Diastolic Stunning Research Articles

Diagnostic accuracy of left ventricular diastolic transverse strain imaging by speckle tracking echocardiography for diagnosing chest pain in diabetic patients

Abstract Background Two-dimensional speckle tracking echocardiography (2D-STE) has been reported to be useful for the diagnosis of myocardial ischemia by detecting delayed relaxation (diastolic stunning) after an episode of angina. 2D-longitudinal strain is not specific besides ischemia such as diastolic dysfunction, and diabetes have been associated with abnormal longitudinal fibers. The aim is to evaluate the diagnostic accuracy of Left ventricular (LV) diastolic transverse strain imaging by STE to detect the presence of acute coronary syndrome (ACS) in diabetic patients with acute chest pain. Methods 385 consecutive patients with acute chest pain and without wall motion abnormality, who were admitted to an emergency department (ED) at 1 of 12 clinical sites in Japan, were enrolled and underwent 2D-STE at ED. Left ventricular (LV) transverse strain values at aortic valve closure (A) and one-third of diastole duration (B) were measured. The strain imaging diastolic index (SI-DI) was value was determined as: (A − B)/A × 100% to assess the LV diastolic strain imaging and was used to identify the regional LV delayed relaxation. All patients underwent coronary CT or coronary angiography to establish the diagnosis of ACS. Clinicians were blinded to the 2D-STE results. Results Out of 385 patients, 2D-STE analysis was possible in 365 patients (94%). 76 patients were diabetic (DM+), and 289 patients were non-diabetic (DM-). With assessment of coronary CT or coronary angiography, ACS was diagnosed in 125 patients (34%). 2D-STE was obtained at a mean of 5.3 hours after chest pain episode. Transverse SI-DI of ischemic segments were significantly lower than those of non-ischemic segments (p value <0.001) in both diabetic and non-diabetic patients, and transverse SI-DI of both diabetic and non-diabetic patients demonstrated high area under curve (AUC) for detection of myocardial ischemia (Figure: RCA; right coronary artery, LAD; left anterior descending artery, LCX; left circumferencial artery). In diabetic patients, sensitivity, specificity, and negative predictive value for ACS of transverse SI-DI are 100%, 95%, 100% in RCA (a cut-off value of 36.2), and 86.4%, 95%, 93% in LAD (a cut-off value of 50.2), and 75%, 85%, 94% in LCX (a cut-off value of 52), respectively. Conclusion LV diastolic transverse strain imaging by 2D-STE at ED increase the sensitivity, specificity and accuracy to predict the presence of ACS in diabetic patients with chest pain, as well as non-diabetic patients. (UMIN000013859). Figure 1. Transverse Strain (SI-DI): AUC (95% CI) Funding Acknowledgement Type of funding source: None

1225 Diagnostic accuracy of left ventricular diastolic strain imaging by speckle tracking echocardiography in detecting ischemic etiology of acute chest pain

