Background
The left atrial appendage (LAA) is a common site of thrombus formation and is the source of systemic thromboembolism in patients with rheumatic mitral stenosis. LAA contractile dysfunction is a common finding in these patients. The aim of this study was to assess immediate and 6-month follow-up LAA function by transesophageal Doppler echocardiography in patients who underwent percutaneous transvenous mitral commissurotomy (PTMC).
Methods
Forty-seven consecutive patients with symptomatic critical mitral stenosis who underwent PTMC were enrolled. All had underwent transthoracic and transesophageal echocardiography before, 24 hours after, and 6 months after PTMC. Pulse Doppler velocities of the LAA were measured, including peak early diastolic (E wave), peak late diastolic (A wave), and peak systolic (S wave). The corresponding tissue Doppler velocities of the LAA, including peak early diastolic (E LAA ), peak late diastolic (A LAA ), and peak systolic (S LAA ), were also measured. LAA ejection fraction was measured using the modified Simpson’s method.
Results
The mean age of the 47 enrolled patients was 31.7 ± 10.26 years. Thirty-eight patients were in sinus rhythm, and the remaining nine were in atrial fibrillation. PTMC was successful in all patients. The pulse Doppler velocities of the LAA at baseline, after PTMC, and at 6-month follow-up were as follows: for the E wave, 15.29 ± 2.26, 17.02 ± 2.25, and 17.97 ± 2.55 cm/sec, respectively ( P < .001); for the A wave 22.45 ± 4.11, 24.19 ± 4.21, and 25.99 ± 4.51 cm/sec, respectively ( P < .001); and for the S wave, 28.52 ± 4.37, 31.45 ± 5.37, and 33.06 ± 4.99 cm/sec, respectively ( P < .001). The corresponding tissue Doppler velocities of LAA were as follows: for E LAA , 4.65 ± 0.91, 5.28 ± 0.85, and 5.80 ± 0.84 cm/sec, respectively ( P < .001); for A LAA , 6.67 ± 1.12, 7.33 ± 1.17, and 7.88 ± 1.22 cm/sec, respectively ( P < .001); and for S LAA , 4.67 ± 1.12, 5.52 ± 1.18, 6.07 ± 1.11 cm/sec, respectively ( P < .001). There was a nonsignificant increase in LAA ejection fraction (48.97 ± 8.14% vs 52.3 ± 13.76% vs 52.11 ± 16.3%, respectively, P = .052). On subgroup analysis between patients in sinus rhythm and those with atrial fibrillation, there was no significant difference for LAA ejection fraction and pulse and tissue Doppler velocities. Very good intraclass correlation of the LAA parameters was also observed for the reproducibility of the data.
Conclusions
The present study shows contractile dysfunction of the LAA in patients with critical mitral stenosis, which significantly improved after PTMC, and a further improvement was observed at 6-month follow-up. Favorable 6-month improvements in LAA parameters suggest continuous structural remodeling of the LAA after PTMC, which is clinically attributed to the absence of thromboembolism. Although there was an improvement in LAA function, it was far below the normal range, suggesting a need for continuous long-term monitoring and management of thromboembolism in these patients.
Patients with critical, rheumatic mitral stenosis (MS) have left atrial (LA) and LA appendage (LAA) dysfunction because of pressure and volume overload. The LAA, being a compliant chamber, plays an important role as a reservoir in this situation. It is also a highly dynamic and contractile structure, which helps prevent local blood stagnation in healthy individuals. Patients with MS have a propensity for local thrombus formation in the LA and LAA because of blood stasis, impaired contractile function, and atrial fibrillation (AF). The LAA is a common site of thrombus formation and a source of systemic embolism in these patients. Its functional status can be assessed by Doppler flow velocities and ejection fraction, measured by two-dimensional transesophageal echocardiography (TEE). Although there have been studies of LAA functional assessment in patients with MS who undergo percutaneous transvenous mitral commissurotomy (PTMC), none have studied short or long-term improvement of LAA function after PTMC. We therefore assessed LAA ejection fraction and pulse and tissue Doppler velocities at baseline, after PTMC, and at 6-month follow-up and conducted a subgroup analysis of LAA function in patients in sinus rhythm and those with AF.
