Assessment of Aortic Insufficiency by Transcutaneous Doppler Ultrasound By DEREK R. BOUGHNER, M.D., Ph.D., F.R.C.P. (C) SUMMARY Using a 2.2 MHz directional Doppler ultrasound unit, the instantaneous peak aortic velocity pattern was recorded transcutaneously in 15 normal persons and 15 patients with aortic insufficiency. The transducer was positioned in the suprasternal notch and aimed posteriorly to cross the descending aortic arch at an angle approximately parallel to blood flow. The electrocardiogram, phonocardiogram, and carotid pulse tracings were recorded simultaneously. In patients with aortic insufficiency there was significant diastolic flow that was not present in normal persons. The planimetered area under the systolic and diastolic velocity tracings represents the distance forward and backward that the stroke volume moves. The ratio was used to approximate the percent regurgitation, which ranged from 9% to 68%. From left ventricular angiograms in the patients with aortic regurgitation single plane ventricular volume measurements were used to calculate ventricular output and when compared with the Fick cardiac output gave an estimate of true percent regurgitation. A strong correlation was obtained with the Doppler estimate (r = 0.91), confirming that this simple ultrasound technique can accurately assess the degree of aortic insufficiency. DESCENDING AORTIC ARCH blood velocity can be obtained from the chest surface using a di- rectional Doppler ultrasound probe.1 This technique has been recommended for the assessment of aortic insufficiency2 but its accuracy has not been es- tablished. In diastole there is normally very little reversed aor- tic blood flow (cm3/sec) and, consequently, little reversed aortic blood velocity (cm/sec). However, in the presence of aortic insufficiency the amount of reversed flow is increased and this increase can be used to calculate the amount of the stroke volume leaking back into the left ventricle. An elec- tromagnetic flowmeter, with the cuff probe positioned around the ascending aorta, gives the best measure of the exact regurgitant volume3 but this is, of course, impractical clinically. Intraaortic blood velocity measurements, using a catheter tip electromagnetic velocity probe4 and a Doppler ultrasound velocity probe,5 have demonstrated the abnormal velocity pattern in aortic regurgitation. In dogs, a good cor- relation has been shown between estimates of the per- cent rerurgitation (rather than volume regurgitation) From the Cardiac Investigation Unit, Department of Medicine, University Hospital and the Department of Biophysics, University of Western Ontario, London, Ontario, Canada. Presented in part at the 47th annual scientific sessions of the American Heart Association, November 1974, Dallas, Texas. Supported by a grant from the Ontario Heart Foundation. Dr. Boughner is an Ontario Heart Foundation Senior Research Fellow. Address for reprints: Dr. Derek Boughner, Cardiac Investigation Unit, University Hospital, 339 Windermere Road, London, On- tario, Canada N6G 2K3. Received April 1, 1975; revision accepted for publication May 28, 1975. 874 calculated from such intraaortic velocity recordings and simultaneous cuff electromagnetic flowmeter recordings.4 Since Doppler ultrasound can be used to record aor- tic blood velocity noninvasively it seemed possible that it could provide a simple and accurate measure of percent regurgitation. Method A 2.2 MHz directional Doppler ultrasound unit manufac- tured by Bach-Simpson Limited was used for this study. The transducer was of the "split D" configuration and was focused at a depth of 8 cm. The technique for obtaining aor- tic blood velocity was described by Light in 1969.1 The transducer was hand held in the suprasternal notch with the patient supine. The ultrasound beam, when aimed downwards and posteriorly towards the tip of the left scapula, crossed the descending arch such that the angle of incidence with the blood flow was approximately zero degrees (fig. 1). In this area of the arch the transluminal velocity profile was relatively flat, i.e., varied little across the lumen, so that the velocity reading obtained approximated the over-all velocity of the stroke volume.6' 7Also since the Doppler shift velocity recording obtained depends upon the cosine of the angle of incidence of the beam with the flowing blood, an angle of 200 from parallel will produce an error of only ± 5% in the instantaneous peak velocity reading. In each case the transducer was rotated in the supraster- nal notch until a peak velocity away from the transducer was obtained. In many patients the carotid vessels lie in the path of the beam and, therefore, simultaneous flow toward the transducer was frequently recorded in systole. The equipment provided three forms of display of the in- formation: 1) The Doppler frequency shift produced by the flowing blood was made audible via stereophonic earphones, with velocities away from and toward the transducer on separate channels. 2) The instantaneous peak velocity away from and toward the transducer was indicated on two calibrated meters on the face of the equipment. 3) An out- put from the Doppler unit was connected directly to a Lit- Circulation, Volume 52, November 1975 by guest on May 13, 2018 http://circ.ahajournals.org/ Downloaded from