Abstract Funding Acknowledgements none OnBehalf A TRAC-SI Multicenter Trial Background Two-dimensional speckle tracking echocardiography (2D-STE) has been reported to be useful for the diagnosis of myocardial ischemia by detecting delay in regional myocardial expansion (diastolic stunning) up to many hours after an episode of angina. The aim is to evaluate the diagnostic accuracy of Left ventricular (LV) diastolic longitudinal, circumferential, transverse and radial strain imaging by STE to detect the presence of acute coronary syndrome (ACS) in patients with acute chest pain. Methods 388 consecutive patients with acute chest pain and without wall motion abnormality, who were admitted to an emergency department (ED) at 1 of 12 clinical sites in Japan, were enrolled and underwent 2D-STE at ED. Left ventricular (LV) longitudinal, circumferential, transverse and radial strain values at aortic valve closure (A) and one-third of diastole duration (B) were measured. The strain imaging diastolic index (SI-DI) was value was determined as: (A-B)/A × 100% to assess the LV diastolic strain imaging and was used to identify the regional LV delayed relaxation. All patients underwent coronary CT or coronary angiography to establish the diagnosis of ACS. Clinicians were blinded to the 2D-STE results. Results Out of 388 patients, 2D-STE analysis was possible in 358 patients (92%). With assessment of coronary CT or coronary angiography, ACS was diagnosed in 118 patients (33%). 2D-STE was obtained at a mean of 5.3 hours after chest pain episode. SI-DI of longitudinal, circumferential, transverse and radial strain of ischemic segments were significantly lower than those of non-ischemic segments (p value < 0.001), and transverse and radial SI-DI demonstrated high diagnostic accuracy, compared with longitudinal SI-DI (Figure 1). Sensitivity, specificity, and negative predictive value for ACS of transverse SI-DI are 87%, 88%, % and 95%, respectively, using a cut-off value of 63.3. Conclusion LV diastolic strain imaging by 2D-STE at ED increase the sensitivity, specificity and accuracy to predict the presence of ACS in patients with chest pain. Compared with longitudinal diastolic strain imaging, transverse diastolic strain imaging is more sensitive marker to detect the myocardial ischemic episode (UMIN000013859). Abstract 1225 Figure 1

Diastolic Stunning as a Marker of Severe Coronary Artery Stenosis: Analysis by Speckle Tracking Radial Strain in the Resting Echocardiogram.

Two-dimensional speckle tracking (2DST) stress echocardiography detects postischemic myocardial diastolic stunning. However, the use of 2DST at rest for detecting diastolic stunning in ischemia is unclear. Thirty-nine patients (age = 65 ± 12 years; male/female = 34/5) with effort angina pectoris that was confirmed by stress myocardial perfusion scintigraphy were enrolled. Ischemic area (I) was determined in the middle LV short axial view using stress myocardial scintigraphy. The area opposite to it was defined as nonischemic area (non-I). Midventricular parasternal short-axis (SAX) radial strains were estimated using 2DST at rest on the following day. LV diastolic function was evaluated using diastolic index (DI, changes in the regional LV radial strain during diastole) and radial strain rate (SR) during early diastolic period. These parameters were compared between I and non-I before and 1 month after percutaneous coronary intervention (PCI) in the I of 3 coronary vessels. For the I, the DI was lower (38 ± 27 vs. 55 ± 27; P = 0.003) and SR was higher (-1.6 ± 0.6 vs. -1.9 ± 0.8; P = 0.007) than in non-I before PCI. One month after PCI, the DI and SR recovered to 53 ± 27 (P = 0.008) and -2.1 ± 0.8 (P = 0.006), respectively. Furthermore, the DI of the LAD and LCX significantly improved (P = 0.0004 and 0.002, respectively); the RCA area showed tendency to improve (P = 0.092), and the SR also improved (P < 0.05) in all areas after PCI. Diastolic stunning in ischemic areas can be detected using 2DST at rest and recover 1 month after PCI.

Acute changes of left ventricular hemodynamics and function during percutaneous coronary intervention in patients with unprotected left main coronary artery disease

Percutaneous coronary interventions (PCIs) are increasingly being used to treat unprotected left main coronary artery (ULMCA) lesions. However, research is sparse on the acute changes of left ventricular (LV) hemodynamics and function during PCI in patients with ULMCA stenosis. We aimed to assess the acute changes of LV function using speckle-tracking imaging during PCI in these patients. Fifteen consecutive patients who underwent elective PCI for ULMCA stenosis were enrolled. Echocardiographic studies and pressure measurement were performed at baseline, during PCI and after PCI. LMCA occlusion with a first balloon inflation induced a marked reduction in the peak positive derivative of LV pressure (dP/dt max), LV global longitudinal strain (GLS), and systolic and diastolic strain rates, and a marked increase in LV end-diastolic pressure (EDP) (all P<0.01). During the second inflation, the degrees of LV hemodynamic and functional changes were similar to those of the first inflation, even with a higher inflation pressure. During the third inflation, the values of GLS and dP/dt max were higher than those of the second inflation (P=0.03 and P=0.05, respectively). After optimal PCI, dP/dt max, LVEDP, and strain parameters were improved to baseline values. LV hemodynamics and function were considerably impaired with the first ballooning during PCI for ULMCA stenosis. However, the degrees of LV hemodynamic and functional changes decreased with each successive balloon inflation, which can be explained by ischemic preconditioning. After all procedures were safely completed, LV systolic function was improved without LV diastolic stunning.