Methods
Fifty patients with critical MS who underwent PTMC from February 2006 to June 2007 were enrolled in our study. Three patients who had grade ≥ 3 mitral regurgitation after PTMC were excluded; hence, analysis of 47 patients was performed. Patients with critical MS with LA or LAA clots, grade ≥ 2 mitral regurgitation, or mitral valve calcification with Wilkins echocardiographic score ≥ 3 did not undergo PTMC and hence were excluded from the study. Patients with LAA clots on TEE were given oral anticoagulant therapy for ≥3 months, and TEE was then repeated to confirm dissolution of the thrombus. Those with dissolved LAA clots underwent PTMC; the remainder were given the option of surgical commissurotomy with clot removal. All enrolled patients underwent standard transthoracic echocardiography (3.5-MHz transducer) and multiplane TEE (5.0-MHz transducer) (Vivid V; GE Vingmed Ultrasound AS, Horten, Norway) before, 24 hours after, and 6-months after PTMC. LA dimensions were measured in the parasternal long-axis view. Transmitral diastolic pressure gradient and mitral valve area by both planimetry and pressure half-time were measured using transthoracic echocardiography.
LAA function was evaluated in the transverse plane using TEE. LAA flow velocities, including peak systolic (S wave), peak early diastolic (E wave), and peak late diastolic (A wave), were measured at the outlet of LAA using electrocardiographically gated pulsed-wave Doppler. LAA end-diastolic and peak systolic volumes were measured using the modified Simpson’s method, and ejection fraction was calculated as [(end-diastolic volume − peak diastolic volume)/end-diastolic volume] × 100% ( Figure 1 ).
LAA tissue Doppler velocities were measured using the spectral mode of myocardial Doppler imaging (5 MHz) in the transverse plane, with positioning of the sample volume in the free wall midway between the tip and the outlet of the LAA. The cursor was placed as parallel as possible to the LAA lateral wall during entire recorded cardiac cycle. Gain setting filter and Nyquist limit were adjusted to optimize Doppler signal. After the electrocardiographic P wave, LAA contraction by tissue Doppler is reflected as a positive wave (A LAA ), which is followed by a negative wave (S LAA ) coinciding with left ventricular contraction and then a positive wave (E LAA ) coinciding with the isometric relaxation phase of left ventricular diastole ( Figure 2 ). Similar to tissue Doppler, the corresponding pulse Doppler velocities of the LAA were recorded, and measurements were made for the A, S, and E waves ( Figure 2 ). The late diastolic waves on both pulse (A wave) and tissue (A LAA ) Doppler were not recorded in patients with AF, because of the absence of P waves on electrocardiography. An average value of three consecutive beats with well-profiled velocity patterns was calculated in all patients with AF and those in sinus rhythm.
After preliminary analysis of our data, it was observed that there was a wide dispersion of various LAA parameters assessed on TEE, necessitating a reproducibility analysis. Because it could not be done in the study cohort retrospectively, we recruited 10 more consecutive patients with critical MS from September to November 2010 and measured pre-PTMC and post-PTMC LAA parameters. Six-month follow-up measurements were not performed in this subgroup. Data acquisition was done by two operators during the same transesophageal echocardiographic examination, and the second operator was blinded to first operator’s data. Intraclass correlation coefficients were calculated for analysis of interobserver variability using a two-way mixed-effects model, whereby patient effects were random and measures effects were fixed, on the basis of an average-measures model using absolute agreement definition. Intraobserver variability could not be assessed, because repeat transesophageal echocardiographic examinations were not ethically permissible.
PTMC Procedure
PTMC was performed using a standard transseptal approach with an Inoue balloon. Pre-PTMC and post-PTMC right-heart catheterization was performed for the calculation of mitral valve area using Gorlin’s formula. Written informed consent was obtained from all subjects before enrollment in the study. The study protocol was approved by the institutional ethics committee.
