Lymph node density predicts lung metastases in oral squamous cell carcinoma.

The association between lymph node density and survival free of lung metastases in oral squamous cell carcinoma (SCC), has not been investigated so far to our knowledge. Lymph node density ≧ 0.07 has been reported by a multicentre international study to be a significant predictor of shorter survival in patients with oral SCC who have invaded nodes. We investigated whether a lymph node density of ≧ 0.07 correlates with shorter overall survival, survival free of distant metastases, and survival free of lung metastases, in patients with oral SCC and invaded lymph nodes. Thirty-five patients with histologically-confirmed invaded lymph nodes werestudied. Their density was calculated as the ratio of the number of invaded lymph nodes:total number of nodes. A density of ≧ 0.07 correlated significantly with shorter overall survival (p<0.02), survival free of distant metastases (p<0.01), and survival free of lung metastases (p<0.01) on log rank testing. On testing by Cox's proportional hazards model of multivariate survival analysis with adjustment for the pathological stage (pstage IV/pstage III), and invaded surgical margins or extracapsular spread, or both, we found that lymph node density ≧ 0.07 was associated with significantly shorter survival (p<0.02). We conclude that lymph node density predicts lung metastases in patients with oral SCC.

Effects of sildenafil citrate (Viagra) combined with nitrate on the heart.

BACKGROUND: Sildenafil citrate (Viagra) is indicated for the treatment of erectile dysfunction. Large and sudden decreases in systemic blood pressure were reported in a substantial number of patients taking sildenafil citrate combined with nitroglycerin. We studied the effect of sildenafil citrate on the relationship between changes in systemic blood pressure and coronary blood flow. METHODS AND RESULTS: Healthy male beagles were used to assess systemic blood pressure, pulmonary arterial pressure, and flow in the left circumflex artery (in which a critical stenosis was established) and left anterior descending coronary artery. After measurement of the hemodynamic variables, 2 mg/kg sildenafil citrate was administered via a nasogastric tube. Hemodynamic changes were monitored for 1 hour. Subsequently, the acute effect of nitrate combined with sildenafil citrate was studied by the bolus injection of 0.2 mg isosorbide dinitrate before and after sildenafil citrate. Systemic blood and pulmonary arterial pressures and circumflex flow did not change during this study; however, left anterior descending coronary arterial flow increased from 16.0+/-5.8 to 24.6+/-8.7 mL/min 1 hour after administration of sildenafil citrate. The prolongation of systemic blood pressure decrease and the circumflex flow decrement induced by isosorbide dinitrate after sildenafil citrate were significantly larger and longer than those before sildenafil citrate. CONCLUSIONS: Sildenafil citrate had the effect of vasodilation in a normal coronary artery; however, a combined effect with nitrate resulted in large and protracted decreases in systemic blood pressure and coronary blood flow in vessels with critical stenosis.

Functional myocardial perfusion abnormality induced by left ventricular asynchronous contraction: experimental study using myocardial contrast echocardiography.

OBJECTIVES: The aim of this study was to clarify how myocardial perfusion is impaired by asynchronous contraction. BACKGROUND: False septal hypoperfusion is noted in some patients with left bundle branch block. METHODS: Eight dogs were examined with epicardial pacing at the left ventricular posterior wall, the right ventricular anterior wall and, as a control, the right atrial appendage. The pacing rate was 80, 110 and 150 beats/min (bpm). Myocardial perfusion was assessed by contrast echocardiography. RESULTS: Left ventricular pacing at 80 and 110 bpm did not change systolic wall thickening or contrast intensity at the pacing site, although an early excitation notch was noted at the pacing site. However, at 150 bpm, systolic thickening was impaired (23.3 +/- 4.2% vs. 37.0 +/- 2.6% during atrial pacing, p < 0.05), and the peak intensity ratio of the pacing site to the ventricular septum was significantly decreased (24.1 +/- 5.7% vs. 37.0 +/- 2.8% at a pacing rate of 80 bpm, p < 0.01). The peak intensity ratio correlated with systolic wall thickening at the pacing site (y = 0.413 x -0.028, r = 0.81, p < 0.0001). However, right ventricular pacing did not change either systolic thickening or the peak intensity ratio at any pacing rate, although an early excitation notch was noted on the ventricular septum. CONCLUSIONS: Wall motion abnormalities after early excitation vary depending on the pacing mode. When tachycardia induces regional wall motion abnormalities, the ventricular wall of the pacing site is functionally hypoperfused.

