- Email: [email protected]
Validation of volumetric flow measurements by means of a Doppler-tipped coronary angioplasty guide wire
Labovitz et al.
Ametican
using phase-contrast tine magnetic resonance imaging. Invest Radio1 1992;27:465-70. 112. Nienaber CA, Von Kodolitsch Y, Nicolas V, Siglow V, Piepho A, Brockhoff C, Koschyk DH, Spielmann RP. The diagnosis of thoracic aortic dissections by noninvasive imaging procedures. N Engl J Med 1993;328:1-9. 113. Mitchell L, Jenkins JPR, Brownlee WC, Isherwood I. Aortic dissection:,morphology and differential flow velocity patterns demonstrated by magnetic resonance imaging. - _ Clin Radio1 1988;39:458-61. 114. Bogren HG, Underwood SR, Firmin DN, Mohfaddin RH, Klipstein RH, Rees RSO, Longmore DB. Magnetic resonance velocity mappinginaortic dissection. Br JRadiol1988;61:456-
117. 118. 119. 120.
62. _-. 115.
116.
Chang J-M, Friese K, Caputo GR, Kondo C, Higgins CB. MR measurement of blood flow in the true and false channel in chronic aortic dissection. J Comput Assist Tomogr 1991;15: 41823. Bradley WG. Magnetic resonance imaging in the evaluation
Arthur J. Labovitz, MD, Dennis M. Anthonis, Morton J. Kern, MD St. Louis, MO.
the Division University.
of Cardiology,
Department
Reprint requests: Arthur J. Labovitz, MD, Center, Division of Cardiology, FDT-14,3635 St. Louis, MO 63110-0250.
AM H~~~~J1993;126:1456-61. Copyright Q 1993 0002~8703/93/$1.00
1456
by Mosby-Year
f.10
4/l/49915
Book,
Inc.
of Internal
Medicine,
122.
of cerebrospinal fluid abnormalities. Magn Reson Q 1992; 8169-96. Martin AJ, Drake JM, Lemaire C, Henkelman RM. Cerebrospinal fluid shunts: flow measurements with MR imaging. Radiology 1989;173:243-7. Firmin DN, Klipstein RH, Hounsfield GL, Paley MP, Longmore DB. Echo-planar high-resolution flow velocity mapnine. Maan Reson Med 1989:12:316-27. GucfoyleDN, Gibbs P, Ordidge RJ, Mansfield P. Real-time flow measurements using echo-planar imaging. Magn Reson Med 1991;18:1-8. Pearlman JD, Moore JR, Lizak MJ. Real-time NMR beamdirected velocity mapping. V-mode NMR. Circulation 1992;86:1433-8. Kaufman L, Mohiaddin RH, Longmore DB. Letter to the editor, J Thorac Cardiovasc Surg 1991;101:1104-6. Coulden R, Liptan MJ. Magnetic resonance imaging and ultrafast computed tomography in cardiac tomography. Curr Opin Cardiol 1992;7:1007-15.
RDCS, Thomas
There has been extensive experience with Doppler velocity measurements in the assessment of volumetric flow in the aorta and pulmonary artery and across each of the cardiac valves. Sequeira et al17 2 initially demonstrated a good correlation between the timeaveraged velocity or velocity integral in the ascending aorta and invasively measured stroke volume. Subsequent investigations3-6 have shown that stroke volume can be calculated as the product of the time velocity integral and the cross-sectional area of the valve or vessel at the sample site. Cardiac output is then obtained by multiplying the stroke volume by the heart rate. The application of these principles has From Louis
121.
St.
St. Louis University Medical Vista Avenue at Grand Blvd.,
Decemkr 1993 H&f Journar
L. Cravens, RN, and
allowed avariety of clinical Doppler measurements of cardiac blood flow, including absolute blood flow, calculation of intracardiac shunts,7> 8 estimation of aortic valve area,g and assessment of regurgitant fractions.lOl l1 In making these measurements a number of assumptions must be accepted, including (1) the laminarity of blood flow; (2) a Doppler interrogation angle of <20 degrees; and (3) a relatively blunt flow profile from which the measurement is calculated. Although these assumptions are generally true for measurement of blood flow in the heart and great vessels, inadequate data exist validating these assumptions in smaller (<6 mm diameter) arteries. Direct measurement of coronary bIood flow in human coronary arteries has been limited, to a large extent, by the available technology. Transcutaneous interrogation of coronary floti has generally been beyond the resolution of commercially available instruments. Recent reports have indicated that the Doppler measurements of coronary flow might be possible in selected patients by using transesophageal
Volume 126, Number 6 American Heart Journal
Labovitz
et al.
1457
design indicating position of Doppler flow wire for volumetric flow measurement in tubing of various sizes.
