Duration of pulmonary venous atrial reversal flow velocity and mitral inflow a ware: New measure of severity of cardiac amyloidosis

Duration of Pulmonary Venous Atrial Reversal Flow Velocity and Mitral Inflow A Wave: N e w Measure of Severity of Cardiac Amyloidosis Ibrahim Abdalla, MD, R. Daniel Murray, PhD, Jar Chi Lee, MS, William I. Stewart, MD, FACC, A. Jamil Tajik, MD, FACC, and Allan L. Klein, MD, FACC, Cleveland, Ohio

Transmitral Doppler flow patterns of patients w i t h cardiac amyloidosis evolve f r o m an early impaired relaxation to a n advanced restrictive pattern. This reflects hicreasing severity of diastolic dysfunction a n d h e n c e left v e n t r i c u l a r filling p r e s s u r e s . The d u r a t i o n o f the p u l m o n a r y venous atrial r e v e r s a l flow was recently show n to exceed that of the mitral inflow A wave in patients w i t h left ve ntr ic a la r enddiastolic p r e s s u r e g r e a t e r t h a n 15 mare Hg. The objective o f this study was to assess the utility of this i n d e x as a m e a s u r e of the severity of c a r dia c amyloidosis. Comprehensive transtiloracic 2-dimensional a nd pulsed-wave D o p p l e r echocardiograms o f t h e p u l m o n a r y venous and t r a n s m i t t a l flows w e r e made of 23 patients (10 w o m e n ) w i t h biopsyproven diagnosis of p r i m a r y systemic amyloidosis and of 49 subjects as age-matched n o r m a l controls. T h e amyloidosis g r o u p was divided into nonrestrictive and restrictive subgroups o n the basis of the patients' t r a n s m i t t a l inflow d e c e l e r a t i o n time (>150 and <150 ms, respectively). The dnratiolls of t h e p u l m o n a r y venous a t r i a l r e v e r s a l a nd m i t r a l inflow A wave w e r e measured, and the differences

between the flow durations were compared with the control and published data in the nonrestrictive and restrictive groups. The m e a n d u r a t i o n o f th e pulmona r y venous atrial reversal was significantly longer in the amyloid t h a n the control group ( P < .01). The m e a n duration of the mitral hiflow A wave was significantly s h o r t e r i n the restrictive g r o u p than both the nonrestrictive and the control groups ( P < .05). The d u r a t i o n of th e p u l m o n a r y venous atrial reversal exceeded that of the mitral inflow A wave i n all patients with cardiac amyloidosis. The difference in duration between p u lmo n a r y venous atrial r e v e r s a l a n d m i t r a l i n f l o w A wave was signif ic a ntly g r e a t e r i n th e amyloidosis g r o u p c o m p a r e d w i t h the n o r m a l group, a n d this index v a r i e d significantly w i t h i n the amyloid g r o u p between the a b n o r m a l relaxation and the restrictive groups. The difference in the duration between the p u l m o n a r y v e n o u s a t r i a l r e v e r s a l a n d th e m i t r a l inflow A wave is a r e l i a b l e i n d e x of diastolic function a nd c a n be used to assess the severity of cardiac amyloidosis. (J Am Soc Echocardiogr 1998;11: 1125 33.)

C a r d i a c amyloidosis is an infiltrative cardiomyopathy characterized by predominantly diastolic dysfunction w i t h relatively p r e s e r v e d left ve ntr ic ula r (LV) systolic fmlction. 1 It has been thought of as a restrictive cardiomyopathy caused by infiltration of ventricular walls with amyloid deposits. 2 However, it has subsequently been shown that on the basis of the degree of this infiltration and hence the LVwaU thickness, a continuum of diastolic abnormalities exist in cardiac amyloidosis. 3 This is reflected by the transmittal Doppler flow pattern evolving from an impaired

relaxation pattern early to a pseudonormalized pattern and ending with the restrictive pattern only late in the course of the disease. T h e s e patterns may reflect gradually increasing severity of diastolic dy~function and accordingly, LV filling pressures. Previous studies have shown the value of Doppler echocardiography in evaluation of LV diastolic function by using b o t h mitral i n f l o w and p u l m o n a r y venous flow. 4d4 Rossvoll and Hatle 15 first described that the difference in the duration b e tw e e n the pulmonary venous atrial reversal and the mitral inflow A wave can p r e d i c t the LV end-diastolic pressure (LVEDP) w i t h a sensitivity of 85% and a specificity of 79%.Appleton et all6 also have shown a significant correlation be twee n the same index and LVEDP in patients w i t h coronary artery disease ( r = 0.77).This n e w index was recently shown 17,m to be minimally influenced by the effect of the aging process. 19