Clinical application of transpulmonary contrast-enhanced Doppler technique in the assessment of severity of aortic stenosis.

OBJECTIVE: The aim of this study was to demonstrate the clinical usefulness of the transpulmonary contrast-enhanced Doppler technique by using it to assess the severity of aortic stenosis. BACKGROUND: Sonicated albumin microbubbles can pass through the pulmonary circulation after peripheral venous injection and have been reported to enhance Doppler signals from the left side of the heart. Therefore, their use to determine aortic flow velocity would facilitate the assessment of the severity of aortic stenosis. METHODS: Twenty-two patients with aortic stenosis and seven normal volunteers were examined. Aortic flow velocity was recorded with continuous wave Doppler technique from an apical window before and after injection of 2 ml of sonicated albumin. RESULTS: In 10 patients with aortic stenosis, the aortic velocity envelope was too indistinct to determine the peak velocity before sonicated albumin was injected. After injection, the aortic flow Doppler signal was enhanced in 9 of the 10 patients and the velocity envelope became clear enough to measure the peak velocity, enabling calculation of the transaortic pressure gradient. In the remaining 12 patients with aortic stenosis and in all 7 normal volunteers, the velocity envelope was clear before injection and became much clearer after injection. The calculated transaortic pressure gradient showed a good agreement with catheterization measurements (y = 1.1x-6.5, r = 0.88, p less than 0.001, SEE = 16 mm Hg, n = 13). Duration of Doppler signal enhancement was measured as the time during which the envelope was clearer than before injection throughout the ejection period. The duration was significantly shorter in patients with aortic stenosis than in normal volunteers (16 +/- 5 vs. 52 +/- 32 s, p less than 0.01). There was a significant correlation between left ventricular systolic pressure measured by catheterization and the duration of signal enhancement (r = -0.69), suggesting that albumin microbubbles were fragile at high pressure. CONCLUSIONS: The transpulmonary contrast-enhanced Doppler technique using sonicated albumin is useful for assessing the severity of aortic stenosis even in patients with poor Doppler recordings, although the duration of signal enhancement might be affected by left ventricular systolic pressure.

Plasma atrial natriuretic peptide response to direct current cardioversion of atrial fibrillation in patients with mitral stenosis.

OBJECTIVES: The purpose of this study was to evaluate the effect of direct current cardioversion therapy on the plasma concentration of atrial natriuretic peptide and to determine the main factors that influence the change in plasma atrial natriuretic peptide levels in patients with atrial fibrillation. BACKGROUND: In atrial arrythmias, whether the fast atrial rate itself or the associated elevation of atrial pressure, or both, contributes to the increase in atrial natriuretic peptide is a subject of debate. METHODS: In 15 patients with mild mitral stenosis, plasma atrial natriuretic peptide levels were measured and transmitral flow pattern was obtained by continuous wave Doppler echocardiography immediately before cardioversion and at 5 min, 4 h, 24 h and 5 days after direct current cardioversion. Mean mitral pressure gradient and atrial filling fraction were calculated on the basis of transmitral flow. RESULTS: In three patients who did not have a successful return to sinus rhythm, plasma atrial natriuretic peptide levels remained elevated after cardioversion. In 12 patients who maintained sinus rhythm, plasma atrial natriuretic peptide levels were significantly reduced from 79 +/- 29 to 36 +/- 11 pg/ml 4 h after cardioversion to sinus rhythm. However, the mitral pressure gradient did not change significantly during the observation period. There were progressive increases in atrial filling fraction throughout the observation period. From 4 h to 5 days after direct current cardioversion, plasma atrial natriuretic peptide levels gradually increased concomitantly with the recovery of atrial mechanical function. CONCLUSION: The reduction of plasma atrial natriuretic peptide levels after direct current cardioversion might be due to recovery from the high rate of atrial firing and not to an alteration in the mitral pressure gradient. Direct current cardioversion itself does not seem to influence atrial natriuretic peptide secretion. The increase in atrial natriuretic peptide levels from 4 h to 5 days after cardioversion concomitantly with an increase in atrial filling fraction may be due to recovery of atrial mechanical function.

Enhancement of Doppler flow signals in the left heart chambers by intravenous injection of sonicated albumin.