Fig. 1. Experimental
echocardiography. l2 Invasively obtained, catheterderived intracoronary flow velocity measurements taken by means of pulsed Doppler, however, have been available for the past several years.13y l4 These directly measured velocity data have been acquired both in the baseline state and during hyperemia induced by a variety of mechanical and pharmacologic stimuli.15-17 Although these velocity measurements have been useful in the clinical assessment of the degree of coronary stenoses, absolute measurements of volumetric coronary flow might be preferable for certain clinical and research questions. The recent availability of a 12 MHz Doppler transducer mounted on an 0.018 inch angioplasty guide wire allows accurate measurement of spectral coronary flow velocities in the distal coronary bed.lse20 By using classic Doppler flow velocity formula, volumetric flow can be computed with appropriate inclusion of the vessel cross-sectional area. Quantitative angiographic and, more recently, intravascular ultrasound imaging21 provides highly accurate determination of the crosssectional area of the coronary vessels. To validate this method for measurement of absolute volumetric blood flow in the coronary vasculature, we examined flaw velocity spectra in an in vitro model applying Doppler formulas commonly applied to large conduits (e.g., aortic, transvalvular flow) for volumetric measurements:obtained in smaller conduits (<6 mm diameter) simulating human coronary arteries. METHODOLOGY
Subselective intracoronaxy flow velocity was measured with an 0.018 inch Doppler angioplasty guide wire (Flowire, Cardiometrics, Inc., Mountain View,
Calif.). As describedIg> 2o and validated by Doucette et a1.,2othe Doppler angioplasty guide wire is a 175 cm long, 0.018 inch diameter flexible steerable guide wire with a 12 MHz piezoelectric ultrasound transducer integrated into the tip. The forward-directed ultrasound beam diverges in a 27-degree arc from the long axis (measured to the -6 decibel round-trip points of the ultrasound beam pattern). The pulse repetition frequency of >40,KHz, pulse duration of +0.83 msec, and sampling delay of 6.5 msec is standard for clinical use. The system is coupled to a real-time spectrum analyzer, videocassette recorder, and video page printer. The Doppler audio signals are processed by the spectrum analyzer with fast Fourier transformation and scrolling gray scale spectral video display. The frequency response of the system calculates approximately 90 spectra per second. Simultaneous ECG and arterial pressures are recorded and displayed on the video monitor. The Doppler guide wire velocity demonstrated excellent correlation with electromagnetic flow velocity and volumetric flow in straight and curved tubed models and in the in vivo testing by using a circumflex canine coronary artery. The Doppler guide wire measures phasic flow velocity patterns and tracks linearly with flow rates in most small, straight coronary ,arteries.lgp 2o Pulsatile flow model. In this experiment a modified two-head roller pump device (Biomedicus Model 540, Minneapolis, Minn.) was connected to a segment of polyurethane tubing (Fig. 1). This tubing was connected to angiographic catheters of various sizes with internal diameters of 6.0 mm, 2.7 mm, 2.1 mm, and 1.9 mm. This system was filled with saline solution mixed with cornstarch echo-reflective particles
1458
Lubovitz et al.
American
December 1993 Hea~Journal
ments for each catheter size; 96 individual measurements were performed. The spectral Doppler tracings were analyzed on custom-developed software in which the outermost portion of the spectral velocity envelope was digitized by using a personal computer and compatible digiting-pad (Fig. 2). Three to five beats were averaged for each flow measurement. Peak and mean velocity measurements and the flow velocity interval were calculated. Total flow was calculated as the product of the cross-sectional area (CSA) of the catheter in which the measurement was made, the time velocity integral (TVI), and the pump cycles per minute (Flow = Rate X TV1 X CSA). Ten flow recordings were traced by two different observers. Interobserver variability for velocity integral measurements was 3.2%. Statistical analysis. Simple linear regression analysis was used to assessthe relationship between the calculated and roller pump blood flow. Results are reported as a mean plus or minus the SD. A probability fp) value of 0.05 was considered to be significant. OBSERVATIONS
Fig. 2. Doppler spectral tracing demonstrating measurement of time velocity integral (WI). Flow wascalculated as product of. velocity integral and cross-sectionalarea of tubing at point of measurementover l-minute period.
(0.5 %) to provide an optimal Doppler spectral display. The Doppler flow wire was introduced into this system through a valved sidearm connector with direct visualization and maintained in a parallel position to flow. Roller pump flow was adjusted incrementally from a minimum of 90 ml/min to a maximal of 550 ml/min for each of the four catheters with different internal luminal diameters. Flow velocity spectra were measured at 10 to 20 ml/min increments after allowing equibration to reach a new steady state. There were approximately 24 flow measure-
There was an excellent correlation between the Doppler calculated absolute flow and pulsed flow model (Fig. 3, Table I). Absolute velocity measurements increased linearly in relation to increases in absolute flow. Volumetric flow output by roller pump varied between 90 and 540 ml/min and by the Doppler flow wire from 71 to 543 ml/min (mean 237 vs 239 ml/min, p = NS). The correlation between roller pump flow and Doppler calculated ff ow for each size tubing was equally strong (r = 0.99). Peak velocities ranged from 4 to 45 cm/set, mean velocity-from 3 to 32 cm/set and total velocity interval from 9 to 30 cm. Doppler flow was related, to pump floti by the equation Doppler flow = 0.94 (pump flow) + 16.4 (r = 0.97, p < 0.0001) CdMMENTS This study showed that accurate assessment of volumetric flow can be obtained by using the Doppler flow wire velocity integral over the flow ranges encountered in clinical situations. Although Doppler velocity measurements have been applied to estimate volumetric flow in the heart and great vessels, absolute floti measurements in the coronary circulation has been much more limited. Only recently have catheters small enough to make measurements within the more distal coronary circulation been developed.
Volume American
126, Number Heart
6
Lab&z
Journal
et al.
1459
600~ y
=
5001
E g
4oo-
#
3001
4 fl
q
0.64x + 16.4
c = 0.67, p c q.6661
2001
0
1.7
.
