doppler color flow mapping in the evaluation of …jacc vol. 13, no. 7 june 1989:1561-71 1561...

JACC Vol. 13, No. 7 June 1989:1561-71

1561

Doppler Color Flow Mapping in the Evaluation of Prosthetic Mitral and Aortic Valve Function

KANWAL K. KAPUR, MD, POHOEY FAN, MD, NAVIN C. NANDA, MD, FACC,

AJIT P. YOGANATHAN, PHD, RAJENDRA G. GOYAL, MD

Birmingham, Alahamtl

Doppler color flow mapping and color-guided conventional Doppler studies were performed on 119 patients with 126 prosthetic valves (mitral alone in 60, aortic alone in 52 and both mitral and aortic in 7 patients) within 2 weeks of the catheterization study or surgery, or both. The mean pressure gradients derived by color-guided continuous wave Doppler ultrasound correlated well with those obtained at catheterization for both the tissue and mechanical mitral and aortic prostheses (r = 0.85 to 0.87). For the effective prosthetic orifice areas, better correlation with catheterization results were obtained with the tissue mitral (r = 0.94) and tissue aortic (r = 0.87) prostheses than with the mechanical mitral (r = 0.79) and mechanical aortic (r = 0.76) prostheses. The maximal width of the color flow signals at their origin from the tissue mitral prostheses also

correlated well with the effective prosthetic orifice area at catheterization (r = 0.81).

Doppler color Row mapping identified prosthetic valvular regurgitation with a sensitivity and specificity of 89% and lOO%, respectively, for the mitral and 92% and 83% for the aortic prostheses. There was complete agreement between the Doppler color flow mapping and angiographic grading of the severity of prosthetic valvular regurgitation in 90% of mitral and 73.5% of the aortic regurgitant prostheses with under- or overestimation by >I grade in only two cases. Valvular and paravalvular regurgitation was correctly categorized by Doppler color flow mapping in relation to the surgical findings in 94% of the mitral and 80.5% of the aortic prostheses.

(J Am Co11 Cardiol1989;13:1561-71)

Differentiation of normally functioning from dysfunctional prosthetic valves by noninvasive techniques is important because such a distinction may not be possible on clinical grounds. Until recently cardiac catheterization was the only definitive method for detecting prosthetic valve dysfunction and quantifying its severity. However, because it is an invasive technique, it cannot be performed repeatedly and entails a small but definite risk. Cinefluoroscopy, echopho- nocardiography and M-mode and two-dimensional echocardiography have been shown to be useful noninvasive modalities for the evaluation of prosthetic valves. These techniques, however, do not provide sufficient sensitivity or specificity in assessing prosthetic valve dysfunction and also do not permit precise quantification of its hemodynamic severity (1).

The use of conventional pulsed and continuous wave

From the Division of Cardiovascular Disease, University of Alabama at Birmingham, Birmingham, Alabama.

Manuscript received August 22, 1988; revised manuscript received No- vember 21, 1988, accepted December 16, 1988.

Address for remi&: Navin C. Nanda, MD, University of Alabama, Heart Station, SWBIWOOI, Birmingham, Alabama 35294.

01989 by the American College of Cardiology

Doppler echocardiography provides a superior noninvasive method for estimating the severity of transprosthetic obstruction and in semiquantifying the severity of regurgitation (24). However, because the flow signals are not visualized by this technique, precise estimation of the transprosthetic velocities entails arbitrary placement of the continuous wave Doppler cursor in the assumed direction of transprosthetic flow (5). Semiquantitative assessment of the prosthetic regurgitation also requires meticulous time-consuming mapping of the left atrium or left ventricular outflow tract in multiple imaging planes. Moreover, the categorization of the prosthetic regurgitation as valvular or paravalvular may be extremely difficult by conventional Doppler echocardiography (6-8).

The recent introduction of Doppler color flow mapping permits direct visualization of transprosthetic flow signals superimposed on almost real-time two-dimensional images (9-11). This technique, therefore, provides a novel method to align the continuous wave Doppler cursor parallel to the transprosthetic flow jet and would thus be expected to obtain transprosthetic velocities reliably and relatively rapidly as has been done for native valve stenosis (12) (Fig. 1). Also,

0735.1097/89/$3.50

1562 KAPUR ET AL. DOPPLER COLOR IMAGING IN PROSTHETIC VALVES

JACC Vol. 13, No. 7 June 1989:1561-71

because the flow signals are visualized in different regions of the heart simultaneously, the presence and severity of the prosthetic regurgitation can be evaluated rapidly and more reliably with the use of multiple imaging planes. The accu- racy of Doppler color flow mapping in assessing the severity of native valve regurgitation has already been shown (10,ll). The purpose of this study was to investigate the usefulness of Doppler color flow mapping combined with continuous wave Doppler ultrasound in evaluating the hemodynamic severity of transprosthetic obstruction and applying the Doppler color flow technique to the accurate assessment of the severity and identification of the site of the prosthetic regurgitation.

Methods Study patients. One hundred nineteen adult patients, 62

men and 57 women, aged 18 to 80 years (mean 52.6) with 126 prosthetic valves formed the basis of this study. They represent consecutive patients who underwent adequate cardiac catheterization and Doppler color flow studies over a period of approximately 3 years. Of these, 60 patients had a mitral prosthesis, 52 an aortic prosthesis and 7 had both aortic and mitral prostheses. The 67 mitral prostheses comprised 39 tissue (24 Carpentier-Edwards, 14 Hancock and 1 Duramater valve) and 28 mechanical valves (16 Bjiirk- Shiley, 7 St. Jude, 3 Starr-Edwards and 2 Braunwald- Cutter). The 59 aortic prostheses comprised 37 tissue (18 Carpentier-Edwards, 13 Hancock, 5 homograft and 1 Iones- cu-Shiley valve) and 22 mechanical valves (14 Bjiirk-Shiley, 3 Starr-Edwards, 2 St. Jude, 1 Medtronic-Hall, 1 Braunwald- Cutter and 1 Lillehei-Kaster valve). Thirty-eight patients had atria1 fibrillation, 4 had a permanent right ventricular pace- maker for heart block and the remaining 77 patients had normal sinus rhythm. The primary cause of valve replacement was rheumatic heart disease in 54, calcific aortic valve disease in 33, bicuspid aortic valve in 8, ventricular septal defect with aortic incompetence in 1, ascending aortic aneu- rysm in 2, infective endocarditis in 6, myxomatous mitral valve degeneration in 7, ischemic heart disease in 5 and corrected transposition, cardiac trauma and hypertrophic cardiomyopathy in 1 patient each.