Statistical Analysis
Continuous variables are expressed as mean ± SD if normally distributed or as medians if their distributions were skewed. Categorical variables are presented as proportions. Comparison of pretreatment and posttreatment observations was performed using paired-samples t tests for continuous variables and McNemar’s test for categorical variables. Trends for continuous variable over 6 months were assessed using Friedman’s test. Reproducibility analysis on the separate cohort of 10 patients was performed using intraclass correlation coefficients, calculated using a two-way mixed-effects model, whereby patient effects were random and measures effects were fixed, on the basis of an average-measures model using absolute agreement definition. Statistical analysis was performed using SPSS version 13.0 (SPSS, Inc., Chicago, IL). All tests were two sided, and P values < .05 were considered statistically significant.
Results
Of 47 enrolled patients, there were 26 female and 21 male patients. The mean age was 31.7 ± 10.26 years (range, 13–60 years). The mean baseline New York Heart Association functional class was 2.57 ± 0.54, with 21 patients in class II, 25 in class III, and one in class IV. Thirty-eight patients were in sinus rhythm, and the remaining nine had AF. The mitral valve Wilkins echocardiographic score was 7.27 (range, 5–11), with 38 patients with scores ≤ 8 and the remaining nine with scores ≥ 9.
PTMC was successful in all enrolled patients. There were significant decreases in mean LA pressure (28.1 ± 8.3 to 13.9 ± 4.8 mm Hg, P < .001), transmitral end-diastolic gradient (18.9 ± 8.0 to 2.7 ± 3.3 mm Hg, P < .001), and peak pulmonary artery systolic pressure (62.2 ± 21.7 to 40.6 ± 13.7 mm Hg, P < .001) after PTMC ( Table 1 ). There were no procedural complications in any of the 47 patients. Mitral valve area significantly increased after PTMC by all three methods: Gorlin’s formula (0.72 ± 0.12 to 1.76 ± 0.17 cm 2 , P < .001), planimetry (0.80 ± 0.14 to 1.73 ± 0.19 cm 2 , P < .001), and pressure half-time (0.78 ± 0.16 to 1.64 ± 0.16 cm 2 , P < .001) ( Tables 1 and 2 ). There was a significant decrease in LA dimension after PTMC, which further decreased at 6-month follow-up (47 ± 5.3 to 44.4 ± 5.8 to 42.5 ± 5.6 mm, P < .001). Both LAA systolic (2.16 ± 0.53 to 1.61 ± 0.63 to 1.39 ± 0.58 mL, P < .001) and diastolic (4.27 ± 0.93 to 3.41 ± 0.97 to 2.96 ± 1.00 mL, P < .001) volumes were significantly decreased at 24 hours and at 6 months, while ejection fraction (48.97 ± 8.14% to 52.3 ± 13.76% to 52.11 ± 16.3%, P = .052) showed a nonsignificant improvement ( Table 3 ). Percentage change in mean LAA ejection fraction at follow-up also showed a nonsignificant improvement. There were significant increases in all pulsed-wave flow velocities of the LAA: for the S wave, 28.52 ± 4.37 to 31.45 ± 5.37 to 33.06 ± 4.99 cm/sec ( P < .001); for the E wave, 15.29 ± 2.26 to 17.02 ± 2.25 to 17.97 ± 2.55 cm/sec ( P < .001); and for the A wave, 22.45 ± 4.11 to 24.19 ± 4.21 to 25.99 ± 4.51 cm/sec ( P < .001) ( Table 3 , Figure 3 ). Similarly, there were significant increases in tissue Doppler velocities of the LAA: for S LAA , 4.67 ± 1.12 to 5.52 ± 1.18 to 6.07 ± 1.11 cm/sec ( P < .001); for E LAA , 4.65 ± 0.91 to 5.28 ± 0.85 to 5.80 ± 0.84 cm/sec ( P < .001); and for A LAA , 6.67 ± 1.12 to 7.33 ± 1.17 to 7.88 ± 1.22 cm/sec ( P < .001) ( Table 3 , Figure 3 ).
