cranial rhythmic impulse related to the traube-hering-mayer oscillation

JAOA Vol 101 No 3 March 2001 163Nelson et al Original contribution

In 1865, Traube1 reported the mea-surement of a fluctuation in pulse pres-sure with the frequency of respirationthat persisted after respiration had beenarrested. These findings were corrobo-rated by Hering 2 in 1869. Independent-ly, Mayer3 identified similar oscillationswith a slower rate in 1876. These phe-

nomena, now known collectively as theTraube-Hering-Mayer (THM) oscilla-tion,4 have been measured in associationwith blood pressure,1,4-10 heart rate, 4,10,11cardiac contractility,12 pulmonary bloodflow,13 cerebral blood flow and move-ment of the cerebrospinal fluid,14,15 andperipheral blood flow, including venousvolume and thermal regulation.4,8,10,16This whole-body phenomenon, whichexhibits a rate typically slightly less thanand independent of respiration, bears astriking resemblance to the primary res-piratory mechanism (PRM).

The PRM was first described bySutherland17 in 1939. In 1961, Woods18coined the term cranial rhythmic impulse(CRI) to describe the palpable sensationof the PRM. The PRM is considered to beintegrally linked to cellular metabolism.Magoun19 stated: This cycle manifests as

the cranial rhythmic impulse and repre-sents a dynamic metabolic interchangein every cell, with each phase of action.The PRM affects all regions of thebody.19-24 The CRI, as a manifestationof the PRM, is most evident to palpationin the head. It is, however, palpable inevery part of the body.24 It is described interms of amplitude and rate, with mostauthors identifying the normal rate as 10to 14 cycles per minute.18,19,21,23,24 ThePRM/CRI, though frequently syn-chronous with respiration, has a period-icity independent of the cardiac and res-piratory cycles.18,20,24-26

The PRM/CRI is a decidedly contro-versial aspect of osteopathic medicine. Itis a subtle enough phenomenon to beeasily overlooked by untrained clinicians.This has led many to doubt its existenceand incorrectly state that it has not beenmeasured. Many authors,18,20,26-30 how-ever, have reported measuring the CRI.

Fryman,20 Upledger and Vrede-voogd,22 Geiger,31 and MacPartland andMein31a have commented on the similar-ity between the THM and the CRI. Fer-nandez and Lecine28 attempted to mea-sure both phenomena simultaneously butdid not demonstrate a statistically signif-icant relationship.

We developed a protocol to measurethe THM and the PRM/CRI simultane-ously.

MethodsSubjectsTwelve healthy subjects over 18 years ofage (6 males; 6 females, none pregnant)were recruited from the faculty and stu-dents of the Chicago College of Osteo-pathic Medicine (Midwestern Universi-ty, Downers Grove, Ill). All signedIRB-approved informed consent forms.

Conditions and protocolCranial examinations were performed byone of the authors (N.S.) in sequence in 1day in a quiet (essential for instrumenta-tion) small office, with low-level illumi-nation provided by incandescent lightingand an ambient temperature of ap-proximately 20C. Prior to each exami-nation, an adhesive Doppler probe wasplaced onto the left earlobe of each sub-

Dr Nelson is a professor in the Department ofOsteopathic Manipulative Medicine, ChicagoCollege of Osteopathic Medicine of Midwest-ern University, Downers Grove, Illinois, whereMss Lipinski and Chapman are medical stu-dents and Dr Glonek is a research professor.Ms Sergueef currently lectures and teachesmanual principles, diagnosis, and treatmentthroughout Europe and maintains a privatemanual therapy practice in Corbas, France.

Correspondence to Kenneth E. Nelson,DO, Family Medicine, Midwestern Universi-ty, 555 31st St, Downers Grove, IL 60515.

E-mail: [email protected]

Cranial rhythmic impulse related tothe Traube-Hering-Mayer oscillation:comparing laser-Doppler flowmetryand palpation

KENNETH E. NELSON, DO; NICETTE SERGUEEF; CELIA M. LIPINSKI, MSII;ARINA R. CHAPMAN, MSII; THOMAS GLONEK, PhD

Original contribution

The primary respiratory mechanism (PRM) as manifested by the cranial rhythmicimpulse (CRI), a fundamental concept to cranial osteopathy, and the Traube-Her-ing-Mayer (THM) oscillation bear a striking resemblance to one another. Becauseof this, the authors developed a protocol to simultaneously measure both phenomena.Statistical comparisons demonstrated that the CRI is palpably concomitant withthe low-frequency fluctuations of the THM oscillation as measured with the Tran-sonic Systems BLF 21 Perfusion Monitor laser-Doppler flowmeter. This opensnew potential explanations for the basic theoretical concepts of the physiologic mech-anism of the PRM/CRI and cranial therapy. Comparison of the PRM/CRI withcurrent understanding of the physiology of the THM oscillation is therefore war-ranted. Additionally, the recognition that these phenomena can be simultaneous-ly monitored and recorded creates a new opportunity for further research into whatis distinctive about the science and practice of osteopathic medicine.

(Key words: osteopathic medicine, cranial osteopathy, primary respiratorymechanism, cranial rhythmic impulse, cardiovascular system, lymphatic system, laser-Doppler flowmetry, Traube-Hering-Mayer oscillation)

ject. Following this, each subject lay qui-etly on the examination table for approx-imately 5 minutes prior to the onset ofdata acquisition. It was essential that theprobe and leads were free of tension sothat earlobe blood flow was not com-promised. Laser-Doppler flowmetry pro-vided measurement of the relative veloc-ities of blood (hemoglobin) while simul-taneously, the examiner, at the head of thetable, palpated the CRI using light touchwith the hands in a biparietal-hold posi-tion. Both examiner and subject wereblinded to activities at the computer screenand keyboard.