Post-exercise diastolic stunning detected by velocity vector imaging is a useful marker for induced ischemia in ischemic heart disease.

Recently, post-exercise diastolic stunning (PES) has been reported as a new clinical marker of induced ischemia. Velocity vector imaging (VVI) has been developed to visualize regional wall motion abnormalities based on vectors detected by the tissue tracking technique, which has the potential to visualize PES. Thus, the aim of this study was to evaluate the usefulness of PES detection by VVI as an objective marker of ischemia compared to stress thallium-201 (Tl-201) single photon emission computed tomography (SPECT). We studied consecutive patients who were scheduled to undergo exercise stress Tl-201 SPECT for the diagnosis of ischemic heart disease. Transthoracic echocardiography was recorded digitally before and 20min after exercise for Tl-201 SPECT, and the data were used subsequently for VVI analysis. We defined PES regions as those with new abnormal vectors observed during isovolumic relaxation. After excluding 14 patients with old myocardial infarction and/or atrial fibrillation, echocardiograms suitable for VVI analysis were obtained from 62 of 65 patients (feasibility, 95%; 44 men; mean age, 64±11years). SPECT revealed induced ischemia in 20 patients, whereas VVI identified PES in 18 patients. VVI detected inducible ischemia with 85% sensitivity and 98% specificity compared to SPECT. VVI detection of PES is a new clinical tool for induced ischemia. Regional diastolic wall motion abnormalities following induced ischemia can be noninvasively detected by VVI.

Abstract 18551: The Detection of Diastolic Stunning by Two-Dimensional Speckle Tracking Imaging Improves Initial Diagnosis in Emergency Department Patients with Suspected Acute Coronary Syndromes

Background: Differential diagnosis of chest pain (CP) in emergency department (ED) is important for the management of acute coronary syndrome (ACS). Regional left ventricular (LV) diastolic dysfunction or diastolic stunning (DS), detected by strain image (SI) derived from two-dimensional speckle tracking technology, may persist after an episode of transient severe myocardial ischemia. We studied the incremental value of the detection of diastolic stunning by SI over the initial diagnosis (InDx) of CP in the ED for identifying patients with ACS. Methods: Consecutive 128 patients (75 men, 61 ±13 years) with CP consistent with ACS were enrolled. SI (ARTIDA, Toshiba) was acquired in the apical long-axis, two- and four-chamber views within 6 hrs in ED. Transverse strain curves were obtained and peak values of strain at the closure of aortic valve (A) and at the one third of diastolic duration (B) were measured. The SI-diastolic index (SI-DI) was determined as (A-B)/A×100% and used to identify regional LV DS. In the 40 normal subjects the SI-DIs were 76.5±10.6, 80.7±8.4, 79.4±11.6 % in the territories of the left anterior descending branch (LAD), the left circumflex branch (LCX) and the right coronary artery (RCA), respectively.Clinical evaluation, serial ECGs, and troponin (Tn) were performed in all patients, with coronary angiography (CAG) performed as clinically indicated. The InDx or “definite ACS” (D-ACS) was based on presenting CP, ECG and Tn. Final diagnosis of D-ACS was defined as elevated Tn and significant disease on CAG (&gt;50% stenosis). Results: 57 patients were diagnosed as having D-ACS. Regional LV DS (SI-DI&lt;40%) had been detected in the perfusion territories of the coronary arteries with culprit lesions in 52 (91%) of D-ACS, whereas it had been noted in 9 (13%) of chest pain syndrome (CPS). SI-DIs were 25.8±15 (n=75), 17.5±14 (n=36), 21.1±11 % (n=14) in the culprit lesions of LAD, LCX and RCA, respectively, and 70±13% (n=217) in the segments corresponding to the non-culprit lesions. For D-ACS, quantitative SI plus InDx increased sensitivity by 28% to 95% (P&lt;0.01), and non-significantly increased NPV by 21% to 97%. Conclusion: Adding the detection of regional LV DS by SI to InDx in ED increased sensitivity and NPV for detecting D-ACS.