2.1
0 2.7
.
z
8.0
mm mm mm mm
loo“l-
0
lid
2hO 360 460 Pump Flow (mllmin)
560
6;O
Fig. 3. Correlation between roller pump and Doppler measured
Moreover, many of these systems have used a zero crossing technique that displays the average rather than differential velocity distribution in the luminal of the vesse1.15-17Most recently, a Doppler guide wire has been introduced that will enable investigators to obtain velocity measurements in the distal coronary bed.lsmzo The velocity signals, which are displayed after spectral analysis, provide a weighted real-time scrolling gray scaie image of the proportion of red cells traveling at a given velocity. Although velocity data have several clinical applicat‘ions,22 volumetric data may be particularly useful when assessing pharmacologii: or mechanical responses of coronary flow. Consideration to changes in coronary vessel diameters and cross-sectional area must be made in the analysis of Doppler velocity. Flow-velocity data alone may be insufficient to predict changes in absolute coronary flow in the setup of active coronary vasomotion. An additional limitation is a subtle problem with &gulation that provides interrogating angles of >20 degrees, which wo6ld also result in somewhat lower average velocity or total velocity interval. lnfluencebf flow velocity profile. In vessels of small diameter such as those of the coronary v&culature, the v&lodity,profile across the lumen may become increasingly paraboiic, with more rapid‘velotiities toward the center of the lumen.23-26 In this case, measurement of velocity with a very sm& sample volume in the coaxial position may falsely elevate the mean
Table
I. Volumetric
flow.
flow correlations
Pump flow (mllmin)
‘Doppler flow (mllmin)
‘Tube diameter (mm)
Mean
Range
Mean
Range
r
1.7’ 2.1 2.7 6.0
240 259 241 213
go-400 go-400 go-400 go-540
225 290 241 211
90-360 100-430 110-350 71-543
0.99 0.99 0.99 0.99
velocity estimation and theoretically would overestimate absolute corontiy blood flow. With this in mind, some investigators routinely use formulas that include a correctiori factor for a fraction of the actual average peak velocity measured in small-caliber vessels in the estimation of volumetric blood flo~.~O The current study demonstrates that this correction factor may not be necessary. The broad 2 mm sample volume that interrogates at least 5 mm beyond the tip of the flow wire is one factor that provides a larger proportion of lower velocity measurements and hence a more accurate representation of absolute flow. Previous in vitro measurement systems simulating normal and stenotic coronary vasculature report conflicting results. Recently, however, two studies have reported experimental models in which Doppler catheters were used. Grayburn used a 4.8F Doppler catheter in an in vitro model and demonstrated a
Deceiber
1460
Labovitz et al.
American
good correlation between the product of the mean velocity and cross-sectional area and coronary flow both with pulsatile and nonpulsatile models.27 The data indicated that in the pulsatile model with 4 mm tubing, the profile was blunt enough to allow direct calculation of flow with a good correlation without any correction factor (r = 0.99). Likewise, Sudhir et alz8 used a 3F Doppler catheter to measure flow in the circumflex coronary artery of an animal model. Flow derived by using the average peak velocity correlated closely with that measured by electromagnetic flow meter. Conclusions. The ability to measure absolute coronary flow with a transducer mounted on the tip of a standard angioplasty guide wire has several important potential clinical applications. The results of the present study indicate that analysis of a spectral Doppler velocity may be accurately applied to estimate volumetric flow in vessels of the same caliber as human coronary arteries. SUMMARY
6.
7.
8.
9.
10.
11.
We used an in vitro model to validate volumetric flow measurements obtained with an 0.018-&h angioplasty guidewire with a 12 MHz transducer mounted on its tip. By using a modified two-head roller pump device, flow was adjusted incrementally from a minimum of 90 ml/min to a maximum of 550 ml/m& Flow was measured with the Doppler guide wire in tubing ranging from 1.9 mm to 6.0 mm internal diameter, as the product of the spectral Doppler velocity integral and the cross-sectional area of the tubing, over a l-minute period. It was an excellent correlation between the Doppler calculated flow rates and actual flow, regardless of tubing diameter (r = 0.99). These results suggest that the Doppler spectral output of this device might be accurately applied to estimates of volumetric flow in human coronary arteries. We thank
5.
Dianne
Reid
for her expert
secretarial
assistance.
12.
13.
14.
15.
16.
17.
REFERENCES
1. Huntsman LL, Gams taneous determination
E, Johnson CC, Fairbanks E. Transcuof aortic blood-flow velocities in man.
18.
AMHEART J 1975;89:605-12. 2. Sequeira RF, Light LH, Cross G, Raftery EB. Transcutaneous aortoveloeranhv: a auantitative evaluation. Br Heart J 1976; 38443-501 - 3. Labovitz AJ, Buckingham TA, Habermehl K, Nelson J, Kennedy HL, Williams GA. The effects of sampling site on the two-dimensional echo-Doppler determination of cardiac output. AM HEART J 1985;109:327-32. 4. Dittmann H, Voelker W, Karsch K, Seipel L. Influence of sampling site and flow area on cardiac output measurements
19.
20.
