- Email: [email protected]
Feasibility of obtaining pulmonary venous flow velocity in cardiac patients using transthoracic pulsed wave Doppler technique
Feasibility of Obtaining Pulmonary Venous Flow Velocity in Cardiac Patients Using Transthoracic Pulsed Wave Doppler Technique Joan L. Jensen, RDMS, Faye E. Williams, Brenda J. Beilby, RDCS, Bonnie L. Johnson, RDCS, RN, Lisa K. Miller, Timothy L. Ginter, RDMS, Gina Tomaselli-Martin, and Christopher P. Appleton, MD, Scotrsdale, Arizona
The purpose of this study was to determine, in an adult population, the percentage of patients in whom high quality pulmonary venous flow velocity recordings can be obtained using current transthoracic pulsed wave Doppler techniques. Pulmonary venous and mitral flow velocity variables obtained with a pulsed wave Doppler method were used for the indirect assessment of left ventricular (LV) diastolic function and LV filling pressures. The general clinical use of these methods, however, remains uncertain because the transthoracic success rate of obtaining all components of pulmonary venous flow velocity has been variable, and sometimes reported to be as low as 30% to 60%. Mitral and pulmonary venous flow velocity variables were obtained using pulsed wave Doppler signals in 200 consecutive adult patients (mean age 68.2 -+ 11.4 years) in normal sinus rhythm who were referred for echocardiographic study. Six cardiac sonographers and five ultrasound systems were used. The success rate for obtaining pulmonary venous systolic and diastolic flow velocity was 95%, reverse flow velocity at atrial contrac-
Mitral
flow velocity and related variables obtained with Doppler ultrasound are being used for the indirect assessment o f left ventricular (LV) diastolic function and LV filling pressures. T M More recently, Doppler pulmonary venous flow velocity variables also have been used as complementary information to this analysis, with the pulmonary venous systolic fraction being related to the mean left atrial pressure 12,~s and the difference between mitral and pulmonary venous flow duration at atrial contraction being re-
From the Division of Cardiovascular Diseases, Mayo Clinic Scottsdale. Reprint requests: Christopher P. Appleton, MD, Cardiovascular Diseases, Mayo Clinic Scottsdale, 13400 East Shea Blvd., Scottsdale, AZ 85259. Copyright 9 1997 by the American Society of Echocardiography. 0894-7317/97 $5.00+0 2 7 / 1 / 7 4 4 9 0
60
tion was 90%, and the duration of reverse flow at atrial contraction was 89%. In the 5% to 11% of patients in whom pulmonary flow velocities could not be adequately recorded, the most common reasons were depth limitations of the pulsed wave Doppler machine, marked cardiac enlargement, or left atrial wall motion artifact. The success rate also was influenced by the ultrasound equipment used, individual variation among sonographers, and even the type (impaired, pseudonormal, restricted) of associated mitral filling pattern. Given current machine technology, sonographer education, and daily practice, high quality, complete recordings of pulmonary venous flow velocity can be obtained in approximately 90% of adult patients using the precordial transthoracic Doppler technique. These results suggest that using these variables as an aid for evaluating LV diastolic function and filling pressures may have broader clinical applicability than previously appreciated. (J Am Soc Echocardiogr 1997; 10:60-6.)
lated to the LV end-diastolic pressure) 3,~4 Recent studies also have shown that transthoracic pulmonary venous recordings, although not o f as high a technical quality as transesophageal recordings, appear to be accurate enough to be used in clinical practice. ~s06 Despite these encouraging clinical results, obtaining high quality recordings o f pulmonary venous flow velocity in all patients has proved to be technically challenging because o f the depth o f the pulmonary veins, wall motion artifacts associated with left atrial motion, and variable body habitus in adult patientsfl 7-19 As a result, even in experienced echocardiographic laboratories, the success rate o f obtaining all measurable pulmonary venous flow velocity variables in adults has been reported to be as low as 30% tO 60%. 15'16"20
It is our impression that success in obtaining pulmonary venous flow velocity variables can be signifi-
Journal of the AmericanSocietyof Echocardiograpby Volume 10 Number 1
Jensen et al.
A
61
B
E
P atA
H Mdt
PVS2 PVv~
Ac
I
H
.d
LVIVRT Adur
v
PVA
M PVAdur
Figure 1 Schematic diagram ofmitral (A) and pulmonary venous (B) flow velocities illustrating the Doppler variables measured in this study. Thesc variables included peak mitral flow velocity in early diastole (E wave), mitral flow velocity at the start of atrial contraction (E at A), peak mitral flow velocity at atrial contraction (A wave), mitral A wave duration (A dur), mitral deceleration time (Mdt), peak pulmonary venous flow velocity during early ventricular systole (PV $1), peak pulmonary venous flow velocity during late ventricular systole (PV $2), peak pulmonary venous flow velocity, during ventricular diastole (PV D), peak reverse pulmonary venous flow velocity at atrial contraction (PVA), and the duration of reverse pulmonary venous flow at atrial contraction (PVAdur). The velocity time integrals of pulmonary venous flow also were measured. cantly increased t h r o u g h a s o n o g r a p h e r e d u c a t i o n p r o g r a m , a t t e n t i o n t o technical detail, 19 a n d o b t a i n ing these r e c o r d i n g s in every p a t i e n t u n d e r g o i n g echo D o p p l e r studies. T h e r e f o r e , the p u r p o s e o f o u r study was to d e t e r m i n e the c u r r e n t success rate in o b t a i n i n g p u l m o n a r y venous flow velocities using transthoracic a p u l s e d wave D o p p l e r t e c h n i q u e in a general a d u l t p o p u l a t i o n .
METHODS Patient Population The study population consisted of 200 consecutive patients in normal sinus rhythm who wcrc referred fbr transthoracic echocardiographic study. Only patients with atrial fibrillation, atrial flutter, or ventricular paced rhythm were excluded from the study. Patients with prosthetic valves were not excluded, and the study included 7 patients with prosthetic aortic valves and 3 patients with prosthetic mitral valves. The average age of the patients was 68 _+ 11.4 years (range 22 to 94 years). The study population comprised 103 women and 97 men. Echocardiography A complete M-mode, two-dimensional (2D) and Doppler echocardiographic study was done by any of six sonographers using one of three different brands of commercially available cardiac ultrasound equipment (Acuson XP-10,
Mountain View, CA; Hewlett-Packard 1500, Andover, MA; or Vingmed CFM-800, Milpitas, CA). Mitral flow velocity was recorded with a 2.0 to 2.5 M H z transducer from an apical transducer position with a pulsed wave Doppler technique using a 1 to 2 mm sample volume placed between the tips of the mitral valve leaflets. 1.19 Color flow Doppler was used to help align the Doppler beam with mitral inflow. In nearly all cases, optimal alignment required the transducer to be moved to a position 15 to 20 degrees lateral to the standard apical imaging position. 1,) Pulmonary venous flow velocities also were obtained from an apical 4-chamber position using a 2 to 3 mm sample volume placed 1 to 2 cm into the right superior pulmonary vein) 9,2~ The Doppler velocity filter (200 to 400 Hz) and gain were set as low as possible to fhcilitate measurement of peak reverse flow velocity and duration at atrial contraction. If the pulmonary vein orifice was difficult to visualize, the use of color flow Doppler helped locate and align the Doppler beam to pulmonary venous inflow. In comparison with the sample volume position used for antegrade flow velocities, the sharpest reverse flow velocity envelope at atrial contraction often was obtained at a position slightly further into the pulmonary vein. 19 If the Doppler signal was weak or the velocity envelope was incomplete, the use of a larger sample volume size (4 to 5 mm), modified apical transducer position, higher Doppler gain, or reimaging with the patient in a supine position were attempted to improve the spectral quality. When a satisfactory signal from the right upper pulmonary vein could not be obtained, or limitations of the machine pulsed
62
Journal of the American Society of Echocardiography January-February 1997
Jensen et al.