Transcript
Assessment of Aortic Insufficiency byTranscutaneous Doppler Ultrasound
By DEREK R. BOUGHNER, M.D., Ph.D., F.R.C.P. (C)
SUMMARYUsing a 2.2 MHz directional Doppler ultrasound unit, the instantaneous peak aortic velocity pattern was
recorded transcutaneously in 15 normal persons and 15 patients with aortic insufficiency. The transducerwas positioned in the suprasternal notch and aimed posteriorly to cross the descending aortic arch at anangle approximately parallel to blood flow. The electrocardiogram, phonocardiogram, and carotid pulsetracings were recorded simultaneously. In patients with aortic insufficiency there was significant diastolicflow that was not present in normal persons. The planimetered area under the systolic and diastolic velocitytracings represents the distance forward and backward that the stroke volume moves. The ratio was used toapproximate the percent regurgitation, which ranged from 9% to 68%. From left ventricular angiograms inthe patients with aortic regurgitation single plane ventricular volume measurements were used to calculateventricular output and when compared with the Fick cardiac output gave an estimate of true percentregurgitation. A strong correlation was obtained with the Doppler estimate (r = 0.91), confirming that thissimple ultrasound technique can accurately assess the degree of aortic insufficiency.
DESCENDING AORTIC ARCH blood velocity canbe obtained from the chest surface using a di-
rectional Doppler ultrasound probe.1 This techniquehas been recommended for the assessment of aorticinsufficiency2 but its accuracy has not been es-tablished.
In diastole there is normally very little reversed aor-tic blood flow (cm3/sec) and, consequently, littlereversed aortic blood velocity (cm/sec). However, inthe presence of aortic insufficiency the amount ofreversed flow is increased and this increase can beused to calculate the amount of the stroke volumeleaking back into the left ventricle. An elec-tromagnetic flowmeter, with the cuff probe positionedaround the ascending aorta, gives the best measure ofthe exact regurgitant volume3 but this is, of course,impractical clinically. Intraaortic blood velocitymeasurements, using a catheter tip electromagneticvelocity probe4 and a Doppler ultrasound velocityprobe,5 have demonstrated the abnormal velocitypattern in aortic regurgitation. In dogs, a good cor-relation has been shown between estimates of the per-cent rerurgitation (rather than volume regurgitation)
From the Cardiac Investigation Unit, Department of Medicine,University Hospital and the Department of Biophysics, Universityof Western Ontario, London, Ontario, Canada.
Presented in part at the 47th annual scientific sessions of theAmerican Heart Association, November 1974, Dallas, Texas.
Supported by a grant from the Ontario Heart Foundation. Dr.Boughner is an Ontario Heart Foundation Senior Research Fellow.
Address for reprints: Dr. Derek Boughner, Cardiac InvestigationUnit, University Hospital, 339 Windermere Road, London, On-tario, Canada N6G 2K3.
Received April 1, 1975; revision accepted for publication May 28,1975.
874
calculated from such intraaortic velocity recordingsand simultaneous cuff electromagnetic flowmeterrecordings.4
Since Doppler ultrasound can be used to record aor-tic blood velocity noninvasively it seemed possiblethat it could provide a simple and accurate measure ofpercent regurgitation.
MethodA 2.2 MHz directional Doppler ultrasound unit manufac-
tured by Bach-Simpson Limited was used for this study. Thetransducer was of the "split D" configuration and wasfocused at a depth of 8 cm. The technique for obtaining aor-tic blood velocity was described by Light in 1969.1 Thetransducer was hand held in the suprasternal notch with thepatient supine. The ultrasound beam, when aimeddownwards and posteriorly towards the tip of the leftscapula, crossed the descending arch such that the angle ofincidence with the blood flow was approximately zerodegrees (fig. 1). In this area of the arch the transluminalvelocity profile was relatively flat, i.e., varied little across thelumen, so that the velocity reading obtained approximatedthe over-all velocity of the stroke volume.6' 7Also since theDoppler shift velocity recording obtained depends upon thecosine of the angle of incidence of the beam with the flowingblood, an angle of 200 from parallel will produce an error ofonly ± 5% in the instantaneous peak velocity reading.