Prediction of coronary artery stenosis using strain imaging diastolic index at rest in patients with preserved ejection fraction

Post-ischemic myocardial diastolic stunning persists for a long time after transient ischemia even after systolic function has recovered. We sought to identify coronary artery stenosis in clinical patients using strain imaging diastolic index (SI-DI) at rest. We retrospectively examined 85 patients with suspected coronary artery disease and preserved ejection fraction (EF; >50%) who underwent both echocardiography and coronary angiography. Speckle tracking strains were measured in 3 apical views and parasternal left ventricular (LV) short-axis views at the papillary muscle level. LV segments with inadequate image quality and deficit segments in the movie were excluded by the blinded observer. After strain analysis, LV segments were classified into no stenosis (≤ 50%), mild stenosis (51-75%), and severe stenosis (>75%) groups on the bases of the coronary angiogram. SI-DI decreased significantly in severe stenosis segments (p<0.05, ANOVA), but none of the peak strains showed significant difference. The area under the curve for predicting severe stenosis in radial, longitudinal, and transverse SI-DI was 0.72, 0.74, and 0.80, respectively. A cut-off value of 49 for transverse SI-DI can predict LV segments with severe stenosis with sensitivity of 0.79 and specificity of 0.73. A screening cut-off value of 63 for transverse SI-DI shows sensitivity of 0.95 and specificity of 0.50. SI-DI at rest is a novel marker in predicting coronary stenosis even in patients with preserved EF. This index can be used to screen patients with suspected coronary artery disease in routine echocardiography and does not require stress provocation.

Abnormal Regional Left Ventricular Systolic and Diastolic Function in Patients With Coronary Artery Disease Undergoing Percutaneous Coronary Intervention: Clinical Significance of Post-Ischemic Diastolic Stunning

This study was designed to characterize both regional left ventricular (LV) systolic and diastolic function after percutaneous coronary intervention by using strain imaging (SI) derived from 2-dimensional speckle-tracking echocardiography. Ischemic insult after coronary occlusion affects not only regional LV systolic but also diastolic function. Regional LV transverse peak strain and strain changes during the first one-third of diastole duration (strain imaging diastolic index [SI-DI]) were monitored in at-risk segments after percutaneous coronary intervention in 30 patients with coronary artery disease. The segments were divided into proximal and distal. Strain data in the at-risk segments were compared with values derived from remote nonischemic segments. Coronary occlusion induced a marked reduction in the systolic strain in both proximal and distal at-risk segments (from 36.9 +/- 6.0% to 12.0 +/- 3.9% and from 31.9 +/- 5.6% to 6.2 +/- 3.3%, respectively, p < 0.0001). Concomitantly, SI-DI values decreased (from 76.6 +/- 5.3% to -21.2 +/- 9.1% and from 72.5 +/- 5.9% to -48.7 +/- 20.8%, respectively, p < 0.0001). Upon reperfusion, systolic deformation parameters returned to near-normal pre-occlusion values. However, SI-DI values in the both proximal and distal at-risk segments decreased (43.2 +/- 9.5%, p < 0.01, and -17.3 +/- 11.1%, p < 0.0001, respectively) 30 min after reperfusion and were still lower (51.5 +/- 9.9%, p < 0.01) in the distal at-risk segment 24 h after reperfusion. SI analysis provides detailed mechanical characterization of regions with myocardial ischemic insult and can demonstrate post-ischemic diastolic stunning despite complete systolic functional recovery after reperfusion.

