- Email: [email protected]
Clinical and echocardiographic predictors of left atrial clot and spontaneous echo contrast in patients with severe rheumatic mitral stenosis: a prospective study in 200 patients by transesophageal echocardiography
International Journal of Cardiology 73 (2000) 273–279 www.elsevier.com / locate / ijcard
Clinical and echocardiographic predictors of left atrial clot and spontaneous echo contrast in patients with severe rheumatic mitral stenosis: a prospective study in 200 patients by transesophageal echocardiography Kewal C. Goswami MD (Med), DM (Card)*, Rakesh Yadav MD (Med), DM (Card), M. Bhaskara Rao MD (Med), DM (Card), Vinay K. Bahl MD (Med), DM(Card), Kewal K. Talwar MD (Med), DM (Card), Subhash C. Manchanda MD (Med), DM (Card) Department of Cardiology, Cardiothoracic Sciences Centre, All India Institute of Medical Sciences, Ansari Nagar, New Delhi-110029, India Received 16 August 1999; received in revised form 11 January 2000; accepted 28 January 2000
Abstract The objective of this study was to prospectively investigate various clinical and echocardiographic variables to predict the left atrial and left atrial appendage clot and spontaneous echo contrast in patients with severe rheumatic mitral stenosis. We studied 200 consecutive patients (112 males and 88 females; mean age 29.669.6 years). Left atrial clot and spontaneous echo contrast were present in 26 and 53.5% of cases, respectively. There were no significant differences in the mitral valve area, mean transmitral diastolic gradient and left ventricular ejection fraction between patients with and without clot. Patients with clot were older (34.4611.4 vs. 28.268.5 years, P,0.001), had longer duration of symptoms (41.4636.0 vs. 28.8622.9 months, P,0.001), more frequent atrial fibrillation and spontaneous echo contrast (69.2 vs. 16.9%, P,0.00001 and 76.9 vs. 45.3%, P,0.00001, respectively) and larger left atrial area and diameter (41.0612.7 vs. 29.967.4 cm 2 , P,0.00001 and 53.968.3 vs. 47.667.4 mm, P,0.0001, respectively) as compared to patients without clot. Similarly patients with spontaneous echo contrast were older (31610.4 vs. 27.868.3 years, P,0.01), had more frequent atrial fibrillation (48.6 vs. 9.7%, P,0.0001), left atrial clot (37.4 vs. 12.9%, P,0.0001), larger left atrial area and diameter (37.6611.2 vs. 28.166.7 cm 2 , P,0.00001 and 52.268.3 vs. 45.966.5 mm, P,0.00001, respectively) and smaller mitral valve area (0.7760.14 vs. 0.8460.13 cm 2 , P,0.01) as compared to patients without spontaneous echo contrast. There were no significant differences in the mean transmitral diastolic gradient and left ventricular ejection fraction. On multiple regression and discriminant function analysis, atrial fibrillation and left atrial area were independent predictors of left atrial clot formation. In a subgroup of patients with sinus rhythm, larger left atrial area and presence of spontaneous echo contrast were significantly associated with the presence of clot in left atrium and appendage. We conclude that in patients with severe mitral stenosis, the presence of atrial fibrillation and in the subgroup of the patients with sinus rhythm the presence of large left atrium ($40 cm 2 ) and spontaneous echo contrast were associated with higher risk of clot formation in the left atrium and might be benefited by prophylactic anticoagulation. 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mitral stenosis; Left atrial clot; Left atrial spontaneous echo contrast; Transesophageal echocardiography
1. Introduction *Corresponding author. Tel.: 191-11-659-3218; fax: 191-11-6862663.
The incidence of thromboembolic complications is higher in patients with rheumatic mitral stenosis and
0167-5273 / 00 / $ – see front matter 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 00 )00235-7
274
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
is a major cause of morbidity and mortality [1–3]. The left atrium and left atrial appendage are well recognised sites of clot formation in these patients. Patients with a clot in the left atrium and left atrial appendage had a higher incidence of systemic embolization [4–6]. Various factors determine the individual risk for the development of left atrial clot and thromboembolization in patients with rheumatic mitral valve disease which include atrial fibrillation [4], left atrial size [7], duration of symptoms [8], older age [6,8] and severity of mitral stenosis [9,10]. However the presence of moderate to severe mitral regurgitation has a negative predictive value for the development of clot in the left atrium [4,11]. The presence of left atrial spontaneous echo contrast which is frequently observed by transesophageal echocardiography in patients with mitral valve disease has been associated with increased risk of clot formation in the left atrium and systemic thromboembolization [4,5,7,9,12,13]. However, previous studies evaluating patients with mitral stenosis were either small [10,13–16] or included a heterogeneous group of mitral valve disease patients consisting of the whole spectrum of severity of mitral stenosis [9,11,14,15] and associated mitral regurgitation [4,11,16]. It has been shown that transesophageal echocardiography is highly sensitive for the detection of left atrial clot, especially in the left atrial appendage [8,17]. The purpose of this study was to prospectively evaluate a large group of patients with severe rheumatic mitral stenosis, who were not receiving anticoagulation or antiplatelet therapy at the time of study, to determine if clinical and echocardiographic variables can predict the presence of clot in the left atrium or left atrial appendage and left atrial spontaneous echo contrast seen by transesophageal echocardiography.
2. Material and methods
2.1. Patients Two hundred consecutive patients with severe mitral stenosis (mitral valve area #1 cm 2 ) being evaluated for percutaneous transvenous mitral commissurotomy, were studied in a tertiary-care referral centre in north India. All clinical and echocardiog-
raphic data were collected prospectively. Patients with more than mild mitral regurgitation, significant aortic valve disease on transthoracic echocardiography, previous closed mitral valvulotomy and on anticoagulation or antiplatelet therapy were excluded from the study.
2.2. Transthoracic echocardiography The transthoracic studies were done by a standard technique using a Hewlett-Packard Sonos 1500 machine with a 3.5-MHz transducer. M-mode measurements were taken according to the recommendations of the American Society of Echocardiography [18]. The mitral valve area was measured by continuous wave Doppler using the pressure half time method. The mean transmitral diastolic pressure gradient was estimated from the maximal transmitral flow velocity using a modified Bernoulli equation. Left atrial diameter was taken in the parasternal long axis view in M-mode at end systole. Left atrial end systolic area on 2D-echocardiography was taken in two views (parasternal long axis view and apical four chamber view) and these two values were averaged. The measurements were made in three beats in patients with normal sinus rhythm and in five beats in atrial fibrillation and the mean was taken for analysis. Mitral regurgitation was graded by colour Doppler echocardiography as recommended by Helmcke et al. [19].
2.3. Transesophageal echocardiography Transesophageal echocardiography was performed after transthoracic echocardiography on the same day in all cases, as described by us previously [8]. A 5-MHz transducer with capacity for pulsed Doppler (ATL biplanar probe for the first 100 patients and a Hewlett-Packard omniplane probe for the next 100 patients) was used. All patients received local pharyngeal anesthesia (1% lidocaine spray) and intravenous diazepam, 2–5 mg. During echocardiography a one-lead electrocardiogram was recorded continuously. The transesophageal probe was introduced with the patient lying supine in the left lateral position. As the probe was advanced into the esophagus, the left atrium was scanned in the short axis view. With a slight counterclockwise rotation of the
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
275
probe at the level of the aortic valve, the left atrial appendage was seen as an outward pouch in the anterolateral region. A left atrial clot was diagnosed by the presence of a clearly defined echogenic intracavitary mass with an echo texture different from that of the underlying endocardium and not due to the pectinate muscle as described by us previously [8]. Left atrial spontaneous echo contrast was diagnosed by the presence of dynamic smoke-like echoes in the left atrial cavity and left atrial appendage with a characteristic swirling motion distinct from a white noise artefact after properly adjusting the gain setting.