1993
H&ahJcliinal
by Doppler echocardiography. J Am Co11 Cardiol1987;10:81823. Miller WE, Richards KL, Crawford MH. Accuracy of mitral Doppler echocardiographic cardiac output determinations in adults. AM HEART J 1990;119:905-10. Lewis JF, Kuo LC, Nelson JG, Limacher MC, Quinones MA. Pulsed Doppler echocardiographic determination of stroke volume and cardiac output: clinical validation of two new methods using the apical window. Circulation 1984;70:425-31. Meijboom EJ, Valdes-Cruz LM, Horowitz S, Sahn DJ, Larson DF, Yqung KA, Oliveira Lima C, Goldberg SJ, Allen HD. A two-dimensional Doppler echocardiographic method for calculation of pulmonary and systemic blood flow in a canine model with a variable-sized left-to-right extracardiac shunt. Circulation 1983;68:437-45. Hori M, Yoshima H, Ohnishi K, Kitabatake A, Inoue M, Asao M, Ito H, Masuyama T, Tanouchi J, Morita T. Noninvasive evaluation of the ratio of pulmonary to systemic flow in atrial septal defect -by duplex Doppler echocardiography. Circulation 1984;69:73-9. Otto CM, Pearlman AS, Comess KA, Reamer RP, Janko CL, Huntsmann LL. Determination of the stenotic aortic valve area in ‘adults using Doppler echocardiography. J Am Co11 Cardiol 1986;7:509-17. Ascah KJ, Stewart WJ, Jiang L, Guerrero JL, Newell JB, Gillam LD, Weyman AE. A Doppler two-dimensional echocardiographic method for quantitation of mitral regurgitation. Circulation 1985;72:377-83. Goldberg SJ, Allen HD. Quantitative assessment by Doppler echocardiography of pulmonary or aortic regurgitation. Am J - Cardiol 1985;56:131-5. Iliceto S. Maranaelli V. Memmola C. Rizzon P. Transesonhageal Doppler echocardiography evaluation of coronary blood flow velocity in baseline conditions and during dipyridamoleinduced coronary vasodilation. Circulation L991;83:61-9. Marcus ML. White CW. Coronarv flow reserve in natients with normal’coronary a.ngiograms.“J Am Co11 Cardioi 1985;6: 1254-6. Wilson RF, Johnson MJ, Marcus ML, Aylward PE, Skorton DJ, Collins S, White CW. The effect of coronary angioplasty on coronary flow reserve. Circulation 1988;76: 873. Kern MJ, Deligonul U, Vandormael M, Labovitz A, Gudipati CV, Gabliani G, Bodet J, Shah Y, Kennedy HL. Impaired coronary vasodilator reserve in the intermediate post coronary angioplasty period: analysis of coronary artery flow velocity indexes and regional cardiac venous afflux. J Am Co11 Cardiol 1989;13:860-72. Wilson RF, Laughlin DE, Ackell PH, Chilian WM, Holida MD. Hartlev CJ. Armstrong ML. Marcus ML. White CW. Transluminal, s&selective measurement of coronary artery blood flow velocity and vasodilator reserve in man. Circulation 1985;72:82. Sibley DH, Millar HD, Hartley CJ, Whitlow PL. Subselective measurement of coronary blood flow velocity using a steerable Doppler catheter. J Am Co11 Cardiol 1986;8:133240. Ofili EO, Kern MJ, Labovitz AJ, St. Vrain JA, Segal J, Aguirre FV, Caste110 R. Analysis of coronary blood flow velocity dynamics in angiographically normal and stenosed arteries before and after endolumen enlargement by angioplasty. J Am Co11 Cardiol 1993;21:308-16. Segal J, Kern MJ, Scott NA, King SB, Doucette JW, Heuser RR, Ofili E, Siegel R. Alterations of phasic coronary artery flow velocity in humans during percutaneous coronary angioplasty. J Am Co11 Cardiol 1992;20:276-86. Doucette JW, Corl PD, Payne HM, Flynn AE, Goto M, Nassi M, Segal J. Validation of a Doppler guide wire for intravascu-
Volume 126, Number 6 American Heart Journal
21.
22. 23.
24.
1a.rmeasurement of coronary artery flow velocity. Circulation 1992;85:1899-911, Tobis JM, Mallery J, Mahon D, Lehmann K, Zalesky P, Griffith J, Gessert J, Moriuchi M, McRae M, Dwyer ML. Intravascular ultrasound imaging of human coronary arteries in vivo. Circulation 1991;83:913-26. Kern MJ, Donohue TJ, Bach RG, Caracciolo EA, Aguirre FV. Clinical applications of the Doppler coronary flow velocity guidewire. Cardiovasc Intervent Radio1 1993;3:10-20. Denardo SJ, Yock PG, Hargrave VK, Srebro JP, Ports TA, Talbot L. Differentiation of abnormal blood flow patterns in coronary arteries based on Doppler catheter recordings. Angiology 1991;42:711-25. Kajiya F, Hoki N, Tomonaga G, Nishihara H. A laser-Dopplervelocimeter using an optical fiber and its application to local velocity measurement in the coronary artery. Experientia 1981;37:1171-3.
Labovitz
et al.
1461
25. Mates RE, Gupta RL, Bell AC, Klocke FJ. Fluid dynamics of coronary artery stenosis. Circ Res 1978;42:152-62. 26. Kilpatrick D, Webber SB. Intravascular blood velocity in simulated coronary artery stenoses. Cathet Cardiovasc Diaan 1986;12:317-23. 27. Grayburn PA, Willard JE, Haagen DR, Brickner ME, Alvarez LG, Eichhorn EJ. Measurement of coronary flow using highfrequency intravascular ultrasound imaging and pulsed Doppler velocimetry: in vitro feasibility studies. J Am Sot Echo 1992;5:5-12. 28. Sudhir K, Hargrave VK, Johnson EL, Aldea G, Mori H, Porte TA, Yock PG. Measurement of volumetric coronary blood flow with a Doppler catheter: Validation in an animal model. AM
H~~~~J1992;124:870-5.