In each case the transducer was rotated in the supraster-nal notch until a peak velocity away from the transducer wasobtained. In many patients the carotid vessels lie in the pathof the beam and, therefore, simultaneous flow toward thetransducer was frequently recorded in systole.The equipment provided three forms of display of the in-
formation: 1) The Doppler frequency shift produced by theflowing blood was made audible via stereophonic earphones,with velocities away from and toward the transducer onseparate channels. 2) The instantaneous peak velocity awayfrom and toward the transducer was indicated on twocalibrated meters on the face of the equipment. 3) An out-put from the Doppler unit was connected directly to a Lit-
ton chart recorder with "away" velocity plotted upright and"toward" velocity inverted.An oscillator contained within the equipment was used
for calibration. When activated it produced a signalequivalent to a frequency shift representing 50 cm/sec. Theelectrocardiogram, phonocardiogram (recorded from theleft sternal border, fourth interspace) and the carotid pulsetracing were recorded simultaneously with the velocity trac-ing at paper speeds of 50 mm/sec and 250 mm/sec.To establish the normal aortic velocity tracing (fig. 2)
recordings were performed in 15 individuals, ranging in agefrom 14-63 years, with no known cardiac disease. Allshowed an essentially parabolic wave form beginning ± 20msec from the onset of the carotid pulse upstroke. Thevelocity tracings showed a rapid rise to peak velocity. Thefall to zero velocity occurred rapidly near the end of systolewith the baseline being reached at the dicrotic notch of thecarotid pulse. The systolic flow in the carotid vessels wasusually recorded simultaneously with the aortic arch bloodvelocity but was readily separated from that of the aorticarch by its opposite direction relative to the transducer. Asmall amount of reversed flow was often seen in earlydiastole. Blood velocity from the innominate vein could berecorded and was of considerable importance to thistechnique's assessment of aortic insufficiency. Flow in thisvessel was toward the transducer, beginning in late systoleand extending through to mid diastole. This velocity patterntherefore could obscure abnormal aortic arch reverse flow inearly diastole and if not carefully excluded could invalidatethe interpretation of the aortic velocity tracing. In mostpatients this vessel was avoided by careful aiming of thetransducer, and the earphone display of the Doppler fre-quency shift aided in this technique as venous and arterialflow tended to sound different. Venous flow produced amore continuous rushing or rumbling sound without therapid rise and fall in amplitude characteristic of the arterialflow.
Occasionally it was not possible to obtain a definite aorticarch velocity signal. In our experience this has been duemost often to the presence of obstructive airways diseasewith air-containing tissue lying between the aortic arch andthe transducer. The ultrasound beam is almost completelyreflected at air-tissue interfaces and penetration to the aortathen becomes impossible. A separate difficulty can be en-countered with the extremes of patient body size. Ourtransducer was focused at 8 cm which is too deep in smallchildren and too shallow in some large adults. In this par-ticular study problems with patient size and emphysema didnot arise.From the instantaneous aortic velocity tracing recorded
on the Litton recorder additional information could be ob-tained. If we were using a cuff electromagnetic flow probearound the aorta our recording would be flow in cm3/secand the area under the systolic portion of that curve wouldyield an accurate measure of the stroke volume (cm3). How-ever, our Doppler equipment provided a velocity record-ing in cm/sec. The area under the systolic portion wastherefore in centimeters, i.e., the distance forward that thestroke volume moved or the amount of aorta occupied bythe stroke volume. It follows that the area under anyreversed velocity curve in diastole represented the distancebackward that the stroke volume moved and the ratiorepresented the percent regurgitation. Since we were thendealing in ratios to obtain this pertinent information itbecame less important for the ultrasound beam to closelyparallel the aortic arch blood flow, i.e., it was not absolutelyessential to obtain the true peak velocity and the techniquewas therefore made simpler.A group of 15 patients with clinical and/or echocar-
diographic findings of aortic insufficiency who were to un-dergo cardiac catheterization were studied prospectivelyusing the Doppler ultrasound technique. In each patient theareas under the systolic and diastolic portions of the aorticvelocity curve were measured with a planimeter. Ten beats
_i~ ECG
Figure 1
The ultrasound probe is positioned in the suprasternal notch andaimed such that the beam crosses the aortic arch approximatelyparallel to blood flow.Circulation, Volume 52, November 1975
L L
50'
25
U~~~~I.t:.' :.''.
' _.',..
U 25.
50-
Figure 2
Phonocardiogram
AorticVelocity
CarotidPulse
Recordings from a normal individual. Blood velocity away from thetransducer is plotted upright and toward the transducer is inverted.Forward aortic velocity is shaded area.
were analyzed from each patient and a mean ratio of the dis-tance forward and backward that the stroke volume moved,i.e., percent regurgitation, was obtained.