Table 1 Comparison of the patients with and without clot a
2.4. Statistical analysis Data are presented as mean6S.D. MEDLINE database as well as relevant reference lists were searched for similar studies. For comparison, unpaired Student’s t-test was used for continuous variables and x 2 -test for categorical variables, and values were considered significant when the P value was ,0.05. Univariate, multiple logistic regression analysis and discriminant function analysis were done for all the variables to determine the factors that independently predict and discriminate the presence of clot and spontaneous echo contrast.
3. Results The mean age of the patients was 29.669.6 years (range 13–64 years). There were 112 males and 88 females. Mean duration of symptoms was 32.0627.4 months (range, 1–180 months). Sixty percent of the patients were in New York Heart Association functional class II and the remaining were in functional classes III (37%) and IV (3%). Twenty-seven patients had mild or less than mild mitral regurgitation and the remaining had no mitral regurgitation. A total of 69.5% of the patients were in normal sinus rhythm, whereas 30.5% were in atrial fibrillation. Clot was present in 26% of the cases and left atrial spontaneous echo contrast was present in 53.5% of the cases. The location of clot was the left atrial cavity in 17%, the left atrial appendage in 59% and the junction of the left atrial appendage with the left atrial cavity in 24%. Twelve percent of patients had a history of thromboembolic episodes.
Variable
With clot (n552)
Without clot (n5148)
P value
Age (years) Duration of symptoms (months) LVEF (%) MVA (cm 2 ) MDG (mmHg) LAD (mm) LA area (cm 2 )
33.4611.4 41.4636.0
28.268.5 28.8622.9
,0.01 ,0.01
68.669.4 0.7860.13 14.265.1 53.968.3 41.0612.7
71.669.2 0.8160.15 15564.9 47.667.4 29.967.4
NS NS NS ,0.0001 ,0.0001
Rhythm: NSR AF
16 (30.8%) 36 (69.2%)
123 (83.1%) 25 (16.9%)
,0.00001
SEC: Present Absent
40 (76.9%) 12 (23.1%)
67 (45.3%) 81 (54.7%)
,0.00001
a LVEF, left ventricle ejection fraction; MVA, mitral valve area; MDG, mean diastolic gradient; LAD, left atrial diameter; LA, left atrium; NSR, normal sinus rhythm; AF, atrial fibrillation; SEC, spontaneous echo contrast.
3.1. Comparison of the patients with and without clot Table 1 shows clinical and echocardiographic variables in these two groups. The patients with clot were older (33.4611.4 vs. 28.268.5 years, P,0.01) and had longer duration of symptoms (41.4636.0 vs. 28.8622.9 months, P,0.01) as compared to the patients without clot. The incidence of atrial fibrillation (69.2%) and left atrial spontaneous echo contrast (76.9%) were significantly higher in patients with clot (P,0.0001 and ,0.0001, respectively). The size of the left atrium, as measured by M-mode in parasternal long axis view, i.e., left atrial diameter (53.968.3 vs. 47.667.4 mm, P,0.0001) and left atrial area (41.0612.7 vs. 29.967.4 cm 2 , P,0.0001) were larger in patients with clot as compared to those without clot. The mitral valve area, left ventricular ejection fraction and mean diastolic gradient were not significantly different between the two groups. Past history of embolization did not correlate with the presence or absence of clot.
3.1.1. Multiple regression analysis On multiple logistic regression analysis only atrial fibrillation (odds ratio 5.502, 95% CI 2.162–14.00, and P,0.001) and left atrial area (odds ratio 1.078,
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
276
95% CI 1.027–1.13 and P,0.01) were found to be independent predictors for the presence of clot.
3.1.2. Discriminant function analysis On discriminant function analysis two variables, i.e., left atrial area and atrial fibrillation, turned out to be the most important variables to discriminate the two groups. The derived discriminate equation is D 5 2 (2.42) 1 (0.06 3 left atrial area) 1 (1.58 3 1 if atrial fibrillation present and 0 if atrial fibrillation absent) where if D .0, the predictive group was with clot and if less than 0 then the predicted group was without clot with 81.6% accuracy. So the upper limit of not having clot for left atrial area was 15 cm 2 in patients with atrial fibrillation and 40 cm 2 in patients with normal sinus rhythm with an accuracy of 82%.
3.1.3. Patients with normal sinus rhythm In the subgroup of the patients with normal sinus rhythm (n5139); longer duration of symptoms (34.8634.5 vs. 27.9622.2 months, P,0.05), larger left atrium (left atrial area, 34.1610.2 vs. 28.666.2 cm 2 , P,0.01 and left atrial diameter: 50.966.4 vs. 46.466.2 mm, P,0.05) and the presence of spontaneous echo contrast (P,0.05) correlated significantly with the presence of clot (Table 2).
Table 2 Comparison of patients with normal sinus rhythm (n5139) with or without clot a Variable
With clot n516
Without clot n5123
P value
Age (years) Duration of symptoms (months): LVEF % MVA (cm 2 ) MDG (mmHg) LAD (mm) LA area (cm 2 ) SEC: Present Absent
27.0610.6 34.8634.5
27.668.4 27.9622.2
NS ,0.05
69.567.7 0.7560.09 17.465.1 50.966.4 34.1610.2
72.168.7 0.8160.15 15.665.0 46.466.2 28.666.2
NS NS NS ,0.05 0.01
9(56.3%) 7 (43.7%)
46(37.4%) 77(62.6%)
,0.05
a
For abbreviations, refer to Table 1.
3.2. Comparison of patients with sinus rhythm and atrial fibrillation When patients with atrial fibrillation and normal sinus rhythm were compared, patients with atrial fibrillation were significantly older (34.2610.1 vs. 27.568.6 years, P,0.001), had longer duration of symptoms (39.7633.1 vs. 28.7623.9 months, P, 0.01) and larger left atrium (left atrial area 41.9611.9 vs. 29.367 cm 2 , P,0.0001 and left atrial diameter 54.669.1 vs. 46.966.4 mm, P,0.0001). Left atrial spontaneous echo contrast was present significantly more in patients with atrial fibrillation. Eighty-five percent (n552) of patients in atrial fibrillation showed spontaneous echo contrast as compared to 39.6% (n555) of patients with normal sinus rhythm (P,0.0001). The patients with atrial fibrillation had higher incidence of clot as compared to the patients with normal sinus rhythm (59 vs. 11.5%, P, 0.00001). There was no difference in the mitral valve area, mean diastolic gradient and left ventricular ejection fraction between patients with atrial fibrillation and sinus rhythm.