OBJECTIVES: The objective of this study was to evaluate the effect of a transpulmonary contrast agent on Doppler flow signals in the left heart chambers. BACKGROUND: Echo contrast agents are good ultrasound reflectors and could be used as Doppler signal enhancers. Sonicated albumin microbubbles are transpulmonary echo contrast agents and could enhance left heart Doppler signals after peripheral venous injection. METHODS: Thirty-one patients with various heart diseases without intracardiac shunts were assessed with Doppler echocardiography before and after injection of sonicated albumin. RESULTS: After an intravenous injection, pulsed Doppler signals of transmitral flow became more intense in all 16 patients examined, although flow velocity itself was not changed. In Doppler color flow imaging, the maximal mitral regurgitant signal area increased from 312 +/- 405 mm2 to 434 +/- 465 mm2, an average increase of 59 +/- 40% in all 17 patients with mitral regurgitation (p < 0.01). These effects were considered to be due to improvement of signal to noise ratio by the enhancement of Doppler flow signals. The duration of enhancement of pulsed Doppler transmitral flow signals was significantly longer than that of the left ventricular echocardiographic opacification (44 +/- 11 s vs. 17 +/- 7 s, p < 0.01). CONCLUSIONS: Intravenous injection of sonicated albumin can enhance the Doppler flow signals in the left heart chambers. This effect may be useful to improve the sensitivity of the Doppler system for detecting abnormalities of left heart blood flow such as mitral regurgitation.

Extension of hemorrhage after reperfusion of occluded coronary artery: contrast echocardiographic assessment in dogs.

OBJECTIVES: The aim of this study was to elucidate the progression of intramural hemorrhage complicated by reperfusion with the use of myocardial contrast echocardiography. BACKGROUND: Although hemorrhagic infarction is known to occur in ischemia followed by reperfusion, its onset and sequence have not been well characterized. METHODS: In 20 anesthetized dogs, 3-h occlusion of the left circumflex-coronary artery was followed by reperfusion. The area at risk during coronary occlusion was approximately 25%. Myocardial contrast echocardiogram was examined, and the time-intensity curves for both ischemic and nonischemic areas were obtained at baseline, at 3 min after reperfusion and then at 15-min intervals until 90 min after reperfusion. The wall thickness of both areas was also measured. RESULTS: Gross hemorrhage in the reperfused areas was observed in five dogs (Group H) but not in seven dogs (Group NH). All wall segments were opacified at 3 min after reperfusion in both groups. However, the contrast defect spread significantly with time after reperfusion in Group H but not in Group NH (18.7 +/- 3.4% and 3.3 +/- 1.8%, respectively, at 90 min after reperfusion p < 0.005). The wall of the risk area at 90 min after reperfusion had thickened to 1.3 times baseline thickness in Group H but was unchanged in Group NH. The other eight dogs were excluded from study because of fatal arrhythmias or the existence of collateral circulation during coronary occlusion. CONCLUSIONS: Both progression of the contrast defect area on myocardial contrast echocardiography and a gradual thickening of the wall with reperfusion are characteristic of hemorrhagic infarction.

Application of Doppler color flow imaging to determine valve area in mitral stenosis.

This study was undertaken to examine whether Doppler color flow imaging could accurately estimate the valve area in mitral stenosis. Doppler color flow assessments were performed in both an in vitro model and in 30 patients with mitral stenosis undergoing cardiac catheterization. In the experimental Doppler study using a circuit model, color jet width correlated well with actual orifice diameter (r = 0.99). In the clinical Doppler study, the mitral valve orifice was assumed to be elliptic and the mitral valve area was calculated from the following equation: (pi/4) (a x b), where a = color jet width at the mitral valve orifice in the apical long-axis view (short diameter) and b = the width in the 90 degrees rotated view (long diameter). Mitral valve area was also determined by two-dimensional echocardiography and the pressure half-time method, and the results for all three noninvasive methods were compared with those obtained at cardiac catheterization. By Doppler color flow imaging, mitral valve area could be determined in all patients and there was a significant correlation between the Doppler jet and catheterization estimates of mitral valve area (r = 0.93). Valve area determined by two-dimensional echocardiography correlated well with catheterization measurements in 26 patients (r = 0.84). However, the area could not be determined in 4 (13%) of the 30 patients because of technical problems. Although there was a fair correlation between the valve area determined by the pressure half-time method and catheterization (r = 0.79), this method tended to overestimate valve area in patients with aortic regurgitation.(ABSTRACT TRUNCATED AT 250 WORDS)

Validation of continuous wave Doppler-determined right ventricular peak positive and negative dP/dt: effect of right atrial pressure on measurement.