Echocardiography Two-dimensional echocardiographic, conventional Dopp-

ler and Doppler color flow examinations were performed in the standard manner with use of an Irex Aloka 880, Coro- metrics Aloka 860, Toshiba SSH65A, Hewlett Packard or Advanced Technology Laboratory Ultra Mark VI system with a 2.0 or 2.5 MHz transducer in all 119 patients. All Doppler color flow measurements were performed by an independent observer who was unaware of cardiac catheterization results.

Pressure gradients and valve areas. For all mitral prostheses, the anterograde transprosthetic flow signals visualized from the apical four chamber view were used as a landmark to guide the parallel placement of the continuous wave Doppler cursor in line with the flow direction. Once the continuous wave cursor was aligned parallel to the visualized anterograde transprosthetic signals, minimal transducer angulation was performed to ensure interrogation of the “core” of the jet so as to obtain the maximal transprosthetic velocities. The peak and mean transmitral prosthetic pressure gradients were computed in all patients in the standard manner from the maximal velocity waveforms utilizing the simplified Bernoulli equation (13,14). The effective prosthetic mitral orifice area was estimated by the pressure half-time method (15). For the aortic prostheses, also, the transvalvular pressure gradients were obtained by using the transprosthetic flow signals as a marker to guide the parallel placement of the continuous wave Doppler cursor, with use of the apical and right parasternal or suprasternal, or both, approaches. In two patients, both with a tissue aortic prosthesis, adequate transprosthetic Doppler spectral waveforms were not obtained from any of the transducer positions and, hence, pressure gradients could not be measured.

The apical window was utilized to place the pulsed Doppler sample volume in the left ventricular outflow tract (LVOT) immediately proximal to the aortic prostheses in the area of maximal brightness of the color flow signals, so as to obtain the maximal left ventricular outflow tract velocities. These were obtained in 54 patients and not measured in 5 patients with an aortic prosthesis. The inner left ventricular outflow tract diameter was measured in the parasternal long-axis view immediately proximal to the prostheses and the cross-sectional area was calculated by assuming a circu- lar outflow tract configuration (16,17). The effective aortic prosthetic orifice area was calculated with the simplified continuity equation: effective orifice area = LVOT cross- sectional area x LVOT peak velocity/peak transprosthetic velocity.

Jet width. For mitral tissue prostheses, the maximal width of the transprosthetic color flow signals was measured at the point of their origin from the prosthesis. The origin of the anterograde flow signals was often not well visualized for the mechanical mitral and both tissue and mechanical aortic prostheses and, therefore, jet width was not measured in patients with these prostheses.

Regurgitation. Prosthetic valve regurgitation was identified by the presence of reversed or mosaic flow signals in the dependent chamber originating from or within 0.5 cm of the prosthesis, during systole for mitral and during diastole for aortic regurgitation. The severity of mitral prosthetic regurgitation was assessed by a method previously described from our laboratory for native mitral valve regurgitation (10). The maximal area of the regurgitant signals was determined by planimetry and expressed as a ratio of the left atria1 area

JACC Vol. 13, No. 7 KAPUR ET AL. June 1989:1561-71 DOPPLER COLOR IMAGING IN PROSTHETIC VALVES

1563

Figure 1. Color-guided continuous wave Doppler estimation of transaortic prosthetic pressure gradient (right parasternal approach). The continuous wave cursor is placed parallel to the color flow signals originating from the Starr-Edwards aortic (AO) prostheses (AP) to obtain a mean pressure gradient of 16 mm Hg, using the Bernoulli equation.

taken in the same frame in which the maximal regurgitant area was noted. Multiple views including the right parasternal imaging plane were utilized to obtain the maximal regurgitant jet area. If the regurgitant signals occupied <20%, 20% to 40% and >40% of the left atrium, the prosthetic valve regurgitation was classified as grade I (mild), grade II (moderate) or grade III (severe), respectively. The severity of prosthetic aortic regurgitation was also graded with use of the criteria previously described from our laboratory for the native aortic valve (11). The ratio of the maximal diameter of the regurgitant jet at its origin to the maximal inner diameter of the left ventricular outflow tract taken at the same point was calculated utilizing standard parasternal, low parasternal and apical long-axis views. Ratios of 1% to 24%, 25% to 46%, 47% to 64% and ~65% were assessed as grade I, grade II, grade III or grade IV/V aortic regurgitation, respectively.

For both mitral and aortic prostheses, the regurgitation was categorized as valvular if the regurgitant signals originated within the confines of the prosthetic stents or the ring (Fig. 2); it was categorized as paravalvular if it was observed to originate beyond the confines of the prosthetic elements (Fig. 3). For the mitral prostheses, the regurgitant jet originating at the level of the stents was also considered to be paravalvular. For the aortic prostheses, if the prosthetic stents or the ring could not be confidently identified, paravalvular regurgitation was diagnosed if the jet originated from the extreme anterior or posterior aspects of the prosthesis, whereas regurgitation originating from more central aspects of the prosthesis was classified as valvular. During the color flow examination for the assessment of prosthetic regurgitation, particular attention was paid to the presence

Figure 2. Doppler color flow assessment of valvular regurgitation in a tissue mitral prosthesis (MP). The apical four chamber view shows mosaic signals of mitral regurgitation (MR) clearly originating within the prosthetic stents (S). The flow acceleration, seen as a small circumscribed greenish area on the left ventricular (LV) aspect of the prosthesis, is also within the stents. LA = left atrium; RA = right atrium; RV = right ventricle.

of any localized area of high velocity signals (flow acceleration) proximal to the origin of the regurgitant jet, that is, on the ventricular side of the mitral and the aortic side of the aortic prosthesis (18). Presence of the flow acceleration signals within the prosthetic stents or the ring categorized the regurgitation as valvular, whereas the location of these signals beyond the confines of the prosthetic elements indi- cated a paravalvular regurgitation. If the aortic prosthetic stents or ring were not well visualized, paravalvular regurgitation was also diagnosed if the flow acceleration signals were located at the extreme anterior or posterior aspects of the prosthesis whereas a more central location of the flow

Figure 3. Doppler color flow assessment of mechanical mitral prosthetic paravalvular regurgitation. The apical four chamber view reveals mosaic signals during systole in the left atrium (LA), originating beyond the elements of the Bjork-Shiley prostheses (MP), indicative of paravalvular prosthetic regurgitation (PVR). Abbreviations as in Figure 2.