A 2-minute-long equilibration-periodrecord was recorded. Following the equi-libration period, the examiner would,using a subdued voice, enunciate f atthe instant a cranial flexion event began,and a second investigator, operating thecomputer console in the examinationroom, would depress a key entering thefirst event mark into the digitized flowme-try record. Thereafter, serial extension(e) and flexion (f) events weremarked into the record. This method ofrecording could introduce a 150-mil-lisecond reaction-time delay into the CRIrecord, 300 milliseconds if the reactiontimes of both the examiner and therecording investigator are considered.32Continuous, unbroken records of approx-imately 5 minutes duration were record-ed for each subject.

Laser-Doppler flowmetryThe perfusion monitor (Transonic Sys-tems Inc, Ithaca, NY) determines theDoppler velocity change of the erythrocyte(hemoglobin) in circulating blood, andthat information is digitized for subse-quent data reduction. The device uses anoptic fiber probe that rests on the skinsurface, which causes no discomfort to thesubject. The probe (type R) has two opticfibers: one sends laser light into the tissue,the other transfers the light reflected fromthose tissues to a photo detector for sub-sequent electronic processing (WinDaqData Acquisition and Playback Software,also from Transonic Systems).

Data reduction and statisticsThe acquisition software was able to

export both the time-domain and fre-quency-domain data in database format,permitting further data processing asrequired. Palpatory and flowmetry time-domain data were compared using thepaired samples t-test statistic (2-tailed sig-nificance),33 with statistical power deter-mined by the method of Dupont andPlummer.34 Comparisons of mean THMfrequencies among subjects were evalu-ated using the one-way analysis of vari-ance in conjunction with the Bonferronirange test for pair-wise comparisons (sig-nificance, P .05).33

ResultsCombined laser-Dopplerflowmetry and CRI records:qualitative overviewFigure 1 presents two different subjectrecords that display all of the featuresobserved in the 12 records of this study.The oscillating waveforms are records ofthe flowmetry relative blood velocitychanges detected at the earlobe (recordedas an output voltage); the vertical eventmarks denote the flexion and extensionevents detected through palpation (enteredby technician). In the 7-minute periodpresented in Figure 1, the high-frequencyoscillations of the Doppler record, such asthose arising from the heartbeat, are com-pressed and appear as spectral noise sothat the low-frequency oscillations may beobserved easily. The earliest event mark(extreme left mark) of each recorded seg-ment denotes a flexion event. Thereafter,the event marks denote alternate extensionand flexion throughout the sequence.

In Figure 2 (top), a portion of subject2s record is expanded along the timeaxis to reveal the fine structure of the car-diac cycle and the relationship of thiscycle to the THM oscillation. Three THMcycles are displayed. The amplitudes of theTHM waves are nearly double those ofcardiac cycle waves. The THM cycle cor-responds to large relative changes in bloodvelocity and, therefore, is of a magnitudethat ought to be readily palpable throughosteopathic techniques, particularly ifthese blood velocity changes are propor-tional to blood volume changes (com-pare with Burch and colleagues35).

Laser-Doppler detection produces

blood velocity records with considerablymore detail than could be obtained withearlier technologies, permitting the obser-vation of signal fine-structure and moreprecise signal quantification. At the timescale presented in Figure 1, several featuresare apparent: (1) The record oscillates ata regular frequency of approximately 0.1cycles/sec (6 cycles/min). (2) The ampli-tude of the signal is not constant butinstead oscillates between relatively high-er and lower values, giving the overallrecord the appearance of waves uponwaves. (3) The waveform exhibits nodeswhere the signal is suppressed or greatlydistorted. These nodes represent beatnotes caused by the destructive interfer-ence of multiple, overlapping componentsignals. Constructive interference givesrise to signals of high amplitude. (4)Because of interference effects, the phaseof the THM signal changes with time.Thus, it is not possible to relate, for exam-ple, a flexion event as always being asso-ciated with a flowmetry maximum. Aftera signal inversion, flexion would corre-spond to a Doppler minimum. What canbe determined and correlated with thepalpatory findings is the periodic rate,defined by the positions in time of themaxima and minima. Wave apices (max-ima/minima) correspond to time pointswhere the change in blood velocity pass-es through zero (Figure 2). (5) The fre-quency of the oscillation is not constantbut, as will be demonstrated later, varieswith time in a regular fashion. [The peak-to-peak variation in time is apparent in therecord of subject 9 (Figure 1).] Signalvariation with time indicates frequencymodulation among interacting signals36and implies the presence of a capacitancein the physiologic mechanism responsiblefor blood velocity oscillation.8,16

The event-mark sequences of Figures1 and 2 exhibit the following features:(1) The sequence mimics the slow-wave[0.1 cycles/sec (6 cycles/min)] Traube-Hering component of the Doppler record.For virtually all of the recorded events, theevent marks correspond in time to eithermaxima or minima in the Doppler record.(2) Event marks do not appear for eachmaximum or minimum; rather, the eventmarks appear to correspond to each full

Nelson et al Original contribution164 JAOA Vol 101 No 3 March 2001

oscillatory cycle of a maximum plus aminimum. That is, event marks denoteadjacent maxima or minima. (3) Althoughevent marks generally mark maxima orminima, there are instances when themark falls between these. Thus, in all ofthe records examined, the correspondenceof marks and signal apices exhibits someirregularity. (4) At nodes in the Dopplerrecord, the CRI still can be palpated, withthe regularity of the event record usuallybeing maintained and with no palpablechange in CRI amplitude. (5) As with theDoppler record, the frequency of eventmarks varies with time in a regular fash-ion, and these variations correspond withthose of the Doppler record.

Laser-Doppler power spectrum:qualitative overviewFigure 3 presents the power spectrum ofa 13-minute flowmetry record segmentusing an input-averaging value of 10. Thepower spectrum is computed from time-domain data (energy vs. time), such aspresented in Figures 1 and 2, through aFourier transformation to generate a fre-quency-domain spectrum (energy vs. fre-quency). The spectrum of Figure 3 revealsthe component signals that are combinedin the original time-domain Dopplerrecord. The frequencies of individual sig-nals are indicated on the abscissa, whilethe relative signal amplitudes (areas) arerelated to the contribution each signalmakes to the original time-domain record.Signals at frequencies greater than approx-imately 0.5 cycles/sec (30 cycles/min) areattributable to the pulse (cardiac cycle)and will not be considered further here.These signals give rise to the rapid beatsseen in Figures 1 and 2.