OBJECTIVES: The present study aimed to validate the peak positive and negative values of the first derivative of right ventricular pressure (dP/dt) using Doppler echocardiography and to determine the impact of right atrial pressure on the measurements. BACKGROUND: A pressure gradient between the right ventricle and the right atrium can be obtained by continuous wave Doppler-derived tricuspid regurgitant velocity using the simplified Bernoulli equation. If right atrial pressure fluctuation during systole and isovolumic diastole were small compared with right ventricular pressure changes, right ventricular pressure could be evaluated, and maximal positive and negative dP/dt could also be determined with Doppler echocardiography. METHODS: We investigated 11 patients with a wide range of right atrial pressure with tricuspid regurgitation using simultaneous examination by Doppler ultrasound and catheterization. Hemodynamic conditions were altered by the Valsalva maneuver, and a total of 40 beats were analyzed. RESULTS: There was good correlation between Doppler-derived and catheterization-derived peak positive dP/dt (y = 1.0x - 15.4, r = 0.98, n = 40), irrespective of the level of right atrial pressure. Doppler-derived peak negative dP/dt also showed good correlation with that determined by catheterization (y = 0.9x + 58.2, r = 0.93, n = 40). However, in patients with high right atrial pressure (v wave pressure > or = 10 mm Hg), Doppler-derived peak negative dP/dt tended to show lower values than those from catheterization measurements, except in patients with pulmonary hypertension. CONCLUSIONS: We conclude that right ventricular dP/dt can be estimated by the Doppler method accurately and noninvasively. However, when right atrial pressure is relatively high compared with corresponding right ventricular pressure changes during isovolumic diastole, Doppler-derived peak negative dP/dt might underestimate catheter-derived measurements.

Smoke-like echo in the left atrial cavity in mitral valve disease: its features and significance.

In some patients with mitral stenosis, a smoke-like echo is observed in the left atrial cavity. The present study in 116 consecutive patients with rheumatic mitral valve disease investigated the echocardiographic features and clinical significance of this echo. The smoke-like echo is characterized by the following echocardiographic features: 1) it is composed of numerous microechoes; 2) it curls up slowly in the enlarged left atrial cavity; and 3) it vanishes as soon as it pours into the ventricular cavity. Hemostasis in the left atrial cavity was considered to be an important underlying condition for development of the echo. Hemorheologic conditions indicated that the shear rate of blood flow in the left atrial cavity was calculated to be low enough for the development of red blood cell aggregation. These conditions suggest that the source of the smoke-like echo might be aggregated cells due to hemostasis in the left atrial cavity. Left atrial thrombi were detected in many patients who had this echo in the left atrial cavity. Although it has not been conclusively determined that the presence of the smoke-like echo is a necessary condition for thrombus formation, this echo appears to be closely related to thrombus formation in the left atrial cavity. It is concluded that the presence of this echo indicates severe left atrial hemostasis and is a warning for thrombus formation.

Left ventricular function and the relationship between left atrial pressure and peak early diastolic filling velocity in dog.

OBJECTIVE: The aim was to clarify the roles of left atrial pressure and ventricular function in the determination of early diastolic filling. METHODS: Various grades of ventricular dysfunction were made in 12 mongrel dogs by coronary microembolization under pentobarbitone anaesthesia. Left atrial pressure was altered by volume loading. Peak early diastolic filling velocity was measured using pulsed Doppler echocardiography. Ventricular fractional shortening was measured using M mode echocardiography. RESULTS: Peak early filling velocity increased as left atrial pressure increased. There was a direct relationship between mean left atrial pressure and the velocity before and after induction of ventricular dysfunction. The slope of the regression line between mean left atrial pressure and peak early filling velocity decreased as the grade of the dysfunction increased. There was a significant correlation between the slope of the regression line and mean left ventricular fractional shortening (r = 0.65, n = 31, p less than 0.01). CONCLUSIONS: Early diastolic filling was affected by both left atrial pressure and left ventricular function. These facts are useful in interpreting the various transmitral flow patterns observed clinically.

Rapidity of progression of aortic stenosis in patients with congenital bicuspid aortic valves.