JACC Vol. 13, No. 7 June 1989:1561-71

acceleration signals was considered indicative of valvular regurgitation.

Cardiac Catheterization

Right and left heart catheterization was performed in 108 (55 with a mitral prosthesis, 46 with an aortic prosthesis and 7 with both mitral and aortic prostheses) of 119 patients studied, within 24 h of the echocardiographic study in 47 and within 2 to 14 days (mean 6.4) in 61 patients. The left ventricle was entered by direct transthoracic puncture in 1, by transeptal approach in 12 and by the retrograde technique in 84 patients; it was not entered in 11 patients. In the 97 patients (55 with a mitral prosthesis, 36 with an aortic prosthesis and 6 with both mitral and aortic prostheses) in whom the left ventricle was entered, mean mitral and mean aortic transprosthetic gradients were obtained from pulmonary capillary wedge or left atria1 and left ventricular and left ventricular and aortic pressure tracings, respectively. Stan- dard biplane left ventricular cineangiography was performed in 54 of 55 patients with a mitral prosthesis and standard biplane aortic root cineangiography was performed in 42 of 46 patients with an aortic prosthesis. Left ventriculography was performed in six of seven and aortography in seven of seven patients with both mitral and aortic prostheses.

Mitral regurgitation, when present, was graded on a scale of 1 to 3 according to the criteria of Nagle et al. (19) and aortic regurgitation, when present, was graded on a scale of 1 to 5 according to the criteria of Hunt et al. (20). Cardiac output was determined in 106 of the 108 patients by either the Fick (63 patients) or the thermodilution technique (43 patients). With use of the standard Gorlin equation (21) with a constant of 38 for the mitral and 44.5 for the aortic valve and excluding patients with angiographically more than grade I mitral or grade II aortic regurgitation, the effective mitral prosthetic area was calculated in 30 patients with a mitral prosthesis and 2 with both mitral and aortic prostheses, whereas the effective aortic prosthetic valve area was calculated in 18 patients with an aortic prosthesis and 5 with both mitral and aortic prostheses.

Surgery

Eighty-five patients with 91 prosthetic valves (45 mitral [31 tissue, 14 mechanical] and 46 aortic [31 tissue, 15 mechanical]) underwent surgery. This group included the 11 patients who did not undergo cardiac catheterization. Pros- thetic obstruction was diagnosed if a large thrombus or vegetations obviously blocked the opening of the prosthetic valve or if the prosthetic valve was virtually immobile because of fibrocalcific degenerative changes. Paravalvular leak was considered to be present if dehiscence of the suture line was noted and an intact suture line excluded it. Torn, perforated, destroyed or retracted leaflets with an intact

2 f 1.5-

ii 2 l.O- n = 10 s y = 1.56 x -0.06 B I = 0.81 0 0.5- F

JiT WioTH BY CO (cm)

Figure 4. Correlation of Doppler color flow (CD) jet widths with effective tissue prosthetic orifice area at cardiac catheterization (CATH) in 10 patients. Data points below the X axis are from 28 patients who did not undergo catheterization or in whom effective orifice area could not be computed at catheterization because of significant prosthetic regurgitation or lack of cardiac output data.

suture line were consistent with significant valvular regurgitation. A large thrombus or vegetations obviously preventing effective closure of a mechanical or tissue prosthetic valve were also considered to be consistent with valvular regurgitation.

Correlation of Doppler color flow findings. The Doppler color flow findings were correlated with cardiac catheterization, angiographic and surgical findings. The mean transprosthetic pressure gradients and effective orifice areas derived by color-guided continuous wave Doppler ultrasound were correlated with cardiac catheterization data separately for tissue and mechanical mitral and aortic prostheses with use of linear regression analysis. The maximal width of the color flow signals at their origin from the tissue mitral prostheses was also correlated with the effective mitral prosthetic area. Assessment of the severity of prosthetic regurgitation by Doppler color flow mapping was compared with angiographic grading of severity for both mitral and aortic prosthesis. Doppler color flow assessment of the type of prosthetic regurgitation was correlated with the surgical findings.

Results Color jet width. The maximal width of the transmitral flow

signals measured at their point of origin from the mitral tissue prostheses correlated well with the effective orifice area estimated at cardiac catheterization by the Gorlin equation (r = 0.81, n = 10) (Fig. 4). However, the number of patients available for this correlation was small because only those with mild or no prosthetic regurgitation by angiography could

JACC Vol. 13, No. 7 June 1989: 1561-71

KAPUR ET AL. 1.565 DOPPLER COLOR IMAGlNG IN PROSTHETIC VALVES

AORTIC PROSTHESES (TISSUE)

MITRAL PROSTHESES (TISSUE) 30-

a no OtlrtnJ~llan 0 Obrmxtmn . no surwn

A MEAN PRESSUAi GAADIE~ GY'CO GthED CW DOPPLER (mmiigb

MITRAL PROSTHESES (MECHANICAL)

0' 5 10 15 20 25 30

C MEAN PRE&lRE GAAOiNT BY CO GUIDE0 CW DOPPLER fmmHgi

be included. A jet width ~0.8 cm was associated with evidence of obstruction at surgery, whereas a larger jet width tended to be associated with a nonobstructed prosthesis.

Mean pressure gradients. The mean pressure gradients obtained by color-guided continuous wave Doppler correlated well with those derived at cardiac catheterization for all types of prostheses: tissue mitral (Fig. 5A): r = 0.86, range 4.2 to 20.0 mm Hg, mean 12.0 ? 4.1, n = 35; tissue aortic (Fig. 5B): r = 0.85, range 3.0 to 51.0 mm Hg, mean 28.9 2 15.1, n = 27; mechanical mitral (Fig. 5C): r = 0.87, range 3.5 to 25.5 mm Hg, mean 9.2 + 4.4, n = 26; mechanical aortic (Fig. 5D): r = 0.87, range 4.0 to 50.0 mm Hg, mean 14.3 ? 8.3, n = 13.