From the Section of Cardiovascular Imaging, Department of Cardiology, The ClevelandClinic Foundation and Mayo Clirfic. Reprint requests: Allan L. Klein, MD, The Cleveland Clinic Foundation, Department of Cardiology, Desk F15, 9500 Euclid Ave, Cleveland, OH 44195. E-mail: [email protected]. Copyright 9 1998 by the American Society of Echoc~rdiography. 0894-7317/98 $5.00 + 0 27/1/93786

1125

I126

Journal of the American Society of Echacardiography December 1998

Abdalla et al

8

E

D

> 0

0 <--

<-Adur

F i g u r e 1 Schematic illustration of pulmonary venous flow (left) and mitral inflow (riglJt). Pulmonary venous atrial reversal ( P V A R ) flow duration and transmittal A-wave flow duration are indicated by ftrrowg.

A l t h o u g h t h i s i n d e x is c u r r e n t l y c o n s i d e r e d b y m o s t i n v e s t i g a t o r s to b e o f s i g n i f i c m l t i m p o r t a n c e i n e v a l u a t i o n o f LV d i a s t o l i c f u n c t i o n b y D o p p l e r e c h o c a r d i o g r a p h y , 2~ its v a l u e i n a s s e s s i n g t h e s e v e r i t y o f cardiac a m y l o i d o s i s h a s n o t b e e n d e s c r i b e d previously. T h e r e f o r e , t h e p u r p o s e o f t h i s s t u d y w a s t o assess t h e utility o f t h i s n e w v a r i a b l e as a m e a s u r e of the severity of diastolic dysfunction in patients with cardiac amyloidosis.

Two-dimensional Echocardiography A phased array system (Hewlett-Packard) w i t h a 2.5- or 3.5MHz transducer was used to provide a comprehensive 2dimensional echocardiographic examination, as previously shown. 25 Parasternal short-axis views at file midventricular level were used to derive the following M-mode measurements: LV end-systolic and end-diastolic dimensions, septaI and p o s t e r i o r LV wall thickness, and left atrial dimension. Mean LV wall thickness (MLVWT) was taken as the mean value of septal wall thickness plus mean posterior LV wall thickness. 3

METHODS Doppler Examination Study Population O u r patient p o p u l a t i o n c o n s i s t e d of 23 n o n s e l e c t e d patients (13 men; m e a n age of 61 -+ 12 years, r a n g i n g from 44 to 8 8 years) w i t h biopsy-proven diagnosis of prim a r y s y s t e m i c amyloidosis. D i a g n o s i s w a s p r o v e d by biopsy of s u b c u t a n e o u s fat, rectum, kidney, or b o n e marrow.All patients had the classic 2-dimensional echocardiographic signs of cardiac amyloidosis 1 and w e r e be i ng treated w i t h m e l p h a l a n and p r edni s one , colchicine, or vitamin E. Patients w h o had clinical s y m p t o m s oF congestive heart laifure and were taking cardiac medications were c o n t i n u e d o n t h e i r s a m e r e g i m e n (diuretics, nitrates, d i g o x i n , or afterload-redueing agents). Patients w i t h h y p e r t e n s i o n ( b l o o d p r e s s u r e > 1 4 0 / 9 0 m m Hg), secondary amyloidosis, or wall m o t i o n abnormalities consistent w i t h co r o n ar y artery disease w e r e e xc l ude d from the study. Patient Subgroups The amyloid g r o u p was subdivided into a nonrestrictive group with a deceleration time (DT) of more than 150 ms (n = 14) and a restrictive group (R) w i t h a DT <150 ms (n = 9).The patients ~dth cardiac amyloidosis were compared w i t h a control group that consisted of 49 age-matched normal volunteers (23 w o m e n ) w h o x~.nged in age from 40 to 84 years (mean 61 -+ 13 years). 19

A complete Doppler evaluation including diastolic filling was performed as previously described. 24 LV inflow velocities were recorded from the apical 4-chamber view w i t h the sample volume at the level of the tips of the mitral leaflets pulmonary venous flow velocities were recorded from the apical 4~chamber view with the sample volume at the orifice and 1 cm into the right u p p e r pulmonary vein. Color f l o w D o p p l e r was helpful in aligning the sample volume parallel to b o t h the pulmonary venous and mitral flows. A heat-sensitive nasal thermistor was used to record the phas e of respiration. A hard c o p y of the D o p p l e r recordings together w i t h the respiratory monitoring was recorded on strip chart paper. 3 Doppler Measurements The peak velocities of both die pulmonary- venous atrial reversal (AR) and the mitral inflow atrial (A) wave were measured from baseline to the maximal flow velocities of both waves. Deceleration time 0DT) of the mitral inflow E wave was measured as the time required for the peak E velocity to return to baseline. The peak velocities w e r e m e a s u r e d w i t h a computer-interfaced digitizing tablet (D ex~da, Inc; Mountain View, Calil) from strip chart tracings. The duration of both the p u l m o n a r y vein AR and the mitral ufflow A waves were measured manually from the Doppler chart recordings. Figure I outlines the Doppler echocardiographic measurements of both mitral and pul-

