How are the Korotkoff sounds produced when do you hear the first Korotkoff sound?

Abstract

Dr. Nikolai Korotkoff described the sounds that bear his name in 1905.1 The five phases that we now recognize as the “Korotkoff phases” are heard by placing a stethoscope over the brachial artery and listening for the changing character of the sounds during deflation of the blood pressure (BP) cuff. The onset of phases I, IV, and V are currently of clinical interest, but the intervening phases have received almost no clinical attention. Even though it is well recognized that normal children may have no phase V,2 it is not known whether the other phases may be absent. The aims of this study were to describe the pattern of Korotkoff phase distribution in a population of normal adults and children; to measure the duration of each of the phases; and to describe the differences in the Korotkoff phases between adults and children.

Methods

Subjects

Two local primary schools agreed to participate in the study. The study was confined to children aged 7 to 8 years. After a visit to the school, children interested in participating in the study took a consent form home, which was approved by the local ethics committee, for their parent or guardian to sign. The study was carried out on a normal school day that did not include a physical educa-tion class. Adult volunteers were recruited from the domestic and clerical staff of the nearby hall of residence. Anyone with a history of hypertension was excluded.

Recording Equipment

The Korotkoff sound measurement system has been described in detail in another study.3 Briefly, the bell of a Littmann cardiology grade stethoscope was attached to a miniature, high sensitivity ceramic microphone. The bell of a pediatric or adult stethoscope was used, depending on the age group being studied. An amplifier was connected to the microphone with the bandwidth of the amplifier extending to 3 kHz, which was chosen to provide realistic and faithful audio quality reproduction on listening. The signal was then passed to a MiniDisc digital recorder (Sony, Japan), which provided high quality sound reproduction with low noise. The operator listened through studio quality audio headphones during recording and playback. To obtain a controlled release of air from the arm cuff, the deflation device from an automatic BP monitor (A&D Blood Pressure Monitor, TM2421 [A&D Co., Ltd., Saitama, Japan]) was used to provide consistent cuff deflation. The average deflation rate was −2.9 mm Hg/beat, which is within the recommendations of the British Hypertension Society for Blood Pressure measurement.4 The cuff deflation was programmed to start at 160 mm Hg in children and rapidly deflate below 30 mm Hg. In adults the cuff deflated from 190 mm Hg and rapidly deflated after 35 mm Hg.

Measurement Procedure

The studies were carried out in a quiet room, in the school for children and in the University campus for adults. Participants sat quietly for at least 10 min before the measurements. A pressure cuff (bladder width at least 40% of arm circumference) was placed around the left arm and supported near heart level using a cushion. A Doppler ultrasound probe (Dopplex, Huntleigh Healthcare; Luton, Bedfordshire, UK) was used to locate the clearest pulsatile signal at the antecubital fossa and the skin marked with a waterproof pencil. Blood pressure was first measured with a mercury sphygmomanometer, using K5 (if present) for diastolic pressure.

Korotkoff Sound Analysis

There are four different types of Korotkoff sounds described when one listens at the antecubital fossa during arm cuff deflation. Each of the four sounds heralds a phase of similar sounds and thereby produces four corresponding Korotkoff phases (phases I, II, III, and IV). The cuff pressure at the onset of phase I sounds (often referred to as K1) is thought to represent systolic BP. The cuff pressure when the Korotkoff sounds disappear (end of phase IV) is clinically important, as it approximates diastolic pressure; this point is traditionally called K5. This point, from disappearance of the sounds (the end of phase IV) until the cuff pressure is zero, can be described as phase V. If the Korotkoff sounds persist to zero cuff pressure (K5 does not exist) there is no phase V. In this situation, the onset of phase IV (often referred to as K4) is taken as the diastolic BP.

The different Korotkoff sounds are identified by the character of the sounds, as outlined by Geddes et al.5 Phase I sounds are loud, with a clear-cut snapping tone; phase II sounds have a murmur-like quality; phase III sounds are similar in character to phase I sounds; and phase IV sounds have a dull or muffled tone. The fifth phase represents the complete disappearance of the sounds. In this study, the Korotkoff sounds were identified as either phase I, II, III, or IV by replaying the recordings of the sounds made during controlled cuff deflation. Because the cuff deflated rapidly below 30 mm Hg in children and 35 mm Hg in adults, it was not possible to say whether phase V was absent. However, for purposes of analysis, if sounds persisted to 30 mm Hg or to 35 mm Hg, respectively, phase V was said to be absent. If the first sounds heard had a murmur-like quality, phase I was said to be absent. The total length of the phases was measured and the number of beats in each phase counted.