| Variable | Before PTMC ( n = 47) | After PTMC ( n = 47) | P |
|---|---|---|---|
| LA pressure (mm Hg) | 28.1 ± 8.3 | 13.9 ± 4.8 | <.001 |
| Transmitral end-diastolic gradient (mm Hg) | 18.9 ± 8.0 | 2.7 ± 3.3 | <.001 |
| Pulmonary artery peak systolic pressure (mm Hg) | 62.2 ± 21.7 | 40.6 ± 13.7 | <.001 |
| Mitral valve area by Gorlin’s formula (cm 2 ) | 0.72 ± 0.12 | 1.76 ± 0.17 | <.001 |
| Variable | Before PTMC ( n = 47) | After PTMC ( n = 47) | At 6 month follow-up ( n = 47) | P (before vs after PTMC) | P (after PTMC vs 6-month follow-up) | P for trend (Friedman’s test) |
|---|---|---|---|---|---|---|
| Mitral valve area (cm 2 ) | ||||||
| Two-dimensional echocardiography | 0.80 ± 0.14 | 1.73 ± 0.19 | 1.70 ± 0.19 | <.001 | <.001 | <.001 |
| Pressure half-time | 0.78 ± 0.16 | 1.64 ± 0.16 | 1.68 ± 0.16 | <.001 | <.001 | <.001 |
| Peak transmitral gradient (mm Hg) | 27.8 ± 7.8 | 17.4 ± 5.3 | 11.3 ± 3.8 | <.001 | <.001 | <.001 |
| Mean transmitral gradient (mm Hg) | 12.2 ± 3.4 | 7.4 ± 2.7 | 6.0 ± 2.3 | <.001 | <.001 | <.001 |
| LA dimension (mm) | 47 ± 5.3 | 44.4 ± 5.8 | 42.5 ± 5.6 | <.001 | <.001 | <.001 |
| Variable | Before PTMC ( n = 47) | After PTMC ( n = 47) | At 6 month follow-up ( n = 47) | P (before vs after PTMC) | P (after PTMC vs 6-month follow-up) | P for trend (Friedman’s test) |
|---|---|---|---|---|---|---|
| LAA end-diastolic volume (mL) | 4.27 ± 0.93 | 3.41 ± 0.97 | 2.96 ± 1.00 | <.001 | <.001 | <.001 |
| LAA peak systolic volume (mL) | 2.16 ± 0.53 | 1.61 ± 0.63 | 1.39 ± 0.58 | <.001 | <.001 | <.001 |
| LAA ejection fraction (%) | 48.97 ± 8.14 | 52.3 ± 13.76 | 52.11 ± 16.3 | .011 | .751 | .052 |
| LAA flow velocity (cm/sec) | ||||||
| Peak systolic (S wave) | 28.52 ± 4.37 | 31.45 ± 5.37 | 33.06 ± 4.99 | <.001 | <.001 | <.001 |
| Peak early diastolic (E wave) | 15.29 ± 2.26 | 17.02 ± 2.25 | 17.97 ± 2.55 | <.001 | <.001 | <.001 |
| Peak late diastolic (A wave) ∗ | 22.45 ± 4.11 | 24.19 ± 4.21 | 25.99 ± 4.51 | <.001 | <.001 | <.001 |
| LAA tissue velocity (cm/sec) | ||||||
| S LAA wave | 4.67 ± 1.12 | 5.52 ± 1.18 | 6.07 ± 1.11 | <.001 | <.001 | <.001 |
| E LAA wave | 4.65 ± 0.91 | 5.28 ± 0.85 | 5.80 ± 0.84 | <.001 | <.001 | <.001 |
| A LAA wave ∗ | 6.67 ± 1.12 | 7.33 ± 1.17 | 7.88 ± 1.22 | <.001 | <.001 | <.001 |
∗ Data for peak late diastolic velocity and A LAA tissue Doppler velocity are from only 38 patients in sinus rhythm; these two velocities were not measured in nine patients with AF.
On subgroup analysis between 38 patients in sinus rhythm and nine with AF, there was no significant difference between mitral valve area ( P = .671) and mean transmitral gradient ( P = .349), suggestive of similar technical success of PTMC in both groups. LAA ejection fraction and pulse-waved and tissue Doppler velocities were not significantly different between the two groups ( Table 4 ).