For the tissue mitral and aortic prostheses, a color-guided continuous wave Doppler-derived mean pressure gradient ~14 mm Hg and ~30 mm Hg, respectively, tended to be associated with an obstructed prosthesis at surgical evalua-

Figure 5. Correlation of color-guided continuous wave Doppler (CD) and cardiac catheterization-derived mean pressure gradients for tissue mitral (A), tissue aortic (B), mechanical mitral (C) and

AORTIC PROSTHESES (MECHANICAL) 50 1 Y

25 i

0 30

130 Ob*lr"clIon ~ B wtrutmn 2 * No Surow

: s

./ /

MEAN PRESSURE GRAOIENT BY CO GUIDED CW DOPPLER (mmHg)

mechanical aortic (D) prostheses. Data points below the X axis are from patients who did not undergo catheterization or in whom the left ventricle could not be entered. Data points outside the Y axis in panel B are from two patients in whom adequate quality Doppler spectral traces could not be obtained.

tion, whereas those with a lower gradient tended to have a nonobstructed valve at surgery. Five nonobstructed tissue mitral and five nonobstructed tissue aortic prostheses had a mean pressure gradient >14 and 230 mm Hg, respectively, whereas three mitral and two aortic tissue prostheses with a mean pressure gradient < 14 and <30 mm Hg, respectively, were found to be obstructed at surgery. The mean pressure gradients for the four obstructed mechanical prostheses

1566 KAPUR ET AL. JACC Vol. 13, No. 7 DOPPLER COLOR IMAGING IN PROSTHETIC VALVES June 1989:1561-71

%

g

i=

ii g 1.5

ii 2 1.0 . g . . = 0.5 . g s L 2.0 ; 0 0.5 1.t

n = 10 y = 0.96 x r = 0.94

1.5 2.0 2.5 3.0

2.0 -

1.0 1’57

::

*I

n 10 = d Oq y = 0.40 x +0.45

. _ I I = 0.87

01 0.5 1.0 1.5 2.0 2.5 3.0 fl !*

A saw T&#m&Qam ma-3 ??

B G A*sc& hi-, L9wl

EFFECTIVE ORIFICE AREA (MITRAL TISSUE) BY CO [cm*) EFFECTIVE ORIFICE AREA (AORTIC TISSUE) BY CD (cm*)

y = 038 x +0.30 r = 0.76

..%E m a EFFECTIVE DRIFiE AREA (ADRTIC MECHAkAL) BY CD (cm’)

Figure 6. Correlation of color-guided conventional Doppler (CD) and cardiac catheterization-derived effective prosthetic orifice areas for A, tissue mitral; B, tissue aortic; C, mechanical mitral and D, mechanical aortic prostheses. Data points below the X axis are from patients who did not undergo cardiac catheterization or in whom the left ventricle could not be entered as well as those in whom effective orifice areas could not be computed because of significant prosthetic regurgitation or lack of cardiac output data. Data points outside the Y axis in panel B are from two patients in whom adequate quality Doppler spectral traces could not be obtained.

were 13.4 and 25.5 mm Hg for the mitral and 34 and 50 mm Hg for the aortic prostheses, whereas lower continuous wave Doppler gradients were found in the remaining patients with nonobstructed mechanical prostheses.

Effective orifice areas. The effective prosthetic orifice areas computed by the pressure half-time method for the mitral prostheses and by the simplified continuity equation method

for the aortic prostheses correlated well with those derived at cardiac catheterization. Better correlations were obtained for the tissue than for the mechanical mitral and aortic prostheses: tissue mitral (Fig. 6A): r = 0.94, range 0.6 to 2.4 cm*, mean 1.35 ? 0.67 cm*, n = 10; tissue aortic (Fig. 6B): r = 0.87, range 0.40 to 1.67 cm*, mean 0.90 + 0.33 cm*, n = 10; mechanical mitral (Fig. 60 r = 0.79, range 0.70 to 3.3 cm*, mean 1.9 2 0.59 cm*, n = 22; mechanical aortic (Fig. 6D): r = 0.76, range 0.68 to 2.46 cm*, mean 1.32 + 0.51 cm*, n = 11.

For the tissue mitral and aortic prostheses, the effective orifice areas by color-guided Doppler ultrasound of I 1.1 and 51 cm*, respectively, tended to be associated with an obstructed prosthesis, whereas those with higher valve orifice areas tended to have a nonobstructed valve at surgery. The effective orifice areas for the four obstructed mechanical prostheses were 0.7 and 1.0 cm* for the mitral

JACC Vol. 13, No. 7 June 1989:1561-71

BO-

60 -

8

MITRAL PROSTHESES (n = 601

z g 40------- - -____________ . . 2 < . .

2 . 20 __________Zi__~.._---___

A AA A*

“*A A ,o ..___ ____._ _ ____ _______

0 I I!

. .

. . ;* **A.* . .A ??.

.* -_____ .

. .

-----

-----.

III ANGlOGRAPHIC GRADING

9 Tmue Proslhesas

A Mechanml Prosthew

A ?? Trace regurg,tatmn by angmgraphy

AORTIC PROSTHESES 907

(n = 49) . .

:.*A :. . __-___ --_-_._---_-~._L___.._~---

60. .

:: A*A .

A

z . . .

3 ---- -_-----_--_--_r_t-------

.

- 40- . .

$

3

& _.--------_-_-~_--_--_--_---_-

220.. . .

S 7 ??

A A*

A ,O____-_-----._----_.--________.

ANGlOGRAPHlCiGRADlNG

. Tissue Prosthem

A Mechanml Prosthesis

6 * Trace regurg,tat,on by angmgraphy

Figure 7. A, Comparison of Doppler color flow mapping (CD) and angiographic grading of mitral prosthetic regurgitation in 60 patients. B, Comparison of Doppler color flow mapping (CD) and angiographic grading of aortic prosthetic regurgitation in 49 patients. JW = regurgitant jet width; LAA = left atria1 area; LVOT = left ventricular outflow tract; RJA = regurgitant jet area.

and 0.68 and 0.87 cm2 for the aortic prostheses. In the majority of patients with a nonobstructed mechanical prosthesis the effective valve area was >1.3 cm2 for the mitral and larger than 1 .O cm2 for the aortic prostheses.