Journal of the American Society of Echocardiography Volume 11 Number 12

Abdalla et al 1 1 2 7

T a b l e 1 Clinical and echocardiographic features o f amyloid g r o u p a n d control g r o u p

Sex ( M / F ) Age (years) Heart rate (bpm) Systolic BP (ram H g ) Diastolic BP (mm Fig) MLVWT Left atrial diameter (ram) Fractional shortening (%) Congestive heart failure (n)

Control

Amylold

Nonrestrictive

Restrictive

(n = 49)

(n = 23)

(n = 14)

(n = 9)

23/26 61.0 9 13.1 65.6 • 10.5 115.4 • 11.9 72.7 -+ 7.7 9.8 • 1.4 36.7 • 4.9 38.0 • 4.2 0/49 (0.6%)

13/10 61.3 • 12.0 83.3 • 12.0 115.6 • 20.6 76.0 • 12.5 14.6 9 1.9 41.0 • 6.7 37.7 • 12.0 7/23 (30.4%)

7/7 63.3 • 12.3 82.6 • 11.6 112.9 +- 19.8 73.4 • 13.8 14.1 • 1.1 40.0 • 7.0 39.9 • 9.4 4/13 (30.8%)

6/3 58.3 -+ 11.6 84.4 e 12.9 119.8 e 22.2 80.0 9 10.1 15.4 • 2.6 42.6 • 5.4 34.1 • 14.1 3/9 (33.3%)

P value* ~

.93 <.001 .97 .26 <.001 .003 .90 <.001

.56 <.001 .66 .86 <.001 .049 .47 .0013

*

.57 <.001 .58 .016 <.001 .002 .43 .0027

.35 .73 .44 .23 .18 .37 .27 .99

BP, Blood pressure; MLVI/V~mean left ventricularwall thickness. *Control vs amyioid, tControl vs nonrestrictive. :)Con~ol vs restrictive. w restrictive.

vs

T a b l e 2 D o p p l e r echocardiographic measurements o f amyloid g r o u p a n d control g r o u p

AR duration (ms) MitralA-wave duration (ms) AR-A difference (ms) Peak AR velocity (m/s) Peak A velociw (m/s) Deceleration time (ms) Mitral E / A ratio pulmonary vein S / D ratio

Control

Amyloid

Nonrestrictive

Restrictive

(n = 49)

(n = 23)

(n ~ 14)

(it = 9)

105.5 126.5 -21.0 0.22 0.52 197.6 1.3 1.5

• 10.2 -+'12.6 -+ i0.2 9 0.04 • 0.16 + 29.4 + 0.7 + 0.7

145.9 125.3 20.6 0.24 0.69 172.0 1.3 1.0

9 16.2 • 16.7 9 23.4 9 0.13 • 0.3i ~+41.3 + 0.7 § 0.5

143.1 132.8 i0.3 0.23 0.79 197.6 1.0 1.3

9 17.2 9 14.4 • 19.2 + 0.16 • 0.31 -+ 14.2 -+ 0.37 • 0.37

150.3 113.7 36.6 0.24 0.54 124.4 1.9 0.6

AR, Atrial reversal. *Control vs ~lyloid. t Control vs no~estricfive. :~Conn'olvs restrictive. w

+ 14.1 + 13.1 -+ 20.8 + 0.09 9 0.26 • 28.8 -+ 0.6 -+ 0.6

_P value_*$

<.001 .73 <.001 .55 .023 .01 .098 <.001

<.001 .12 <.001 .82 .007 .99 .039 .16

<.001 .007 <.001 .53 .83 <.001 .019 <.00i

.31 .0042 .0053 .89 .07 <.001 <.001 .002

vs restrictive.

m o n a r y venous flows. If t h e p u l m o n a r y venous AR w a v e

RESULTS

was n o t clearly recorded, its duration w a s estimated as t h e difference b e t w e e n t h e e n d o f t h e diastolic a n d t h e start o f t h e systolic p u l m o n a r y v e n o u s f l o w s , as previously