The rapidity of progression of aortic stenosis in patients with congenital bicuspid aortic valves, and its relation to aging and valve anatomy are not well known. To elucidate these aspects, 75 patients aged 15 to 76 years were examined by echocardiography. Aortic valve sclerosis began from the second decade, the sclerotic index progressing with age (r = 0.72; p < 0.0001). Aortic valve calcium was noted from the fourth decade. Aortic valve pressure gradient increased approximately 18 mm Hg each decade, concomitant with progression of valve sclerosis (r = 0.78; p < 0.0001). Progression of cusp sclerosis was faster in patients with anteroposteriorly located cusps than in those with right-left-located cusps (p < 0.005), and was faster in those with eccentric cusps (width ratio of major and minor cusps > or = 1.2) than in those with symmetric cusps (p < 0.05). In patients with eccentric and anteroposteriorly located cusps, aortic valve pressure gradient increased 27 mm Hg per decade. In patients with congenital bicuspid aortic valves, the progression of aortic stenosis is rapid, and the rapidity depends to some extent on the position and eccentricity of the cusps.

Echocardiographic study of abnormal position and motion of the posterobasal wall of the left ventricle in cases of giant left atrium.

In 35 of 70 patients with rheumatic mitral valve disease, two dimensional echocardiography revealed the posterobasal wall of the left ventricle to be entrapped between the left ventricular and atrial cavities and bent inward. The motion of the bending segment was paradoxical. This abnormality was assumed to be induced by the left atrial dilatation extending inferiorly behind the left ventricle, because the length of the bending segment correlation with the left atrial dimension. There was no correlation between the degree of abnormal bending and left atrial pressure, mitral valve pressure gradient or left ventricular dimension. The systolic excursion of the posterobasal wall of the left ventricle was reduced according to the length of the bending segment. This abnormal feature was also observed in five postmortem heart specimens with an extremely dilated left atrium. The macroscopic and microscopic findings in the myocardium of the bending segment were not different from those of the remaining segment of the left ventricle. Therefore, the asynergic motion of the bending segment is assumed to be caused by the abnormal spatial orientation of the left ventricle and the left atrium. It should be considered that the giant left atrium not only oppresses the surrounding organs but also affects the left ventricle.

Left main coronary flow velocity associated with stenosis. Evaluation by transesophageal color-guided pulsed Doppler technique.

To quantitatively estimate the extent of left main coronary artery (LMCA) stenosis, flow velocity of the LMCA in 33 patients was analyzed by a transesophageal color-guided pulsed Doppler technique. In 11 of 20 patients with LMCA stenosis, coronary flow velocity could be measured. The peak diastolic flow velocity at the stenotic segments was 90 +/- 32 (SD) cm/s which was significantly greater than that at the nonstenotic segments (n = 13; 34 +/- 8 cm/s; p < 0.01), and was correlated with the angiographically determined percentage of diameter stenosis of the vessel which ranged from 52 to 90 percent (r = 0.77; y = 6.34 square root of x + 10.4; p < 0.01). These results suggest that acceleration of flow velocity at the point of stenosis may be correlated with the severity of the stenosis. Measurement of flow at the point of stenosis by transesophageal color-guided pulsed Doppler technique may facilitate the quantitative assessment of LMCA stenosis, although its sensitivity requires improvement.

Angiographic volume estimation of right ventricle. Re-evaluation of the previous methods.

Previous angiographic methods for volume estimation of the right ventricle were re-evaluated by ten cast studies as far as the positions of the right ventricular casts were concerned. The spatial orientation of the right ventricular chamber cavity varies clockwise or counterclockwise, in the normal heart. For right ventricular volume estimation, it is important to analyze the right ventricular casts with rotated positions. The casts were studied by biplane cineangiocardiograms with four clockwise rotated positions, which were 0 degrees, 30 degrees, 40 degrees, 45 degrees, and 60 degrees. Simpson's rule and the area-length methods were used for angiocardiographic determination of the right ventricular volume. In four positions, four linear regression equations comparing true volumes with calculated volumes had similar values to each other in the two methods. Regression equation comparing true volumes with calculated volumes which contained a total of 40 casts in four positions, yielded a high correlation coefficient (r = +0.98, p less than 0.001) and small errors in both Simpson's rule and the area-length methods. This study indicates that the right ventricular chamber volume can be accurately estimated with only one regression equation from biplane cineangiocardiograms, even when the right ventricular cavity would have been rotated clockwise or counterclockwise.

Visualization of coronary blood flow by color Doppler imaging with a transesophageal approach.

