152 JMC Vol. 24, No. I July 1994: 152-8

RICK A. NISHl URA, MD, FACC, CHARANJIT S. RI ,

DAVID R. HOLMES, JR., MQ FACC

Rochester, Minnesota

for

rement of the tranmPrd gmtient, but its accuracy has remained u&!ar.

Accurate measurement of the mean gradient across the mitral valve during diastole is of paramount importance in the evaluation of the patient with mitral stenosis. The severity of obstruction is reflected by the absolute transmi- tml gradient in the rest state and with exercise. Calculation of valve area by the Gorlin equation (I) requires accurate knowledge of this transmitral gradient.

Simultaneous measurement of left atrial and left ventric- ular pressures is the most accurate method by which the mean mitral valve gradient can be derived in patients with mitral stenosis. This method requires transseptal catheter- ization, a procedure with known inherent risks, For direct measurement of left atrial pressure (23. Nearly five decades ago, measurement of the pulmonary capillary wedge pres-

From the Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester. Minnesota.

Manuscript received November 8. 1993; revised manuscript received Febmary 8. 1994, accepted February 11, 1994.

Address.far Dr. Rick A. Nishimura. Mayo Clinic, 200 Fit Street SW, Rochester, Minnesota SSWS.

0 1994 by the American College of Cardiology

-.

sure was proposed as a means for indirectly dete~m~u~ng left atrial pressure (3-5). Since then, pulmonary capillary wedge pressure has been used in many invasive laboratories For determination of mean transmitral gradient (6). However, investigators have recently cautioned against the use of pulmonary capillary wedge pressure in patients with mitral stenosis and prosthetic mitral valves because a clinically significant overestimation of the true transmitra~ gradient may result (7-9).

Doppler echocardiography has been proposed as a non- in-gasive method For measuring the transmi!ral gradient in patients with mitral stenosis (9-l I). The initial studies, however. compared the ~ppler-delved mean gmdient with that obtained by cardiac catheterization with use of pulmo- nary capillary wedge pressure (9-l 1). Because of the poten- tial inaccuracies associated with this catheterization method (74% the true accuracy of Doppler echocardiography has remained unclear.

The purpose of this study was to examine prosp.*ctively the relative accuracy of Doppler echocardiography and conventional cardiac catheterization (using pulmonary cap-

073s1097194/$7.00

JACC Vol. 24, No. 1 July IW4: 152-8

onary artery tree until ressure contour was obtaine

heart catbete~zatioo with the turations were not obtained be-

cause of concern for potential pulolo~ary artery erosion (12). rockenbrough technique (13) was used to perform 1 catheterization in all patients. An 8F transse

sheath was used, and the transseptal puncture was per- formed under fluoroscopic biplane imaging. After copfir tion [hat the trausseptal catheter tip was in the left atrium, 5,OtXI U of heparin was given, and left atria1 pressure

were made directly from tbe sheath placed in of the lef& atrium. All three pressures (left

nary capillary wedge and left ventricular) were

y. A continuous wave Doppler ure was performed during the

with use of either a Hewlett-Packard or an rdiographic instrument. A small, nonimaging,

2.0-MHz continuous wave transducer was placed at the left ventricular apex, and the continuous wave Doppler beam

ressures were recorde

of the u wave on bot

I velocity curves were se of the tra~smitral Dop

pressures were traced. (pressure gradient = 4 instantaneous velocities to derive instantaneous pressures at

edge of the pressure curves. §tat~st~ca~ analysis. Comparison of the mea

the different metkods was done by use of t land and Altman (U), in which the di~ere~c

m~asure~~e~ts is plott e average of the two measurements for each linear regression anal- ysis was performed to e mean gradients. Corn- parison of the mean absolute pressures and tbe slope of the v wave was made with a paired ? test, and p < 0.85 was

considered significant. Intraobserver variability was deter ined for the digitiza-

tion of the transmitral Qoppler How velocity curves. Two random beats from 12 consecutive patients were digitized

154 NI§~tMURA ET AL. saw Vol. 24, No. I ACCURATE TRANSMTRAL GRADIENT MEASUREMENT July 1 : m-E-8

30.0 2 7.0 39.0 t 11.0 22.0 + 7.0 17.0 + 83.0* &@I f O.OW

m m ~~~~ a m v&s c I SD. *p < 0.01 compared with left atrial (LA) pressure. tp < O.aOl compared with left atrial pressure. plllmmy capiky wed&s pressure.