Clinical and Echocardiographic Characteristics

described.15, 24 Mean values for t h e measurements described

There was no significant difference regarding age,

above w e r e obtained by averaging t h e values of these mea-

s y s t o l i c a n d d i a s t o l i c b l o o d p r e s s u r e s , o r LV systolic function (fractional shortening) when comparing the amyloid group with the control population. H o w e v e r , t h e h e a r t r a t e , MLVWT, a n d t h e l e f t atrial

surements for 9 cardiac cycles. Statistical M e f l l o d s Continuous data s u c h as age, h e a r t late, blood pressure, weight, and Doppler measures are expressed as m e a n +_SD. Dichotomous variables such as sex, mitml regurgitation, o r congestive heart failure are expressed as f r e q u e n c y / n (%). Two group comparisons (normal versus amyloid or restrictive versus nonrestrictive) of continuous data w e r e m a d e w i t h a 2-sample t test. Fisher's exact test was used to analyze differences in t h e p r o p o r t i o n of dichotomous data b e t w e e n 2 groups.Three group comparisons of continuous data w e r e made w i t h a 1-way analysis of variance a n d pairwise contrasts. A value of P < .05 was considered statistically significant. Sensitivities and specificities for various values of the AlIA i n d e x w e r e calculated to d e t e r m i n e t h e best cutoff value for separating t h e amyloid fi-om the n o r m a l group and t h e nonrestrictive flora t h e restrictive subgroup.

dimension were significantly greater for the amyloid g r o u p ( T a b l e 1). On dividing the amyloidosis group into nonrestrictive a n d r e s t r i c t i v e s u b g r o u p s , t h e M L V W T t e n d e d t o be higher in the restrictive group compared with that i n t h e n o n r e s t r i c t i v e s u b g r o u p ( P = .18) (Table 1). Doppler

Measurements

Doppler recordings from the control population and restrictive and nonrestrictive subgroups are shown i n F i g u r e s 2 t h r o u g h 4. Arnyloidosis

Versus Control

A l t h o u g h t h e p u l m o n a r y v e n o u s atrial r e v e r s a l d u r a tion was significantly longer in the amyloidosis

Journal of the American Society of Echocardiography December 1998

1128 Abdalla et al

$

140 msec

~e

i

Figure 2 Pulmonary venous and mitral inflow recordings of 47 year old woman from normal group. Pulmonary venous atrial reversal flow duration (AFDUR) is less than the transmitral A-wave flow duration by 40 ms.

group compared w i t h the control group (P < .001), no significant difference was fotmd between these in the duration of the mitral inflow A wave (Table 2). T h e p u l m o n a r y venous atrial reversal duration e x c e e d e d that of the mitral i n f l o w A wave in all patients w i t h cardiac amyloidosis (100%). In contrast, in normal volunteers, the pulmonary venous atrial reversal duration was less than the duration of the mitral inflow except in 1 (2%) normal volunteer. The difference between both durations was signifi-

Figure 3 Pulmonary venous and mitral inflow recordings of 64-year-old woman from nonrestrictive group. Pulmonary venous atrial reversal (AR) flow duration exceeds transmitral A-wave flow duration by 40 ms.

cantly higher for the amyloifi group than for the normal group (1~ < .001). A cutoff value of -4.2 ms for the difference between the AR and A durations was found to separate the amyloid group from the control group, with a sensitivity of 91% and a specificity of 96% (Figure 5). Nonrestrictive and Restrictive Subgroups The mean duration of the pulmonary venous atrial reversal was significantly longer in both nonrestrictive and restrictive subgroups than in the control group (P < .001). Although this duration tended to be

Journal of the American Societyof Echocardiography Volume 11 Number 12

longer in the restrictive group than in tile nonrestrictive group, the difference was not statistically significant. No sigilificant difference was found in the mean duration of the mitral A wave between the control and the nonrestrictive groups. However, this duration was significantly longer in both these groups than in the restrictive group (P < .004) (Table 2). The difference between the pulmonary venous atrial reversal duration and the duration of the ntitral inflow A wave was significantly higher in both the nonrestrictive and restrictive subgroups compared w i t h that in the control group (P < .001). Furthermore, this difference was significantly higher in tile restrictive compared with the nonrestrictive subgroup (P < .001) (Table 2).A cutoff value of 26 ms for the difference between the AR and A durations demonstrated a sensitivity of 78% and a specificity of 79% for separating the nonrestrictive from the restrictive subgroup. No significant difference was found between the 3 groups on comparing the peak velocity of the pulmonary venous atrial reversal. However, the peak velocity of the mitral inflow A wave tended to be higher in the nonrestrictive group than in the restrictive group (P = .07) and the normal group (P = .007).