3.3. Comparison of the patients with and without spontaneous echo contrast ( Table 3) When patients with and without spontaneous echo
Table 3 Comparison of the patients with and with out spontaneous echo contrast (SEC)a Variable
With SEC n5107
Without SEC n593
P value
Age (years) Duration of symptom (months) MVA cm 2 LVEF % MDG (mmHg) LAD (mm) LA area (cm 2 )
31.0610.4 32.6625.4
27.868.3 31.3629.6
0.01 NS
0.7760.14 69.567.9 15.164.8 52.268.3 37.6611.2
0.8460.13 72.469.2 15.265.1 45.966.5 28.166.7
,0.01 NS NS ,0.0001 ,0.0001
Rhythm: NSR AF
55 (51.4%) 52 (48.6%)
84 (90.3%) 9 (9.7%)
,0.0001
Clot: Present Absent
40 (37.4%) 67 (62.6%)
12 (12.9%) 81 (87.1%)
,0.0001
a
For abbreviations, refer to Table 1.
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
contrast were compared, patients with spontaneous echo contrast were older (31.0610.4 vs. 27.868.3 years, P,0.01), had smaller mitral valve area (0.7760.14 vs. 0.8460.13 cm 2 , P,0.01), larger left atrium (left atrial area 37.6611.2 vs. 28.166.7 cm 2 , P,0.0001 and left atrial diameter 52.268.3 vs. 45.966.5 mm, P,0.0001) and were more frequently in atrial fibrillation (48.6 vs. 9.7%, P,0.0001). The patients with spontaneous echo contrast had higher incidence of clot in the left atrium (37.4 vs. 12.9%, P,0.0001). However, there was no significant difference with respect to the duration of symptoms, left ventricular ejection fraction and mean diastolic pressure between these groups.
3.3.1. Multiple regression analysis After multiple logistic regression analysis, atrial fibrillation (odds ratio 7.67, 95% CI 2.56–22.98 and P,0.001) and left atrial area (odds ratio 1.08, 95% CI 1.02–1.14 and P,0.01) were found to be independently related to the presence of spontaneous echo contrast. 3.3.2. Discriminate function analysis On discriminate function analysis, left atrial area, atrial fibrillation and mitral valve area were the most important variables to discriminate between the groups. The discriminate equation derived is D 5 1.19 1 (0.044 3 left atrial area) 1 (23.81 3 mitral valve area) 1 (1.513 3 1 if atrial fibrillation present and 0 if atrial fibrillation absent), where if D .0 the predictive group was with spontaneous echo contrast and if less than 0, the predictive group was without spontaneous echo contrast, with an accuracy of 75%.
277
neous echo contrast. In this subgroup of patients; age, duration of symptoms, mean diastolic gradient, and left ventricle ejection fraction were not different between patients with and without spontaneous echo contrast.
4. Discussion In this prospective study of patients with rheumatic severe mitral stenosis we evaluated several clinical and echocardiographic variables which correlated with the presence of left atrial and left atrial appendage clot and spontaneous echo contrast. Left atrial clot formation is a major complication in patients with rheumatic mitral valve disease [1–3] and these patients with clot had higher chance of systemic embolization [4–6]. Besides the presence of clot in the left atrium, spontaneous echo contrast is also found to be an important independent factor of systemic embolization in patients with nonrheumatic atrial fibrillation and rheumatic heart disease [4,5,9,12,20,21]. Mitral stenosis is more commonly associated with clot formation in the left atrium as compared to mitral regurgitation [11] and its increasing severity was found to be significantly correlated with increased incidence of clot in the left atrium and left atrial appendage [10]. In the present prospective study we evaluated a large number of patients only with severe mitral stenosis, without significant mitral regurgitation or other valvular lesion. Our patient population was younger and the majority were in normal sinus rhythm as compared to previous studies reported. The incidence of left atrial clot in our study was 26%. In various earlier studies the incidence of left atrial clot in patients with mitral stenosis varied from 5 to 28% [4,7,8,10,11,14,15,22,23].
4.1. Clinical predictors of left atrial clot 3.3.3. Patients with normal sinus rhythm In the subgroup of patients with normal sinus rhythm, left atrial size was significantly larger in patients with spontaneous echo contrast as compared to patients without echo contrast (left atrial area 32.266.7 vs. 27.666.6 cm 2 , P,0.005; and left atrial diameter 49.366.4 vs. 45.465.9 mm, P,0.001) and mitral valve area was significantly less (0.7560.15 vs. 0.8460.14 cm 2 , P,0.01) in patients with sponta-
We found that older patients, longer duration of symptoms, and presence of atrial fibrillation were positively correlated with the presence of clot. On multiple regression analysis and discriminate function analysis, atrial fibrillation along with left atrial size was found to be an independent predictor of clot in the left atrium. A total of 69.2% of the patients with clot were in atrial fibrillation. Earlier studies have
278
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
also observed atrial fibrillation as an independent factor associated with clot in the left atrium [4,11,15]. Ozkan [11], in 169 patients with rheumatic mitral valve disease (52 with predominant mitral stenosis), also found atrial fibrillation to be independently related to the formation of clot in all subgroups of patients with mitral valve disease (96.4 vs. 51.8%, P,0.005). Vigna et al. [15], in a retrospective study of 59 patients, showed atrial fibrillation to be present in 91.6% of the cases with clot and on multivariate analysis it was independently associated with the presence of clot (P,0.00001). Even in the absence of atrial fibrillation, when a subgroup of patients with normal sinus rhythm were analysed, larger left atrial size and presence of spontaneous echo contrast were positively correlated with left atrial clot. We found that in patients with normal sinus rhythm, larger left atrial size and smaller mitral valve area were significantly correlated with the presence of spontaneous echo contrast, but not with age, duration of symptoms, mean diastolic gradient and left ventricular ejection fraction.
4.2. Echocardiographic predictors of left atrial clot Transesophageal echocardiography has been used widely for the detection of left atrial clot and has been shown to be superior to transthoracic echocardiography. In the present study all patients underwent a routine transesophageal echocardiography following transthoracic echocardiography. We found that left atrial size and spontaneous echo contrast were positively correlated with the presence of left atrial clot, while there was no significant difference in mitral valve area and mean diastolic gradient between patients with and without clot. We measured the left atrial size by both M-Mode and 2-D echocardiography and both correlated positively with the presence of clot. However, on multivariate analysis and discriminant function analysis, the left atrial area was analysed (left atrial area is more accurate for estimation of left atrial size as compared to left atrial diameter alone because of uneven dilatation of the left atrium) and has correlated significantly with left atrial clot, which was also observed in earlier studies [7,15]. Conradie et al. [7] showed that left atrial size .4.8 cm was an independent factor for the clot formation in the left atrium. Small mitral valve area has been shown to be associated with increased
incidence of left atrial clot in several earlier studies [4,12]. However we did not find any significant difference between mitral valve area in patients with and without left atrial clot, although patients with left atrial spontaneous echo contrast had significantly smaller mitral valve area. This discrepancy could be attributed to the fact that all our patients had severe mitral stenosis without associated regurgitation. While earlier studies included patients with mitral stenosis of all grades of severity, we studied a homogeneous group of patients with severe mitral stenosis who as such were already at high risk for the formation of the clot in the left atrium and left atrial appendage.