from the Doppler strip recordings. The mean observer V ility was J-1 mm H

The absolute mean pressure, maximal and rn~n~~~ pres- e in maximal and minimal pressures and fk left atrkl and pM~rnou~y ca~~~~ary

pressures arc shown in Table 1. Then was no tween the absolute mean left atrial

Simultaneous Doppler-derived and pressure curves nts with mitral stenosis. In each example, the left

illustrates the pulmonary capillary wedge (PCWP) and left ve ular (LV) pressures with the transmitral LIopplcr velocity curve, and

Qustrates the left atrial (LA) and left ventricular the simultaneous transmitra! Doppler velocity. The

izaGon is very close.

illary wedge pressure/left ventric less scatter than the uncorrected pressure/left ventricular ~~~~e~t compared with the kft

e a *

“l__-______

-20’ ’ J 0 5 10 15 20

(IA+ PCWP gradient)/;!

25

E 20

Q 5 $ 15

r

r-O.77 y-0.67x+4.6

0 5 10 15 al

WLV gradient

0 5 10 15 20

IALV gradient

abbreviations as in Figure 1.

average 53% overesti

ences between the illary wedge and left atrial pressures when a catheter or an end-bode catheter was used

ence between the Doppler gradient and the gradient using us left atrial/left ventricular prassures was 0.2 k

nt of the transmitral gradien

to left ventricular inflow in patients with mitral stenosis. Although direct measurement of left ventricular and left atrial pressures simultaneously is the most accurate method for determination of the mean mitral diastolic gradient, this

1.56 NlSHlMlJRA ET AL. ACCURATE TRANSMITRAL GRADIENT MEASUREMENT

SACC Vol. 24, No. I July 1994: m-8

-0 5 10 15 20

t compared with difference of the ler (Dope) velocity or left atlas (LA)

demonstrating the relation af the Doppler-derived transmitral gradient with the transmitral gradient from left atria! pressure. LV = left ventricular.

necessitates transseptal catheterization with its i~be~~t ks (2). These risks are increased in the telnosis who frequently has a distorte

reased incidence of lefi atrial thrombi. apikuy pressure is used by many ion labo s as an indirect measure-

ment of left atria! pressure (61. The mean the pulmonary capillary wedge and left ven both at rest and with exercise has thus become a well accepted measurement of the transmit with mitral stenosis. Recent reports, h tioned the reliability of this technique for assessment of the transmiti gradient both in patients with native mitral ste- nosis and in those with prosthetic mitral valves (7-9). These reports have shown a consistent overestimation of the true mitral ient when the pulmonary capillary wedge and left ventricular pressures are used, and this may result in erro- neous clinical assessment of the severity of mitral stenosis. On the basis of these studies, it has been recommended that transseptal catheterization be used for the evaluation of transmittal gradients when critical pressure measurements are required (9). Others have suggested that these errors could be reduced by I) use of an end-hole catheter for mea- surement of pulmonary capillary wedge pressure, 2) oxime-

try confirmation, and 3) use of transmission of pressure

ressure com-

contrast to other erron that could

left ve~t~cut~ pressures from catbetel~zation, ~tboMgb ent Mnderestimation of the t~nsrnit~l ech~~dio~~phy (NL11.17).

ith the inherent problems of pulmo~~y cap~t~ary pressure in measurement of the transmitral gradie cussed earlier, the true accuracy of Doppler echoc phy has been unclear. Holen et al. (10) suggested that “the ultrasound technique may well determine the actual gradient more accurately than catheterization,” and a subsequent study showed an excellent correlation of the transmitrat Doppler gradient with catheter gradients when a direct left atrial pressure measurement was used (18). As demonstrated in this prospective, simultaneous study, the Doppler-derived t~~smitral tidiest is eed more accurate than conven- tional catheterization. e have previously shown that this superior accuracy of Doppler echoc iography over con- ventional cardiac catheterization in determination of the transmitral gradient is also seen in patients with prosthetic mitral valves (19).

Liiitation of pulmon surements. The use of the pulmonary capillary wedge pres- sure for indirect assessment of left atria1 pressure was first introduced almost five decades ago (3-5). Early studies demonstrated excellent correlation between the mean pul-

JAW Vol. 24, No. Ii July I :152-a

atrial pressure (21,26,30). In the current study, this change in contour and dampening of the pulmonary capillary wedge pressure was noted to be the cause of the greatest discrep- ancy in the transmitral gradient when it was compared directly with the gradient measured by left atria1 pressure. There was a smaller difference between the maximal and the

pulmonary capillary wedge pressures than between ct left atrial maximal and minimal pressures. The

of the decrease in pressure after mitral valve opening, the slope of the Y descent, was less rapid in the

ary capillary wedge than in the left atrial pressure, resulting in an overestimation of the transmitral gradient. These errors did not occur in ail instances in which the pulmonary capillary wedge pressure was used (Fig. 21, but it was not possible to differentiate which were accurate from the pressure contour alone.