Exercise-Induced Post-Ischemic Left Ventricular Delayed Relaxation or Diastolic Stunning: Is it a Reliable Marker in Detecting Coronary Artery Disease?

The aim of this study was to determine whether post-ischemic left ventricular (LV) delayed relaxation could be detected by using strain imaging (SI) derived from 2-dimensional speckle-tracking echocardiography in patients with stable effort angina. Regional LV delayed relaxation during early diastole is a sensitive sign of acute myocardial ischemia and may persist beyond recovery of exercise-induced ischemia. Regional LV transverse strain changes during the first one-third of diastole duration (strain imaging diastolic index [SI-DI]) were determined at baseline and 5 and 10 min after the exercise test in 162 patients with stable effort angina. The ratio of SI-DI before and after exercise (SI-DI ratio) was used to identify regional LV delayed relaxation. A total of 117 patients had significant (> or =50% of luminal diameter) coronary stenoses. The mean SI-DI decreased from 78.0 +/- 9.7% to 27.6 +/- 16.0% (p < 0.0001) in 191 territories perfused by coronary arteries with significant stenoses 5 min after the treadmill exercise, whereas it remained unchanged in 280 territories perfused by arteries with nonsignificant stenoses. Ten minutes after exercise, regional delayed relaxation was still observed in 85% of territories perfused by stenotic coronary arteries. An SI-DI ratio with a cutoff value of 0.74 had a sensitivity of 97% and a specificity of 93% to detect significant coronary stenosis in the receiver-operator characteristic curve. Detection of post-ischemic regional LV delayed relaxation or diastolic stunning after treadmill exercise using SI is a sensitive and reliable method for the detection of coronary artery disease.

Detection of Postischemic Regional Left Ventricular Delayed Outward Wall Motion or Diastolic Stunning After Exercise-induced Ischemia in Patients with Stable Effort Angina by Using Color Kinesis

To determine whether postischemic diastolic stunning could be detected using color kinesis, we evaluated regional left ventricular (LV) diastolic wall motion in 36 patients with stable effort angina and a coronary stenosis (> or = 70% of luminal diameter), and in 30 control subjects. Regional LV filling fraction in the short-axis view during the first 30% of the LV filling time (color kinesis diastolic index) was determined before, 20 minutes, 1 hour, and 24 hours after the treadmill exercise test. In 33 of the 36 patients (92%), new regional LV delayed outward motion during early diastole (color kinesis diastolic index < or = 40%) was detected at 20 minutes after exercise. The regional LV delayed diastolic wall motion showed significant improvement but persisted 1 hour afterward in 20 of 36 patients (56%), and disappeared 24 hours after exercise. Detection of regional stunned myocardium with impaired diastolic function may be a useful tool for the diagnosis of coronary artery disease.