There is clearly a subjective element in allocating the sounds to the various phases. We addressed this problem in a previous study in which we demonstrated, using blinded reanalysis, that the recorded sounds could be consistently allocated to the different Korotkoff phases by an experienced observer.3 In the preliminary work we also found that when two other doctors were asked to assess the sounds (after being educated on the character of the sounds), there was excellent agreement in identifying the various phases. The only discrepancy evident was in the length of phase IV; this was due to the gradual disappearance of phase IV sounds, making it sometimes difficult to decide whether the sound had disappeared or was barely audible. Therefore, to ensure consistency in the comparison among subjects, the same experienced observer made all measurements.

Statistical Methods

The differences in phase lengths between adults and children were assessed using the Mann-Whitney U test. The relationship between the BP and arm circumference was analyzed using standard regression plots. The difference in phases present or absent in the two groups was assessed using the χ2 test.

Results

Table 1 provides details of BP, arm measurements, and age of the individuals studied. Table 2 gives the number of children and adults in whom the different phases were detected. All phases were detected in 40% of children and in 41% of adults. Phase I was virtually always (98%) present in children but in only 80% of adults. Phases II and III were more common in adults than in children (P = .001), but phase IV was more common in children (P = .018). If phase II was present, phase III was likely to be present also (P < .0001). The different permutations of phases detected are outlined in Table 3. The most common patterns in both adults and children were for all phases to be present, but a number of other patterns were also found.

Table 1

Data on subjects (median and interquartile range)

Table 1

Data on subjects (median and interquartile range)

Table 2

Children and adults in whom the different Korotkoff phases were detected

Table 2

Children and adults in whom the different Korotkoff phases were detected

Table 3

Phase pattern of children (aged 7 to 8 years) and adults

Plus signs denote presence of phases.

Table 3

Phase pattern of children (aged 7 to 8 years) and adults

Plus signs denote presence of phases.

The differences in the length of the phases can be seen in Fig. 1. Phases I and IV were significantly longer in children (median 3.9 [interquartile range, or IQR, 2.1 to 6.7] and median 6.7 [IQR 3.2 to 9.8] sec, respectively) when compared with adults (1.3 [IQR 0.7 to 2.7], P < .001, and 1.7 (0.3 to 2.6), P < .001, sec, respectively). Phases II and III were significantly shorter in children (1.7 [IQR 0.0 to 2.9] and 0 [IQR 0.0 to 3.0] sec, respectively) when compared with adults (3.6 [IQR 2.1 to 4.7], P < .001 and 3.1 [IQR 0.8 to 4.3], P < .005, sec, respectively). The longest phases were IV in children and phase II in adults.

a to c). Difference in phase length between children and adults. Fig. presents the length of each phase in children (▾) and adults (▴) in terms of time (in seconds) (a), number of sounds in each phase (b), and proportion (% of total duration of sounds) occupied by each phase (c). Median and interquartile ranges are shown. *P < .01; **P < .005; ***P < .001.

Fig. 1.

Arm circumference was significantly correlated with BP (r = 0.55, P < .01, for children; r = 0.4, P < .01, for adults). Phase I was longer when phases II and III were absent (mean 13.7 v 1.7 beats, P < .0001 for adults and 12.2 v 3.4 beats, P < .001 for children). In the adult group, the length of phase I had a significant positive correlation (r = 0.32, P < .05) and phase II a significant negative correlation (r = 0.5, P < .01) with heart rate.

Discussion

In this study we investigated the pattern of the Korotkoff phases in a group of normal children and adults. At present, the identification of the Korotkoff sounds and their allocation to the various Korotkoff phases is a subjective assessment. We found that recording the sounds and replaying them allowed us to consistently allocate the sounds to the different phases.3 It is possible, however, that other investigators would get slightly different results. Nonetheless, to ensure a valid comparison between subject groups, the same experienced investigator made all of the measurements. Errors can be minimized by making recordings in a quiet room and by ensuring an optimal environment when listening to the recordings, as some of the sounds are of low intensity and often fade away rather than abruptly stop. In our study we were meticulous in ensuring that the recordings were made in a very quiet setting.

When systolic BP is measured in the clinic setting using the auscultatory (Korotkoff) method, the first sound heard on deflation of the cuff is taken as the systolic BP and is presumed to be the start of phase I. This is not always the case, however, as in 20% of the adults we studied phase I was not detected, and the start of phase II had to be taken as systolic BP. This is not something that clinicians would take note of, and it may be of no clinical significance; but it highlights the fact that little attention is paid to the character of the sounds heard when measuring systolic BP.