Abdalla et al 1129

y ...t....~....***..o.........j..H.

rasp

.

......... .....~....,....a....v

A

DISCUSSION It has previously been shown that in patients with cardiac amyloidosis, the degree of infiltration of the cardiac muscle w i t h amyloid deposits and hence MLVWT determines largely the severity of diastolic dysfunction hi this condition) Thus cardiac amyloidosis exhibits a wide spectrum of diastolic filling abnormalities. At one end of file spectrum is impaired relaxation that predominates early in the course of the disease with relatively normal or mildly elevated LV filling pressures.The other end of the spectrum is represented by the restrictive pattern that predominates late in the course of the disease with a more significant increase in MLVWT and higher LV fiUing pressures.The transition between these two stages is marked by pseudonormal Doppler flow patterns caused by nloderately elevated pressures and impaired filling. A new index of diastolic dysfunction that was described by Rossvoli and Hatie 15 and confirmed by Appleton et all6 showed that a pulmonary venousAR duration that exceeds the transmittal A duration can predict an LVEDP greater than 15 m m Hg. Furthen~lore, 2 recent studies have shown that this index is minimallyinfluenced by the aging process, 17,18

HIll tttl

120

msec

~

,

!

l~igure 4 Pulmonary venous and mitral inflow recordings of 40-year-old female patient from restrictive group. Pulmonary venous atrial reversal (AR) flow duration exceeds transmittal A wave flow duration by 60 ms. contrary to most other Doppler variables of diastolic function. 19 Therefore, age should not affect its reliability in identifying patients with abnormal diastolic function. The current study describes the utility of this new index as a measure of tile severity of diastolic dysflmction in nonrestrictive (impaired relaxation and psendonormal) and restrictive cardiac amyloidosis. The difference in the duration between the pulmonary venous AR wave and the mitral inflow A wave, a new index of diastolic function, was highly sensitive in separating patients with normal diastolic function from those with cardiac amyloidosis.

lournal of the American Socie~ of Echocardiography December I998

1130 Abdalia r al

7 0 84 60 50 40

9

30 26 20 10 9

0

.. :t

.

.

4.2

T

-10 ~

20

9

9

9~ -30 40

9

ooo el 9 6 0

!

9

9

9

9

|

9

coo o o9 9 9 9 9

Control

Non-Restrictive

Restrictive

Figure 5 Distribution of values of difference between pulmonary vein atrial reversal flow duration and mitral inflow A-wave duration of normal group and nonrestrictive and restrictive subgroups, separated by corresponding cutoff values.

E f f e c t o f LV F i l l i n g P r e s s u r e o n D u r a t i o n o f P t d m o n a r y V e n o u s All a n d M i t r a l A Wave Our results show that the mean dilration of the pulmonary venous AR wave is significantly longer in patients w i t h cardiac arnyioidosis than in normal subjects (P < .001). Within the amyloidosis group, t h e r e was no significant difference b e t w e e n the restrictive and the nonrestrictive groups. At the same time, it was shown that the duration of the mitral inflow A wave is significantly shorter in the restrictive group c o m p a r e d w i t h both the nonrestrictive and control groups (Table 2). These results suggest that in the earlier stages of cardiac amyloidosis, prolongation of the duration of the pulmonary venous AR wave is the p r e d o m i n a n t m e c h a n i s m causing a difference b e t w e e n both dm~tions. With further progression of the disease into the restrictive stage and a more significant elevation of LV filling prcssm'es, no further prolongation of the pulmonary venous AiR wave was noted. Hence, shortening of the duration of the mitral inflow A wave was the predominant m e c h a n i s m that a c c e n t u a t e d the difference in duration b e t w e e n both waves. E f f e c t o f LV F i l l i n g P r e s s u r e o n N e w I n d e x This study also shows that the difference in the duration b e t w e e n the puhlaonary venous All wave and

the mitral inflow A wave, a n e w index of diastolic ftmction, was highly sensitive in separating patients with normal diastolic function from those with cardiac amyloidosis (P < .01). T h e AR duration was longer in only 1 (2%) o f the 49 n o r m a l subjects, whereas the A wave was longer in the rest of this group (98%). On the other hand, all patients (100%) w i t h cardiac amyloidosis had a longer All wave duration than that of the mitral inflow A wave. In fact, a cutoff value of -4.2 ms separated the control from the amyloid group, with a sensitivity of 91% and a specificity of 96%. Within the amyloid group, the same index was highly reliable in assessing the degree or severity of diastolic dysfunction. Patients w i t h the restrictive pattern of mitral inflow (DT - 150 ms, MLVWT = 14.1 _+ 1.1 ram) and a relatively milder degree of LV filling pressure eievation. A cutoff value of 26 ms separated both groups, with a sensitivity of 78% and a specificity of 79% (Figure 5). The lower sensitivity and specificity of this index in separating nonrestrictive from restrictive patients c o m p a r e d w i t h separating patients w i t h amyioid