Ametican
using phase-contrast tine magnetic resonance imaging. Invest Radio1 1992;27:465-70. 112. Nienaber CA, Von Kodolitsch Y, Nicolas V, Siglow V, Piepho A, Brockhoff C, Koschyk DH, Spielmann RP. The diagnosis of thoracic aortic dissections by noninvasive imaging procedures. N Engl J Med 1993;328:1-9. 113. Mitchell L, Jenkins JPR, Brownlee WC, Isherwood I. Aortic dissection:,morphology and differential flow velocity patterns demonstrated by magnetic resonance imaging. - _ Clin Radio1 1988;39:458-61. 114. Bogren HG, Underwood SR, Firmin DN, Mohfaddin RH, Klipstein RH, Rees RSO, Longmore DB. Magnetic resonance velocity mappinginaortic dissection. Br JRadiol1988;61:456-
117. 118. 119. 120.
62. _-. 115.
116.
Chang J-M, Friese K, Caputo GR, Kondo C, Higgins CB. MR measurement of blood flow in the true and false channel in chronic aortic dissection. J Comput Assist Tomogr 1991;15: 41823. Bradley WG. Magnetic resonance imaging in the evaluation
Arthur J. Labovitz, MD, Dennis M. Anthonis, Morton J. Kern, MD St. Louis, MO.
the Division University.
of Cardiology,
Department
Reprint requests: Arthur J. Labovitz, MD, Center, Division of Cardiology, FDT-14,3635 St. Louis, MO 63110-0250.
AM H~~~~J1993;126:1456-61. Copyright Q 1993 0002~8703/93/$1.00
1456
by Mosby-Year
f.10
4/l/49915
Book,
Inc.
of Internal
Medicine,
122.
of cerebrospinal fluid abnormalities. Magn Reson Q 1992; 8169-96. Martin AJ, Drake JM, Lemaire C, Henkelman RM. Cerebrospinal fluid shunts: flow measurements with MR imaging. Radiology 1989;173:243-7. Firmin DN, Klipstein RH, Hounsfield GL, Paley MP, Longmore DB. Echo-planar high-resolution flow velocity mapnine. Maan Reson Med 1989:12:316-27. GucfoyleDN, Gibbs P, Ordidge RJ, Mansfield P. Real-time flow measurements using echo-planar imaging. Magn Reson Med 1991;18:1-8. Pearlman JD, Moore JR, Lizak MJ. Real-time NMR beamdirected velocity mapping. V-mode NMR. Circulation 1992;86:1433-8. Kaufman L, Mohiaddin RH, Longmore DB. Letter to the editor, J Thorac Cardiovasc Surg 1991;101:1104-6. Coulden R, Liptan MJ. Magnetic resonance imaging and ultrafast computed tomography in cardiac tomography. Curr Opin Cardiol 1992;7:1007-15.
RDCS, Thomas
There has been extensive experience with Doppler velocity measurements in the assessment of volumetric flow in the aorta and pulmonary artery and across each of the cardiac valves. Sequeira et al17 2 initially demonstrated a good correlation between the timeaveraged velocity or velocity integral in the ascending aorta and invasively measured stroke volume. Subsequent investigations3-6 have shown that stroke volume can be calculated as the product of the time velocity integral and the cross-sectional area of the valve or vessel at the sample site. Cardiac output is then obtained by multiplying the stroke volume by the heart rate. The application of these principles has From Louis
121.
St.
St. Louis University Medical Vista Avenue at Grand Blvd.,
Decemkr 1993 H&f Journar
L. Cravens, RN, and
allowed avariety of clinical Doppler measurements of cardiac blood flow, including absolute blood flow, calculation of intracardiac shunts,7> 8 estimation of aortic valve area,g and assessment of regurgitant fractions.lOl l1 In making these measurements a number of assumptions must be accepted, including (1) the laminarity of blood flow; (2) a Doppler interrogation angle of <20 degrees; and (3) a relatively blunt flow profile from which the measurement is calculated. Although these assumptions are generally true for measurement of blood flow in the heart and great vessels, inadequate data exist validating these assumptions in smaller (<6 mm diameter) arteries. Direct measurement of coronary bIood flow in human coronary arteries has been limited, to a large extent, by the available technology. Transcutaneous interrogation of coronary floti has generally been beyond the resolution of commercially available instruments. Recent reports have indicated that the Doppler measurements of coronary flow might be possible in selected patients by using transesophageal
Volume 126, Number 6 American Heart Journal
Labovitz
et al.
1457
design indicating position of Doppler flow wire for volumetric flow measurement in tubing of various sizes.
