisability of treating people with diastolic fifth phase blood pressures
of 390 mm Hg and the Pediatric National Advisory Task Force recommendation
of routine annual blood pressure measurements for children (7) there is a great

’ Supported in part by NHLBI Grants HL-19877 and HL-00827.

2 To whom reprints requests should be addressed.

71.5
0091-7435/83 $3.00
Copyright 0 1983 by Academic Press, Inc.
All rights of reproduction in any fom reserved.
716 PRINEAS AND JACOBS

need to standardize the methods of measurement. Such recommendations have

been made (3). Some of the recommendations have been misinterpreted by some
manufacturers of blood pressure measuring equipment as exampled by incorrect
bladder cuff sizes (5). Other recommendations rest on theoretical assumptions
not tested experimentally.
It is important to understand the basis for standard recommendations so pos-
sible causes for systematic bias can be recognized. Two elements of indirect
measurement of blood pressure that have been made on theoretical grounds have
not, to our knowledge, been put to the test of experiment. These are use of the
bell rather than the diaphragm of the stethoscope and placement of the stetho-
scope head, not in the cubital fossa, but more proximally and medially over the
maximal pulsation of the brachial artery. There is certainly conflict in daily usage
of these elements in clinical practice and clinical trials. The purpose of the present
study was to compare the systematic effects on the indirect measurement of blood
pressure by use of the bell versus the diaphragm and by placement of the steth-
oscope head in the cubital fossa versus over the brachial artery in the upper arm
immediately superior to the radial medial epicondyle and medial to the biceps
tendon.
METHODS
Blood pressure measurements were all made by the same two trained, expe-
rienced, and repeatedly tested technicians. The technicians had no prior expec-
tation of the outcome of the experiment.
Forty-eight adult men and women with a 20+ year age span and a variety of
arm sizes and blood pressure levels were examined. All subjects lay supine for 5
min with both arms level and flat on a bed with hands pronated. Each had a
series of four blood pressure measurements made on each arm simultaneously
by the two technicians. The cuffs on each arm were connected separately to two
Hawksley random-zero sphygmomanometers (11) and by a Y-connection to a
single bulb. The stethoscopes used were by Littman. Each observer used a dif-
ferent combination of stethoscope-head/arm placement for each of the four blood
pressure measurements. Either the bell or diaphragm of the stethoscope was used
and placed over the cubital fossa (Fig. 1) or over the previously palpated brachial
artery pulse on the upper arm superior to the medial epicondyle and medial to
the biceps tendon (Fig. 2). This latter position will be referred to hereafter as the
“brachial artery.” The combinations bell-brachial artery (BB), bell-cubital fossa
(BC), diaphragm-brachial artery (DB), and diaphragm-cubital fossa (DC) were
applied in random order by one observer. The order was determined by use of
random latin squares. The second technician, for any simultaneous measurement,
used both a different stethoscope head and a different arm-placement combination
from the other technician. At every other measurement the bulb was controlled
by alternate technicians. Further, midway through the experiment (after mea-
surement on 24 subjects) stethoscopes were exchanged for the remaining mea-
surements. For each alternate patient each technician was assigned to measure-
ments on alternate arms of the subjects. After each blood pressure measurement
the subjects raised their arms for 5 set and at least 30 set elapsed before the next
IMPARING KOROTKOFF SOUND CLARITY 717

FIGURE I

p
FIGURE 2
718 PRINEAS AND JACOBS

measurement was made. Systolic blood pressure, diastolic fourth phase, and di-
astolic fifth phase sounds were recorded at each measurement (8). For each of
the 48 subjects, each observer measured the blood pressure with four separate
stethoscope-head/arm-placement combinations.
Thus, measurements for all stethoscope-head/arm-placement combinations
were fully balanced with respect to which of the two stethoscopes was used,
which technician controlled the bulb, and which arm of the subject was used for
blood pressure measurement. In addition, each stethoscope-head/arm-placement
combination was used an equal number of times. The primary question addressed
was are there systematic differences between DC and BB? This was answered
by paired comparisons of simultaneous measurements between observers and for
each observer separately. It was also possible to examine DB vs BC in this way.
RESULTS
The comparison between the stethoscope-head/arm-placement combinations
DC and BB is summarized in Table 1. It can be seen that the BB gave statistically
significant, systematically higher readings for both systolic blood pressure and
diastolic fourth phase blood pressure than those obtained with the DC combi-
nation The BB gave the lowest readings for diastolic fifth phase blood pressure
but the difference was not statistically significant. There were no systematic dif-
ferences between DB and BC combinations for either systolic or fourth or fifth
phase diastolic blood pressures. Because different stethoscope heads were used
at each measurement by each observer, comparisons of BC vs BB and DC vs
DB could not be made.
DISCUSSION
Korotkoff sounds are low frequency sounds and, therefore, one would expect
to hear them more clearly with the bell of the stethoscope (6), and hence to detect
systolic blood pressure slightly earlier (higher) and fifth phase diastolic blood
pressure later (lower) because onset and offset sounds are at lower amplitude.
The effect on fourth phase diastolic blood pressure is more difficult to predict,
but a soft onset of muMing would be heard earlier. The stethoscope head is closer
to the brachial artery in the B position than in the cubital fossa. This again would