KAPUR ET AL. 1567 DOPPLER COLOR IMAGING IN PROSTHETIC VALVES

Prosthetic Regurgitation

Detection of regurgitation (Fig. 7). Of the 109 prosthetic valves (60 mitral, 49 aortic) studied angiographically, 40 mitral and 34 aortic prostheses had evidence of regurgitation by both Doppler color flow mapping and angiography; 16 mitral and 10 aortic prostheses did not demonstrate evidence of regurgitation by both techniques. Four mechanical mitral prostheses and three mechanical aortic prostheses had trace regurgitation by angiography but none by Doppler color flow mapping. One mechanical mitral prosthesis with angiographic grade I mitral regurgitation had none by Doppler color flow mapping. Also, two aortic prostheses (one tissue, one mechanical) with mild regurgitation by Doppler color flow mapping had no evidence of leakage during angiography. Thus, the sensitivity of Doppler color flow mapping in the detection of prosthetic mitral regurgitation was 89% and the specificity was 100%; for the detection of prosthetic aortic regurgitation, the sensitivity of the Doppler color flow mapping was 92% and the specificity was 83%.

Grading of regurgitation (Fig. 7). There was complete agreement between the Doppler color flow mapping and angiographic grading of the severity of prosthetic valvular regurgitation in 36 (90%) of the 40 mitral and 2.5 (73.5%) of the 34 aortic regurgitant prostheses. Underestimation by one grade occurred in three mitral and four aortic prostheses and by two grades in one aortic prosthesis and none of the mitral prostheses. Overestimations by one grade occurred in six aortic prostheses and none of the mitral prostheses. Over- estimation by two grades did not occur in any of the mitral or aortic prostheses.

Type of regurgitation. Doppler color flow mapping diagnosed valvular regurgitation in 20, paravalvular regurgitation in 3 and combined paravalvuiar and valvular regurgitation in 3 of the 32 tissue mitral prostheses studied at surgery. Thirteen of 20 prostheses with valvular regurgitation were found to have torn, perforated. destroyed or retracted leaflets at surgery, whereas evidence of leaflet thickening and degeneration only was noted in the remaining 7. All three prostheses with paravalvular regurgitation had suture line dehiscence with no obvious evidence for leaflet destruction. One of the three prostheses with combined paravalvular and valvular regurgitation demonstrated both valve destruction and suture line dehiscence, whereas in the remaining 2, valve destruction with intact suture line was observed. In both these cases, the additional regurgitant jet by Doppler color flow mapping was noted to originate at the level of the stents, although in all others with paravalvular regurgitation the regurgitant jet originated beyond the level of the stents. Doppler color flow mapping diagnosed paravalvular regurgitation in 7 and valvular regurgitation in none of the 13 mechanical mitral prostheses studied at surgery. All seven demonstrated dehiscence of the suture line without any evidence of occluder dysfunction.


JACC Vol. 13, No. 7 June 1989:1561-71

Doppler color flow mapping diagnosed valvular regurgitation in 19, paravalvular regurgitation in 6 and combined paravalvular and valvular regurgitation in 3 of the 31 tissue aortic prostheses assessed at surgery. Thirteen of 18 tissue prostheses with valvular regurgitation showed evidence at surgery of torn, destroyed, perforated or retracted leaflets, whereas the remaining 5 demonstrated only leaflet thickening and degeneration. Two of the six prostheses with paravalvular regurgitation had sewing ring dehiscence with no obvious evidence for leaflet destruction, three demonstrated evidence of leaflet destruction with an intact suture line and both valve leaflet destruction and suture line dehiscence were noted in the remaining one. Two of the three prostheses with both paravalvular and valvular regurgitation demonstrated both suture line dehiscence and leaflet tissue destruction, whereas the remaining one demonstrated only suture line dehiscence without any obvious evidence of leaflet tissue destruction.

Doppler color flow mapping diagnosed valvular regurgitation in 3, paravalvular regurgitation in 3 and combined paravalvular and valvular regurgitation in 2 of the 15 mechanical aortic prostheses studied at surgery. Two of three prostheses with valvular regurgitation demonstrated, at surgery, thrombus preventing effective closure of the mechanical occluder without suture line dehiscence, whereas the remaining one showed suture line dehiscence with a normally functioning occluder. All three prostheses with paravalvular regurgitation showed sewing ring dehiscence without obvious occluder dysfunction. One of the two prostheses with combined valvular and paravalvular regurgitation demonstrated both suture line dehiscence and incomplete closure of the occluder due to infective vegetations; the other one had sewing ring dehiscence with a normally functioning occluder. In regard to paravalvular aortic prosthetic regurgitation, a correct diagnosis was made by Doppler color flow mapping in 2 of 2 prostheses in which the regurgitant jet was observed to originate beyond the level of the sewing ring and in 9 of 12 prostheses with the jet origin noted to be at the extreme anterior or posterior aspects of the prostheses but not clearly beyond the level of the prosthetic elements. For the valvular aortic prosthetic regurgitation, correct diagnosis was made by Doppler color flow mapping in 24 of the 27 prostheses in which the jet was observed to originate from the central aspect of the prostheses. None of the mitral or aortic prostheses without evidence of regurgitation by Dopp- ler color flow mapping demonstrated leaflet destruction, retraction, occluder dysfunction or suture line dehiscence at surgery.

Doppler color flow mapping demonstrated a localized area of flow acceleration in 3.5 prostheses (29 mitral [25 tissue, 4 mechanical]; and 6 aortic [4 tissue, 2 mechanical]). With use of this criterion, the type of regurgitation was correctly characterized in relation to the surgical findings in all mitral prostheses (22 valvular, 6 paravalvular and 1

combined valvular and paravalvular) and 5 of 6 aortic prostheses (2 valvular, 2 paravalvular and 1 combined valvular and paravalvular). In one tissue aortic prosthesis valvular regurgitation was incorrectly diagnosed as paravalvular.