Fig. 1. Experimental
echocardiography. l2 Invasively obtained, catheterderived intracoronary flow velocity measurements taken by means of pulsed Doppler, however, have been available for the past several years.13y l4 These directly measured velocity data have been acquired both in the baseline state and during hyperemia induced by a variety of mechanical and pharmacologic stimuli.15-17 Although these velocity measurements have been useful in the clinical assessment of the degree of coronary stenoses, absolute measurements of volumetric coronary flow might be preferable for certain clinical and research questions. The recent availability of a 12 MHz Doppler transducer mounted on an 0.018 inch angioplasty guide wire allows accurate measurement of spectral coronary flow velocities in the distal coronary bed.lse20 By using classic Doppler flow velocity formula, volumetric flow can be computed with appropriate inclusion of the vessel cross-sectional area. Quantitative angiographic and, more recently, intravascular ultrasound imaging21 provides highly accurate determination of the crosssectional area of the coronary vessels. To validate this method for measurement of absolute volumetric blood flow in the coronary vasculature, we examined flaw velocity spectra in an in vitro model applying Doppler formulas commonly applied to large conduits (e.g., aortic, transvalvular flow) for volumetric measurements:obtained in smaller conduits (<6 mm diameter) simulating human coronary arteries. METHODOLOGY
Subselective intracoronaxy flow velocity was measured with an 0.018 inch Doppler angioplasty guide wire (Flowire, Cardiometrics, Inc., Mountain View,
Calif.). As describedIg> 2o and validated by Doucette et a1.,2othe Doppler angioplasty guide wire is a 175 cm long, 0.018 inch diameter flexible steerable guide wire with a 12 MHz piezoelectric ultrasound transducer integrated into the tip. The forward-directed ultrasound beam diverges in a 27-degree arc from the long axis (measured to the -6 decibel round-trip points of the ultrasound beam pattern). The pulse repetition frequency of >40,KHz, pulse duration of +0.83 msec, and sampling delay of 6.5 msec is standard for clinical use. The system is coupled to a real-time spectrum analyzer, videocassette recorder, and video page printer. The Doppler audio signals are processed by the spectrum analyzer with fast Fourier transformation and scrolling gray scale spectral video display. The frequency response of the system calculates approximately 90 spectra per second. Simultaneous ECG and arterial pressures are recorded and displayed on the video monitor. The Doppler guide wire velocity demonstrated excellent correlation with electromagnetic flow velocity and volumetric flow in straight and curved tubed models and in the in vivo testing by using a circumflex canine coronary artery. The Doppler guide wire measures phasic flow velocity patterns and tracks linearly with flow rates in most small, straight coronary ,arteries.lgp 2o Pulsatile flow model. In this experiment a modified two-head roller pump device (Biomedicus Model 540, Minneapolis, Minn.) was connected to a segment of polyurethane tubing (Fig. 1). This tubing was connected to angiographic catheters of various sizes with internal diameters of 6.0 mm, 2.7 mm, 2.1 mm, and 1.9 mm. This system was filled with saline solution mixed with cornstarch echo-reflective particles
1458
Lubovitz et al.
American
December 1993 Hea~Journal
ments for each catheter size; 96 individual measurements were performed. The spectral Doppler tracings were analyzed on custom-developed software in which the outermost portion of the spectral velocity envelope was digitized by using a personal computer and compatible digiting-pad (Fig. 2). Three to five beats were averaged for each flow measurement. Peak and mean velocity measurements and the flow velocity interval were calculated. Total flow was calculated as the product of the cross-sectional area (CSA) of the catheter in which the measurement was made, the time velocity integral (TVI), and the pump cycles per minute (Flow = Rate X TV1 X CSA). Ten flow recordings were traced by two different observers. Interobserver variability for velocity integral measurements was 3.2%. Statistical analysis. Simple linear regression analysis was used to assessthe relationship between the calculated and roller pump blood flow. Results are reported as a mean plus or minus the SD. A probability fp) value of 0.05 was considered to be significant. OBSERVATIONS
Fig. 2. Doppler spectral tracing demonstrating measurement of time velocity integral (WI). Flow wascalculated as product of. velocity integral and cross-sectionalarea of tubing at point of measurementover l-minute period.
(0.5 %) to provide an optimal Doppler spectral display. The Doppler flow wire was introduced into this system through a valved sidearm connector with direct visualization and maintained in a parallel position to flow. Roller pump flow was adjusted incrementally from a minimum of 90 ml/min to a maximal of 550 ml/min for each of the four catheters with different internal luminal diameters. Flow velocity spectra were measured at 10 to 20 ml/min increments after allowing equibration to reach a new steady state. There were approximately 24 flow measure-
There was an excellent correlation between the Doppler calculated absolute flow and pulsed flow model (Fig. 3, Table I). Absolute velocity measurements increased linearly in relation to increases in absolute flow. Volumetric flow output by roller pump varied between 90 and 540 ml/min and by the Doppler flow wire from 71 to 543 ml/min (mean 237 vs 239 ml/min, p = NS). The correlation between roller pump flow and Doppler calculated ff ow for each size tubing was equally strong (r = 0.99). Peak velocities ranged from 4 to 45 cm/set, mean velocity-from 3 to 32 cm/set and total velocity interval from 9 to 30 cm. Doppler flow was related, to pump floti by the equation Doppler flow = 0.94 (pump flow) + 16.4 (r = 0.97, p < 0.0001) CdMMENTS This study showed that accurate assessment of volumetric flow can be obtained by using the Doppler flow wire velocity integral over the flow ranges encountered in clinical situations. Although Doppler velocity measurements have been applied to estimate volumetric flow in the heart and great vessels, absolute floti measurements in the coronary circulation has been much more limited. Only recently have catheters small enough to make measurements within the more distal coronary circulation been developed.
Volume American
126, Number Heart
6
Lab&z
Journal
et al.
1459
600~ y
=
5001
E g
4oo-
#
3001
4 fl
q
0.64x + 16.4
c = 0.67, p c q.6661
2001
0
1.7
.