The low-frequency components thatgive rise to the THM oscillation in theDoppler record lie in the frequency rangeof 0 to 0.5 cycles/sec (30 cycles/min). Forreference, the signal at 0.32 cycles/sec (19cycles/min) has been assigned to the res-piration rate,4,37-41 while the signal at 0.02cycles/sec (1.2 cycles/min) has been at-tributed by various authors to the thermalregulatory component.8,16,42,43 Of inter-est to this study is the high-amplitude sig-nal at 0.10 cycles/sec (6 cycles/min) whoseorigins are not fully understood,1-4,44 but

Nelson et al Original contribution JAOA Vol 101 No 3 March 2001 165

Figure 1. Compressed laser-Doppler flowmeter relative blood velocity (waveform)and flexion-extension records (vertical event marks) from two subjects (N, interfer-ence nodes).

Figure 2. Expanded laser-Doppler flowmeter relative blood velocity record of sub-ject 2. TopFlowmeter trace, revealing fine structure of the cardiac cycle. The dou-ble-headed arrow indicates the wavelength of one Traube-Hering-Mayer cycle. Bot-tomTraube-Hering-Mayer wave form component only of the top trace. Thebottom waveform was created from the top waveform by filtering (removing) the high-frequency cardiac component, leaving only the low-frequency baro and respiratorycomponents. Inverse Fourier transformation of this filtered data generates the bot-tom trace. Both traces are in register with respect to time, and the event markers indi-cate the positions of the palpatory findings.

which has been assigned to baroreceptorfunction.45 The signal amplitudes shownin Figure 3 indicate that these three signalscomprise the principal components of theslow-wave oscillations observed in theoriginal flowmetry time-domain record.This is illustrated in Figure 4 (bottom),which is an inverse Fourier transformspectrum of the data of Figure 3 afterhigh-frequency filtering. This filtering pro-cess removed those components of theoriginal record that were above 0.5cycles/sec (30 cycles/min).

Minor signals also are observed in thepower spectra that have not been as-signed. Of particular note is a signal at0.39 cycles/sec (23.4 cycles/min). Oscil-lations in this frequency range have beenattributed to lymphatic vasomotion.46-52Some of the minor signals undoubtedlyrepresent harmonics of the major signals;others may represent fundamental phys-iologic functions. Each, however, willcontribute added complexity to the time-domain record.

Combined record: quantification oftime-domain dataDescriptive statistics. Twelve subjects par-ticipated in the study. Of these, 11 pro-vided high-quality data for analysis. Forsubject 12, the signal-to-noise ratioobserved in the laser-Doppler (time-

domain) output was too low for precisequantitative measurement. However, theFourier transform (frequency-domain)record of subject 12 included all of the fea-tures observed for the other 11 subjects.

Recorded frequencies for maxima andminima were distributed uniformlyamong the 11 test records (n 613;mean, 56; range, 39 to 77). There were166 flexion events and 162 extensions(n 328) associated equally betweenmaxima (n 164) and minima (n 164), with no correlation between theoccurrence of a maximum or minimumand the palpation of a flexion or extensionevent (Pearsons R value, 0.085; approx.sig., 0.123).Laser-Doppler flowmetry compared withpalpation. The time at which a maxi-mum or minimum occurred in theflowmetry record was compared with thetime recorded for the nearest flexion orextension event. For almost all flexionor extension events, association of theevent with the nearest maximum or min-imum presented no difficulty (Figure 2).For 2% of events, signal assignment wasproblematic because the flexion/exten-sion event mark fell either approximate-ly midway between maxima or minima orbecause the signal-to-noise ratio at thatpoint was insufficient for precise maxi-mum or minimum location. It was pos-

sible, however, to resolve all assignmentambiguities through the use of spectralfiltering and expansion routines.

Considering all flexion or extensionevents recorded for the 11 subjects withhigh-quality data, and their correspond-ing flowmetry maxima or minima, thetwo groups are the same. By the paired t-test there was no statistical differencebetween the time values recorded for pal-pated flexion or extension events and thecorresponding flowmetry maxima or min-ima (N 328; mean difference betweenpairs, flowmetry time-palpation time,0.078; SD, 1.361; 2-tailed sig., 0.303).

As may be anticipated from Figures1 and 2, both groups of time values werehighly correlated (N 328 data pairs;correlation, 1.000; sig., 0.000). The sta-tistical comparison was repeated afterintroducing a 150-millisecond reactiontime delay for the technician into the flex-ion/extension record. This computationcaused a slight convergence of the twomeans with a change in sign of the meandifference (mean difference between pairs,0.072; SD, 1.361; 2-tailed sig., 0.336).The computed statistical power34 was0.527 (alpha, 0.336; difference, 0.0724;sigma, 1.3614; N 328). The additionof a second 150-millisecond reaction timedelay for the examiner (300 millisecondtotal introduced delay), however, caused


Figure 3. Power spectrumof the laser-Doppler flow-meter record presented inFigure 1 subject 2. A, ther-mal signal. B, baro signal.C, respiration.


Figure 4. Inverse Fourier transform time-domain spectra from subject 2; top spectrum, all frequency domain data used in theinverse computation; bottom spectrum, only the frequency component lying below 0.5 cycles/sec (30 cycles/min) used. Thebottom spectrum is the THM oscillation. The insert box shows that portion of the Fourier transform spectrum used to com-pute the inverse THM spectrum (see Figure 3).