TABLE 1
COMPARISON OF USE OF DIAPHRAGM IN CUBITAL FOSSA (DC) vs COMBINATION OF BELL OVER
BRACHIAL ARTERY (BB) BY Two OBSERVERS MAKING SIMULTANEOUS MEASUREMENTS ON 48 ADULT
MEN AND WOMEN

Blood Mean (BB-DC)

pressure0 k SD (mm Hg) t P

S 1.22 k 0.60 2.04 (91 df) 0.044

D4 1.59 * 0.66 2.41 (91 df) 0.018
D5 -0.26 k 0.61 -0.43 (91 df) 0.668
(1S = systolic blood pressure; D4 = fourth phase (onset of muffled sounds) diastolic blood pressure;
D5 = fifth phase (disappearance of sounds) diastolic blood pressure.
 COMPARING KOROTKOFF SOUND CLARITY 719

increase the chance of hearing Korotkoff sounds earlier for systolic blood pres-
sure to give a higher reading and for the fifth phase diastolic blood pressure to
be heard later and thus give a lower reading in the B position compared to the
cubital fossa.
The observation of systematic differences between BB and DC support the
hypothesized increase in clarity of Korotkoff sounds using the bell and placing
the stethoscope head more directly over the brachial artery. The observation of
no difference between BC and DB combinations suggests that the gain in use of
the bell is canceled if it is placed further away from the brachial artery, that is,
in the cubital fossa. Which combination BB, BC, DB, or DC gives the “truest”
estimation of blood pressure cannot be determined without comparison with di-
rect measurements of blood pressure. However, in our opinion, the most “desir-
able” technique is that which provides the clearest sounds. We, therefore, rec-
ommend use of the bell of the stethoscope placed over the brachial artery pul-
sation (superior and medial to the cubital fossa and medial to the biceps tendon,
below the blood pressure cuff).

REFERENCES
1. Goodman, E. H., and Howell, A. A. Further clinical studies in the auscultatory method of de-
termining blood pressure. Amer. Med. Sci. 142, 334-353 (1911).
2. Hypertension Detection and Follow-up Program Cooperative Group. Five-year findings of the
Hypertension Detection and Follow-up Program. I. Reduction in mortality of persons with high
blood pressure, including mild hypertension. JAMA 242, 2652-2571 (1979).
3. Kirkendall, W. M., Feinleib, M., Freis, E. D., and Mark, A. L. Recommendations for human
blood pressure determination by sphygmomanometers. Circulation 62, 1145A-1155A (1980).
4. Korotkoff, N. C. On methods of studying blood pressure. Bull. Imperial Military Med. Acad.
(St. Petersburg) 11, 365 (1905).
5. Maxwell, M. H., Waks, A. U., Schroth, P. C., Karam, M., and Domfeld, L. P. Error in blood-
pressure measurement due to incorrect cuff size in obese patients. Lancer 2, 33-35 (1982).
6. McCutcheon, E. P., and Rushmer, R. E Korotkoff sounds: An experimental technique. Circu-
lation Res. 20, 149-161 (1967).
7. NHLBI Task Force on Blood Pressure Control in Children. Report of the task force on blood
pressure control in children. Pediatrics 59 (Suppl.) (1977).
8. Prineas, R. J. “Blood Pressure Sounds: Their Measurement and Meaning.” Gamma Medical
Products Corp., Philadelphia, 1978.
9. Prineas, R. J., Gillum, R. E, Horibe, H., and Hannan, P. J. The Minneapolis Children’s Blood
Pressure Study. Part 2. Multiple determinants for children’s blood pressure. Hypertension
(Dallas) 2 (Suppl. I), 1-24-I-28 (1980).
10. Rose, G. Standardization of observers in blood pressure measurement. Lancer 1, 673-674 (1965).
11. Wright, B. M., and Dore, C. F. A random-zero sphygmomanometer. Lancet 1, 337-339 (1970).