2.1
0 2.7
.
z
8.0
mm mm mm mm
loo“l-
0
lid
2hO 360 460 Pump Flow (mllmin)
560
6;O
Fig. 3. Correlation between roller pump and Doppler measured
Moreover, many of these systems have used a zero crossing technique that displays the average rather than differential velocity distribution in the luminal of the vesse1.15-17Most recently, a Doppler guide wire has been introduced that will enable investigators to obtain velocity measurements in the distal coronary bed.lsmzo The velocity signals, which are displayed after spectral analysis, provide a weighted real-time scrolling gray scaie image of the proportion of red cells traveling at a given velocity. Although velocity data have several clinical applicat‘ions,22 volumetric data may be particularly useful when assessing pharmacologii: or mechanical responses of coronary flow. Consideration to changes in coronary vessel diameters and cross-sectional area must be made in the analysis of Doppler velocity. Flow-velocity data alone may be insufficient to predict changes in absolute coronary flow in the setup of active coronary vasomotion. An additional limitation is a subtle problem with &gulation that provides interrogating angles of >20 degrees, which wo6ld also result in somewhat lower average velocity or total velocity interval. lnfluencebf flow velocity profile. In vessels of small diameter such as those of the coronary v&culature, the v&lodity,profile across the lumen may become increasingly paraboiic, with more rapid‘velotiities toward the center of the lumen.23-26 In this case, measurement of velocity with a very sm& sample volume in the coaxial position may falsely elevate the mean
Table
I. Volumetric
flow.
flow correlations
Pump flow (mllmin)
‘Doppler flow (mllmin)
‘Tube diameter (mm)
Mean
Range
Mean
Range
r
1.7’ 2.1 2.7 6.0
240 259 241 213
go-400 go-400 go-400 go-540
225 290 241 211
90-360 100-430 110-350 71-543
0.99 0.99 0.99 0.99
velocity estimation and theoretically would overestimate absolute corontiy blood flow. With this in mind, some investigators routinely use formulas that include a correctiori factor for a fraction of the actual average peak velocity measured in small-caliber vessels in the estimation of volumetric blood flo~.~O The current study demonstrates that this correction factor may not be necessary. The broad 2 mm sample volume that interrogates at least 5 mm beyond the tip of the flow wire is one factor that provides a larger proportion of lower velocity measurements and hence a more accurate representation of absolute flow. Previous in vitro measurement systems simulating normal and stenotic coronary vasculature report conflicting results. Recently, however, two studies have reported experimental models in which Doppler catheters were used. Grayburn used a 4.8F Doppler catheter in an in vitro model and demonstrated a
Deceiber
1460
Labovitz et al.
American
good correlation between the product of the mean velocity and cross-sectional area and coronary flow both with pulsatile and nonpulsatile models.27 The data indicated that in the pulsatile model with 4 mm tubing, the profile was blunt enough to allow direct calculation of flow with a good correlation without any correction factor (r = 0.99). Likewise, Sudhir et alz8 used a 3F Doppler catheter to measure flow in the circumflex coronary artery of an animal model. Flow derived by using the average peak velocity correlated closely with that measured by electromagnetic flow meter. Conclusions. The ability to measure absolute coronary flow with a transducer mounted on the tip of a standard angioplasty guide wire has several important potential clinical applications. The results of the present study indicate that analysis of a spectral Doppler velocity may be accurately applied to estimate volumetric flow in vessels of the same caliber as human coronary arteries. SUMMARY
6.
7.
8.
9.
10.
11.
We used an in vitro model to validate volumetric flow measurements obtained with an 0.018-&h angioplasty guidewire with a 12 MHz transducer mounted on its tip. By using a modified two-head roller pump device, flow was adjusted incrementally from a minimum of 90 ml/min to a maximum of 550 ml/m& Flow was measured with the Doppler guide wire in tubing ranging from 1.9 mm to 6.0 mm internal diameter, as the product of the spectral Doppler velocity integral and the cross-sectional area of the tubing, over a l-minute period. It was an excellent correlation between the Doppler calculated flow rates and actual flow, regardless of tubing diameter (r = 0.99). These results suggest that the Doppler spectral output of this device might be accurately applied to estimates of volumetric flow in human coronary arteries. We thank
5.
Dianne
Reid
for her expert
secretarial
assistance.
12.
13.
14.
15.
16.
17.
REFERENCES
1. Huntsman LL, Gams taneous determination
E, Johnson CC, Fairbanks E. Transcuof aortic blood-flow velocities in man.
18.
AMHEART J 1975;89:605-12. 2. Sequeira RF, Light LH, Cross G, Raftery EB. Transcutaneous aortoveloeranhv: a auantitative evaluation. Br Heart J 1976; 38443-501 - 3. Labovitz AJ, Buckingham TA, Habermehl K, Nelson J, Kennedy HL, Williams GA. The effects of sampling site on the two-dimensional echo-Doppler determination of cardiac output. AM HEART J 1985;109:327-32. 4. Dittmann H, Voelker W, Karsch K, Seipel L. Influence of sampling site and flow area on cardiac output measurements
19.
20.