Table 1Descriptive Statistics (11 Subjects) for the Relative Blood Velocity Oscillation

Determined by Laser-Doppler Flowmetry

10.208.59

10.879.979.94

13.477.59

12.0311.759.63

11.67

10.35

1.890.932.192.222.051.381.041.742.051.842.46

2.42

0.250.120.320.290.300.200.120.230.33 0.210.32

0.099

9.698.34

10.219.379.32

13.067.33

11.5511.069.21

11.01

10.15

10.708.85

11.5310.5710.5513.897.85

12.5112.4310.0612.32

10.54

4.806.367.205.087.70

11.525.607.048.006.245.12

4.80

14.8812.0014.4016.0016.8016.6410.1817.8016.2013.4417.92

17.92

123 56789101112

Total

5755455546456653377557

591

No. ofcycles

Mean cycleperiod

(sec/cycle)Standarddeviation

Standarderror

95% Confidenceinterval (mean)

Lowerbound

Upperbound

Minimum MaximumSubjectNo.


Table 2Multiple Comparisons of the Traube-Hering-Mayer Component of the Relative Blood Velocity

Oscillation Laser-Doppler Record (sec/cycle; One-way Analysis of Variance, Range Test: Bonferroni)

1

2

3

5

1.60*-0.670.220.26

-3.27*2.60*-1.83*-1.55*0.56-1.47*

-2.27*-1.37*-1.34*-4.88*1.00-3.44*-3.15*-1.03-3.07*

0.890.93-2.60*3.27*-1.16-0.881.23*-0.80

0.032-3.50*2.37*-2.06*-1.78*0.33-1.70*

0.3490.3680.3490.3660.3680.3340.3520.3900.3240.346

0.3710.3520.3690.3710.3370.3550.3920.3280.349

0.3710.3870.3890.3570.3740.4100.3480.368

0.3690.3710.3370.3550.3920.3280.349

0.0001.0001.0001.0000.0000.0000.0000.0041.0000.001

0.0000.0060.0160.0000.1680.0000.0000.0880.000

0.8580.8950.0000.0000.1051.0000.0231.000

1.0000.0000.0000.0000.0001.0000.000

0.44-1.89-0.93-0.95-4.501.49-3.01-2.85-0.51-2.62

-3.51-2.54-2.57-6.11-0.12-4.62-4.46-2.13-4.23

-0.33-0.35-3.902.08-2.41-2.240.076-2.02

-1.19-4.741.25-3.25-3.08-0.76-2.86

2.760.551.391.48-2.053.71-0.66-0.251.64-0.32

-1.03-0.20-0.11-3.642.12-2.25-1.840.053-1.91

2.132.22-1.304.460.0810.482.390.42

1.26-2.263.50-0.88-0.471.42-0.53

2356789

101112

356789

101112

56789

101112

6789

101112

(J)Subject

No.

Meandifference

(I-J)Standard

errorSignificance 95% Confidence interval

Lowerbound

Upperbound

(I)Subject

No.

*Mean difference is significant at the .05 level.

the two means to diverge significantly(mean difference between pairs, 0.222;SD, 1.361; 2-tailed sig., 0.003).Mean frequency variance among sub-jects. The mean oscillatory frequenciesdetermined by both laser-Doppler flow-metry and palpation (these are equiva-lent) differed among subjects even thoughthe experimental circumstances (supine,awake, and at rest) were essentially iden-tical for all (Table 1). Pair-wise compar-isons of computed means (Table 2) alsowere tested for significance by the one-

way analysis of variance (P .05). Of the55 possible combinations of study subjectpairs, 38 (69%) differed significantly.Frequency variability. As mentioned pre-viously, examination of either the flow-metry or the palpation records of all sub-jects revealed the presence of both higher-frequency and lower-frequency segmentsthat alternated in a periodic manner, sug-gesting the presence of a frequency-mod-ulated component.53 Moreover, these fre-quency changes were not related in anyobvious way to analogous amplitude

changes that also occurred in all records.To document the periodic frequencychanges and to assess their magnitudeswithin and among individual subject re-cords, two contiguous groups of approx-imately 10 complete cycles each, the short-est group and the longest group, wereselected from each subject record. Allselected sequences were separated withineach subject record by at least 2 minutes.Frequencies for each cycle (in cycles/min)were computed from the time data, andpairs of short and long segments within


Table 2 (continued)Multiple Comparisons of the Traube-Hering-Mayer Component of the Relative Blood Velocity

Oscillation Laser-Doppler Record (sec/cycle; One-way Analysis of Variance, Range Test: Bonferroni)

6

7

8

9

10

11

-3.53*2.34*-2.09*-1.81*0.30-1.73*

5.88*1.44*1.72*3.84*1.80*

-4.44*-4.15*-2.04*-4.07*

0.282.40*0.36

2.11*0.079

-2.03*

0.3870.3540.3720.4080.3460.366

0.3570.3740.4100.3480.368

0.3400.3790.3110.334

0.3950.3310.352

0.3710.390

0.324

0.0000.0000.0000.0011.0000.000

0.0000.0070.0020.0000.000

0.0000.0000.0000.000

1.0000.0001.000

0.0001.000

0.000

-4.821.16-3.33-3.17-0.84-2.95

4.690.190.362.680.57

-5.58-5.42-3.07-5.19

-1.031.29-0.80

0.88-1.21

-3.11

-2.243.52-0.85-0.451.45-0.51

7.072.683.095.003.03

-3.30-2.89-1.00-2.96

1.603.501.53

3.351.37

-0.95

789

101112

89

101112

9101112

101112

1112

12

(J) Subject

No.

Meandifference

(I-J)Standard

errorSignificance 95% Confidence interval

Lowerbound

Upperbound

(I) Subject

No.

*Mean difference is significant at the .05 level.

each subject record were compared by t-test for equality of means. Two exam-ples, one from a higher-frequency (subject8) and one from a lower-frequency (sub-ject 11), are presented in Table 3. Themean differences between higher- andlower-frequency sets within each subjectrecord are statistically significant (P .01), with the high-frequency (subject 8)exhibiting a frequency difference of 1.51cycles/min and the low-frequency (subject11) exhibiting a difference of 1.73 cy-cles/min.