DELAYED PRECONDITIONING VIA ANGIOTENSIN‐CONVERTING ENZYME INHIBITION: PROS AND CONS FROM AN EXPERIMENTAL STUDY

1. Bradykinin B(2) receptor activation confers preconditioning from ischaemic injury. In the present study, we tested whether an angiotensin-converting enzyme (ACE) inhibitor (captopril) could mediate delayed preconditioning and, thus, cardioprotection. 2. New Zealand white rabbits received 15 mL infusion of either saline (control group; n = 7) or drugs (0.3 mg/kg captopril (CAP group; n = 7) or 0.3 mg/kg captopril + 0.1 mg/kg HOE 140 (CAPHOE group; n = 7)) via a marginal ear vein over 30 min. After 24 h, hearts were connected to a Langendorff apparatus and buffer perfused. The experimental protocol consisted of 20 min global normothermic hypoxia, followed by 120 min reperfusion. 3. Compared with baseline, the mean (SEM) contractile state (= dP/dt(max)) at 120 min reperfusion was decreased to 42 +/- ;23, 72 +/- ;16 (*P < 0.05 vs control) and 49 +/- ;22% in the control, CAP and CAPHOE groups, respectively. Early relaxation (= dP/dt(min)) was reduced to 55 +/- ;28, 73 +/- ;15 (*P < 0.05 vs control) and 52 +/- ;19% in the control, CAP and CAPHOE groups, respectively. The estimate for myocardial oxygen consumption (MVO(2)= rate-pressure product) was decreased to 52 +/- ;15, 69 +/- ;24 (*P < 0.05 vs control) and 56 +/- ;15% in the control, CAP and CAPHOE groups, respectively. Similarly, coronary flow was decreased in the control, CAP and CAPHOE groups to 49 +/- ;20, 67 +/- ;18 and 46 +/- ;19%, respectively. In contrast, ventricular extrasystoles during reperfusion were significantly elevated in both the CAP and CAPHOE groups (1.3 +/- ;0.2 and 1.1 +/- ;0.3 /min, respectively) compared with control (0.4 +/- ;0.2 /min). 4. Captopril confers delayed preconditioning against stunning via a B(2) receptor-mediated pathway. This pharmacological preconditioning protects against systolic and diastolic stunning, against vascular stunning and preserves cardiac metabolism. In addition to its accepted cardioprotective effects in early preconditioning, captopril should induce delayed preconditioning (e.g. for routine interventional cardiology or in elective cardiac surgery).

Serial Changes in Regional Diastolic Left Ventricular Function After a First Acute Myocardial Infraction

After acute myocardial infarction (AMI), regional diastolic function may be abnormal even though regional systolic function appears normal. However, it is not known whether this represents a transient or permanent phenomenon, and the relation to myocardial viability is not known. To determine this we assessed regional left ventricular function during contraction and filling after AMI in 106 patients with a first AMI. Echocardiography with color kinesis was performed on day 1 and 1, 3, and 6 months after AMI. For both left ventricular systole and diastole the percentage of segments with abnormal wall motion was calculated. During the first 6 months, the area of diastolic wall-motion abnormality decreased (38 +/- 16%-30 +/- 27%, P = .001) whereas no overall change in area of systolic wall-motion abnormality was seen (18 +/- 15%-19 +/- 19%, P = .66). However, for patients with no viable myocardium no significant change in diastolic wall-motion abnormality occurred (baseline 45 +/- 12% vs 44 +/- 27% at 6 months, P = .93). In contrast, a significant decrease was seen for patients with viability (33 +/- 16%-22 +/- 23%, P < .001). This was almost exclusively caused by normalization of regions where only diastolic wall-motion abnormalities were present (19 +/- 11%-7 +/- 15%, P < .0001). In a multivariable regression model, myocardial viability (P = .01) and N-terminal pro-brain natriuretic peptide concentration on day 3 (P = .003) were associated with changes in regional diastolic wall-motion abnormalities. Thus, abnormal diastolic wall motion during diastole is frequently present after AMI, and it seems to resolve to a greater extent than abnormal systolic wall-motion abnormality especially for patients with myocardial viability. This suggests diastolic stunning.

Endothelin-1 exacerbates diastolic stunning in conscious dogs.

The purpose of this study was to examine the effects of endothelin-1 (ET-1) on "diastolic stunning" in the postischemic myocardium. In 14 conscious dogs receiving either placebo (n = 7) or ET-1 (2.5 ng.kg-1.min-1, n = 7), left ventricular (LV) hemodynamics and regional wall motion (systolic segmental shortening by sonomicrometry and the ischemic-nonischemic regional asynchrony during isovolumic relaxation) were assessed at baseline, during 10 min of left anterior descending coronary artery occlusion (CAO) and at 60 min after reflow (R-60). During CAO, the ischemic segment shortening was severely depressed and both regional asynchrony and LV relaxation time constant were significantly increased in the placebo and ET-1 groups. At R-60, this LV diastolic dysfunction recovered to baseline conditions in the placebo group but was still present in the ET-1 group. Because coronary and myocardial blood flow returned to the baseline level at R-60 in both groups, the deleterious effects of ET-1 on diastolic stunning are probably mediated by its direct action on the myocardium.