Table 1
Mitral flow velocity variables in the patient study group Variable
Peak E ( c m / s ) Peak A ( c m / s ) E at A ( c m / s ) E / A ratio M d t (ms) A dur (ms) LV IVRT (ms) Mitral-PV Adur (ms)
Mean values
82.3 83.2 18.9 1.09 226.8 129.7 93.8 -23.8
+- 25.4 + 28.5 + 15.6 + 0.46 + 71.5 + 17.8 _+ 20.7 _+ 27.3
All values are _+ SD. Peak E, peak mitral flow velocity in early diastole; Peak A, peak mitral flow velocity at atrial contraction; E at A, mitral flow velocity just before atrial contraction; Mdt, mitrat deceleration time; A dur, duration of mitral flow at atrial contraction; LV IVRT, left ventricular isovolumetric relaxation time; Mitral-PVAdur, difference in duration of mitral as compared to pulmonary venous flow velocity at atrial contraction.
wave Doppler depth existed, parasternal or suprasternal views also were attempted. Because mitral, and especially pulmonary venous, flow varies with respiration, ~7-19measurements were made from recordings taken during apnea at a recording speed of 100 mm/s. All data also were recorded using a super VHS video cassette recorder.
Data Processing and Analysis Echocardiographic variables. The echocardiographic variables were measured over three consecutive beats and were averaged. The LV ejection fraction was calculated from M-mode dimensions and the left atrial ejection fraction by the area-length method. The mitral and pulmonary venous flow velocity variables measured are shown in Figure 1. The duration of reverse flow at atrial contraction was measured directly from the width of the velocity signal, or by measuring the discontinuation between pulmonary venous diastolic and systolic antegrade flow. 22 When the latter method was used, the end of pulmonary venous diastolic flow was checked with the end of the P wave on the electrocardiogram to make sure it was coincident with, and not before, mechanical atrial systole. In cases where the pulmonary venous flow in systole was biphasic,12 14,17both peak velocities and velocity time integrals were measured. Statistical analysis All v a l u e s are e x p r e s s e d as m e a n -+ 1 S D . T h e g r o u p m e a n s for h e m o d y n a m i c a n d D o p p l e r variables w e r e calcul a t e d u s i n g a statistical p a c k a g e (SAS I n c . , C a r y , N C ) for a personal computer.
RESULTS H e m o d y n a m i c Variables The mean heart was 69 --- 13 b e a t s / m i n u t e and the mean PR interval was 180 + 28 ms. Thirty-three
Table 2
Pulmonary venous flow velocity variables in the patient study group Variable
Mean values
PV Sl ( c m / s ) PV S 1 VTI (cm) PV S 2 ( c m / s ) PV S2 VTI (crn) PV D ( c m / s ) PV D VTI (cm) PV A (crn/s) P V A V T I (cm)
46.6 6.60 58.7 14.14 44.1 9.95 30.8 3.20
_+ 13.6 + 2.89 + 13.5 + 4.29 + 13.6 -+ 3.72 -+ 7.4 -+ 0.95
PV Adur (ms)
153.7 + 26.7
All values are _+ SD. PV St, peak velocity of pulmonary venous flow in early systole; PV S~ velocity of pulmonary venous flow in late systole; PV D, peak velocity of pulmonary venous flow in diastole; PVA, peak pulmonary venous reverse flow velocity at atrial contraction; PV Adur, duration of reverse pulmonary venous flow at atrial contraction; VTI, velocity time integral of specified flow.
patients (17%) had a first degree (greater than 200 ms) atrioventricular block.
Echocardiographic F i n d i n g s Thirteen patients (6.5%) had normal echocardiogram findings, whereas the remaining 187 patients had cardiac abnormalities such as LV hypertrophy, valvular stenosis or regurgitation, LV systolic dysfunction, or elevated filling pressures. The mean LV ejection fraction was 60.3% _ 13.3% and the mean LA ejection fraction was 29.2% + 9.8%. The mean values for selected mitral flow velocity variables are shown in Table I and pulmonary venous flow velocity variables in Table 2. The mitral E wave velocity was measurable in 199 o f the 200 patients. In 55 patients (28%), mitral flow velocity at the start o f atrial contraction (E at A velocity) was greater than 20 c m / s , indicating that the E / A ratio was affected by a fast heart, a prolonged P R interval, or delayed mitral valve opening. 2a As a result o f this, the mitral deceleration time could not be measured in 7 patients. The mitral A wave duration was measurable in all 200 patients. The left ventricular isovolumic relaxation time (IVRT) was recorded in 196 patients (98%) using continuous wave Doppler technique in 126 patients (63%) and pulsed wave Doppler technique in 70 patients ( 3 5 % ) . 19 By previously defined criteria, l 9 o f 200 (5%) mitral patterns were normal, 103 (52%) showed impaired LV relaxation, 67 (34%) were pseudonormal, 10 (5%) were restricted, and 10 (5%) were unclassifiable (five A wave only or near complete E / A fusion and five with prosthetic mitral valves). Table 3 shows the success rate o f obtaining the pulmonary venous flow velocity variables. Biphasic pulmonary venous systolic flow with early and late systolic components was seen in 64 patients (32%). In
Journal of the American Socie~ of Echocardiography Volume 10 Number 1
10 patients, no pulmonary flow velocities could be adequately recorded. This was because of depth limitations of the pulsed wave Doppler signal in 3 patients ( 30% ), cardiac enlargement in 2 patients (20%), excessive left atrial wall motion noise in 2 patients (20%), ultrasound interference from mechanical prosthetic valves (one mitral and one aortic) in 2 patients (20%), and chronic pulmonary disease in 1 patient (10%). In 14 patients, antegrade pulmonary venous flows were well recorded but reverse flow velocity (4 patients), flow duration (6 patients), or both (4 patients) at atrial contraction could not be measured. In 8 patients, the signals were technically inadequate; 2 patients had severe lung disease, 2 patients had morbid obesity, 2 patients had long (greater than 250 ms) first degree atrioventricular block, and 1 patient had a markedly dilated heart. In 1 patient, reverse flow at atrial contraction was well recorded although antegrade flows were inadequate for measurement. In patients in whom inadequate pulmonary venous signals were present from a standard left lateral apical transducer position, a more supine apical position or parasternal position resulted in an improved signal in 2 patients (1%). Examples of pulmonary venous flow velocities from 3 study patients are shown in Figure 2. Of the 24 patients who had incomplete pulmonary venous Doppler assessment, 4 (17%) were noted to have normal mitral patterns (1 with depth limitation), 12 (50%) had impaired relaxation mitral patterns, 3 (13%) had pseudonormal patterns (1 because of depth limitation of the pulsed wave Doppler signal), 3 (13%) had restricted patterns (2 with depth limitation), 1 had a prosthetic valve signal, and 1 had a mitral A wave only. Excluding the 6 patients in whom a pulsed wave Doppler depth limitation, a prosthetic valve, or E / A fusion were present, an incomplete pulmonary venous Doppler assessment occurred most often in patients with an impaired relaxation mitral pattern (8 of 18) and less frequently in patients with pseudonormal and restricted (1 of 18 each) mitral patterns. Variability in success rates also was present among individual sonographers and the three types of ultrasound machines. The success rate of obtaining all components of the pulmonary venous flow velocity signal (except PV S 1) among the individual sonographers ranged from 80% to 93% (individual values = 93%, 92%, 92%, 88%, 88%, and 80%; average 89%). Inability to obtain any measurable pulmonary venous signal varied among machines from 2.7% to 15%, whereas failure to obtain the A wave components varied between 8% and 25%.
J e n s e n et al.