CommentsThe statistical comparisons demonstratethat the CRI is concomitant with low-frequency fluctuations in blood velocity asmeasured with the Transonic SystemsBLF 21 Perfusion Monitor laser-Dopplerflowmeter. Moreover, the observed con-comitance is sustained at all points in therecord even though blood-velocity-changefrequency may vary, with periodicity, asmuch as 20% (Table 3). These findingsimply that the PRM/CRI and the THMoscillation are simultaneous, if not thesame phenomenon. This opens new pos-sible explanations for the basic theoreti-cal concepts of the physiologic mecha-nism of the PRM/CRI and cranialtherapy. Comparison of the PRM/CRIwith current understanding of the phys-iologic mechanism of the THM oscillationis, therefore, warranted.Primary respiratory mechanism. ThePRM is described as the driving forceassociated with the activity of cellularmetabolism.17,19,21,24 The THM oscilla-tions are intimately involved in the regu-lation of peripheral blood flow and, con-sequently, tissue perfusion.4,8,10,16Circulatory and body core temperaturehomeostasis are considered to be a resultof the THM oscillation.8 A hypotheticalexplanation for the PRM can be devisedby employing our understanding of theTHM oscillation.

The THM oscillation affects all tis-sues of the body through its impact on theentire circulatory system. Blood pumpedfrom the heart, the rhythm of which fluc-tuates under the influence of the THMoscillation,11,12 arrives in all of the capil-lary beds in the body via arteries and

arterioles whose walls are contractingsynchronously at the THM frequen-cy.8,10,16 Blood pressure and capillaryblood velocity are consequently oscillat-ing at the THM frequency.

Cellular metabolic exchange occurswithin the interstitial space. As much as99.9% of interstitial fluid exists in a gel-like state. Fluid cannot move as freelythrough this gel as it can in a purely liq-uid medium. The interstitial gel, howev-er, demonstrates elasticity.54 Somethingmust facilitate fluid movement in theextracellular space.

The THM blood flow and pressureoscillation in the thicker-walled arterialsystem are less constrained as the bloodpasses through the capillaries and entersthe venous system. The capacitance ofthese thin-walled veins allows for signif-icantly greater volume fluctuations withproportionate displacement of adjacentstructures. In the capillary beds, local con-tractile responses to distention have beendemonstrated in some tissues.5(p127)

It has been suggested that the arterio-lar and venular vasomotion and bloodpressure fluctuation that result from theTHM oscillation aid in the distributionand mixing of all of the extravascularfluids and may mechanically facilitate thepassage of fluid through capillary andlymphatic walls.55 Thus, local (direct)and neural (indirect) control mechanismsact synergistically to satisfy the metabol-ic demands of the peripheral tissues.Locally, the activity of the musculature ofthe vascular bed is modified and inte-grated by changes in the composition ofthe extracellular fluid. Neural control isexercised via specialized sensory endingsof peripheral afferent cells within the inte-grative centers of the central nervous sys-tem. Response occurs to varying levelsof oxygen, carbon dioxide, and hydro-gen ion concentration, as well as tem-perature of the blood and extracellularfluid.56 Or, as Magoun19 proposed of thePRM, the THM oscillation facilitatesdynamic metabolic interchange in everycell, with each phase of action.Whole-body phenomenon. The PRM/CRI and the THM oscillation share thequality of being demonstrable throughoutthe body. Multiple authors describe the

PRM/CRI as palpable in all areas of thebody.19-24 The THM oscillations are rec-ognized to occur throughout the body.57They have been measured simultaneous-ly in the right index finger, right secondtoe, and pinna of the right ear.35Rate of the PRM/CRI. The PRM/CRIhas a generally agreed-upon rate of 0.17to 0.23 cycles/sec (10 to 14 cy-cles/min).17,18,21,23,24 Many researchershave measured the CRI, using variousmethods, on human subjects18,20,25-29 andanimals,30,58 identifying rates between0.083 and 0.23 cycles/sec (5 and 14 cy-cles/min).

Further, Becker59 describes two com-ponents to the PRM/CRI. They are thefast tide, at a rate of 0.13 to 0.20cycles/sec (8 to 12 cycles/min), and theslow tide, at a rate of 0.01 cycles/sec(0.6 cycles/min). Fryman20 attempted todetermine what relation might existbetween peripheral volumetric changesand the rhythmic cycles of the cranium byusing a pressure transducer applied tothe head to monitor the CRI, in con-junction with plethysmography applied tothe arm or finger. In so doing, she notedthat cranial motion recordings coincid-ed with appendicular volume changes.Plethysmography also demonstratedlong slow cycles from 50 to 60 secondsduration, which were thought not to berelated to cranial changes.20 Upledger andKarni,29 using plethysmography to mon-itor the CRI, identified a frequency of(0.15 to 0.18 cycles/sec) 9 to 11 cpm andan even slower frequency of (0.02 to 0.03cycles/sec) 1 to 2 cpm.

Within the THM oscillation, two dis-tinct frequencies exist that are significantto this discussion. The Traube-Heringcomponent of the THM oscillationdemonstrates a frequency range of 0.09 to0.17 cycles/sec,8-10,16,35,60 or 5 to 10 cycles/min. The Mayer component demonstratesa frequency range from 0.01 to 0.09 cy-cles/sec,9,10 or 0.6 to 5.4 cycles/min. Itcould be argued that a single frequency ofoscillation shared between the PRM/CRIand the THM oscillation is coincidental;however, the occurrence of two distinctfrequencies shared between the PRM/CRIand the THM oscillation renders coinci-dence a highly improbable explanation.