| Variable | Normal sinus rhythm ( n = 38) | AF( n = 9) | P | ||||
|---|---|---|---|---|---|---|---|
| Before PTMC | After PTMC | At 6-month follow-up | Before PTMC | After PTMC | At 6-month follow-up | ||
| MV area (cm 2 ) | |||||||
| Two-dimensional echocardiography | 0.83 ± 0.13 | 1.72 ± 0.19 | 1.70 ± 0.19 | 0.7 ± 0.13 | 1.75 ± 0.20 | 1.71 ± 0.20 | .671 |
| Pressure half-time | 0.79 ± 0.17 | 1.64 ± 0.16 | 1.68 ± 0.16 | 0.70 ± 0.10 | 1.65 ± 0.16 | 1.67 ± 0.16 | .936 |
| Peak transmitral gradient (mm Hg) | 27.8 ± 8.2 | 17.5 ± 5.4 | 11.3 ± 3.6 | 27.6 ± 6.1 | 16.8 ± 5.1 | 11.3 ± 4.5 | .367 |
| Mean transmitral gradient (mm Hg) | 12.4 ± 3.6 | 7.5 ± 2.6 | 6.0 ± 2.3 | 11.2 ± 2.9 | 7.1 ± 3.0 | 6.0 ± 2.4 | .349 |
| LA dimension (mm) | 46.9 ± 5.17 | 44.2 ± 5.7 | 42.6 ± 5.7 | 47.6 ± 6.2 | 44.8 ± 6.5 | 42.4 ± 5.9 | .730 |
| LAA end-diastolic volume (mL) | 4.24 ± 0.93 | 3.39 ± 1.00 | 2.93 ± 1.01 | 4.39 ± 0.99 | 3.46 ± 0.89 | 3.10 ± 1.04 | .514 |
| LAA peak systolic volume (mL) | 2.16 ± 0.54 | 1.64 ± 0.64 | 1.42 ± 0.58 | 2.15 ± 0.56 | 1.48 ± 0.63 | 1.28 ± 0.59 | .522 |
| LAA ejection fraction (%) | 48.7 ± 7.1 | 51.1 ± 14 | 50.2 ± 17.2 | 49 ± 12.2 | 57.8 ± 10.3 | 59.9 ± 8.5 | .181 |
| LAA flow velocity (cm/sec) | |||||||
| Peak systolic (S) | 28.51 ± 3.75 | 31.40 ± 4.55 | 33.06 ± 4.32 | 28.54 ± 6.68 | 31.68 ± 8.37 | 33.05 ± 7.56 | .890 |
| Peak early diastolic (E) | 15.52 ± 2.30 | 17.18 ± 2.35 | 18.20 ± 2.60 | 14.31 ± 1.88 | 16.32 ± 1.72 | 17.00 ± 2.21 | .308 |
| LAA tissue velocity (cm/sec) | |||||||
| S LAA | 4.49 ± 1.02 | 5.34 ± 1.17 | 5.915 ± 1.11 | 5.42 ± 1.30 | 6.26 ± 0.92 | 6.75 ± 0.87 | .025 |
| E LAA | 4.60 ± 0.95 | 5.24 ± 0.90 | 5.73 ± 0.89 | 4.82 ± 0.72 | 5.49 ± 0.59 | 6.12 ± 0.53 | .432 |
Nine patients (18%) had residual iatrogenic atrial septal defects on TEE at 6-month follow-up. All had predominantly left-to-right shunting on color Doppler. We did not perform the Valsalva maneuver for demonstration of right-to-left shunting across the atrial septal defect. On subgroup analysis, variables such as mitral valve area, transmitral gradient, LA dimension, and the presence of AF were comparable in two groups ( P = NS; Table 5 ).
| Variable | No ASD at 6 months ( n = 38) | ASD at 6 months ( n = 9) | P |
|---|---|---|---|
| Mitral valve area (cm 2 ) | |||
| Two-dimensional echocardiography | 1.70 ± 0.20 | 1.73 ± 0.14 | .681 |
| Pressure half-time | 1.67 ± 0.17 | 1.70 ± 0.13 | .634 |
| Peak transmitral gradient (mm Hg) | 11.2 ± 3.5 | 11.8 ± 4.8 | .725 |
| Mean transmitral gradient (mm Hg) | 5.9 ± 2.2 | 6.5 ± 2.4 | .486 |
| LA dimension (mm) | 43.00 ± 5.87 | 40.80 ± 4.77 | .302 |
| Grade 2 mitral regurgitation | 3 | 0 | <.001 |
| AF | 5 (13.1%) | 4 (44.4%) | .466 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