4.3. Spontaneous echo contrast The incidence of spontaneous echo contrast in mitral stenosis varies from 21 to 67% [4,11,12,15]. Previous reports have shown that in majority [4,12,13,24], or all [9], left atrial clot and systemic thromboembolization were associated with spontaneous echo contrast in left atrium on transesophageal echocardiography. Spontaneous echo contrast was also found to be an important predictor of systemic embolization, independent of the presence of clot in the left atrium [9,20,21]. Fatkin [21], in a group of 140 patients who underwent transesophageal echocardiography, found that the presence of spontaneous echo contrast, not the presence of clot in left atrium, correlated significantly with previous history of thromboembolization; however, Acarturk et al. [5] found that the presence of both (spontaneous echo contrast and left atrial clot) correlated significantly with thromboembolism. In our study, left atrial spontaneous echo contrast was present in 53.5% of cases and was significantly correlated with older age, large left atrium and atrial fibrillation. Although on univariate analysis, spontaneous echo contrast was found to be significantly correlated with the presence of clot, when all significant variables were included in multivariate analysis this variable was not significant, suggesting a more powerful effect of other variables, such as atrial fibrillation and left atrial area which were found to be independent predictors for left atrial clot. Similar results were also shown by Castello et al. [13]. in a small series of 17 patients with mitral stenosis. Though patients with left atrial clot had signifi-
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
cantly longer duration of symptoms than patients without clot, the duration of symptoms did not correlate with the presence of left atrial spontaneous echo contrast. This may mean that stasis of blood in the left atrium occurs relatively early in patients with mitral stenosis, thus creating a milieu for clot formation. Stasis of blood has been widely thought to be responsible for the appearance of left atrial spontaneous echo contrast [9,12,13,20]. However, whether the presence of left atrial spontaneous echo contrast in the absence of clot should be an indication for initiating anticoagulation is debatable.
5. Conclusion We concluded that in patients with severe mitral stenosis, atrial fibrillation and left atrial size were found to be independent predictors of clot and spontaneous echo contrast formation in left atrium and left atrial appendage. In a subgroup of the patients with normal sinus rhythm, patients with larger left atrium (.40 cm 2 ) and spontaneous echo contrast have a higher risk of clot formation in the left atrium and left atrial appendage. We postulated that in patients with severe mitral stenosis, the presence of atrial fibrillation, and a subgroup of patients with normal sinus rhythm and left atrial area $40 cm 2 with spontaneous echo contrast, had a significantly higher chance of having a clot and might benefit from prophylactic anticoagulation.
References [1] Wolf PA, Dawber TR, Thomas E, Kannel WB. Epidemiological assessment of chronic atrial fibrillation and the risk of stroke: the Framingham study. Neurology 1978;28:973–7. [2] Rowe JC, Bland EF, Sprague HB, White PD. The course of mitral stenosis without surgery: ten years and twenty years perspectives. Ann Intern Med 1960;52:741–9. [3] Deverall PB, Olley PM, Smith DR, Watson DA, Whitaker W. Incidence of systemic embolism before and after mitral valvotomy. Thorax 1968;23:530–6. [4] Hwang JJ, Kuan P, Lin SC et al. Reappraisal by transesophageal echocardiography of the significance of left atrial clot in prediction of systemic embolization in rheumatic mitral valve disease. Am J Cardiol 1992;70:769–73. [5] Acarturk E, Usal A, Demir M, Akgul F, Ozeren A. Thromboembolic risk in patients with mitral stenosis. Jpn Heart J 1997;38:669–75. [6] Chiag CW, Lo SK, Ko YS, Cheng NJ, Lin PJ, Chang CH. Predictor of systemic embolization in patients with mitral stenosis. A prospective study. Ann Intern Med 1998;128:885–9.
279
[7] Conradie C, Schall R, Marx JD. Left atrial size — a risk factor for left atrial thrombi in mitral stenosis. Clin Cardiol 1995;18:518–20. [8] Goswami KC, Narang R, Bahl V K, Talwar KK, Manchanda SC. Comparative evaluation of transthoracic and transesophageal echocardiography in the detection of left atrial clot before percutaneous transvenous mitral commissurotomy: do all patients need transesophageal examination? Int J Cardiol 1997;62:237–49. [9] Daniel WG, Nellessen U, Schroder E et al. Left atrial contrast in mitral valve disease: an indicator for an increased thromboembolic risk. J Am Coll Cardiol 1988;11:1204–11. [10] Kronzon I, Tunick PA, Glassman E, Slater J, Schwinger M, Freedberg RS. Transesophageal echocardiography to detect atrial clot in candidates for percutaneous transseptal mitral balloon valvuloplasty. J Am Coll Cardiol 1990;16:1320–2. [11] Ozkan M, Kaymaz C, Kirma C et al. Predictors of left atrial clot and spontaneous echo contrast in rheumatic valve disease before and after mitral valve replacement. Am J Cardiol 1998;82:1066–70. [12] Beppu S, Nimura Y, Sakakibara H, Nagata S, Park YD, Izumi S. Smoke like echo in the left atrial cavity in mitral valve disease, it’s features and significance. J Am Coll Cardiol 1985;6:744–9. [13] Castello R, Pearson AC, Labovitz AJ. Prevalence and clinical implication of atrial spontaneous contrast in patients undergoing transesophageal echocardiography. Am J Cardiol 1990;65:1149–53. [14] Bernstein NE, Demopoulos LA, Tunick PA, Rosenzweig BP, Kronzon I. Correlation of spontaneous echo contrast in patients with mitral stenosis and normal sinus rhythm. Am Heart J 1994;128:287– 92. [15] Vigna C, de Rito V, Criconia GM et al. Left atrial clot and spontaneous echo contrast in nonanticoagulated mitral stenosis. A transesophageal echocardiography study. Chest 1993;103:348–52. [16] Li YH, Hwang JJ, Ko YL, Lin JL, Tseng YZ, Kuan P, Lein WP. Left atrial spontaneous echo contrast in patients with rheumatic mitral valve disease in sinus rhythm. Implication of an altered left atrial appendage function in its formation. Chest 1995;108(l):99–103. [17] Fatkin D, Scalia G, Jacobs N et al. Accuracy of biplaner transesophageal echocardiography for the detection of left atrial clot. Am J Cardiol 1996;77:321–3. [18] Sahn DJ, DeMaria A, Kisslo J et al. Recommendation regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurement. Circulation 1978;58:1072–83. [19] Helmcke F, Nanda N, Hsiung M, Soto B, Adet CK, Goyal RG, Gate wood RP. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175–83. [20] Black IW, Hopkins AP, Lee LL, Walsh WF, Jacobson BM. Left atrial spontaneous echo contrast: a clinical and echocardiographic analysis. J Am Coll Cardiol 1991;18:398–404. [21] Fatkin D, Kelly RP, Feneley MP. Relation between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994;23:961–9. [22] Cormier B, Vahanian A, Lung B et al. Influence of percutaneous mitral commissurotomy on left atrial spontaneous contrast of mitral stenosis. Am J Cardiol 1993;71:842–7. [23] Rao AS, Murthy RS, Naidu PB, Raghu K, Anjaneyulu AV, Raju PK, Naik SR. Transesophageal echocardiography for the detection of left atrial clot. Indian Heart J 1994;46:37–40. [24] Chiang CW, Lin FC, Fang BR, Kuo CT, Lee YS, Chang CH. Sand drift, echoes and clot formation in left atrium. Am Heart J 1988;115:908–11.