63
Table 3 Successrates of obtaining pulmonary venous flow velocity variables in the patient study group (n = 200)
PV S2 PV S2 VTI PV D PV D VTI PV A PV A VTI PV Adur
No. pts
Percent
190/200 190/200
95 95
189/200 189/200
95 95
180/200 178/200 178/200
90 89 89
Variable abbreviations are the same as in Table 2.
DISCUSSION
The noninvasive evaluation of LV diastolic function and filling pressures using pulsed wave Doppler ultrasound has become a part of routine clinical examinations in many echocardiographic laboratories. As detailed in recent reviews, 17 ~9the analysis of pulmonary venous flow velocity recordings aids this evaluation and provides information that can be directly related to left heart filling pressures. However, the clinical use of these methods in a general adult echocardiographic population is uncertain because previous studies have reported a wide variation in obtaining pulmonary venous flow velocities using transthoracic techniques. The main findings of this transthoracic Doppler study are that using current ultrasound equipment, antegrade (95%) and retrograde (89%) pulmonary venous flow velocity variables are obtainable in a high percentage of elderly patients with heart disease who are undergoing echocardiographic study. Technically inadequate recordings were present in only 5% to 11% of patients and were caused by depth limitations of the pulsed wave Doppler signal, cardiac enlargement, excessive left atrial wall motion noise, obesity, and pulmonary disease. As reflected by the lower success rate, pulmonary venous reverse flow at atrial contraction was more difficult to record than forward pulmonary venous flow velocities. The most limiting technical factor was excessive wall motion noise associated with atrial contraction. The results of this study also indicate that success in obtaining recordings of pulmonary venous flow velocity is dependent on the capabilities of individual sonographers, the ultrasound equipment used, and even the type of mitral flow velocity pattern present. For instance, the success of obtaining all components of pulmonary venous flow velocity varied by 13%
64
Journal of the American Society of Echocardiography January February 1997
J e n s e n et al.
A
0.8 -
B
~
~
0.4 -
C
PV S 2
1
,,t t [~t;"
0.2 -
~, l~ ~
'!
,
' :
0
PV Adur
Figure 2 Pulmonary venous flow velocity recordings obtained with pulsed wave Doppler in three individual study patients. Panels A and B show high quality recordings in which all pulmonary venous flow velocity variables shown in Figure 1 can bc measured. Panel C shows a patient with a technically inadequate recording, with incomplete systolic and diastolic flow velocity envelopes and no distinct reverse flow at atrial contraction. Low velocity wall motion artifact associated with left atrial filling and contraction is present in C (arrows).
among the individual sonographers, from 80% to 93%. Similarly, the inability to obtain any measurable pulmonary venous flow velocity varied among machines from 3.5% to 8%, and failure to obtain the A wave components varied between 8% and 25%. The reduced success rate o f one type o f machine appeared to be related to excessive noise on the spectral display, especially in the A wave signal, whereas failures in another machine mostly were related to depth limitations o f the pulsed wave Doppler signal in large patients or patients with marked cardiomegaly. Finally, pulmonary venous flow velocity was more difficult to record in patients with impaired patterns in relaxation mitral flow velocity as compared with patients who had pseudonormal or restrictive physiology mitral patterns. Excessive wall motion artifact caused by normal size pulmonary veins and vigorous atrial contraction was most noticeable in the impaired relaxation group.
Comparison w i t h Previous Studies In transthoracic studies in pediatric populations o f both healthy subjects and patients with anomalous pulmonary venous rcturn, a nearly 100% success rate in recording pulmonary vcnous flow has been reported. 24'25 Presumably this reflects the smaller body size o f these individuals and the greater sensitivity o f pulsed wave Doppler ultrasound at lesser depths. In contrast, the percentages o f technically satisfactory recordings for both antegrade and retrograde pulmo-
nary venous flow velocities are lower in the adult population, with reported ranges from 37% to 86% in healthy i n d i v i d u a l s , lsA6,26"27 In older patients with coronary artery disease, systolic and diastolic pulmonary venous flow has been successfully recorded in 72% to 98% o f the patients, with lower success rates (39% to 92%) reported for recording reverse flow at atrial contraction.~3,14,2~ Excluding machine factors, the higher success rate in obtaining pulmonary venous flow velocities in this study are attributable to several factors. A typical 'learning' curve with initially lower success rates was not present because the study was undertaken after completion o f a 1-year educational program on obtaining optimal Doppler flow velocity recordings for evaluating LV diastolic function? 9 The sonographer group had an average o f 1 5 years scanning experience and was already familiar with routinely obtaining pulsed wave Doppler signals. Furthermore, both mitral and pulmonary venous recordings were obtained in every patient undergoing an echo Doppler study. Thcrefore, this study likely reflects the best results obtainable with current equipment and transthoracic technique. We found that attention to several technical details markedly increased the quality o f the pulmonary venous flow velocity recordings. The best pulmonary venous flow velocity signals were obtained by placing a 2 to 3 mm pulsed wave sample volume 1 to 2 cm into the fight upper pulmonary vein, using an apical
Journal of the American Society of Echocardiography Volume 10 Number I
Table 4 Technique used for obtaining pulmonary venous flow velocity Transducer location
Apical 4-chamber
Identify right upper PV orifice Alignment Sample volume size Sample volume placement* Doppler filter setting Final adjustments
Color flow Doppler Parallel to flow 2-3 mm 1-2 cm into pulmonary vein Low (200-400 MHz) Clarity of audio/spectral signal
*May vary slightly depending on antegrade and retrograde PV flow.
4-chamber view with a slight anterior angulation. The proper sample volume location often was found most rapidly by first directly visualizing flow entering the left atrium with a color flow Doppler method. With the velocity filter setting as low as possible (200 to 400 Hz), final small adjustments in sample volume position and beam angulation were made while listening to the clarity of the audio signal and observing the spectral display. When the pulmonary flow velocity signal was suboptimal, a larger sample volume size (4 to 5 mm), higher Doppler gain or reimaging with the patient in the supine position often improved the spectral quality. A higher, modified apical transducer position or the left upper pulmonary vein was tried if the recordings remained unsatisfactory. Additional imaging positions from which pulmonary venous recordings could occasionally be made include the parasternal short axis, suprasternal notch, and subcostal positions. Tables 4 and 5 summarize the suggested technique for obtaining pulmonary venous flow velocity and the common pitfalls. Although they are not the primary focus of this study, several other results are noteworthy. Using a precordial technique, 32% of successful pulmonary venous recordings showed distinct early (PV $1 ) and late (PV $2) systolic components. An early component was more commonly observed with pulsed wave sample volumes located closer to the pulmonary venous orifice or in patients with prolonged PR intervals. In 28% of patients, the mitral flow velocity at the start of atrial contraction (E at A velocity) was greater than 20 cm/s, indicating that the A wave velocity was affected (incrcased) by a 'relatively' fast heart, a prolonged PR interval, or a delayed mitral valve opening. 2s Therefore, the E / A ratio (which usually is used for the initial classification of LV filling pattern) was decreased by 'partial' fusion of E wave and A wave filling peaks in almost one third of thc elderly population studied. Futurc studies using LV filling patterns to assess diastolic function in the
Jensen et al.