Relationship between the PRM/CRI andpulmonary respiration. Pulmonary res-piration has always been recognized asclosely associated with, yet independentof, the PRM/CRI. The respiratory coop-eration of the patient is often employedin association with cranial treatment.17,59Cranial manipulation has been said toaffect respiration,61 and spontaneous deepsighing respiration has been reportedcoincidental with the therapeutic endpoint.23 As shown by the Fourier analy-sis (Figures 3 and 4), the component partsof the total THM oscillation, that is,Mayer (0.01 to 0.09 cycles/sec), Traube-Hering (0.09 to 0.17 cycles/sec), and res-piration (0.2 to 0.3 cycles/sec), are dis-tinctly separate.4,9,10,53,62 The componentsat 0.01 to 0.09 cycles/sec and 0.09 to0.17 cycles/sec, however, are closelylinked to, and may be modulated by, thepulmonary respiration component at 0.2to 0.3 cycles/ sec.39,40Fluctuation of the cerebrospinal fluidand motion of the central nervous system.The PRM/CRI is described as consistingof five distinct component parts.17,19,21-24Of these, the fluctuation of the cere-brospinal fluid (CSF) and the inherentmobility of the central nervous system(CNS) are of particular interest in thecontext of the THM oscillation. Motionof the brain63 and motion of the CSF insynchrony with the cardiac cycle hasbeen demonstrated using magnetic reso-nance velocity imaging.63,64 For a briefperiod during systole, there is a net inflowof blood into the brain, causing it toexpand in volume. This causes the centralportion of the brain and the brainstem tobe displaced caudally. An amount of CSFapproximating the volume change of the

brain is displaced from the ventricles andintracranial subarachnoid space.

During systole, the CSF moves medi-ally from the lateral ventricles of the cere-bral cortex into the third ventricle andin a craniocaudad direction from the thirdventricle into the fourth ventricle and inthe subarachnoid space surrounding thespinal cord. During diastole, with thereturn of the caudal displacement of thebrain, CSF motion reverses direction. Thefourth ventricle acts as a mixing cham-ber to allow for CSF oscillation. Thesystem must demonstrate capacitance.The spinal dural sac acts as the requiredcapacitor.

Oscillations of cortical metabolism(9.58 cycles/min in an awake, nonanes-thetized cat) with associated fluctuationsin blood volume have been demonstrat-ed by reflectance spectrophotometry ofthe cortical cytochrome oxidase redoxstate. The data65 suggest that the cyclicincreases in cortical oxidative metabolismrepresent the primary oscillatory process,followed by reflex hemodynamic changesthat affect intracranial blood volume.Volume oscillations representing pre-sumed THM waves have been measuredin the brains of conscious healthy humansusing ultrasound.66

As the intracranial blood volume in-creases, CSF is displaced from the interi-or of the brain case into the extracranialsubarachnoid space, increasing the amountof CSF in the spinal dural sac capacitor. Asintracranial blood volume decreases, thetension of the spinal dural sac facilitates thereturn of CSF into the skull. Because of thissynchronicity with and relation to THMoscillation, the CSF could be described asebbing and flowing.

Entrainment and manipulation of thePRM/CRI. Manipulative treatmentdirected at affecting the PRM/CRI isoften used to modulate the rate (fre-quency), amplitude, and direction of thewave.19(p339) Respiratory cooperation ofthe patient can be used in associationwith cranial treatment.17,59 Spontaneousdeep sighing respiration has been report-ed coincidental with the therapeutic endpoint.23

Entrainment of frequency occurswhen two nonlinear oscillatory systemsare coupled and operating at close butdifferent frequencies.8,16,36,45 The cou-pling causes the two oscillators to lockinto a common frequency. The THMoscillation has been entrained using rhyth-mic alteration of body position,45 expo-sure to fluctuating temperature,16 andrespiratory activity.4,38,39,45 Entrainmentof THM has been accomplished usingbaroreceptors and vasomotor reflexes;the lower limit of the entrainment band-width is 0.0841 (SD, 0.0030) cycles/sec,and the upper limit is 0.1176 (SD,0.0013) cycles/sec.49 Entrainment of theTHM by the respiratory rate specifical-ly occurs over a similar frequency rangeof 5 breaths/min (0.083 cycles/sec) to 7breaths/min (0.12 cycles/sec).40 Althoughcranial manipulation involves more com-plexity of intervention than merely mod-ulating the PRM/CRI, the concept ofoscillatory entrainment offers an inter-esting explanation for this one aspect oftreatment, as has been proposed by Mac-Partland and Mein.31a

ConclusionThe results of this study indicate that thePRM/CRI and the THM oscillation occur


Table 3Comparisons of Shorter and Longer Oscillations

Within Individual Subject Records (t-test for Equality of Means)

ShortLongShortLong

99

1110

6.498.00

8.8810.61

1.51

1.73

0.001

0.006

8

11

0.630.93

0.721.68

Short orlong cycles

Mean(sec/cycle)

Standarddeviation

N Mean difference

Significance(2-tailed)

SubjectNo.

simultaneously, though they may or maynot represent the exact same phenom-enon. Like blood pressure, serum glu-cose, and all other diagnostically signifi-cant parameters, the PRM/CRI and theTHM are unique among individuals(Table 2). While the clinical norms forthe PRM/CRI have been measured, thesignificance of deviations from norma-tive values currently exists only in anec-dotal form. Similarly, the normative rangeof the THM oscillation has been mea-sured, but little information exists as to itsrelevance regarding pathologic condi-tions. The use of laser-Doppler flowme-try provides a quantifiable, noninvasive,instrumental method for documentingnormative blood velocity values and theirclinical relevance. As such, it may alsoprovide a method to study the PRM/CRI,thereby expanding our knowledge of theTHM oscillation and our basic under-standing of the complexity of the PRM.Laser-Doppler flowmetry also may pre-sent new opportunities to study thedynamics of cranial manipulation.