Clinical and echocardiographic predictors of left atrial clot and spontaneous echo contrast in patients with severe rheumatic mitral stenosis: a prospective study in 200 patients by transesophageal echocardiography Kewal C. Goswami MD (Med), DM (Card)*, Rakesh Yadav MD (Med), DM (Card), M. Bhaskara Rao MD (Med), DM (Card), Vinay K. Bahl MD (Med), DM(Card), Kewal K. Talwar MD (Med), DM (Card), Subhash C. Manchanda MD (Med), DM (Card) Department of Cardiology, Cardiothoracic Sciences Centre, All India Institute of Medical Sciences, Ansari Nagar, New Delhi-110029, India Received 16 August 1999; received in revised form 11 January 2000; accepted 28 January 2000
Abstract The objective of this study was to prospectively investigate various clinical and echocardiographic variables to predict the left atrial and left atrial appendage clot and spontaneous echo contrast in patients with severe rheumatic mitral stenosis. We studied 200 consecutive patients (112 males and 88 females; mean age 29.669.6 years). Left atrial clot and spontaneous echo contrast were present in 26 and 53.5% of cases, respectively. There were no significant differences in the mitral valve area, mean transmitral diastolic gradient and left ventricular ejection fraction between patients with and without clot. Patients with clot were older (34.4611.4 vs. 28.268.5 years, P,0.001), had longer duration of symptoms (41.4636.0 vs. 28.8622.9 months, P,0.001), more frequent atrial fibrillation and spontaneous echo contrast (69.2 vs. 16.9%, P,0.00001 and 76.9 vs. 45.3%, P,0.00001, respectively) and larger left atrial area and diameter (41.0612.7 vs. 29.967.4 cm 2 , P,0.00001 and 53.968.3 vs. 47.667.4 mm, P,0.0001, respectively) as compared to patients without clot. Similarly patients with spontaneous echo contrast were older (31610.4 vs. 27.868.3 years, P,0.01), had more frequent atrial fibrillation (48.6 vs. 9.7%, P,0.0001), left atrial clot (37.4 vs. 12.9%, P,0.0001), larger left atrial area and diameter (37.6611.2 vs. 28.166.7 cm 2 , P,0.00001 and 52.268.3 vs. 45.966.5 mm, P,0.00001, respectively) and smaller mitral valve area (0.7760.14 vs. 0.8460.13 cm 2 , P,0.01) as compared to patients without spontaneous echo contrast. There were no significant differences in the mean transmitral diastolic gradient and left ventricular ejection fraction. On multiple regression and discriminant function analysis, atrial fibrillation and left atrial area were independent predictors of left atrial clot formation. In a subgroup of patients with sinus rhythm, larger left atrial area and presence of spontaneous echo contrast were significantly associated with the presence of clot in left atrium and appendage. We conclude that in patients with severe mitral stenosis, the presence of atrial fibrillation and in the subgroup of the patients with sinus rhythm the presence of large left atrium ($40 cm 2 ) and spontaneous echo contrast were associated with higher risk of clot formation in the left atrium and might be benefited by prophylactic anticoagulation. 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mitral stenosis; Left atrial clot; Left atrial spontaneous echo contrast; Transesophageal echocardiography
1. Introduction *Corresponding author. Tel.: 191-11-659-3218; fax: 191-11-6862663.
The incidence of thromboembolic complications is higher in patients with rheumatic mitral stenosis and
0167-5273 / 00 / $ – see front matter 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 00 )00235-7
274
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
is a major cause of morbidity and mortality [1–3]. The left atrium and left atrial appendage are well recognised sites of clot formation in these patients. Patients with a clot in the left atrium and left atrial appendage had a higher incidence of systemic embolization [4–6]. Various factors determine the individual risk for the development of left atrial clot and thromboembolization in patients with rheumatic mitral valve disease which include atrial fibrillation [4], left atrial size [7], duration of symptoms [8], older age [6,8] and severity of mitral stenosis [9,10]. However the presence of moderate to severe mitral regurgitation has a negative predictive value for the development of clot in the left atrium [4,11]. The presence of left atrial spontaneous echo contrast which is frequently observed by transesophageal echocardiography in patients with mitral valve disease has been associated with increased risk of clot formation in the left atrium and systemic thromboembolization [4,5,7,9,12,13]. However, previous studies evaluating patients with mitral stenosis were either small [10,13–16] or included a heterogeneous group of mitral valve disease patients consisting of the whole spectrum of severity of mitral stenosis [9,11,14,15] and associated mitral regurgitation [4,11,16]. It has been shown that transesophageal echocardiography is highly sensitive for the detection of left atrial clot, especially in the left atrial appendage [8,17]. The purpose of this study was to prospectively evaluate a large group of patients with severe rheumatic mitral stenosis, who were not receiving anticoagulation or antiplatelet therapy at the time of study, to determine if clinical and echocardiographic variables can predict the presence of clot in the left atrium or left atrial appendage and left atrial spontaneous echo contrast seen by transesophageal echocardiography.
2. Material and methods
2.1. Patients Two hundred consecutive patients with severe mitral stenosis (mitral valve area #1 cm 2 ) being evaluated for percutaneous transvenous mitral commissurotomy, were studied in a tertiary-care referral centre in north India. All clinical and echocardiog-
raphic data were collected prospectively. Patients with more than mild mitral regurgitation, significant aortic valve disease on transthoracic echocardiography, previous closed mitral valvulotomy and on anticoagulation or antiplatelet therapy were excluded from the study.
2.2. Transthoracic echocardiography The transthoracic studies were done by a standard technique using a Hewlett-Packard Sonos 1500 machine with a 3.5-MHz transducer. M-mode measurements were taken according to the recommendations of the American Society of Echocardiography [18]. The mitral valve area was measured by continuous wave Doppler using the pressure half time method. The mean transmitral diastolic pressure gradient was estimated from the maximal transmitral flow velocity using a modified Bernoulli equation. Left atrial diameter was taken in the parasternal long axis view in M-mode at end systole. Left atrial end systolic area on 2D-echocardiography was taken in two views (parasternal long axis view and apical four chamber view) and these two values were averaged. The measurements were made in three beats in patients with normal sinus rhythm and in five beats in atrial fibrillation and the mean was taken for analysis. Mitral regurgitation was graded by colour Doppler echocardiography as recommended by Helmcke et al. [19].
2.3. Transesophageal echocardiography Transesophageal echocardiography was performed after transthoracic echocardiography on the same day in all cases, as described by us previously [8]. A 5-MHz transducer with capacity for pulsed Doppler (ATL biplanar probe for the first 100 patients and a Hewlett-Packard omniplane probe for the next 100 patients) was used. All patients received local pharyngeal anesthesia (1% lidocaine spray) and intravenous diazepam, 2–5 mg. During echocardiography a one-lead electrocardiogram was recorded continuously. The transesophageal probe was introduced with the patient lying supine in the left lateral position. As the probe was advanced into the esophagus, the left atrium was scanned in the short axis view. With a slight counterclockwise rotation of the
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
275
probe at the level of the aortic valve, the left atrial appendage was seen as an outward pouch in the anterolateral region. A left atrial clot was diagnosed by the presence of a clearly defined echogenic intracavitary mass with an echo texture different from that of the underlying endocardium and not due to the pectinate muscle as described by us previously [8]. Left atrial spontaneous echo contrast was diagnosed by the presence of dynamic smoke-like echoes in the left atrial cavity and left atrial appendage with a characteristic swirling motion distinct from a white noise artefact after properly adjusting the gain setting.