Following cardiac catheterization the left ventricularcineangiograms for each patient, recorded in the rightanterior oblique projection, were analyzed. The end-systolicand end-diastolic frames were selected by referring to asimultaneously recorded electrocardiogram on which the ex-posures of the angiograms had been marked automatically.These two frames were then projected and traced, the areasand long axis were measured manually with a planimeterand ruler and finally the corrected ventricular volumes atend systole and end diastole were calculated using the singleplane technique outlined by Dodge.8 Care was taken toavoid extrasystolic and postextrasystolic beats. The strokevolume was thus obtained as the difference between the twovolumes and when multiplied by the heart rate gave leftventricular output. The Fick cardiac output was measuredby the rebreathing technique prior to the angiogram and thedifference between the two measurements of outputprovided a close approximation of the percent aorticregurgitation.9The estimates of the regurgitant fraction obtained from
the noninvasive Doppler ultrasound technique and the car-diac catheterization technique were then compared. Noneof the patients were found to have evidence of mitral in-sufficiency.
ResultsThe aortic velocity patterns from normal persons
were similar to the invasive patterns obtained byBenchimol et al. using a directional intraaorticDoppler catheter.5 In addition, our recordings werecomparable to the noninvasive Doppler aortic velocitytracings obtained by Light,10 whose equipmentdiffered from ours primarily in the recording system.
In our normals, the aortic velocity pattern did notregularly demonstrate the small amount of low
velocity reverse flow normally present in earlydiastole. This failure was due primarily to the passband of the Doppler instrument. The unit is sensitiveto velocities from 7-100 cm/sec; frequency shiftsrepresenting velocities below or above these limits arenot detected. Thus very low velocity diastolic reversedflow was not recorded. This characteristic of the unitcan introduce a small degree of error in the estimationof the amount of reversed flow in aortic insufficiencysince the final low velocities will be lost.The problem of innominate vein flow was more
significant. In that vessel, flow occurs toward thetransducer in late systole and early diastole and canmask the presence of early diastolic reversed flow inthe aorta if both vessels fall simultaneously in the ul-trasound beam. If the velocity patterns from the twovessels cannot be separated adequately by carefulaiming of the transducer this technique cannot beapplied reliably. Two patients were excluded fromthis study for this reason (fig. 3).
In each of the 15 patients with aortic insufficiencyan abnormal diastolic aortic velocity pattern towardthe transducer (i.e., reversed flow), was recorded bythe Doppler technique. This pattern began at thedicrotic notch and extended throughout varying por-tions of diastole (figs. 4 and 5). It was not present innormals (fig. 2) and was differentiated from in-nominate vein flow by the methods outlined above.The systolic portion of the velocity tracing showed nofeatures that would correspond to the aortic pressurepulse abnormalities (such as the bifid aortic pulse)seen in patients with aortic insufficiency. This was notsurprising since the interrelation between the two
ECG
0' k_ Phonocard iogram
Aortic bloodvelocity
Venous bloodvelocity
CarotidPulse
Figure 3
Recordings from a patient with angiographicallymild aortic insufficiency. The reversed aorticvelocity pattern in diastole is obliterated by therecording from the innominate vein of venous
blood velocity toward the transducer (shadedarea).
Recordings from patient with mild aortic insufficiency, regurgitantfraction 29%. Blood velocity away from the transducer is plottedupright and toward the transducer is inverted. Shaded area insystole is forward aortic velocity and in diastole is reversed aorticvelocity in aortic arch. Unshaded area is velocity of blood in neckvessels toward transducer in systole.