Journal of the American Societyof Echocardiography Volume 11 Number 12

from normal patients can be partly explained by the presence of a few patients with the pseudonormal pattern (and relatively" higher LV filling pressures) among the nonrestrictive group. Further explanation can be given by distinguishing between the effect of abnormal LV compliance occurring relatively early in diastole (represented by the deceleration time) and elevated filling pressures occurring at end-diastole (represented by the difference in the duration of the pulmonary venous AR and mitral inflow A wave). Mechanisms The mechanism for this close relation between this index and LV filling pressures has been previously explained by the observations of Matsuda et al. 26 They showed that with an elevated LVEDE the left atrial pressure curve becomes biphasic, with the first peak being responsible for the positive atrioventricufar pressure gradient, and is represented by the mitml A wave.The second peak is a reflection of the pressure buildup in a stiff ventricle. 26 Therefore, an elevated LVEDP is expected in patients with a prolonged pulmonary venous AR and a large difference between the pulmonary venous AR duration and the mitral inflow A wave durations. However, our results and previous studies27 suggest that this mechanism operates only with mild to moderate elevation of LV tilling pressures and that with marked elevation, shortening of the mitral inflow A wave is the responsible mechanism for the larger difference between both durations. This may be explained by the assumption that in the early nonrestrictive stage of cardiac amyloidosis, a mild decrease in LV compliance occurs and is compensated for by an increase in left atrial contractility resulting from the increase in left atrial preload (Starling law).This increase in left atrial contractility maintains the same duration of positive atrioventricular pressure gradient and hence the normal duration of forward flow- (mitral inflow A wave) in the face of the increased LV diastolic pressure. However, as expected, it results in an increase ha the duration of the backward flow (pulmonary venous AR wave). On the other hand, in the late restrictive stage of diastolic dysfunction, the chronic significant decrease in LV compliance results in an abnormally rapid rise of LV diastolic pressure during atrial contraction and hence a shorter duration of the forward flow (mitral inflow A wave). This is even enhanced by the decrease in left atrial contractility (atrial systolic failure) resulting from the chronic elevation in left atrial preload (ie, an overstretched left atrium operating on the descending part of its Starling curve) or atrial infiltration.

Abdalla et al 1131

The importance of recognizing the sequence described above is to realize that the LV filling pressure cannot be predicted on the duration of either wave alone. The pulmonary venous AR duration is significantly prolonged only with mild to moderate but not severe elevation of LV tilling pressures. Conversely; the rnltral A wave is significantly shortened only with severe elevation of these pressures. P r e v i o u s Studies This mechanism of events is consistent with the results of Rossvoll and Hatle, 15 who showed that significant shortening of the duration of the mitral inflow- A wave was noted only- with a significantly elevated LV filling pressure before atrial contraction (ie, pre-A pressure >15 mm Hg). However,*this was not noted with milder elevation of the pre-A pressure (10 to 15 mm Hg).This was explained by an overstretched left atrium operating on the descending part of its Starling curve with a higher pre-A pressure. A more recent study also demonstrated that with a mild to moderate decrease in LV compliance, prolongation of the pulmonary venous atrial reversal occurs. 27 However, with further decrease in LV cornprance to a severe degree (LV pre-A pressure >15 mm Hg), shortening of the mitral inflow A wave was the only significant change that resulted in a larger difference between the duration of both waves. 27 Effect o f LV F i l l i n g P r e s s u r e s o n Left Atrial Size a n d C o n t r a c t i l i t y Previous studies5,7,12A3,27 have shown that left atrial enlargement and systolic failure is a common finding in patients with restrictive LV physiology and chronically high tilting pressures. This is consistent with our finding of a tendency of the left atrial size to be higher in the restrictive group than in the nonrestrictive subgroup (42.6 + 1.8 vs 40 -+ 1.9 ram). In fact, Appleton et all6 have shown that in the absence of mitral valve disease, left atrial size and ejection fraction are independent detern~nants of LV filling pressures. Further studies are needed to evaluate these parameters in patients with restrictive versus nonrestrictive physiology. Peak Velocity o f P u l n l o n a r y Venous AR Wave Rossvoli and Hatle 15 reported a significant correlation between peakAR velocity and both the amount of LV pressure elevation at the time of atrial contraction (IV A-wave pressure) and the LVEDI~By using transesophageal Doppler echocardiography, Nisifimlwa et a111 reported a close correlation (r = 0.81) between peak AR velocity and the increase in pulmonary capillary wedge pressure. However, our