Table 1 Comparison of the patients with and without clot a
2.4. Statistical analysis Data are presented as mean6S.D. MEDLINE database as well as relevant reference lists were searched for similar studies. For comparison, unpaired Student’s t-test was used for continuous variables and x 2 -test for categorical variables, and values were considered significant when the P value was ,0.05. Univariate, multiple logistic regression analysis and discriminant function analysis were done for all the variables to determine the factors that independently predict and discriminate the presence of clot and spontaneous echo contrast.
3. Results The mean age of the patients was 29.669.6 years (range 13–64 years). There were 112 males and 88 females. Mean duration of symptoms was 32.0627.4 months (range, 1–180 months). Sixty percent of the patients were in New York Heart Association functional class II and the remaining were in functional classes III (37%) and IV (3%). Twenty-seven patients had mild or less than mild mitral regurgitation and the remaining had no mitral regurgitation. A total of 69.5% of the patients were in normal sinus rhythm, whereas 30.5% were in atrial fibrillation. Clot was present in 26% of the cases and left atrial spontaneous echo contrast was present in 53.5% of the cases. The location of clot was the left atrial cavity in 17%, the left atrial appendage in 59% and the junction of the left atrial appendage with the left atrial cavity in 24%. Twelve percent of patients had a history of thromboembolic episodes.
Variable
With clot (n552)
Without clot (n5148)
P value
Age (years) Duration of symptoms (months) LVEF (%) MVA (cm 2 ) MDG (mmHg) LAD (mm) LA area (cm 2 )
33.4611.4 41.4636.0
28.268.5 28.8622.9
,0.01 ,0.01
68.669.4 0.7860.13 14.265.1 53.968.3 41.0612.7
71.669.2 0.8160.15 15564.9 47.667.4 29.967.4
NS NS NS ,0.0001 ,0.0001
Rhythm: NSR AF
16 (30.8%) 36 (69.2%)
123 (83.1%) 25 (16.9%)
,0.00001
SEC: Present Absent
40 (76.9%) 12 (23.1%)
67 (45.3%) 81 (54.7%)
,0.00001
a LVEF, left ventricle ejection fraction; MVA, mitral valve area; MDG, mean diastolic gradient; LAD, left atrial diameter; LA, left atrium; NSR, normal sinus rhythm; AF, atrial fibrillation; SEC, spontaneous echo contrast.
3.1. Comparison of the patients with and without clot Table 1 shows clinical and echocardiographic variables in these two groups. The patients with clot were older (33.4611.4 vs. 28.268.5 years, P,0.01) and had longer duration of symptoms (41.4636.0 vs. 28.8622.9 months, P,0.01) as compared to the patients without clot. The incidence of atrial fibrillation (69.2%) and left atrial spontaneous echo contrast (76.9%) were significantly higher in patients with clot (P,0.0001 and ,0.0001, respectively). The size of the left atrium, as measured by M-mode in parasternal long axis view, i.e., left atrial diameter (53.968.3 vs. 47.667.4 mm, P,0.0001) and left atrial area (41.0612.7 vs. 29.967.4 cm 2 , P,0.0001) were larger in patients with clot as compared to those without clot. The mitral valve area, left ventricular ejection fraction and mean diastolic gradient were not significantly different between the two groups. Past history of embolization did not correlate with the presence or absence of clot.
3.1.1. Multiple regression analysis On multiple logistic regression analysis only atrial fibrillation (odds ratio 5.502, 95% CI 2.162–14.00, and P,0.001) and left atrial area (odds ratio 1.078,
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
276
95% CI 1.027–1.13 and P,0.01) were found to be independent predictors for the presence of clot.
3.1.2. Discriminant function analysis On discriminant function analysis two variables, i.e., left atrial area and atrial fibrillation, turned out to be the most important variables to discriminate the two groups. The derived discriminate equation is D 5 2 (2.42) 1 (0.06 3 left atrial area) 1 (1.58 3 1 if atrial fibrillation present and 0 if atrial fibrillation absent) where if D .0, the predictive group was with clot and if less than 0 then the predicted group was without clot with 81.6% accuracy. So the upper limit of not having clot for left atrial area was 15 cm 2 in patients with atrial fibrillation and 40 cm 2 in patients with normal sinus rhythm with an accuracy of 82%.
3.1.3. Patients with normal sinus rhythm In the subgroup of the patients with normal sinus rhythm (n5139); longer duration of symptoms (34.8634.5 vs. 27.9622.2 months, P,0.05), larger left atrium (left atrial area, 34.1610.2 vs. 28.666.2 cm 2 , P,0.01 and left atrial diameter: 50.966.4 vs. 46.466.2 mm, P,0.05) and the presence of spontaneous echo contrast (P,0.05) correlated significantly with the presence of clot (Table 2).
Table 2 Comparison of patients with normal sinus rhythm (n5139) with or without clot a Variable
With clot n516
Without clot n5123
P value
Age (years) Duration of symptoms (months): LVEF % MVA (cm 2 ) MDG (mmHg) LAD (mm) LA area (cm 2 ) SEC: Present Absent
27.0610.6 34.8634.5
27.668.4 27.9622.2
NS ,0.05
69.567.7 0.7560.09 17.465.1 50.966.4 34.1610.2
72.168.7 0.8160.15 15.665.0 46.466.2 28.666.2
NS NS NS ,0.05 0.01
9(56.3%) 7 (43.7%)
46(37.4%) 77(62.6%)
,0.05
a
For abbreviations, refer to Table 1.
3.2. Comparison of patients with sinus rhythm and atrial fibrillation When patients with atrial fibrillation and normal sinus rhythm were compared, patients with atrial fibrillation were significantly older (34.2610.1 vs. 27.568.6 years, P,0.001), had longer duration of symptoms (39.7633.1 vs. 28.7623.9 months, P, 0.01) and larger left atrium (left atrial area 41.9611.9 vs. 29.367 cm 2 , P,0.0001 and left atrial diameter 54.669.1 vs. 46.966.4 mm, P,0.0001). Left atrial spontaneous echo contrast was present significantly more in patients with atrial fibrillation. Eighty-five percent (n552) of patients in atrial fibrillation showed spontaneous echo contrast as compared to 39.6% (n555) of patients with normal sinus rhythm (P,0.0001). The patients with atrial fibrillation had higher incidence of clot as compared to the patients with normal sinus rhythm (59 vs. 11.5%, P, 0.00001). There was no difference in the mitral valve area, mean diastolic gradient and left ventricular ejection fraction between patients with atrial fibrillation and sinus rhythm.