The THM oscillation may well rep-resent one aspect of the complex clinicalarena of Sutherlands discovery. It mayexplain the rate and rhythm of the CRIand offer insight into the physiologicmechanism of the PRM. It may repre-sent a component of the inherent mobil-ity of the CNS. It does not yet explainthe complex patterns of motion observedwhen palpating the CRI, and it offers noexplanation of membranous strain dys-functions. Nonetheless, the recognitionthat the CRI, one aspect of the PRM,may be monitored and recorded throughthe observation of the THM oscillationprovides an opportunity for furtherresearch into that which is distinctive inosteopathic medicine.

AcknowledgmentsThe authors thank James & Associates, Inc.,of Barrington, Illinois, for use of the BLF 21Transonic Laser-Doppler Flowmeter, andEdgar F. Allin, MD, Professor in Anatomy atthe Chicago College of Osteopathic Medicineof Midwestern University, for his wit, wis-dom, and willingness to always champion thenull hypothesis.

References1. Traube L. Uber periodische Tatigkeitsanderun-gen des Vasomotorischen und Hemmungs-Ner-venzentrums. Cbl Med Wiss 1865;56:881-885.

2. Hering E. Uber Athembewegungen des Ge-fasssystems. Sitzungb d k Akad d W math naturw1869;60:829-856.

3. Mayer S. Uber spontane Blutdruckschwankun-gen. Sitzungb d k Akad d W math naturw 1876;67:281-305.

4. Penaz J. Mayer waves: history and methodol-ogy. Automedica 1978;2:135-141.

5. Best and Taylors Physiologic Basis of MedicalPractice. 12th ed. Baltimore, Md: Williams &Wilkins, 1990.

6. Bykov KM. Textbook of Physiology. Moscow:Foreign Languages Publishing House, 1958; 142.

7. Guyton AC. Textbook of Physiology. 3rd ed.Philadelphia, Pa: WB Saunders Co., 1966; 293.

8. Hyndman BW. The role of rhythms in home-ostasis. Kybernetik 1974;15:227-236.

9. Akselrod S, Gordon D, Ubel FA, Shannon DC,Barger AC, Cohen RJ. Power spectral analysisof heart rate fluctuation: a quantitative probe ofbeat-to-beat cardiovascular control. Science1981;213:220-221.

10. Akselrod S, Gordon D, Madwed JB, SnidmanNC, Shannon DC, Cohen RJ. Hemodynamic reg-ulation: investigation by spectral analysis. Am JPhysiol 1985;249:H867-H875.

11. Fuller BF. The effects of stress-anxiety andcoping styles on heart rate variability. Int J Psy-chophysiol 1992;12:81-86.

12. Negoescu R, Filcescu V, Boanta F, Dinca-Panaitescu S, Popovici C. Hypobaric hypoxia:dual sympathetic control in the light of RR and QTspectra. Rom J Physiol 1994;31:47-53.

13. Szidon JP, Cherniack NS, Fishman AP.Traube-Hering waves in the pulmonary circula-tion of the dog. Science 1969;164:75-76.

14. White DN. The early development of neu-rosonology: III. Pulsatile echoencephalographyand Doppler techniques. Ultrasound Med Biol1992;18:323-376.

15. Clarke MJ, Lin JC. Microwave sensing ofincreased intracranial water content. Invest Radi-ol 1983;18:245-248.

16. Kitney RI. An analysis of the nonlinear behav-ior of the human thermal vasomotor control system.J Theor Biol 1975;52:231-248.

17. Sutherland WG. The Cranial Bowl. Mankato,Minn: Free Press Co; 1939.

18. Woods JM, Woods RH. A physical findingrelating to psychiatric disorders. JAOA 1961;60:988-993.

19. Magoun HI. Osteopathy in the Cranial Field.2nd ed. Kirksville, Mo: The Journal Printing Com-pany, 1966.

20. Frymann VM. A study of the rhythmic motionsof the living cranium. JAOA 1971;70:928-945.

21. Becker RE. Craniosacral trauma in the adult.Osteopathic Annals 1976;4:43-59.

22. Upledger JE, Vredevoogd JD. CraniosacralTherapy. Chicago, Ill: Eastland Press, 1983.

23. Sergueef N. Le B.A.BA du crnien. Paris,France: SPEK, 1986.

24. Lay E. Cranial field. In: Ward RC, ed. Foun-dations for Osteopathic Medicine. Baltimore, Md:Williams & Wilkins; 1997:901-913.

25. Lay EM, Cicora RA, Tettambel M. Recordingof the cranial rhythmic impulse. JAOA 1978;78:149.

26. Wirth-Pattullo V, Hayes KW. Interrater reliabilityof craniosacral rate measurements and their rela-tionship with subjects and examiners heart andrespiratory rate measurements. Phys Ther 1994;74:908-916.

27. Zanakis MF, Cebelenski RM, Dowling D,Lewandoski MA, Lauder CT, Kirchner KT, et al.The cranial kinetogram: objective quantification ofcranial mobility in man. JAOA 1994;93:759.

28. Fernandez D, Lecine A. Lenregistrement delonde de Traube-Herring et de la palpation crani-enne simultanee. Kinesitherapie Scientifique 1990;292:33-40.

29. Upledger JE, Karni Z. Strain plethysmogra-phy and the cranial rhythm. Proc XII Internat ConfMed Biol Eng, Jerusalem, Israel, Aug 19-24, 1979,Part IV, p 69.

30. Michael DK, Retzlaff EW. A preliminary studyof cranial bone movements in the squirrel monkey.JAOA 1975;74:866-869.

31. Geiger AJ. Generating dialogue concerningcranial plate motion [letter]. JAOA 1992;92:1088-1093.

31a. MacPartland J, Mein EA. Entrainment of thecranial rhythmic impulse. Altern Ther Health Med1997;3:40-45.

32. Takagi M, Watanabe S. Two different com-ponents of contingent negative variation (CNV)and their relation to reaction time under hypobar-ic hypoxic conditions. Aviat Space Environ Med1999;70:30-34.