To examine whether CBF can be imaged by color Doppler technique, we visualized CBF using a transesophageal color Doppler imaging system. In 36 of 39 patients with normal coronary arteriograms after heart surgery, the LMT and the proximal LAD were clearly imaged by 2-dimensional echocardiography. Among them, CBF was coded mainly in blue in 32 patients, showing that the flow runs from proximal to distal. In some cases, yellowish and/or reddish components were seen, suggesting the occurrence of aliasing associated with augmentation of flow velocity or flow turbulence. The fast Fourier transformation spectrogram of the pulsed Doppler sampled in the colored flow showed a typical flow velocity pattern of the LAD, which mainly consisted of a predominantly diastolic component. These results indicate that the transesophageal color Doppler technique is advantageous in noninvasively imaging CBF. It should be further sought whether flow abnormality due to coronary stenosis could be determined by this technique.

Surgical treatment of giant left atrium combined with mitral valvular disease. Plication procedure for reduction of compression to the left ventricle, bronchus, and pulmonary parenchyma.

Giant left atrium associated with mitral valvular disease frequently produce postoperative problems with regard to hemodynamical and respiratory management. A new procedure of para-annular and superior plication combined with the conventional right-side plication with trimming of the left atrial wall was devised to relieve compression induced by the giant left atrium. A total of 40 patients with giant left atrium underwent operation. Ten had the valvular procedure only and 30 had the valvular and plication procedures. The plication procedure resulted in a significant decrease in the incidence of low-output syndrome and respiratory failure postoperatively, as well as a marked decrease in mortality. We conclude that the plication procedure is an effective means of relieving compression in the presence of giant left atrium.

Resection of leiomyosarcoma originating in internal iliac vein and extending into heart via inferior vena cava.

A patient's condition was initially diagnosed as constrictive pericarditis, but an echocardiogram revealed the presence of an intracardiac tumor. The tumor was a slow-growing leiomyosarcoma that originated in the left internal iliac vein, extended into the inferior vena cava, and reached the right ventricle. This tumor was considered to be a recurrent uterine leiomyosarcoma that had been resected 4 years before. After a second resection of this tumor, the patient was living a normal life 61/2 years postoperatively, in spite of recurrence in the pelvis and lung. We reviewed 12 cases of leiomyosarcoma in the world literature that reached the heart through the inferior vena cava. We added our case and evaluated the clinical features of all 13. This is the first successfully resected case of leiomyosarcoma of this nature so far as we could determine.

Early diagnosis of stuck Starr-Edwards disc valve in mitral position: inconstant and interrupted mitral regurgitation.

The patient was a 40-year-old man with Starr-Edwards disc valve for 21 years without symptoms. Newly developed systolic murmur at the apex was noticed at the monthly medical examination. Doppler color flow mapping indicated severe mitral regurgitation. However, the appearance of mitral regurgitation was not constant. Continuous wave Doppler recording indicated that some of the mitral regurgitations were abruptly terminated at early or mid systole with the disc closing signal. Mitral regurgitation was observed only at bradycardia and on the left lateral position. Cardiac surgery demonstrated no abnormality on the disc valve except small sulci on the rim of the disc. When the disc was tilted in the cage, it was stuck and did not move. Echocardiography was useful in detecting the early stage of stuck artificial valve that might be fatal in the mitral position.

Effect of pericardium on left ventricular early filling assessed by pulsed Doppler echocardiography.

Pericardial effect on left ventricular early filling was studied in six dogs by use of pulsed Doppler echocardiography. Rapid dextran infusion was used to manipulate left atrial pressure. Left ventricular peak early filling velocities before and after pericardiectomy were measured at different levels of left atrial pressure with right atrial and left ventricular pressures. Peak early filling velocity correlated with mean left atrial pressure before and after pericardiectomy. Before pericardiectomy, when mean right atrial pressure exceeded 12 mm Hg, further increase of the velocity was not observed in spite of volume loading. After pericardiectomy, peak early filling velocity became higher and the slope of the relation line between mean left atrial pressure and the velocity became steeper than before pericardiectomy. When the filling pressure was denoted by the transmural mean left atrial pressure, the points on the relation line before pericardiectomy fell along the relation line after pericardiectomy. We conclude the pericardium exerts an external constraint on left ventricular early filling. Therefore we should not consider the absolute but the transmural filling pressure when estimating left ventricular early diastolic function with use of Doppler-derived left ventricular early filling velocity.