3.3. Comparison of the patients with and without spontaneous echo contrast ( Table 3) When patients with and without spontaneous echo
Table 3 Comparison of the patients with and with out spontaneous echo contrast (SEC)a Variable
With SEC n5107
Without SEC n593
P value
Age (years) Duration of symptom (months) MVA cm 2 LVEF % MDG (mmHg) LAD (mm) LA area (cm 2 )
31.0610.4 32.6625.4
27.868.3 31.3629.6
0.01 NS
0.7760.14 69.567.9 15.164.8 52.268.3 37.6611.2
0.8460.13 72.469.2 15.265.1 45.966.5 28.166.7
,0.01 NS NS ,0.0001 ,0.0001
Rhythm: NSR AF
55 (51.4%) 52 (48.6%)
84 (90.3%) 9 (9.7%)
,0.0001
Clot: Present Absent
40 (37.4%) 67 (62.6%)
12 (12.9%) 81 (87.1%)
,0.0001
a
For abbreviations, refer to Table 1.
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
contrast were compared, patients with spontaneous echo contrast were older (31.0610.4 vs. 27.868.3 years, P,0.01), had smaller mitral valve area (0.7760.14 vs. 0.8460.13 cm 2 , P,0.01), larger left atrium (left atrial area 37.6611.2 vs. 28.166.7 cm 2 , P,0.0001 and left atrial diameter 52.268.3 vs. 45.966.5 mm, P,0.0001) and were more frequently in atrial fibrillation (48.6 vs. 9.7%, P,0.0001). The patients with spontaneous echo contrast had higher incidence of clot in the left atrium (37.4 vs. 12.9%, P,0.0001). However, there was no significant difference with respect to the duration of symptoms, left ventricular ejection fraction and mean diastolic pressure between these groups.
3.3.1. Multiple regression analysis After multiple logistic regression analysis, atrial fibrillation (odds ratio 7.67, 95% CI 2.56–22.98 and P,0.001) and left atrial area (odds ratio 1.08, 95% CI 1.02–1.14 and P,0.01) were found to be independently related to the presence of spontaneous echo contrast. 3.3.2. Discriminate function analysis On discriminate function analysis, left atrial area, atrial fibrillation and mitral valve area were the most important variables to discriminate between the groups. The discriminate equation derived is D 5 1.19 1 (0.044 3 left atrial area) 1 (23.81 3 mitral valve area) 1 (1.513 3 1 if atrial fibrillation present and 0 if atrial fibrillation absent), where if D .0 the predictive group was with spontaneous echo contrast and if less than 0, the predictive group was without spontaneous echo contrast, with an accuracy of 75%.
277
neous echo contrast. In this subgroup of patients; age, duration of symptoms, mean diastolic gradient, and left ventricle ejection fraction were not different between patients with and without spontaneous echo contrast.
4. Discussion In this prospective study of patients with rheumatic severe mitral stenosis we evaluated several clinical and echocardiographic variables which correlated with the presence of left atrial and left atrial appendage clot and spontaneous echo contrast. Left atrial clot formation is a major complication in patients with rheumatic mitral valve disease [1–3] and these patients with clot had higher chance of systemic embolization [4–6]. Besides the presence of clot in the left atrium, spontaneous echo contrast is also found to be an important independent factor of systemic embolization in patients with nonrheumatic atrial fibrillation and rheumatic heart disease [4,5,9,12,20,21]. Mitral stenosis is more commonly associated with clot formation in the left atrium as compared to mitral regurgitation [11] and its increasing severity was found to be significantly correlated with increased incidence of clot in the left atrium and left atrial appendage [10]. In the present prospective study we evaluated a large number of patients only with severe mitral stenosis, without significant mitral regurgitation or other valvular lesion. Our patient population was younger and the majority were in normal sinus rhythm as compared to previous studies reported. The incidence of left atrial clot in our study was 26%. In various earlier studies the incidence of left atrial clot in patients with mitral stenosis varied from 5 to 28% [4,7,8,10,11,14,15,22,23].
4.1. Clinical predictors of left atrial clot 3.3.3. Patients with normal sinus rhythm In the subgroup of patients with normal sinus rhythm, left atrial size was significantly larger in patients with spontaneous echo contrast as compared to patients without echo contrast (left atrial area 32.266.7 vs. 27.666.6 cm 2 , P,0.005; and left atrial diameter 49.366.4 vs. 45.465.9 mm, P,0.001) and mitral valve area was significantly less (0.7560.15 vs. 0.8460.14 cm 2 , P,0.01) in patients with sponta-
We found that older patients, longer duration of symptoms, and presence of atrial fibrillation were positively correlated with the presence of clot. On multiple regression analysis and discriminate function analysis, atrial fibrillation along with left atrial size was found to be an independent predictor of clot in the left atrium. A total of 69.2% of the patients with clot were in atrial fibrillation. Earlier studies have
278
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
also observed atrial fibrillation as an independent factor associated with clot in the left atrium [4,11,15]. Ozkan [11], in 169 patients with rheumatic mitral valve disease (52 with predominant mitral stenosis), also found atrial fibrillation to be independently related to the formation of clot in all subgroups of patients with mitral valve disease (96.4 vs. 51.8%, P,0.005). Vigna et al. [15], in a retrospective study of 59 patients, showed atrial fibrillation to be present in 91.6% of the cases with clot and on multivariate analysis it was independently associated with the presence of clot (P,0.00001). Even in the absence of atrial fibrillation, when a subgroup of patients with normal sinus rhythm were analysed, larger left atrial size and presence of spontaneous echo contrast were positively correlated with left atrial clot. We found that in patients with normal sinus rhythm, larger left atrial size and smaller mitral valve area were significantly correlated with the presence of spontaneous echo contrast, but not with age, duration of symptoms, mean diastolic gradient and left ventricular ejection fraction.
4.2. Echocardiographic predictors of left atrial clot Transesophageal echocardiography has been used widely for the detection of left atrial clot and has been shown to be superior to transthoracic echocardiography. In the present study all patients underwent a routine transesophageal echocardiography following transthoracic echocardiography. We found that left atrial size and spontaneous echo contrast were positively correlated with the presence of left atrial clot, while there was no significant difference in mitral valve area and mean diastolic gradient between patients with and without clot. We measured the left atrial size by both M-Mode and 2-D echocardiography and both correlated positively with the presence of clot. However, on multivariate analysis and discriminant function analysis, the left atrial area was analysed (left atrial area is more accurate for estimation of left atrial size as compared to left atrial diameter alone because of uneven dilatation of the left atrium) and has correlated significantly with left atrial clot, which was also observed in earlier studies [7,15]. Conradie et al. [7] showed that left atrial size .4.8 cm was an independent factor for the clot formation in the left atrium. Small mitral valve area has been shown to be associated with increased
incidence of left atrial clot in several earlier studies [4,12]. However we did not find any significant difference between mitral valve area in patients with and without left atrial clot, although patients with left atrial spontaneous echo contrast had significantly smaller mitral valve area. This discrepancy could be attributed to the fact that all our patients had severe mitral stenosis without associated regurgitation. While earlier studies included patients with mitral stenosis of all grades of severity, we studied a homogeneous group of patients with severe mitral stenosis who as such were already at high risk for the formation of the clot in the left atrium and left atrial appendage.