Does the myocardium become “stunned” after episodes of angina at rest, angina on effort, and coronary angioplasty?

To assess whether myocardial stunning occurs after brief periods of ischemia, global and regional ventricular function assessed by radionuclide angiography was studied in 52 patients. Patients were divided into 3 groups according to the type of ischemic episodes. Group 1 consisted of 15 patients studied before, during and after episodes of angina at rest. Seventeen patients studied immediately before and after coronary angioplasty constituted group 2. Group 3 consisted of 20 patients with stable angina studied before, during and after exercise-induced ischemia. Medical therapy was discontinued 48 hours before the study in all patients except those undergoing coronary angioplasty who were receiving diltiazem 180 mg/day. No difference in baseline ejection fraction was found between groups, whereas peak filling rate was statistically tower in group 3 patients. Evidence of left ventricular dysfunction during ischemia was seen in patients in groups 1 and 3, whereas transient ischemia was documented by ST-segment displacement and/or typical chest pain during balloon inflation in group 2. Persistence of systolic or diastolic dysfunction was not observed in any of the 3 groups in the recovery phase after ischemia. In conclusion, transient ischemia caused either by a primary reduction in oxygen supply (angina at rest, coronary angioplasty) or by an increase in oxygen demand (angina on effort) did not reproduce the phenomenon of systolic and diastolic stunning observed in animal experiments, although in all patients the ischemia was of sufficient duration and severity to induce marked ventricular dysfunction. The search for stunned myocardium should be extended to other different clinical models characterized by prolonged ischemia such as unstable angina or myocardial infarction.

Transient left ventricular filling abnormalities (diastolic stunning) after acute myocardial infarction

A variety of experimental studies suggest that diastolic left ventricular (LV) function changes after acute myocardial infarction (AMI), but limited data exist on these changes in humans. To assess diastolic filling after AMI, 60 patients underwent Doppler echocardiographic examination within 24 hours of AMI. Of 54 patients who also underwent catheterization, 45 (83%) were successfully reperfused. A subgroup of 17 patients underwent a follow-up Doppler examination at 7 days after infarction, whereas 15 patients with stable exertional angina served as control subjects. There was no significant difference in age, gender, incidence of systemic hypertension or diabetes mellitus, heart rate, mean arterial pressure or severity of coronary artery disease between the infarct and control groups. The infarct group had a lower velocity time integral total (9.9 ± 0.4 cm vs 12.0 ± 0.9 cm, p < 0.001), a lower velocity time integral E (5.8 ± 0.3 cm vs 6.8 ± 0.5 cm, p < 0.01) and a lower velocity time integral 0.333 (3.5 ± 0.4 cm vs 6.1 ± 0.5 cm, p < 0.01) than the control group. In addition, velocity time integral A/total was significantly greater in the infarction group (0.44 ± 0.03 vs 0.35 ± 0.04, p < 0.01) compared to the control group. The follow-up subgroup showed an increase in velocity time integral total (p < 0.01), velocity time integral E (p < 0.05) and velocity time integral 0.333/total (p < 0.05) over the first 7 days after infarction. The final recovery values at 7 days were not significantly different from those of the coronary artery disease group. Patients with initial ejection fractions <40% or anterior infarction had the greatest recovery during the 7-day period. In conclusion, LV filling is further impaired during acute infarction compared to patients with stable coronary artery disease with the predominant impairment in early diastole. After infarction, filling parameters improve over the first 7 days after AMI, suggesting a recovery of diastolic stunning.