Discussion

Prosthetic Obstruction

Conventional Doppler echocardiography. Pressure gradient. The introduction of conventional continuous wave Doppler echocardiography allows estimation of transprosthetic pressure gradients. For both tissue and mechanical mitral prostheses several Doppler echocardiographic studies have shown excellent correlations (r = 0.93 to 0.96) with cardiac catheterization for the mean pressure gradients (6,7,13,14). However, the number of patients studied in each of the series was relatively small, ranging from 8 to 19 patients. As far as the aortic prostheses is concerned, only one study (7) has been published so far comparing the mean pressure gradients by conventional Doppler echocardiography with cardiac catheterization. The investigators obtained a high correlation coefficient of 0.93 in 10 patients (8 with porcine, 2 with Bjork-Shiley prostheses). Because transprosthetic pressure gradients are dependent on cardiac output as well as the type and size of the prosthesis, relatively high gradients have been obtained even with clinically normal prosthetic valve function (22,23), thus limiting the usefulness of this variable.

Effective or&e area. Recently, attempts have been made to noninvasively assess the effective mitral prosthetic orifice area with use of the Doppler pressure half-time method (6,7,13,24), but conflicting results have been obtained when these are correlated with the Gorlin formula-derived effective orifice area at catheterization. In small numbers of patients with both tissue and mechanical mitral prostheses, Sagar et al. (7) (12 patients) and Alam et al. (6) (8 patients) obtained excellent correlation coefficients of 0.97 and 0.83, respectively, although Wilkins et al. (13) obtained a poor correlation in the 8 patients they studied (r = 0.65, p = NS). Although aortic prosthetic areas can also be potentially measured noninvasively with the simplified continuity equation (24), no systematic study has been published document- ing its validity with cardiac catheterization results.

Surgical correlation. It is also important to correlate both the catheterization and the noninvasive findings with the presence of obvious obstruction at surgery because all manufactured tissue and mechanical prosthetic valves are mildly obstructive (25,26) and may demonstrate a significant pressure gradient and reduced effective orifice area at catheterization. However, very sparse data exist on the use of Doppler echocardiography in predicting the presence or absence of transprosthetic obstruction as assessed at surgery. Goldrath et al. (27) in a study of seven Bjork-Shiley

JAW Vol. Ii. No. 7

June 1989:1561-71 KAPUR ET AL. 1569

DOPPLER COLOR IMAGING IN PROSTHETIC VI\L,VES

and Starr-Edwards mitral prostheses found a mean transprosthetic pressure gradient of > 10 mm Hg and an effective orifice area <I cm’ to be associated with definite prosthetic obstruction at surgery (thrombosis, tissue ingrowth). whereas Alam et al. (6) observed a mean pressure gradient > 11 mm Hg and a valve area 4 1.1 cm’ to be associated with transprosthetic obstruction in the 10 surgically evaluated tissue mitral prostheses.

Doppler color flow mapping. A major limitation of conventional Doppler echocardiography in assessing prosthetic obstruction is that the transprosthetic flow is not visualized. Therefore, no landmark exists for the parallel alignment of the continuous wave cursor with the direction of transprosthetic flow. Assessing transprosthetic pressure gradients without the guidance of color flow signals is likely to be difficult. especially for the mechanical aortic prosthesis. because of the eccentric direction of the transprosthetic jets. On the other hand. Doppler color flow mapping. by allowing visualization of transprosthetic signals, provides a reference mark for the placement of the continuous wave Doppler cursor (12.24.28). Using this approach, our present study has demonstrated in a relatively large number of patients fairly good correlations with catheterization results for the mean transprosthetic pressure gradients derived by color-guided continuous wave Doppler ultrasound for both tissue and mechanical mitral and aortic prostheses. Fairly good correlations were also obtained for the effective prosthetic orifice areas calculated by, Doppler technique but the results should be interpreted with caution because the number of patients in each prosthesis group is relatively small and recent studies have raised some doubts about the validity of the Gorlin equation in the catheterization estimation of prosthetic orifice area, especially for aortic prostheses (29-31). A limitation of our study is that Doppler color flow examinations were not done simultaneous with cardiac catheterization and. in more than half of the cases, the mean interval between the two procedures was 6.4 days. This may partly explain some of the discrepancies between Doppler color flow and cardiac catheterization findings.

Jet widrlz. Doppler color flow mapping also provides another index for the assessment of mitral tissue prosthetic obstruction. In our experience, the maximal color jet width at its origin from the tissue mitral prosthesis correlated well with the effective orifice area measured at catheterization.

Prosthetic Vahe Regurgitation

Conventional Doppler echocardiography. Although conventional pulsed Doppler echocardiography has a high sensitivity and specificity in detecting native and prosthetic valve regurgitation, it has been found to be less reliable in assessing of the severity of such regurgitation, with several series reporting significant underestimation by conventional Doppler technique (6,32). A major reason for this underes-

timation is the inability of the technique to visualize the regurgitant signals in real time, making it difficult to assess the area of turbulence in multiple planes which is essential to provide the full extent and size of the three-dimensional regurgitant jets. This limitation also makes it difficult to categorize the type of regurgitation. Doppler color flow mapping overcomes many of the limitations of the pulsed Doppler technique in the evaluation of prosthetic regurgitation because it allows simultaneous visualization of flow patterns superimposed on practically real-time two- dimensional images, making it easier to visualize the flow disturbance in multiple imaging planes. Proper assessment of the severity of prosthetic regurgitation is important because minimal to mild regurgitation is seen in many apparently normally functioning tissue and mechanical prosthetic valves (6,7.22,23).

Doppler color flow mapping. In the relatively large number of patients we studied, Doppler color flow mapping was found to be highly sensitive and specific in detecting both tissue and mechanical mitral or aortic prosthetic regurgitation; however, the technique failed to detect trace regurgitation seen by angiography in some patients. Doppler color flow mapping was also found to be reliable in assessing regurgitation severity when the technique was compared with standard contrast angiography; underestimation by more than one grade occurred in only one patient and overestimation by more than one grade occurred in none. In addition, Doppler color flow mapping also proved to be useful in evaluating the type of prosthetic regurgitation, which has important surgical implications because significant valvular regurgitation entails valve replacement, whereas a paravalvular leak may be repaired without having to replace the valve itself. Because in many instances the type of prosthetic regurgitation cannot be delineated by contrast angiography (33,34), the Doppler echocardiographic findings on the type of prosthetic regurgitation need to be compared with surgical and anatomic findings. In 94% of the mitral and 80.5% of aortic prostheses studied at surgery, the Doppler color flow mapping findings of the type of regurgitation correlated well with surgical findings.