The character of the sounds is known to be important in measuring diastolic BP, as this is how the onset of phase IV (muffling, K4) is recognized. There is some debate about whether the onset of phase IV or the disappearance of sounds (end of phase IV or onset of phase V) best represents diastolic BP, particularly in children. In this study we found the median length of phase IV was nine beats (6.7 sec, IQR 3.2 to 9.8 sec) in children, and two beats (1.7 sec, IQR 0.3 to 2.6 sec) in adults. Under the conditions of this study the measurement of diastolic BP in children, using the onset of phase IV or phase V, would provide quite different results. It was also noticeable that phase IV sounds often faded rather than abruptly terminating, reinforcing the importance of low background noise when listening for phase V.

Phases II and III have received almost no attention in the medical literature. We found that 85% of adults have phase II and 76% have phase III. These phases were detected in only approximately 56% and 47%, respectively, of the children (aged 7 to 8 years) studied, which is similar to our previous findings in 11-year-old children, of whom 61% had phase II detected and 51% had phase III.3 Because the mechanism responsible for the Korotkoff sounds is not completely understood,6,7 we can only speculate as to the significance of these differences between children and adults in Korotkoff phase distribution. The difference in heart rate between adults and children may be an important factor in determining the relative duration of the phases, but differences in vessel compliance and sympathetic nervous system activity are just some of the other factors that likewise may be important. This question was not formally addressed in this study and is clearly an area in which further study is needed.

Our study shows that all five Korotkoff phases are present in approximately 40% of normal adults and children, but a number of different patterns of Korotkoff phase distribution are found. There is a need for research into the relationship between the various phases and vessel compliance, as this may be the major factor accounting for the differences in phases II and III between children and adults observed in our study. The relationship between diastolic BP and the Korotkoff sounds needs further study; in particular, it would be interesting to investigate whether the pattern of phases is significantly different in patients with hypertension.

To our knowledge the only other clinical studies of the Korotkoff phases were published more than 80 years ago.8,9,10 As we approach the centenary of the original description of the Korotkoff sounds, we should refocus attention on the forgotten Korotkoff phases.

Acknowledgments

We are indebted to Cragside Primary School and Benton Park Primary School, Newcastle upon Tyne, UK, for their cooperation and enthusiasm. We thank the staff of Castle Leazes Halls of Residence for participating in the study and Mrs. Toni Allen, research assistant, for her patient organization of the study.

References

1.:

On the subject of methods of determining blood pressure [in Russian]

,

Bulletin of the Imperial Military Medical Academy, St Petersburg

1905

;

11

:

365

.

2., :

Blood pressure measurement in childhood epidemiological studies

.

Circulation

1995

;

92

:

1049

–

1057

.

3., , :

Korotkoff phase distribution in children

.

J Hum Hypertens

2001

;

15

:

197

–

201

.

4., , :

Blood pressure measurement. Conventional sphygmomanometry: technique of auscultatory blood pressure measurement

.

Br Med J

2001

;

322

:

1043

–

1047

.

5., , :

Introduction of the auscultatory method of measuring blood pressure—including a translation of Korotkoff’s original paper

.

Cardiovasc Res Cent Bull

1966

;

5

:

57

–

74

.

6., , :

The Korotkoff sound

.

Ann Biomed Eng

1989

;

17

:

325

–

359

.

7., :

Korotkoff sounds. An experimental critique

.

Circ Res

1967

;

20

:

149

–

161

.

8.:

The auscultatory method of blood pressure determination: a clinical study

.

Internat Clin

1914

;

4

:

130

–

190

.

9., :

Clinical studies in the auscultatory method of determining blood pressure

.

Univ Penn Med Bull

1910

;

23

:

469

–

475

.

10., :

Further clinical studies in the auscultatory method of determining blood pressure

.

Am J Med Sci

1911

;

142

:

334

–

353

.

Author notes

*

This research was supported by a grant from the Trustees of Freeman Hospital, Newcastle upon Tyne, UK.

© 2002 by the American Journal of Hypertension, Ltd.

American Journal of Hypertension, Ltd.

© 2002 by the American Journal of Hypertension, Ltd.

How are Korotkoff sounds produced?

What causes the first korotkoff sound?

What does the 1st sound heard when taking a BP represent?

What is the 1st sound or the korotkoff sound that you hear in taking blood pressure?