65
Table 5
Common technical pitfalls negatively influencing the quality of pulmonary venous flow velocity recordings Doppler PW depth limitations (machine/patient) Improper machine Dopplcr settings SV size, location Doppler gain Filter settings Left atrial wall motion artifact Failure to try multiple transducer positions Variation in pulmonary venous flow with respiration
elderly may need to take this information into account. The mean mitral A wave duration was 130 ms whereas the mean duration of reverse pulmonary venous flow at atrial contraction was 154 ms. An increased (greater than 30 ms) duration of reverse compared to forward flow at atrial contraction was seen in 36% of the patients, indicating that the LV end-diastolic pressure was elevated. 13,14 This likely reflects the high prevalence of cardiac disease in the current study population. The smaller standard deviation of mitral A wave duration compared with reverse pulmonary venous flow at atrial contraction suggests that reverse pulmonary venous flow changes more between healthy and diseased states. Limitations
The results of this study reflect the three types of ultrasound equipment used and the expertise of the sonographers who performed the examinations. Similarly, it is possible that using more than one sonographer to obtain pulmonary venous flow velocity in difficult cases may have slightly increased the percentage of successful recordings. Patients with atrial fibrillation were excluded from this study because they lacked pulmonary venous A wave and early systolic flow velocity components, making a comparison of success rates with patients in normal sinus rhythm difficult. It has bccn our experience that obtaining signals ofantegrade diastolic flow velocity in patients in atrial fibrillation is actually easier because the wall motion artifact is reduced without normal atrial contraction and a higher volume of flow exists during diastole. Obtaining the pulmonary venous late systolic component can be challenging, however, because of reduced volume and a lower peak flow velocity. Conclusions
Given current machine technology, sonographer education, and daily practice, high quality pulmonary venous flow velocity recordings can be obtained in
66
Journal of the American Societyof Echocardiography January-February 1997
Jensen et al.
m o s t ( a p p r o x i m a t e l y 90%) a d u l t p a t i e n t s u s i n g precordial t r a n s t h o r a c i c D o p p l e r t e c h n i q u e s . T h e s e results s u g g e s t t h a t u s i n g p u l m o n a r y v e n o u s f l o w vel o c i t y variables for e v a l u a t i n g L V diastolic f u n c t i o n a n d filling pressures m a y h a v e b r o a d e r clinical applicability t h a n p r e v i o u s l y a p p r e c i a t e d .
14.
We thank Marvin Ruona for his technical assistance in preparing the figures and Amy Weaver for statistical help. 15. REFERENCES
1. Appleton CP, Hatle LK, Popp RL. Relation oftransmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coil Cardiol 1988;12:426-40. 2. Nishimura R, Housmans PR, Hatle LK, Tajik AJ. Assessment of diastolic function of the heart: background and current applications of Doppler echocardiography. Part II: Clinical studies. Mayo Clinic Proc 1989;64:181-204. 3. Vanoverschelde JJ, Raphael DA, Robert AR, Cosyns JR. Left ventricular filling in DCM: relation to functional class and hemodynamics. J Am Coil Cardiol 1990;15:1288-95. 4. Nishimura RA, Abel MD, Hatle LK, Tajik AJ. Relation of pulmonary vein to mitral flow velocities by transesophageal echocardiography. Effect of different loading conditions. Circulation 1990;81 :1488-97. 5. Klein AL, Hatle LK, Taliercio CP, et al. Prognostic significance of Doppler measures of diastolic function in cardiac amyloidosis. A Doppler echocardiographic study. Circulation 1991;83:808-16. 6. Mulvagh S, Quinones MA, Kleiman NS, Cheirif J, Zoghbi WA. Estimation of left ventricular end-diastolic pressure from Doppler transmitral flow velocity in cardiac patients independent of systolic performance. J Am Coil Cardiol 1992;20:112-9. 7. Oh JK, Ding ZP, Gersh BJ, Bailey KR, Tajik AJ. Restrictive left ventricular diastolic filling identifies patients with heart failure after acute myocardial infarction. J Am Soc Echocardiogr 1992;5:497-503. 8. Pinamonti B, Di Lenardo A, Sinagra G, Camerini F. Restrictive left ventricular filling pattern in dilated cardiomyopathy assessed by Doppler echocardiography: clinical, echocardiographic and hemodynamic correlations and prognostic implications. J Am Coil Cardiol 1993;22:808-15. 9. Ortiz J, Matsumoto AY, Ghefter C, et al. Prognosis in dilated myocardial disease: influence of diastolic dysfunction and anatomic changes. Echocardiography 1993;10:247-53. 10. Giannuzzi P, Imparato A, Temporelli PL, et al. Dopplerderived mitral deceleration time of early filling as a strong predictor of pulmonary capillary wedge pressure in postinfarction patients with left ventricular systolic dysfunction. J Am Coil Cardiol 1994;23:1630-7. 11. Rihal CS, Nishimura RA, Hatle LK, Bailey KR, Tajik AJ. Systolic and diastolic dysfunction in patients with clinical diagnosis of dilated cardiomyopathy. Relation to symptoms and prognosis. Circulation 1994;90:2772-9. 12. Kuercherer HF, Muhiudeen IA, Kusumoto FM, et al. Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow. Circulation 1990;82:1127-39. 13. RossvoU O, Hatle LK. Pulmonary venous flow velocities re-
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
corded by transthoracic Doppler ultrasound: relations to left ventricular diastolic pressures. J Am Coll Cardiol 1993;21: 1687-96. Appleton CP, Galloway JM, Gonzalez MS, Gaballa M, Basnight MA. Estimation of left ventricular filling pressures using two-dimensional and Doppler echocardiography in adult cardiac patients: additional value of analyzing left atrial size, left atrial ejection fraction and the difference in the duration of pulmonary venous and mitral flow velocities at atrial contraction. J Am Coll Cardiol 1993;22:1972-82. Castello R, Pearson A, Lenzen P, Labovitz A. Evaluation of pulmonary venous flow by transesophageal echocardiography in subjects with a normal heart: comparison with transthoracic echocardiography. J Am Coil Cardiol 1991;18:65-71. Masuyama T, Nagano R, Nariyama K, et al. Transthoracic Doppler echocardiographic measurements of pulmonary venous flow velocity patterns: comparison with transesophageal measurements. J Am Soc Echocardiogr 1995;8:61-9. Klein AL, TajikAJ. Doppler assessment of pulmonary venous flow in healthy subjects and patients with heart disease. J Am Soc Echocardiogr 1991 ;4:379-92. Meijburg HWJ, Visser CA. Pulmonary venous flow as assessed by Doppler echocardiography. Echocardiography 1995;12: 425-40. Appleton CP, Hargrove JL, Hatle LK, Oh JK. Doppler evaluation of left and right ventricular diastolic function: a technical guide for obtaining optimal flow velocity recordings. J Am Soc Echocardiogr. In Press. Brunazzi MC, Chirillo F, Pasqualini M, et al. Estimation of left ventricular diastolic pressures from precordial pulsedDoppler analysis of pulmonary venous and mitral flow. Am Heart J 1994;128:293-300. Basnight MA, Gonzalez MS, Kershenovich SC, Appleton CP. Pulmonary venous flow velocity: relation to hemodynamics, mitral flow velocity and left atrial volume and ejection fraction. J Am Soc Echocardiogr 1991;4:547-58. Appleton CP. The Doppler assessment of left ventricular diastolic function: the refinements continue. J Am Coil Cardiol 1993;21:1697-700. Appleton CP. Influence of incremental changes in heart rate on mitral flow velocity: assessment in lightly sedated dogs. J Am Coil Cardiol 1991;17:227-36. Smallhorn J, Freedom R. Pulsed Doppler echocardiography in the preoperative evaluation of total anomalous pulmonary venous connection. J Am Coil Cardiol 1986;8:1413-20. Smallhorn J, Freedom R, Olley P. Pulsed Doppler echocardiographic assessment of extraparenchymal pulmonary vein flow. J Am Coil Cardiol 1987;9:573-9. Masuyama T, Lee J-M, Tamai M, Tanouchi J, Kitabatake A, Kamada T. Pulmonary venous flow velocity pattern as assessed with transthoracic pulsed Doppler echocardiography in subjects without cardiac disease. Am J Cardiol 1991;67:1396404. Klein AL, Burstow DJ, Tajik AJ, Zachariah PK, Bailey KR, Seward JB. Effects of age on left ventricular dimensions and filling dynamics in 117 normal persons. Mayo Clin Proc 1994;69:212-24. Masuyama T, Ye J-M, Yamamoto I4, Tanouchi J, Hori M, Kamada T. Analysis of pulmonary venous flow velocity patterns in hypertensive hearts: its complementary value in the interpretation of mitral flow velocity patterns. Am Heart J 1992;124:983-94.