4.3. Spontaneous echo contrast The incidence of spontaneous echo contrast in mitral stenosis varies from 21 to 67% [4,11,12,15]. Previous reports have shown that in majority [4,12,13,24], or all [9], left atrial clot and systemic thromboembolization were associated with spontaneous echo contrast in left atrium on transesophageal echocardiography. Spontaneous echo contrast was also found to be an important predictor of systemic embolization, independent of the presence of clot in the left atrium [9,20,21]. Fatkin [21], in a group of 140 patients who underwent transesophageal echocardiography, found that the presence of spontaneous echo contrast, not the presence of clot in left atrium, correlated significantly with previous history of thromboembolization; however, Acarturk et al. [5] found that the presence of both (spontaneous echo contrast and left atrial clot) correlated significantly with thromboembolism. In our study, left atrial spontaneous echo contrast was present in 53.5% of cases and was significantly correlated with older age, large left atrium and atrial fibrillation. Although on univariate analysis, spontaneous echo contrast was found to be significantly correlated with the presence of clot, when all significant variables were included in multivariate analysis this variable was not significant, suggesting a more powerful effect of other variables, such as atrial fibrillation and left atrial area which were found to be independent predictors for left atrial clot. Similar results were also shown by Castello et al. [13]. in a small series of 17 patients with mitral stenosis. Though patients with left atrial clot had signifi-
K.C. Goswami et al. / International Journal of Cardiology 73 (2000) 273 – 279
cantly longer duration of symptoms than patients without clot, the duration of symptoms did not correlate with the presence of left atrial spontaneous echo contrast. This may mean that stasis of blood in the left atrium occurs relatively early in patients with mitral stenosis, thus creating a milieu for clot formation. Stasis of blood has been widely thought to be responsible for the appearance of left atrial spontaneous echo contrast [9,12,13,20]. However, whether the presence of left atrial spontaneous echo contrast in the absence of clot should be an indication for initiating anticoagulation is debatable.
5. Conclusion We concluded that in patients with severe mitral stenosis, atrial fibrillation and left atrial size were found to be independent predictors of clot and spontaneous echo contrast formation in left atrium and left atrial appendage. In a subgroup of the patients with normal sinus rhythm, patients with larger left atrium (.40 cm 2 ) and spontaneous echo contrast have a higher risk of clot formation in the left atrium and left atrial appendage. We postulated that in patients with severe mitral stenosis, the presence of atrial fibrillation, and a subgroup of patients with normal sinus rhythm and left atrial area $40 cm 2 with spontaneous echo contrast, had a significantly higher chance of having a clot and might benefit from prophylactic anticoagulation.
References [1] Wolf PA, Dawber TR, Thomas E, Kannel WB. Epidemiological assessment of chronic atrial fibrillation and the risk of stroke: the Framingham study. Neurology 1978;28:973–7. [2] Rowe JC, Bland EF, Sprague HB, White PD. The course of mitral stenosis without surgery: ten years and twenty years perspectives. Ann Intern Med 1960;52:741–9. [3] Deverall PB, Olley PM, Smith DR, Watson DA, Whitaker W. Incidence of systemic embolism before and after mitral valvotomy. Thorax 1968;23:530–6. [4] Hwang JJ, Kuan P, Lin SC et al. Reappraisal by transesophageal echocardiography of the significance of left atrial clot in prediction of systemic embolization in rheumatic mitral valve disease. Am J Cardiol 1992;70:769–73. [5] Acarturk E, Usal A, Demir M, Akgul F, Ozeren A. Thromboembolic risk in patients with mitral stenosis. Jpn Heart J 1997;38:669–75. [6] Chiag CW, Lo SK, Ko YS, Cheng NJ, Lin PJ, Chang CH. Predictor of systemic embolization in patients with mitral stenosis. A prospective study. Ann Intern Med 1998;128:885–9.
279
[7] Conradie C, Schall R, Marx JD. Left atrial size — a risk factor for left atrial thrombi in mitral stenosis. Clin Cardiol 1995;18:518–20. [8] Goswami KC, Narang R, Bahl V K, Talwar KK, Manchanda SC. Comparative evaluation of transthoracic and transesophageal echocardiography in the detection of left atrial clot before percutaneous transvenous mitral commissurotomy: do all patients need transesophageal examination? Int J Cardiol 1997;62:237–49. [9] Daniel WG, Nellessen U, Schroder E et al. Left atrial contrast in mitral valve disease: an indicator for an increased thromboembolic risk. J Am Coll Cardiol 1988;11:1204–11. [10] Kronzon I, Tunick PA, Glassman E, Slater J, Schwinger M, Freedberg RS. Transesophageal echocardiography to detect atrial clot in candidates for percutaneous transseptal mitral balloon valvuloplasty. J Am Coll Cardiol 1990;16:1320–2. [11] Ozkan M, Kaymaz C, Kirma C et al. Predictors of left atrial clot and spontaneous echo contrast in rheumatic valve disease before and after mitral valve replacement. Am J Cardiol 1998;82:1066–70. [12] Beppu S, Nimura Y, Sakakibara H, Nagata S, Park YD, Izumi S. Smoke like echo in the left atrial cavity in mitral valve disease, it’s features and significance. J Am Coll Cardiol 1985;6:744–9. [13] Castello R, Pearson AC, Labovitz AJ. Prevalence and clinical implication of atrial spontaneous contrast in patients undergoing transesophageal echocardiography. Am J Cardiol 1990;65:1149–53. [14] Bernstein NE, Demopoulos LA, Tunick PA, Rosenzweig BP, Kronzon I. Correlation of spontaneous echo contrast in patients with mitral stenosis and normal sinus rhythm. Am Heart J 1994;128:287– 92. [15] Vigna C, de Rito V, Criconia GM et al. Left atrial clot and spontaneous echo contrast in nonanticoagulated mitral stenosis. A transesophageal echocardiography study. Chest 1993;103:348–52. [16] Li YH, Hwang JJ, Ko YL, Lin JL, Tseng YZ, Kuan P, Lein WP. Left atrial spontaneous echo contrast in patients with rheumatic mitral valve disease in sinus rhythm. Implication of an altered left atrial appendage function in its formation. Chest 1995;108(l):99–103. [17] Fatkin D, Scalia G, Jacobs N et al. Accuracy of biplaner transesophageal echocardiography for the detection of left atrial clot. Am J Cardiol 1996;77:321–3. [18] Sahn DJ, DeMaria A, Kisslo J et al. Recommendation regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurement. Circulation 1978;58:1072–83. [19] Helmcke F, Nanda N, Hsiung M, Soto B, Adet CK, Goyal RG, Gate wood RP. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175–83. [20] Black IW, Hopkins AP, Lee LL, Walsh WF, Jacobson BM. Left atrial spontaneous echo contrast: a clinical and echocardiographic analysis. J Am Coll Cardiol 1991;18:398–404. [21] Fatkin D, Kelly RP, Feneley MP. Relation between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994;23:961–9. [22] Cormier B, Vahanian A, Lung B et al. Influence of percutaneous mitral commissurotomy on left atrial spontaneous contrast of mitral stenosis. Am J Cardiol 1993;71:842–7. [23] Rao AS, Murthy RS, Naidu PB, Raghu K, Anjaneyulu AV, Raju PK, Naik SR. Transesophageal echocardiography for the detection of left atrial clot. Indian Heart J 1994;46:37–40. [24] Chiang CW, Lin FC, Fang BR, Kuo CT, Lee YS, Chang CH. Sand drift, echoes and clot formation in left atrium. Am Heart J 1988;115:908–11.