In our experience, if the flow signals were clearly visualized as originating beyond the prosthetic stents or ring, the regurgitation was always paravalvular, whereas if the regurgitant signals were distinctly visualized as originating well within the confines of the prosthetic stents or ring, the regurgitation was always valvular. On the other hand, if the regurgitant flow signals originated at the level of the mitral or aortic prosthetic stents or from the extreme anterior or posterior aspects of the aortic prostheses (when the stents or the ring could not be confidently identified), the regurgitation tended to be paravalvular rather than valvular, but a confi- dent diagnosis of the type of the prosthetic regurgitation was not possible.

F/OK’ ac.c.eleration. Flow acceleration (18.35). which is an

1570 KAPUR ET AL. JACC Vol. 13, No. 7 DOPPLER COLOR IMAGING IN PROSTHETIC VALVES June 1989:1561-71

area of localized increase in flow velocity proximal to the origin of the regurgitant jet, when clearly seen between and beyond the level of the prosthetic stents or ring, represented another independent variable in predicting the site of the prosthetic regurgitation. Visualization of the flow acceleration site served to increase the confidence with which the diagnosis of the type of prosthetic regurgitation could be made. However, it is less useful than the site of the regurgitant jet because it is not visualized in many patients, especially those with an aortic prosthesis.

Limitations

Doppler color flow mapping. This technique has several limitations in assessing prosthetic valve function. As for conventional Doppler echocardiography, the metallic com- ponents of the prosthetic valves impede the transmission of the ultrasound beam, thus creating an acoustically silent area behind the prostheses and making the detection of regurgitant flow signals difficult (36). Moreover, the prosthetic reverberations from both the aortic and mitral prostheses often clutter the left atrium, obscuring the presence of true regurgitant signals. These limitations may be obviated in many patients by the use of multiple two-dimensional echocardiographic planes including the right parasternal approach, which allows interrogation of the left atrium by the ultrasound beam without the interposition of the prosthesis, thus facilitating identification of regurgitant signals. In three of our patients with a mitral prosthesis, the maximal regurgitant jet area was visualized from the right parasternal transducer position. However, adequate quality examinations may not be obtained in some patients with small acoustic windows. Because Doppler color flow mapping is an operator-dependent technique, it requires examination by a highly trained and experienced echocardiographer who is fully conversant with the instrument settings and technical pitfalls. This is an important limitation of the Doppler color how technique.

Cardiac catheterization. Although we correlated our Doppler color flow mapping findings with cardiac catheterization, the latter technique also has several limitations in the assessment of prosthetic valve function. The presence of a mechanical prosthesis in the aortic position makes retrograde left heart catheterization almost impossible and the more invasive transeptal approach or a direct left ventricular puncture, to assess transprosthetic pressure gradients and native or prosthetic mitral valve function, becomes neces- sary. The use of fluid-filled catheter systems also creates inaccuracies in the estimation of pressure gradients. Further errors are introduced in the computation of valve area by the use of the Fick or thermodilution method for the estimation of cardiac output. Moreover, as mentioned previously, there are documented limitations in the direct application of the Gorlin formula to the estimation of prosthetic orifice area

(29-31). In some patients with a prosthetic mitral valve, the pulmonary capillary wedge pressure may occasionally over- estimate the left atria1 pressure, resulting in an erroneous effective orifice area (37). In addition, the angiographic criteria used for the semiquantitative assessment of regurgitation severity are rather subjective and could be affected by the amount of contrast medium injected and the position of the catheter tip in the left ventricle or aortic root (38).

Conclusions

Our study of a relatively large number of patients with prosthetic valves demonstrates the usefulness of Doppler color flow mapping in their evaluation, especially in the assessment of prosthetic regurgitation. Greater experience in a larger number of patients with different types of obstructed prosthesis is required to evaluate the reliability of color-guided continuous wave Doppler echocardiography in assessing effective prosthetic orifice area.

We appreciate the secretarial assistance of Marcy Ashburn and Lindy Chapman in the preparation of this manuscript. -

1.

2.

3.

4.

5.

6.

I.

8.

9.

10.

II.

References Kotler MN, Mintz GS, Panidis IP, Morganroth J, Segal BL, Ross J. Noninvasive evaluation of normal and abnormal prosthetic valve function. J Am Coil Cardiol 1983;2:151-73.

Gross CM, Wann LS. Doppler echocardiography in the assessment of prosthetic cardiac valves. In: Nanda NC, ed. Doppler Echocardiography. New York: Igaku-Shoin, 1985:293-335.

Ramirez ML, Wong M. Reproducibility of stand-along continuous wave Doppler recordings of aortic flow velocity across bioprosthetic valves. Am J Cardiol 1985;55: 1197-g.

Ferrara RP, Labovitz AJ, Wiens RD, Kennedy HL, Williams GA. Prosthetic mitral regurgitation detected by Doppler echocardiography. Am J Cardiol 1985;55:22%30.

Switzer DF, Nanda NC. Limitations of pulsed and continuous wave Doppler echocardiography. Echocardiography: A Review of Cardiovas- cular Ultrasound 1985;1:207-11.

Alam M, Rosman HS, Lakier JB, et al. Doppler and echocardiographic features of normal and dysfunctioning bioprosthetic valves. J Am Co11 Cardiol 1987;10:851-8.

Sagar KB, Wann LS, Paulsen WHJ, Romhilt DW. Doppler echocardiographic evaluation of Hancock and Bjork-Shiley prosthetic valves. J Am Coil Cardiol 1986;7:681-7.

Come PC. Pitfalls in the diagnosis of periprosthetic valvular regurgitation by pulsed Doppler echocardiography. J Am Coil Cardiol 1987;9: 1176-g.

Omoto KR, Yokote Y, Takamoto S, et al. The development of real time two-dimensional Doppler echocardiography and its clinical significance in acquired valvular diseases with special references to the evaluation of valvular regurgitation. Jpn Heart J 1984;25:32540.