The purpose of this study was to determine, in an adult population, the percentage of patients in whom high quality pulmonary venous flow velocity recordings can be obtained using current transthoracic pulsed wave Doppler techniques. Pulmonary venous and mitral flow velocity variables obtained with a pulsed wave Doppler method were used for the indirect assessment of left ventricular (LV) diastolic function and LV filling pressures. The general clinical use of these methods, however, remains uncertain because the transthoracic success rate of obtaining all components of pulmonary venous flow velocity has been variable, and sometimes reported to be as low as 30% to 60%. Mitral and pulmonary venous flow velocity variables were obtained using pulsed wave Doppler signals in 200 consecutive adult patients (mean age 68.2 -+ 11.4 years) in normal sinus rhythm who were referred for echocardiographic study. Six cardiac sonographers and five ultrasound systems were used. The success rate for obtaining pulmonary venous systolic and diastolic flow velocity was 95%, reverse flow velocity at atrial contrac-
Mitral
flow velocity and related variables obtained with Doppler ultrasound are being used for the indirect assessment o f left ventricular (LV) diastolic function and LV filling pressures. T M More recently, Doppler pulmonary venous flow velocity variables also have been used as complementary information to this analysis, with the pulmonary venous systolic fraction being related to the mean left atrial pressure 12,~s and the difference between mitral and pulmonary venous flow duration at atrial contraction being re-
From the Division of Cardiovascular Diseases, Mayo Clinic Scottsdale. Reprint requests: Christopher P. Appleton, MD, Cardiovascular Diseases, Mayo Clinic Scottsdale, 13400 East Shea Blvd., Scottsdale, AZ 85259. Copyright 9 1997 by the American Society of Echocardiography. 0894-7317/97 $5.00+0 2 7 / 1 / 7 4 4 9 0
60
tion was 90%, and the duration of reverse flow at atrial contraction was 89%. In the 5% to 11% of patients in whom pulmonary flow velocities could not be adequately recorded, the most common reasons were depth limitations of the pulsed wave Doppler machine, marked cardiac enlargement, or left atrial wall motion artifact. The success rate also was influenced by the ultrasound equipment used, individual variation among sonographers, and even the type (impaired, pseudonormal, restricted) of associated mitral filling pattern. Given current machine technology, sonographer education, and daily practice, high quality, complete recordings of pulmonary venous flow velocity can be obtained in approximately 90% of adult patients using the precordial transthoracic Doppler technique. These results suggest that using these variables as an aid for evaluating LV diastolic function and filling pressures may have broader clinical applicability than previously appreciated. (J Am Soc Echocardiogr 1997; 10:60-6.)
lated to the LV end-diastolic pressure) 3,~4 Recent studies also have shown that transthoracic pulmonary venous recordings, although not o f as high a technical quality as transesophageal recordings, appear to be accurate enough to be used in clinical practice. ~s06 Despite these encouraging clinical results, obtaining high quality recordings o f pulmonary venous flow velocity in all patients has proved to be technically challenging because o f the depth o f the pulmonary veins, wall motion artifacts associated with left atrial motion, and variable body habitus in adult patientsfl 7-19 As a result, even in experienced echocardiographic laboratories, the success rate o f obtaining all measurable pulmonary venous flow velocity variables in adults has been reported to be as low as 30% tO 60%. 15'16"20
It is our impression that success in obtaining pulmonary venous flow velocity variables can be signifi-
Journal of the AmericanSocietyof Echocardiograpby Volume 10 Number 1
Jensen et al.
A
61
B
E
P atA
H Mdt
PVS2 PVv~
Ac
I
H
.d
LVIVRT Adur
v
PVA
M PVAdur
Figure 1 Schematic diagram ofmitral (A) and pulmonary venous (B) flow velocities illustrating the Doppler variables measured in this study. Thesc variables included peak mitral flow velocity in early diastole (E wave), mitral flow velocity at the start of atrial contraction (E at A), peak mitral flow velocity at atrial contraction (A wave), mitral A wave duration (A dur), mitral deceleration time (Mdt), peak pulmonary venous flow velocity during early ventricular systole (PV $1), peak pulmonary venous flow velocity during late ventricular systole (PV $2), peak pulmonary venous flow velocity, during ventricular diastole (PV D), peak reverse pulmonary venous flow velocity at atrial contraction (PVA), and the duration of reverse pulmonary venous flow at atrial contraction (PVAdur). The velocity time integrals of pulmonary venous flow also were measured. cantly increased t h r o u g h a s o n o g r a p h e r e d u c a t i o n p r o g r a m , a t t e n t i o n t o technical detail, 19 a n d o b t a i n ing these r e c o r d i n g s in every p a t i e n t u n d e r g o i n g echo D o p p l e r studies. T h e r e f o r e , the p u r p o s e o f o u r study was to d e t e r m i n e the c u r r e n t success rate in o b t a i n i n g p u l m o n a r y venous flow velocities using transthoracic a p u l s e d wave D o p p l e r t e c h n i q u e in a general a d u l t p o p u l a t i o n .
METHODS Patient Population The study population consisted of 200 consecutive patients in normal sinus rhythm who wcrc referred fbr transthoracic echocardiographic study. Only patients with atrial fibrillation, atrial flutter, or ventricular paced rhythm were excluded from the study. Patients with prosthetic valves were not excluded, and the study included 7 patients with prosthetic aortic valves and 3 patients with prosthetic mitral valves. The average age of the patients was 68 _+ 11.4 years (range 22 to 94 years). The study population comprised 103 women and 97 men. Echocardiography A complete M-mode, two-dimensional (2D) and Doppler echocardiographic study was done by any of six sonographers using one of three different brands of commercially available cardiac ultrasound equipment (Acuson XP-10,
Mountain View, CA; Hewlett-Packard 1500, Andover, MA; or Vingmed CFM-800, Milpitas, CA). Mitral flow velocity was recorded with a 2.0 to 2.5 M H z transducer from an apical transducer position with a pulsed wave Doppler technique using a 1 to 2 mm sample volume placed between the tips of the mitral valve leaflets. 1.19 Color flow Doppler was used to help align the Doppler beam with mitral inflow. In nearly all cases, optimal alignment required the transducer to be moved to a position 15 to 20 degrees lateral to the standard apical imaging position. 1,) Pulmonary venous flow velocities also were obtained from an apical 4-chamber position using a 2 to 3 mm sample volume placed 1 to 2 cm into the right superior pulmonary vein) 9,2~ The Doppler velocity filter (200 to 400 Hz) and gain were set as low as possible to fhcilitate measurement of peak reverse flow velocity and duration at atrial contraction. If the pulmonary vein orifice was difficult to visualize, the use of color flow Doppler helped locate and align the Doppler beam to pulmonary venous inflow. In comparison with the sample volume position used for antegrade flow velocities, the sharpest reverse flow velocity envelope at atrial contraction often was obtained at a position slightly further into the pulmonary vein. 19 If the Doppler signal was weak or the velocity envelope was incomplete, the use of a larger sample volume size (4 to 5 mm), modified apical transducer position, higher Doppler gain, or reimaging with the patient in a supine position were attempted to improve the spectral quality. When a satisfactory signal from the right upper pulmonary vein could not be obtained, or limitations of the machine pulsed
62
Journal of the American Society of Echocardiography January-February 1997
Jensen et al.