1132 Abdalla et al

results did not confirm these findings but w e r e consistent with those of Appleton et a116 and those of a previous study12 that used the transesophageal technique. Discordant results on this relation may be explained by left atrial systolic failure limiting the increase in peak pulnlonary venousAR velocity even in the p r e s e n c e of significantly elevated LV filling pressures.Therefore, the current consensus is that in the a b sence of left atrial systolic failure or atrial arrhythmias, a peak pulmonary venous AR velocity g r e a t e r than 35 c m / s p r e d i c t s abnormal diastolic function w i t h an elevated LVEDE20 Technical Considerations Some studies have r e p o r t e d a low success rate of obtaining the pulmonary venous flow- velocity variables by the transthoracic technique, 28-3~ and the reliability of these m e a s u r e m e n t s has b e e n questioned. 18 However, w e and others recently reported a high degree of reproducibility of these measurements by using the transthoracic technique17,18,31 (in our study interobserver variability: interclass correlation coefficient = 0.97 for the pulmonary venous All duration and 0.99 for measuring the mitral inflowA wave duration). 17 T he success rate of the more r e c e n t c o m p a r e d w i t h earlier studies seems to d e p e n d on several factors, including attention to technical details, experience and capabilities of the individual sonographers, the ultrasound equipment used, and the mitml inflow pattern. The technical details include the sample volume size and location, Doppler gain and filter settings, avoidance of left atrial wall motion artifacts, and respiratory variation. A t e c h n i c al guide for optimal r e c or ding of the pub monaty venous flows and the common technical pitfails involved in obtaining them has recently b e e n published.32 T h e transesophageal and the contrast-enhanced techniques also have b e e n sh own to improve the t e c h n i c a l quality of the pulsed D o p p l e r signals obtained by interrogating the pulmonary veins. 3335 Tiffs can certainly minimize errors in measurement and would be expected to show a higher correlation of this n e w index with LV filling pressures. Limitations The major limitation in this study is the absence of invasive hemodynamic data.Although the restrictive mitral i n f l o w p a t t e r n w as shown by many studies 5,7,12,13 to reflect a more significant elevation of LV filling pressures than other patterns, invasive data would have added significant weight to the results. Another limitation is the relatively small number

Journal of the American Society of Echocardiography December 1998

of patients. This limitation p r e v e n t e d f u r th e r subgrouping of the nonrestrictive g r o u p into an "impaired relaxation" group and a "psendonormai" group to study the reliability of this index in separating those groups.These small numbers also prevented differences b e t w e e n the nonrestrictive and restrictive subgroups in terms of congestive heart failure incidence and class from being evident.Also, w e did not use n e w D o p p l e r echocardingraphic applications of color M-mode and tissue Doppler to assess LV filling pressures. 36 Conclusions T h e difference in the duration b e t w e e n th e pulmonary venous All w a v e and the mitral i n f l o w A wave is a new, highly reliable index of diastolic function.This index can be used to assess the severity of cardiac amyloidosis and to separate patients w i t h restrictive LV physiology and significantly elevated LV filling pressures from those w i t h milder degrees of diastolic dysfunction. Further studies are needed to evaluate the utility of this index in other disease processes characterized by diastolic dysfunction.

REFERENCES

1. Cueto Garcia L, Reeder GS, Kyle RA, et al. Echocardiographic findings in systemic anayldidosis: spectrum of cardiac involvementand relation to survival. J Am Coil Cardiol 1985;6:737 d-3. 2. Roberts WC, WallerBF. Cardiac amyloidosis causing cardiac dysfunction: analysis of 54 necropsy patients. Am J Cardidi I983;52:i37 46. 3. Klein AL, Hare LK. Burstow DJ, et el. Doppler characterization of LV filnction in cardiac amyldidosis. J Am Coil Cardidi 1989;13:1017 26. 4. Appleton CP, Hatle LK, Popp RL. Demonstration of restrictive ventricdiar physiology by Doppler echocardiography. J Am Coll Cardiol 1988;12:757-68. B. Appleton CP, Hade LK, Popp RL. Relation of transmitraI flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coli Cardiol 1988;12:426 40. 6. Stoddard MF, Pearson AC, Kern MJ, RatcliffJ, Mrosek DG, Labovitz AJ. Left ventricular diastolic function: comparison of pulsed Doppler echocardJographic and hemodynamic indexes in subjects with mad without coronary artery disease. J Am Coil Cardiol 1989;13:327-36. 7. NishimuraRA, Abel MD, Hare LY-,TajikAI. Assessment of diastolic function of the heart: background and current applications of Doppler echocardiography, II: clinical studies. Mayo Clinic Proc 1989;64:181-204. 8. ICdein AL, Hatle LI4, Taliercio CP, et el. Serial Doppler echocardiographJc follow-up of left ventriculardiastolicfimction in cardiac amyldidosis. J Am Coil Cardidi 1990;16: 1135-41. 9. Vanovershelde JJ, Pdaphael DA, Robert AR, Cosyns IR. Left