I.

2.

3.

4.

5.

6.

1.

8.

9.

IO.

11.

Gorlin R. Gorlin SC. Hydraulic formula for calculation of the area of the slenotic mitral valve. other cardiac valves, and central circulatory shunts. I. Am Heart J 1951;41:1-29. Conti CR, Ross RS. The risks of cardiac catheterization. Am Heart 9 1%9;78:289-91. Hellems UK, Haynes FW. Dexter L, Kianey TD. Pulmonary capillary pressure in animals estimated by venous and arterX catheterization. Am J Physiol 1948:155:98-105. Hellems NK, Haynes FW. Dexter L. Pulmonary “capillary” pressure in man. J Appl Physiol 1049;2:24-9. Lagerlof PI. We&ii L. Studies on the circulation of blood in man. VI. The pulmonary capillary venous pressure pulse in man. Stand J Clin Lab Invest 1949;l: 147-61. Alpert JS. The lessons of history as reflected in the pulmonary capillary wedge pressure. J Am Coil Cardiol 1989:13:830-l. Schoenfeld MH, Paiacios IF, Hutter AM Jr. Jacoby SS, Block PC. LJnderestimation of prosthetic mitral valve areas: role of traasseptal catheterization in avoiding unnecessary repeat mitral valve surgery. 3 Am Coil Cardiol 1985;5:1387-Y2. Hosenpud JD, McAnuhy JH, Morton MJ. Overeslimation of mitral valve gradients obtained by phasic pulmonary capillary wedge pressure. Cathet Cardiovasc Diagn 1983;8:283-90. Dunn M. Is transseptal catheterization necessary? J Am Coil Cardiol 1985;5: 1393-4. Holen 1. Aaslid R, Landmark K, Simonsen S. Determination of pressure gradient in mitral stenosis with a non-invasive ultrasound Doppler tech- nique. Acta bled Stand 1976;199:455-60. Hatle L. Bnrbakk A, Tromsdal A, Angelsen B. Noninvasive assessment

158 NISHIMURA ET AL. JACC Vol. 24. No. 1 ACCURATE TRANSMITRAL GRAPIENT MEASUREMENT July 1994~ 152-8

12.

13.

14.

1%

16.

II.

18.

19.

26.

21.

22.

of pressure drop in mitral stenosis by Doppler ultrasound. Br Weart J 1978;40:131-46. Wpe LA, HaRajee CI, Markis JE, et al. Fatal pulmonary hemorrhage atter use of the bw-directed balloon-tipped catheter. Ann Intern Med 1979$8544-l. Brockenbrough EC, Braunwald E. A new technic for left ventricular angiocardiography and transseptal left heart catheterization. Am J Cardiol 196&6: 1862-4. Nishimura RA, Schwartz RS, T#k AI, Holmes DR Jr. Noninvasive measurement of tate of left ventricular relaxation by Doppler echocardi-

validation with simuhaneous cardii catheterization. Circula-

Bland JM, Altman DG. Statistical methods for assessing agreement between two methods ofclinieal measurement. Lancet 1~1507-10.

RA, Moore DM Jr, C RG, Hillis LD. Use of pulmonary verity of mitral stenosis: is ttuc IeR

ler ultrasound. J Am Coil Cardi

pressures. Proc Std Meet Mayo Clin 1953;28:72-8. Connolly DC, Kirklin JW, Wood EW. The relationship between pulmo- ~~~~~~~ pressure and left atriai pressure in man. Circ Res

.’ _ Werk6 L, Vamauskas E, Elliasch H, Lagerl6f H, Sennin8 A, Thomasson B. Further evidence that the pulmonary capillary venous pressure pulse in

23. Luchsinger PC, Seipp HW Jr, Fate! DJ

man reflects cyclic pressure changes in the left atrium.

artery-wedge pressure to left atriai pressu

C 337-9.

315-8. 24. Ankeney JL. Further ex evidence that ~u~rn~a capi~~

pressures do not reflect cychc changes in left atria! pressure (mitral lesions and pulmonary embolism). Circ Res 1953;1:58-61.

VO, Malmstriim G LG. Left auricular pressme measure- ts in man. Ann Su

iamond 0, Chonette D. a flow-di~e~tc~ bal~~o-