waveforms is complex. Pressure and velocitywaveforms are propagated at different rates and evenin normals they have quite different contours.7By planimetry the areas under the systolic and
diastolic portions of the aortic velocity curves were ob-tained and thus estimates of the percent regurgitationwere calculated, varying from 9% to 68%. Eachestimate was then compared to the percent regurgita-tion calculated from the left ventricular cineangio-grams and Fick cardiac output (fig. 6). A strongcorrelation was demonstrated with an r value of 0.91.An aortic valve gradient was present in nine of the
patients indicating the presence of aortic stenosis inaddition to the insufficiency. Although this abnor-mality could alter the aortic velocity pattern (fig. 7) itdid not interfere appreciably with the estimation ofthe percent regurgitation by the Doppler technique.We observed that small valve gradients tended toproduce a minor delay in the rise to peak aorticvelocity while more severe aortic stenosis producedprolonged delays but this finding was not consistent.In our patients forward flow in the aorta still felltowards zero at the dicrotic notch and the regurgitantpattern could still be distinguished. It seems likelythat such will not always be the case and that accuratedifferentiations of the pertinent forward and reversedvelocity patterns in aortic stenosis might be disruptedby jet flow from the stenosed valve. If the jet were to
Circulation, Volume 52, November 1975
1k A A ECG
Phonocardiogram
so 1
25 t ::
U,, I - Aortic
:E1.11 20 -Velocity
50-
CarotidPulse
Figure 5
Recordings from patient with moderately severe aortic insufficiency,regurgitant fraction 54%. Shaded areas represent aortic archvelocity in systole and diastole. Planimetry of areas provided es-timate of regurgitation.
extend around the arch the velocity profile would notbe flat and the peak velocity indicated would not berepresentative of the mean velocity of the stroke
100-
Z 80-0
0 60
0 40- *0.
W ~0 0LU.2JL~ 20-
0 20 40 60 80 100ANGIOGRAPHIC % REGURGITATION
Figure 6
Correlation between percent regurgitation calculatedfrom Dopplervelocity recordings and regurgitant percentage calculated fromangiographic left ventricular output (from left ventricular volumemeasurements) minus Fick cardiac output.
Delay in rise to peak aortic velocity as recorded in some patientswith aortic stenosis. The percent regurgitation can still becalculated.
volume. Fortunately such jets are usually eccentricand tend to strike the aortic wall as it bendsposteriorly. However, they are probably responsiblefor the variable delay in the aortic velocity upstroke,since a variety of forward and reversed patterns are setup in the ascending aorta.
DiscussionThe need exists for simpler and more accurate
means of assessing aortic insufficiency. This lesion ismost commonly assessed invasively by cineaor-tography, a technique of limited accuracy. Originallythat method was proposed by Lehman et al. as aprocedure that could give only a rough estimate of thedegree of insufficiency. " Subsequent studies haveborne this out.4' 9, 12 For example, Mennel et al.3recently measured the amount of regurgitation with acuff electromagnetic flowmeter probe around theascending aorta at operation and compared the resultswith cineaortography. From that study it was ap-parent that cineaortography can distinguish mild fromsevere regurgitation but between those extremes it haslittle quantitative value. Similar results were obtainedby Hunt et al.9 who used as their standard thetechnique that we have applied, i.e., Fick cardiac out-put versus left ventricular volume measurements.Their correlation coefficient was 0.56 between thecineaortographic estimate and true regurgitant frac-tion.
Noninvasive techniques presently employed haveabout the same degree of accuracy. The physical ex-amination, the electrocardiogram and the chest X-ray
have only limited abilities to quantify the abnor-mality.'2 '3 Echocardiography provides little informa-tion regarding severity of the aortic lesion although itcan indirectly indicate the presence of regurgitationby recording mitral valve anterior leaflet flutter duringdiastole. Transcutaneous directional Doppler ul-trasound has been used to record peripheral arterialvelocity patterns in patients with aortic valve dis-ease.14 Although reversed velocity in diastole was nor-mally present in radial and femoral artery tracings, itwas of significantly increased magnitude and durationin aortic insufficiency. Correlation with the degree ofregurgitation was not attempted.
Incorrect assessment of aortic insufficiencyproduces important difficulties in the management ofpatients with coincident aortic and mitral valve dis-ease. Accurate information regarding the presenceand degree of aortic regurgitation is necessary for theproper management of the mitral lesion since the aor-tic lesion can prove a serious technical handicap dur-ing open operative procedures on the mitral valve.'2The aortic Doppler technique can provide the infor-mation with greater ease and accuracy than other non-invasive methods.
In addition, patients with aortic prostheses maydevelop a degree of aortic regurgitation. Theappearance of a diastolic murmur can be worrisomeand may represent either a minor, small volumeperiprosthetic leak or a significant degree of regurgita-tion. The physical findings and noninvasive studies
ECG
Phonocardiogrom
50-
25
Aortic\ 0~~~~~~~~ ~Velocity25]
50
CarotidPulse
Figure 8
Leaking aortic prosthetic valve. The arrows indicate the openingand closing artifacts produced by the prosthesis.
are often not helpful.'` The Doppler technique whichcan detect and grade the regurgitation (fig. 8) mayprovide a suitable method for following thesepatients.Thus the transcutaneous measurement of aortic
arch blood velocity by directional Doppler ultrasoundappears to have a useful place in patient management.It has the advantage of being noninvasive and can bedone rapidly and safely. When a good tracing is ob-tained the results provide an accurate assessment ofthe degree of aortic regurgitation.