Helmcke F, Nanda NC, Hsiung MC, et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175-83.

Perry GJ, Helmcke F, Nanda NC, Byard C, Soto B. Evaluation of aortic insufficiency by Doppler color flow mapping. J Am Co11 Cardiol 1987;9: 952-g.

JACC Vol. 13, No. 7 June 1989:1561-71

KAPUR ET AL. 1571 DOPPLER COLOR IMAGING IN PROSTHETIC VALVES

12. Fan PH, Kapur KK, Nanda NC. Color-guided Doppler echocardiographic assessment of aortic valve stenosis. J Am Coll Cardiol 1988;12: 441-9.

13.

14.

IS.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

Wilkins GT, Gillam LD, Kritzer GL, Levin RA, Palacios IF, Weyman AE. Validation of continuous-wave Doppler echocardiographic measurements of mitral and tricuspid prosthetic valve gradients: a simultaneous Doppler-catheter study. Circulation 1986;74:786-95.

Holen J, Simonsen S, Froysaker T. An ultrasound Doppler technique for the noninvasive determination of the pressure gradient in the Bjork-Shiley mitral valve. Circulation 1979;59:436-42.

Hatle L, Angelsen B, Tromsdal A. A noninvasive assessment of atrio- ventricular pressure half-time by Doppler ultrasound. Circulation 1979; 60:1096-IO4

Skjaserpe T, Hegrenaes L, Hatle L. Noninvasive estimation of valve area in patients with aortic stenosis by Doppler ultrasound and two- dimensional echocardiography. Circulation 1985;12:810-8.

Zoghbi WA, Farmer KL, Soto JG, Nelson JG, Quinones MA. Accurate noninvasive quantification of stenotic aortic valve area by Doppler echocardiography. Circulation 1986;73:452-9.

Fan PH, Kapur KK, Aggarwal KK, Jain S, Goyal R, Nanda NC. Utility of color flow acceleration in differentiating valvar from paravalvar prosthetic mitral regurgitation (abstr). Circulation 1987;76(suppl IV):IV-448.

Nagle RE, Walker D, Grainger RG. The angiographic assessment of mitral incompetence. Clin Radio1 1%8;19:154-64.

Hunt D, Baxley WA, Kennedy JW, Judge TP, Williams JE, Dodge HT. Quantitative evaluation of cineaortography in the assessment of aortic regurgitation. Am J Cardiol 1973;31:69&700.

Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenotic mitral valve, other cardiac valves and central circulatory shunts. Part I. Am Heart J 1951;41:1-20.

Williams GA, Labovitz AJ. Doppler hemodynamic evaluation of prosthetic (Starr-Edwards and Bjork-Shiley) and bioprosthetic (Hancock and Carpentier-Edwards) cardiac valves. Am J Cardiol 1985;56:325-32.

Cooper DM, Steward WJ, Schiavone WA, et al. Evaluation of normal prosthetic valve function by Doppler echocardiography. Am Heart J 1987;114:576-82.

Gupta A, Helmcke F, Pandey BJ, Nanda NC, Aggarwal KK, Yoganathan AP. Color guided continuous wave Doppler assessment of effective prosthetic valve area (abstr). J Am Coll Cardiol 3988:2:377A.

Rahimtoola SH. The problem of valve prosthesis-patient mismatch. Circulation 1978:58:2(cl.

26. Holen J, Hoie J. Semb B. Obstructive characteristics of Bjork-Shiley, Hancock, and Lillehei-Kaster prosthetic mitral valves in the immediate postoperative period. Acta Med Stand 1978;204:5-10.

27. Goldrath N, Zimes R, Vered Z. Analysis of Doppler-obtained velocity curves in functional evaluation of mechanical prosthetic valves in the mitral and aortic positions. J Am Sot Echo 1988;1:21 I-25.

28. Goyal RG, Kan MN, Soto B, et al. Color Doppler assessment of prosthetic valve function and its limitations (abstr). Circulation 1986; 74(suppl IV):IV-389.

29. Cannon SR, Richards KL, Crawford M. Hydraulic estimation of stenotic orifice area: a correction of the Gorlin formula. Circulation 1985;71: 1170-8.

30. Cannon SR, Richards KL, Crawford MH, et al. Inadequacy of the Gorlin formula for predicting prosthetic valve area. Am J Cardiol 1988;62: 113-6.

31. Ubago JL, Figueroa A, Colman T, Ochoteco A, Duran C. Hemodynamic factors that affect calculated orifice areas in the mitral Hancock xenograft valve. Circulation 1980;61:388-96.

32. Panidis IP, Ross J, Mintz GS. Normal and abnormal prosthetic valve function as assessed by Doppler echocardiography. J Am Coll Cardiol 1986:8:317-26.

33. Meister SG, Wolf NM. Post operative management of patients with implanted valvular prostheses. In: Morse D, Steiner RM, Femandez J, eds. Guide to Prosthetic Cardiac Valves. New York: Springer-Verlag, 1985:179-89.

34. Linhart JW, Wheat MW. Cineangiographic studies in patients with Starr-Edwards aortic valves. Circulation 1966;34:343-5.

35. Simpson IA, Sahn DJ, Valdes-Cruz LM, Chung KJ, Sherman FS, Swensson RE. Color Doppler flow mapping in patients with coarctation of the aorta: new observations and improved evaluation with color flow diameter and proximal acceleration as predictors of severity. Circulation 1988:4:73ti4.

36. Sprecher DL, Adamick R, Adams D, Kisslo J. In vitro color flow, pulsed and continuous wave Doppler ultrasound masking of flow by prosthetic valves. J Am Coll Cardiol l985;9: 1306-10.

37. Schoenfeld MH, Palacios IF, Hutter AM Jr, Jacoby SS, Block PC. Underestimation of prosthetic mitral valve areas: role of transseptal catheterization in avoiding unnecessary repeat mitral valve surgery. J Am Coll Cardiol 1985;5: 1387-92.

38. Croft CH, Lipscomb K, Mathis K, et al. Limitations of qualitative angiographic grading in aortic or mitral regurgitation. Am J Cardiol 1984:53:359s.

doppler color flow mapping in the evaluation of …jacc vol. 13, no. 7 june 1989:1561-71 1561...

Documents