33. Nie NH, Hayadlai Hull C, Jenkins JG, Stein-brenner K, Bent DH. SPSS: Statistical Package forthe Social Sciences. 2nd ed. New York, NY:McGraw-Hill; 1975:181-202.

34. Dupont WD, Plummer Jr WD. Power and sam-ple size calculations. A review and computer pro-gram. Controlled Clin Trials 1990;11:116-128.


35. Burch GE, Cohn AE, Neumann C. A study byquantitative methods of the spontaneous varia-tions in volume of the finger tip, toe tip, and pos-terior-superior portion of the pinna of resting nor-mal white adults. Am J Physiol 1942;136:433-447.

36. Hyndman BW, Kitney RI, Sayers BM. Spon-taneous rhythms in physiological control systems.Nature 1971;233:339-341.

37. Dornhorst AC, Howard P, Leathart GL. Res-piratory variations in blood pressure. Circulation1952;6:553-558.

38. Sayers BM. Analysis of heart rate variability.Ergonomics 1973;16:17-32.

39. Barman SM, Gebber GL. Basis for synchro-nization of sympathetic and phrenic nerve dis-charges. Am J Physiol 1976;231:1601-1607.

40. Ahmed AK, Harness JB, Mearns AJ. Respi-ratory control of heart rate. Eur J Appl Physiol1982;50:95-104.

41. Bachoo M, Polosa C. Properties of the inspi-ration-related activity of sympathetic preganglion-ic neurones of the cervical trunk in the cat. J Phys-iol (London) 1987;385:545-564.42. Burton AC, Taylor RM. A study of the adjust-ment of peripheral vascular tone to the require-ments of the regulation of body temperature. AmJ Physiol 1940;129:565-577.

43. Bornmyr S, Svensson H, Lilja B, Sundkvist G.Skin temperature changes and changes in skinblood flow monitored with laser Doppler flowmetryand imaging: a methodological study in normalhumans. Clin Physiol 1997;17:71-81.

44. Lewis T. Studies of the relationship betweenrespiration and blood-pressure. Part II. Facts bear-ing on the relationship of different factors in theproduction of respiratory curves of blood-pressure.J Physiol (London) 1908;37:233-255.45. Nasimi SG, Harness JB, Marjanovic DZ, KnightT, Mearns AJ. Periodic posture stimulation of thebaroreceptors and the local vasomotor reflexes. JBiomed Eng 1992;11:307-312.

46. McHale NG, Roddie IC. The effect of trans-mural pressure on pumping activity in isolatedbovine lymphatic vessels. J Physiol 1976;261:255-269.

47. Olszewski WL, Engeset A. Lymphatic con-tractions. N Engl J Med 1979;300:316.

48. Ohhashi T, Azuma T, Sakaguchi M. Activeand passive mechanical characteristics of bovinemesenteric lymphatics. Am J Physiol 1980;239:H88-H95.

49. McHale NG, Roddie IC, Thornbury KD. Ner-vous modulation of spontaneous contractions inbovine mesenteric lymphatics. J Physiol 1980;309:461-472.

50. Olszewski WL, Engeset A. Intrinsic contractil-ity of prenodal lymph vessels and lymph flow inhuman leg. Am J Physiol 1980;239:H775-H783.

51. Mislin H. The lymphangion. In: Foldi M, Casley-Smith JR, eds. Lymphangiology. Stutgart, NY:F.K. Schatauer-Verlag, 1983;165-175.

52. Johnston MG, Kanalec A, Gordon JL. Effectsof arachidonic acid and its cyclo-oxygenase andlipoxygenase products on lymphatic vessel con-tractility in vitro. Prostaglandins 1983;25:85-98.

53. Kobayashi M, Musha T. 1/f Fluctuation of heart-beat period. IEEE Trans Biomed Eng 1982;BME-29:456-457.

54. Johnston MG, ed. Experimental Biology of theLymphatic Circulation. Amsterdam, The Nether-lands: Elsevier; 1985.

55. Parsons RK, McMaster PD. The effect of thepulse upon the formation and flow of lymph. J ExpMed 1938;68:353-376.

56. Laborit H. Stress and Cellular Function.Philadelphia, Pa: JB Lippincott Co. 1959;19-21.

57. Burton AC, Taylor RM. A study of the adjust-ment of peripheral vascular tone to the regulationof body temperature. Am J Physiol 1940;129:565-577.

58. Retzlaff EW, Michael DK, Roppel RM. Cra-nial bone mobility. JAOA 1975;74:869-873.

59. Becker RE. Life in Motion. Portland, Ore: RudraPress; 1997.

60. Chess GF, Tam RMK, Calaresu FR. Influenceof cardiac neural inputs on rhythmic variations ofheart period in the cat. Am J Physiol 1975;228:775-780.

61. Younozai R, Fryman VM, Nardell BE, Pryor MJ,Senicki M. Effects of temporal manipulation onrespiration. JAOA 1981;80:751.

62. Barron DH. Vasomotor regulation. In: RuchTC, Fulton JF, ed. Medical Physiology and Bio-physics. 18th ed. Philadelphia, Pa: WB SaundersCo, 1960:691-707.

63. Enzmann DR, Pelc NJ. Normal flow patternsof intracranial and spinal cerebrospinal fluid definedwith phase-contrast cine MR imaging. Radiology1991;178:467-474.

64. Feinberg DA, Mark AS. Human brain motionand cerebrospinal fluid circulation demonstratedwith MR velocity imaging. Radiology 1987;163:793-799.

65. Vern BA, Schuette WH, Leheta B, Juel VC,Radulovacki M. Low-frequency oscillations of cor-tical oxidative metabolism in waking and sleep. JCereb Blood Flow Metab 1988;8:215-226.

66. Jenkins CO, Campbell JK, White DN. Modu-lation resembling Traube-Hering waves recordedin the human brain. Eur Neurol 1971;5:1-6.


cranial rhythmic impulse related to the traube-hering-mayer oscillation

Documents