Table 1
Mitral flow velocity variables in the patient study group Variable
Peak E ( c m / s ) Peak A ( c m / s ) E at A ( c m / s ) E / A ratio M d t (ms) A dur (ms) LV IVRT (ms) Mitral-PV Adur (ms)
Mean values
82.3 83.2 18.9 1.09 226.8 129.7 93.8 -23.8
+- 25.4 + 28.5 + 15.6 + 0.46 + 71.5 + 17.8 _+ 20.7 _+ 27.3
All values are _+ SD. Peak E, peak mitral flow velocity in early diastole; Peak A, peak mitral flow velocity at atrial contraction; E at A, mitral flow velocity just before atrial contraction; Mdt, mitrat deceleration time; A dur, duration of mitral flow at atrial contraction; LV IVRT, left ventricular isovolumetric relaxation time; Mitral-PVAdur, difference in duration of mitral as compared to pulmonary venous flow velocity at atrial contraction.
wave Doppler depth existed, parasternal or suprasternal views also were attempted. Because mitral, and especially pulmonary venous, flow varies with respiration, ~7-19measurements were made from recordings taken during apnea at a recording speed of 100 mm/s. All data also were recorded using a super VHS video cassette recorder.
Data Processing and Analysis Echocardiographic variables. The echocardiographic variables were measured over three consecutive beats and were averaged. The LV ejection fraction was calculated from M-mode dimensions and the left atrial ejection fraction by the area-length method. The mitral and pulmonary venous flow velocity variables measured are shown in Figure 1. The duration of reverse flow at atrial contraction was measured directly from the width of the velocity signal, or by measuring the discontinuation between pulmonary venous diastolic and systolic antegrade flow. 22 When the latter method was used, the end of pulmonary venous diastolic flow was checked with the end of the P wave on the electrocardiogram to make sure it was coincident with, and not before, mechanical atrial systole. In cases where the pulmonary venous flow in systole was biphasic,12 14,17both peak velocities and velocity time integrals were measured. Statistical analysis All v a l u e s are e x p r e s s e d as m e a n -+ 1 S D . T h e g r o u p m e a n s for h e m o d y n a m i c a n d D o p p l e r variables w e r e calcul a t e d u s i n g a statistical p a c k a g e (SAS I n c . , C a r y , N C ) for a personal computer.
RESULTS H e m o d y n a m i c Variables The mean heart was 69 --- 13 b e a t s / m i n u t e and the mean PR interval was 180 + 28 ms. Thirty-three
Table 2
Pulmonary venous flow velocity variables in the patient study group Variable
Mean values
PV Sl ( c m / s ) PV S 1 VTI (cm) PV S 2 ( c m / s ) PV S2 VTI (crn) PV D ( c m / s ) PV D VTI (cm) PV A (crn/s) P V A V T I (cm)
46.6 6.60 58.7 14.14 44.1 9.95 30.8 3.20
_+ 13.6 + 2.89 + 13.5 + 4.29 + 13.6 -+ 3.72 -+ 7.4 -+ 0.95
PV Adur (ms)
153.7 + 26.7
All values are _+ SD. PV St, peak velocity of pulmonary venous flow in early systole; PV S~ velocity of pulmonary venous flow in late systole; PV D, peak velocity of pulmonary venous flow in diastole; PVA, peak pulmonary venous reverse flow velocity at atrial contraction; PV Adur, duration of reverse pulmonary venous flow at atrial contraction; VTI, velocity time integral of specified flow.
patients (17%) had a first degree (greater than 200 ms) atrioventricular block.
Echocardiographic F i n d i n g s Thirteen patients (6.5%) had normal echocardiogram findings, whereas the remaining 187 patients had cardiac abnormalities such as LV hypertrophy, valvular stenosis or regurgitation, LV systolic dysfunction, or elevated filling pressures. The mean LV ejection fraction was 60.3% _ 13.3% and the mean LA ejection fraction was 29.2% + 9.8%. The mean values for selected mitral flow velocity variables are shown in Table I and pulmonary venous flow velocity variables in Table 2. The mitral E wave velocity was measurable in 199 o f the 200 patients. In 55 patients (28%), mitral flow velocity at the start o f atrial contraction (E at A velocity) was greater than 20 c m / s , indicating that the E / A ratio was affected by a fast heart, a prolonged P R interval, or delayed mitral valve opening. 2a As a result o f this, the mitral deceleration time could not be measured in 7 patients. The mitral A wave duration was measurable in all 200 patients. The left ventricular isovolumic relaxation time (IVRT) was recorded in 196 patients (98%) using continuous wave Doppler technique in 126 patients (63%) and pulsed wave Doppler technique in 70 patients ( 3 5 % ) . 19 By previously defined criteria, l 9 o f 200 (5%) mitral patterns were normal, 103 (52%) showed impaired LV relaxation, 67 (34%) were pseudonormal, 10 (5%) were restricted, and 10 (5%) were unclassifiable (five A wave only or near complete E / A fusion and five with prosthetic mitral valves). Table 3 shows the success rate o f obtaining the pulmonary venous flow velocity variables. Biphasic pulmonary venous systolic flow with early and late systolic components was seen in 64 patients (32%). In
Journal of the American Socie~ of Echocardiography Volume 10 Number 1
10 patients, no pulmonary flow velocities could be adequately recorded. This was because of depth limitations of the pulsed wave Doppler signal in 3 patients ( 30% ), cardiac enlargement in 2 patients (20%), excessive left atrial wall motion noise in 2 patients (20%), ultrasound interference from mechanical prosthetic valves (one mitral and one aortic) in 2 patients (20%), and chronic pulmonary disease in 1 patient (10%). In 14 patients, antegrade pulmonary venous flows were well recorded but reverse flow velocity (4 patients), flow duration (6 patients), or both (4 patients) at atrial contraction could not be measured. In 8 patients, the signals were technically inadequate; 2 patients had severe lung disease, 2 patients had morbid obesity, 2 patients had long (greater than 250 ms) first degree atrioventricular block, and 1 patient had a markedly dilated heart. In 1 patient, reverse flow at atrial contraction was well recorded although antegrade flows were inadequate for measurement. In patients in whom inadequate pulmonary venous signals were present from a standard left lateral apical transducer position, a more supine apical position or parasternal position resulted in an improved signal in 2 patients (1%). Examples of pulmonary venous flow velocities from 3 study patients are shown in Figure 2. Of the 24 patients who had incomplete pulmonary venous Doppler assessment, 4 (17%) were noted to have normal mitral patterns (1 with depth limitation), 12 (50%) had impaired relaxation mitral patterns, 3 (13%) had pseudonormal patterns (1 because of depth limitation of the pulsed wave Doppler signal), 3 (13%) had restricted patterns (2 with depth limitation), 1 had a prosthetic valve signal, and 1 had a mitral A wave only. Excluding the 6 patients in whom a pulsed wave Doppler depth limitation, a prosthetic valve, or E / A fusion were present, an incomplete pulmonary venous Doppler assessment occurred most often in patients with an impaired relaxation mitral pattern (8 of 18) and less frequently in patients with pseudonormal and restricted (1 of 18 each) mitral patterns. Variability in success rates also was present among individual sonographers and the three types of ultrasound machines. The success rate of obtaining all components of the pulmonary venous flow velocity signal (except PV S 1) among the individual sonographers ranged from 80% to 93% (individual values = 93%, 92%, 92%, 88%, 88%, and 80%; average 89%). Inability to obtain any measurable pulmonary venous signal varied among machines from 2.7% to 15%, whereas failure to obtain the A wave components varied between 8% and 25%.
J e n s e n et al.
63
Table 3 Successrates of obtaining pulmonary venous flow velocity variables in the patient study group (n = 200)
PV S2 PV S2 VTI PV D PV D VTI PV A PV A VTI PV Adur
No. pts
Percent
190/200 190/200
95 95
189/200 189/200
95 95
180/200 178/200 178/200
90 89 89
Variable abbreviations are the same as in Table 2.