Acknowledgment
Appreciation is expressed for the excellent technical assistance ofMrs. Joan Persaud and the continued interest and assistance of Dr.S. P. Ahuja and Dr. W. J. Kostuk. We also wish to thank the Bach-Simpson Co. Ltd., London, Ontario for supplying the ultrasoundunit and Mr. Victor Carriere for his technical advice.
References1. LIGHT LH: Non-injurious ultrasonic technique for observing
flow in the human aorta. Nature 224: 1119, 19692. THOMPSON PD, MENNEL RG, MACVAUGH H, JOYNER CR:
Evaluation of aortic insufficiency in humans with atranscutaneous Doppler velocity probe. (abstr) Ann InternMed 72: 781, 1970
3. MENNEL RG, JOYNER CR, THOMPSON PD, PYLE RR, MACVAUGHH: The preoperative and operative assessment of aorticregurgitation. Am J Cardiol 29: 360, 1972
4. NOLAN SP, FISHER RD, DIXON SH, MORROW AG: Quantificationof aortic regurgitation with a catheter tip velocimeter.Surgery 65: 876, 1969
5. BENCHIMOL A, DESSER KB, GARTLAN JL JR: Bidirectional bloodflow velocity in the cardiac chambers and great vesselsstudied with the Doppler ultrasonic flowmeter. Am J Med52: 467, 1972
6. SEED WA, WOOD NB: Velocity patterns in the aorta. CardiovascRes 5: 319, 1971
7. SCHULTZ D: Pressure and flow in large arteries: InCardiovascular Fluid Dynamics, vol. 1, edited by BERGELDH. New York, Academic Press, 1972, p 287
8. DODGE HT: Determination of left ventricular volume and mass.Radiol Clin North Am 9: 459, 1971
9. HUNT D, BAXLEY WA, KENNEDY JW, JUDGE TP, WILLIAMS JE,DODGE HT: Quantitative evaluation of cineaortography inthe assessment of aortic regurgitation. Am J Cardiol 31: 696,1973
10. LIGHT LH: Initial evaluation of transcutaneous aortovelog-raphy - A new noninvasive technique for haemodynamicmeasurements in the major thoracic vessels, In Cardiovas-cular Applications of Ultrasound, edited by RENEMAN RS.New York, American Elsevier Co, 1974, p 325
11. LEHMAN JS, BOULE JH, DEBBAS JH: Quantitation of aorticvalvular insufficiency by catheter thoracic aortography.Radiology 79: 361, 1962
12. COHN LH, MASON DT, Ross JR, MORROW AG, BRAUNWALD E:Preoperative assessment of aortic regurgitation in patientswith mitral valve disease. Am J Cardiol 19: 177, 1967
13. HIRSHFELD JW JR, EPSTEIN SE, ROBERTS AJ, GLANCY DL,MORROW AG: Indices predicting long-term survival aftervalve replacement in patients with aortic stenosis. Circula-tion 50: 1190, 1974
14. RITTENHOUSE EA, STRANDNESS DE: Oscillatory flow patterns inpatients with aortic valve disease. Am J Cardiol 28: 568, 1971
15. HILDNER FJ: Detection of prosthetic valve dysfunction bybedside and laboratory evaluation. In CardiovascularClinics, vol 5, number 2, "Valvular Heart Disease,' editedby LIKOFF W. Philadelphia, F. A. Davis Co, 1973, p 290
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.52.5.874
1975;52:874-879Circulation.
http://circ.ahajournals.org/content/52/5/874the World Wide Web at:
The online version of this article, along with updated information and services, is located on
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.
Permissions and Rights Question and Answer Further information about this process is available in therequested is located, click Request Permissions in the middle column of the Web page under Services.the Editorial Office. Once the online version of the published article for which permission is being
can be obtained via RightsLink, a service of the Copyright Clearance Center, notCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:
![Transcutaneous electrical nerve stimulation for acute painuir.ulster.ac.uk/22796/1/TENS_acute_pain_cochrane_review.pdf · [Intervention Review] Transcutaneous electrical nerve stimulation](https://static.documents.pub/doc/80x56/5b5932df7f8b9a4e1b8cebc7/transcutaneous-electrical-nerve-stimulation-for-acute-intervention-review.jpg)