1993
H&ahJcliinal
by Doppler echocardiography. J Am Co11 Cardiol1987;10:81823. Miller WE, Richards KL, Crawford MH. Accuracy of mitral Doppler echocardiographic cardiac output determinations in adults. AM HEART J 1990;119:905-10. Lewis JF, Kuo LC, Nelson JG, Limacher MC, Quinones MA. Pulsed Doppler echocardiographic determination of stroke volume and cardiac output: clinical validation of two new methods using the apical window. Circulation 1984;70:425-31. Meijboom EJ, Valdes-Cruz LM, Horowitz S, Sahn DJ, Larson DF, Yqung KA, Oliveira Lima C, Goldberg SJ, Allen HD. A two-dimensional Doppler echocardiographic method for calculation of pulmonary and systemic blood flow in a canine model with a variable-sized left-to-right extracardiac shunt. Circulation 1983;68:437-45. Hori M, Yoshima H, Ohnishi K, Kitabatake A, Inoue M, Asao M, Ito H, Masuyama T, Tanouchi J, Morita T. Noninvasive evaluation of the ratio of pulmonary to systemic flow in atrial septal defect -by duplex Doppler echocardiography. Circulation 1984;69:73-9. Otto CM, Pearlman AS, Comess KA, Reamer RP, Janko CL, Huntsmann LL. Determination of the stenotic aortic valve area in ‘adults using Doppler echocardiography. J Am Co11 Cardiol 1986;7:509-17. Ascah KJ, Stewart WJ, Jiang L, Guerrero JL, Newell JB, Gillam LD, Weyman AE. A Doppler two-dimensional echocardiographic method for quantitation of mitral regurgitation. Circulation 1985;72:377-83. Goldberg SJ, Allen HD. Quantitative assessment by Doppler echocardiography of pulmonary or aortic regurgitation. Am J - Cardiol 1985;56:131-5. Iliceto S. Maranaelli V. Memmola C. Rizzon P. Transesonhageal Doppler echocardiography evaluation of coronary blood flow velocity in baseline conditions and during dipyridamoleinduced coronary vasodilation. Circulation L991;83:61-9. Marcus ML. White CW. Coronarv flow reserve in natients with normal’coronary a.ngiograms.“J Am Co11 Cardioi 1985;6: 1254-6. Wilson RF, Johnson MJ, Marcus ML, Aylward PE, Skorton DJ, Collins S, White CW. The effect of coronary angioplasty on coronary flow reserve. Circulation 1988;76: 873. Kern MJ, Deligonul U, Vandormael M, Labovitz A, Gudipati CV, Gabliani G, Bodet J, Shah Y, Kennedy HL. Impaired coronary vasodilator reserve in the intermediate post coronary angioplasty period: analysis of coronary artery flow velocity indexes and regional cardiac venous afflux. J Am Co11 Cardiol 1989;13:860-72. Wilson RF, Laughlin DE, Ackell PH, Chilian WM, Holida MD. Hartlev CJ. Armstrong ML. Marcus ML. White CW. Transluminal, s&selective measurement of coronary artery blood flow velocity and vasodilator reserve in man. Circulation 1985;72:82. Sibley DH, Millar HD, Hartley CJ, Whitlow PL. Subselective measurement of coronary blood flow velocity using a steerable Doppler catheter. J Am Co11 Cardiol 1986;8:133240. Ofili EO, Kern MJ, Labovitz AJ, St. Vrain JA, Segal J, Aguirre FV, Caste110 R. Analysis of coronary blood flow velocity dynamics in angiographically normal and stenosed arteries before and after endolumen enlargement by angioplasty. J Am Co11 Cardiol 1993;21:308-16. Segal J, Kern MJ, Scott NA, King SB, Doucette JW, Heuser RR, Ofili E, Siegel R. Alterations of phasic coronary artery flow velocity in humans during percutaneous coronary angioplasty. J Am Co11 Cardiol 1992;20:276-86. Doucette JW, Corl PD, Payne HM, Flynn AE, Goto M, Nassi M, Segal J. Validation of a Doppler guide wire for intravascu-
Volume 126, Number 6 American Heart Journal
21.
22. 23.
24.
1a.rmeasurement of coronary artery flow velocity. Circulation 1992;85:1899-911, Tobis JM, Mallery J, Mahon D, Lehmann K, Zalesky P, Griffith J, Gessert J, Moriuchi M, McRae M, Dwyer ML. Intravascular ultrasound imaging of human coronary arteries in vivo. Circulation 1991;83:913-26. Kern MJ, Donohue TJ, Bach RG, Caracciolo EA, Aguirre FV. Clinical applications of the Doppler coronary flow velocity guidewire. Cardiovasc Intervent Radio1 1993;3:10-20. Denardo SJ, Yock PG, Hargrave VK, Srebro JP, Ports TA, Talbot L. Differentiation of abnormal blood flow patterns in coronary arteries based on Doppler catheter recordings. Angiology 1991;42:711-25. Kajiya F, Hoki N, Tomonaga G, Nishihara H. A laser-Dopplervelocimeter using an optical fiber and its application to local velocity measurement in the coronary artery. Experientia 1981;37:1171-3.
Labovitz
et al.
1461
25. Mates RE, Gupta RL, Bell AC, Klocke FJ. Fluid dynamics of coronary artery stenosis. Circ Res 1978;42:152-62. 26. Kilpatrick D, Webber SB. Intravascular blood velocity in simulated coronary artery stenoses. Cathet Cardiovasc Diaan 1986;12:317-23. 27. Grayburn PA, Willard JE, Haagen DR, Brickner ME, Alvarez LG, Eichhorn EJ. Measurement of coronary flow using highfrequency intravascular ultrasound imaging and pulsed Doppler velocimetry: in vitro feasibility studies. J Am Sot Echo 1992;5:5-12. 28. Sudhir K, Hargrave VK, Johnson EL, Aldea G, Mori H, Porte TA, Yock PG. Measurement of volumetric coronary blood flow with a Doppler catheter: Validation in an animal model. AM
H~~~~J1992;124:870-5.