lournal of the American Society of Echocardiography Volume 11 Ntmaber 12

10. 11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

vennicular filling in dilated cardiomyopathy: relation to functional class and hemodynamics. J Am Coil Cardiol 1990;15: 1288-95. Skagseth E. Pulmonary vein flow pattern in man during thoracotomy. Scand J Thorac Cardiovasc Surg 1976;10:36-42. Nishimura RA, Abel MD, Hatie LK, Taiik AJ. Relation of pulmonary vein to mitsal flow velocities by transesophageal Doppler echocardiography: effect of differing loading condi tions. Circulation 1990;81:1488-97. Kuercherer I-IF, 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. Basnight MA, Gonzales MS, I(erslmnovich SC, Appleton C1~ Pulmonary venous flow velocity: relation to hemodynamics, mitral flow velocity and left atrial volume and ejection fraction. J Am Soc Echocardiogr 1991;4:379 92. Klein AL, Tajik AJ. Doppler assessment of pulmonary venous flow in healthy subjects and patients with heart disease, l Am Soc Echocardiogr 1991;4:379 92. gossvoll O, Hatle LI(. Pulmonaryvenous flowvelocities recorded by" rransthoracic Doppler ultrasound: relation to left venrric ular diastolic pressures. J Am Cull Cardio11993;21:1687 96. Appleton CP, Galloway IM, Gonzales MS, Gaballa M, Basnight MA. Estimation of left ventricular filling pressures using two dimensional and Doppler echocardiography in adult patients with cardiac disease. J Am Coil Cardiol 1993; 22:1972-82. Klein AL, Abdalta I, Lee JC, Murray RE), Appleton CP, Tajik I. Effect of aging on duration of pulmonary venous atrial reversal and ntitral inflow A wave: a new index of diastolic function. J Am Soc Echocardingr 1994;7(suppl 3):$38. Malkowski MJ, Guo R, Gray PG, Pearson AC. Is the pul monary venous transmittal A wave duration difthrence altered by age and hypertension? Am J Cardiol 1995;76:722-4. Klein AL, gurstow DJ, Tajik AJ, et al. Effects of age on left vcntricular dimensions and filling dynamics in 117 norma~ persons. Mayo Clin Proc 1994;69:212-24. Rakowsld H, Appleton C, Chan K-L, et al. Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiograpli}~ J Anl Soc Echocardiogr 1996;9:736 760. Cohen GL, Pietrulungo J1a, Thomas JD, Klein AL. A practical guide to assessment ofventlicdiar diastolic function using Doppler echocardiography. J Am Coil Cardiol I996;27: 1753-60. Oh JK, Appleton CP, Hatle LK, Nishfuaura RA, Seward JB, Tajik AJ. The nodinvasive assessment of left ventricular dias tolic function with two-dimensinnal and Doppler echocardiography. J Am Soc Echocardiogr 1997;I0:246 70. Tajik AJ, Seward JB, HaNer DJ, Mair DD, Lie JT. Two dimensional real-time ultrasonic imaging of the heart and great vessels: technique, image orientation, structure, iden-

AbdaUa et al 1 1 3 3

tification and validation. Mayo Clinic Proc I978;53:271303. 24. Appleton CR Doppler assessment of left ventricular diastolic function: the refinements continue [editorial]. ~ Am Coil Cardio11993;21:I697-1700. 25. Little WC, Ohno M, Kltzman DW, Thomas JD, Cheng CP. Determination of left ventricular chamber sfiffuess from the time for deceleration of early left ventticular filling. Circulation 1995;92:1933-9. 26. Marsuda Y, Toma Y, Matsuzaki M, et al. Change in left atrial systolic pressure wavefornl in relation to left ventricular enddiastolic pressure. Circulation 1990;82:1659-67. 27. Yamamoto K, Nisfumura RA, Burnett JC, I