DISCUSSION
The noninvasive evaluation of LV diastolic function and filling pressures using pulsed wave Doppler ultrasound has become a part of routine clinical examinations in many echocardiographic laboratories. As detailed in recent reviews, 17 ~9the analysis of pulmonary venous flow velocity recordings aids this evaluation and provides information that can be directly related to left heart filling pressures. However, the clinical use of these methods in a general adult echocardiographic population is uncertain because previous studies have reported a wide variation in obtaining pulmonary venous flow velocities using transthoracic techniques. The main findings of this transthoracic Doppler study are that using current ultrasound equipment, antegrade (95%) and retrograde (89%) pulmonary venous flow velocity variables are obtainable in a high percentage of elderly patients with heart disease who are undergoing echocardiographic study. Technically inadequate recordings were present in only 5% to 11% of patients and were caused by depth limitations of the pulsed wave Doppler signal, cardiac enlargement, excessive left atrial wall motion noise, obesity, and pulmonary disease. As reflected by the lower success rate, pulmonary venous reverse flow at atrial contraction was more difficult to record than forward pulmonary venous flow velocities. The most limiting technical factor was excessive wall motion noise associated with atrial contraction. The results of this study also indicate that success in obtaining recordings of pulmonary venous flow velocity is dependent on the capabilities of individual sonographers, the ultrasound equipment used, and even the type of mitral flow velocity pattern present. For instance, the success of obtaining all components of pulmonary venous flow velocity varied by 13%
64
Journal of the American Society of Echocardiography January February 1997
J e n s e n et al.
A
0.8 -
B
~
~
0.4 -
C
PV S 2
1
,,t t [~t;"
0.2 -
~, l~ ~
'!
,
' :
0
PV Adur
Figure 2 Pulmonary venous flow velocity recordings obtained with pulsed wave Doppler in three individual study patients. Panels A and B show high quality recordings in which all pulmonary venous flow velocity variables shown in Figure 1 can bc measured. Panel C shows a patient with a technically inadequate recording, with incomplete systolic and diastolic flow velocity envelopes and no distinct reverse flow at atrial contraction. Low velocity wall motion artifact associated with left atrial filling and contraction is present in C (arrows).
among the individual sonographers, from 80% to 93%. Similarly, the inability to obtain any measurable pulmonary venous flow velocity varied among machines from 3.5% to 8%, and failure to obtain the A wave components varied between 8% and 25%. The reduced success rate o f one type o f machine appeared to be related to excessive noise on the spectral display, especially in the A wave signal, whereas failures in another machine mostly were related to depth limitations o f the pulsed wave Doppler signal in large patients or patients with marked cardiomegaly. Finally, pulmonary venous flow velocity was more difficult to record in patients with impaired patterns in relaxation mitral flow velocity as compared with patients who had pseudonormal or restrictive physiology mitral patterns. Excessive wall motion artifact caused by normal size pulmonary veins and vigorous atrial contraction was most noticeable in the impaired relaxation group.
Comparison w i t h Previous Studies In transthoracic studies in pediatric populations o f both healthy subjects and patients with anomalous pulmonary venous rcturn, a nearly 100% success rate in recording pulmonary vcnous flow has been reported. 24'25 Presumably this reflects the smaller body size o f these individuals and the greater sensitivity o f pulsed wave Doppler ultrasound at lesser depths. In contrast, the percentages o f technically satisfactory recordings for both antegrade and retrograde pulmo-
nary venous flow velocities are lower in the adult population, with reported ranges from 37% to 86% in healthy i n d i v i d u a l s , lsA6,26"27 In older patients with coronary artery disease, systolic and diastolic pulmonary venous flow has been successfully recorded in 72% to 98% o f the patients, with lower success rates (39% to 92%) reported for recording reverse flow at atrial contraction.~3,14,2~ Excluding machine factors, the higher success rate in obtaining pulmonary venous flow velocities in this study are attributable to several factors. A typical 'learning' curve with initially lower success rates was not present because the study was undertaken after completion o f a 1-year educational program on obtaining optimal Doppler flow velocity recordings for evaluating LV diastolic function? 9 The sonographer group had an average o f 1 5 years scanning experience and was already familiar with routinely obtaining pulsed wave Doppler signals. Furthermore, both mitral and pulmonary venous recordings were obtained in every patient undergoing an echo Doppler study. Thcrefore, this study likely reflects the best results obtainable with current equipment and transthoracic technique. We found that attention to several technical details markedly increased the quality o f the pulmonary venous flow velocity recordings. The best pulmonary venous flow velocity signals were obtained by placing a 2 to 3 mm pulsed wave sample volume 1 to 2 cm into the fight upper pulmonary vein, using an apical
Journal of the American Society of Echocardiography Volume 10 Number I
Table 4 Technique used for obtaining pulmonary venous flow velocity Transducer location
Apical 4-chamber
Identify right upper PV orifice Alignment Sample volume size Sample volume placement* Doppler filter setting Final adjustments
Color flow Doppler Parallel to flow 2-3 mm 1-2 cm into pulmonary vein Low (200-400 MHz) Clarity of audio/spectral signal
*May vary slightly depending on antegrade and retrograde PV flow.
4-chamber view with a slight anterior angulation. The proper sample volume location often was found most rapidly by first directly visualizing flow entering the left atrium with a color flow Doppler method. With the velocity filter setting as low as possible (200 to 400 Hz), final small adjustments in sample volume position and beam angulation were made while listening to the clarity of the audio signal and observing the spectral display. When the pulmonary flow velocity signal was suboptimal, a larger sample volume size (4 to 5 mm), higher Doppler gain or reimaging with the patient in the supine position often improved the spectral quality. A higher, modified apical transducer position or the left upper pulmonary vein was tried if the recordings remained unsatisfactory. Additional imaging positions from which pulmonary venous recordings could occasionally be made include the parasternal short axis, suprasternal notch, and subcostal positions. Tables 4 and 5 summarize the suggested technique for obtaining pulmonary venous flow velocity and the common pitfalls. Although they are not the primary focus of this study, several other results are noteworthy. Using a precordial technique, 32% of successful pulmonary venous recordings showed distinct early (PV $1 ) and late (PV $2) systolic components. An early component was more commonly observed with pulsed wave sample volumes located closer to the pulmonary venous orifice or in patients with prolonged PR intervals. In 28% of patients, the mitral flow velocity at the start of atrial contraction (E at A velocity) was greater than 20 cm/s, indicating that the A wave velocity was affected (incrcased) by a 'relatively' fast heart, a prolonged PR interval, or a delayed mitral valve opening. 2s Therefore, the E / A ratio (which usually is used for the initial classification of LV filling pattern) was decreased by 'partial' fusion of E wave and A wave filling peaks in almost one third of thc elderly population studied. Futurc studies using LV filling patterns to assess diastolic function in the
Jensen et al.
65
Table 5
Common technical pitfalls negatively influencing the quality of pulmonary venous flow velocity recordings Doppler PW depth limitations (machine/patient) Improper machine Dopplcr settings SV size, location Doppler gain Filter settings Left atrial wall motion artifact Failure to try multiple transducer positions Variation in pulmonary venous flow with respiration
elderly may need to take this information into account. The mean mitral A wave duration was 130 ms whereas the mean duration of reverse pulmonary venous flow at atrial contraction was 154 ms. An increased (greater than 30 ms) duration of reverse compared to forward flow at atrial contraction was seen in 36% of the patients, indicating that the LV end-diastolic pressure was elevated. 13,14 This likely reflects the high prevalence of cardiac disease in the current study population. The smaller standard deviation of mitral A wave duration compared with reverse pulmonary venous flow at atrial contraction suggests that reverse pulmonary venous flow changes more between healthy and diseased states. Limitations
The results of this study reflect the three types of ultrasound equipment used and the expertise of the sonographers who performed the examinations. Similarly, it is possible that using more than one sonographer to obtain pulmonary venous flow velocity in difficult cases may have slightly increased the percentage of successful recordings. Patients with atrial fibrillation were excluded from this study because they lacked pulmonary venous A wave and early systolic flow velocity components, making a comparison of success rates with patients in normal sinus rhythm difficult. It has bccn our experience that obtaining signals ofantegrade diastolic flow velocity in patients in atrial fibrillation is actually easier because the wall motion artifact is reduced without normal atrial contraction and a higher volume of flow exists during diastole. Obtaining the pulmonary venous late systolic component can be challenging, however, because of reduced volume and a lower peak flow velocity. Conclusions
Given current machine technology, sonographer education, and daily practice, high quality pulmonary venous flow velocity recordings can be obtained in
66
Journal of the American Societyof Echocardiography January-February 1997
Jensen et al.
m o s t ( a p p r o x i m a t e l y 90%) a d u l t p a t i e n t s u s i n g precordial t r a n s t h o r a c i c D o p p l e r t e c h n i q u e s . T h e s e results s u g g e s t t h a t u s i n g p u l m o n a r y v e n o u s f l o w vel o c i t y variables for e v a l u a t i n g L V diastolic f u n c t i o n a n d filling pressures m a y h a v e b r o a d e r clinical applicability t h a n p r e v i o u s l y a p p r e c i a t e d .
14.
We thank Marvin Ruona for his technical assistance in preparing the figures and Amy Weaver for statistical help. 15. REFERENCES
1. Appleton CP, Hatle LK, Popp RL. Relation oftransmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coil Cardiol 1988;12:426-40. 2. Nishimura R, Housmans PR, Hatle LK, Tajik AJ. Assessment of diastolic function of the heart: background and current applications of Doppler echocardiography. Part II: Clinical studies. Mayo Clinic Proc 1989;64:181-204. 3. Vanoverschelde JJ, Raphael DA, Robert AR, Cosyns JR. Left ventricular filling in DCM: relation to functional class and hemodynamics. J Am Coil Cardiol 1990;15:1288-95. 4. Nishimura RA, Abel MD, Hatle LK, Tajik AJ. Relation of pulmonary vein to mitral flow velocities by transesophageal echocardiography. Effect of different loading conditions. Circulation 1990;81 :1488-97. 5. Klein AL, Hatle LK, Taliercio CP, et al. Prognostic significance of Doppler measures of diastolic function in cardiac amyloidosis. A Doppler echocardiographic study. Circulation 1991;83:808-16. 6. Mulvagh S, Quinones MA, Kleiman NS, Cheirif J, Zoghbi WA. Estimation of left ventricular end-diastolic pressure from Doppler transmitral flow velocity in cardiac patients independent of systolic performance. J Am Coil Cardiol 1992;20:112-9. 7. Oh JK, Ding ZP, Gersh BJ, Bailey KR, Tajik AJ. Restrictive left ventricular diastolic filling identifies patients with heart failure after acute myocardial infarction. J Am Soc Echocardiogr 1992;5:497-503. 8. Pinamonti B, Di Lenardo A, Sinagra G, Camerini F. Restrictive left ventricular filling pattern in dilated cardiomyopathy assessed by Doppler echocardiography: clinical, echocardiographic and hemodynamic correlations and prognostic implications. J Am Coil Cardiol 1993;22:808-15. 9. Ortiz J, Matsumoto AY, Ghefter C, et al. Prognosis in dilated myocardial disease: influence of diastolic dysfunction and anatomic changes. Echocardiography 1993;10:247-53. 10. Giannuzzi P, Imparato A, Temporelli PL, et al. Dopplerderived mitral deceleration time of early filling as a strong predictor of pulmonary capillary wedge pressure in postinfarction patients with left ventricular systolic dysfunction. J Am Coil Cardiol 1994;23:1630-7. 11. Rihal CS, Nishimura RA, Hatle LK, Bailey KR, Tajik AJ. Systolic and diastolic dysfunction in patients with clinical diagnosis of dilated cardiomyopathy. Relation to symptoms and prognosis. Circulation 1994;90:2772-9. 12. Kuercherer HF, Muhiudeen IA, Kusumoto FM, et al. Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow. Circulation 1990;82:1127-39. 13. RossvoU O, Hatle LK. Pulmonary venous flow velocities re-
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
corded by transthoracic Doppler ultrasound: relations to left ventricular diastolic pressures. J Am Coll Cardiol 1993;21: 1687-96. Appleton CP, Galloway JM, Gonzalez MS, Gaballa M, Basnight MA. Estimation of left ventricular filling pressures using two-dimensional and Doppler echocardiography in adult cardiac patients: additional value of analyzing left atrial size, left atrial ejection fraction and the difference in the duration of pulmonary venous and mitral flow velocities at atrial contraction. J Am Coll Cardiol 1993;22:1972-82. Castello R, Pearson A, Lenzen P, Labovitz A. Evaluation of pulmonary venous flow by transesophageal echocardiography in subjects with a normal heart: comparison with transthoracic echocardiography. J Am Coil Cardiol 1991;18:65-71. Masuyama T, Nagano R, Nariyama K, et al. Transthoracic Doppler echocardiographic measurements of pulmonary venous flow velocity patterns: comparison with transesophageal measurements. J Am Soc Echocardiogr 1995;8:61-9. Klein AL, TajikAJ. Doppler assessment of pulmonary venous flow in healthy subjects and patients with heart disease. J Am Soc Echocardiogr 1991 ;4:379-92. Meijburg HWJ, Visser CA. Pulmonary venous flow as assessed by Doppler echocardiography. Echocardiography 1995;12: 425-40. Appleton CP, Hargrove JL, Hatle LK, Oh JK. Doppler evaluation of left and right ventricular diastolic function: a technical guide for obtaining optimal flow velocity recordings. J Am Soc Echocardiogr. In Press. Brunazzi MC, Chirillo F, Pasqualini M, et al. Estimation of left ventricular diastolic pressures from precordial pulsedDoppler analysis of pulmonary venous and mitral flow. Am Heart J 1994;128:293-300. Basnight MA, Gonzalez MS, Kershenovich SC, Appleton CP. Pulmonary venous flow velocity: relation to hemodynamics, mitral flow velocity and left atrial volume and ejection fraction. J Am Soc Echocardiogr 1991;4:547-58. Appleton CP. The Doppler assessment of left ventricular diastolic function: the refinements continue. J Am Coil Cardiol 1993;21:1697-700. Appleton CP. Influence of incremental changes in heart rate on mitral flow velocity: assessment in lightly sedated dogs. J Am Coil Cardiol 1991;17:227-36. Smallhorn J, Freedom R. Pulsed Doppler echocardiography in the preoperative evaluation of total anomalous pulmonary venous connection. J Am Coil Cardiol 1986;8:1413-20. Smallhorn J, Freedom R, Olley P. Pulsed Doppler echocardiographic assessment of extraparenchymal pulmonary vein flow. J Am Coil Cardiol 1987;9:573-9. Masuyama T, Lee J-M, Tamai M, Tanouchi J, Kitabatake A, Kamada T. Pulmonary venous flow velocity pattern as assessed with transthoracic pulsed Doppler echocardiography in subjects without cardiac disease. Am J Cardiol 1991;67:1396404. Klein AL, Burstow DJ, Tajik AJ, Zachariah PK, Bailey KR, Seward JB. Effects of age on left ventricular dimensions and filling dynamics in 117 normal persons. Mayo Clin Proc 1994;69:212-24. Masuyama T, Ye J-M, Yamamoto I4, Tanouchi J, Hori M, Kamada T. Analysis of pulmonary venous flow velocity patterns in hypertensive hearts: its complementary value in the interpretation of mitral flow velocity patterns. Am Heart J 1992;124:983-94.