Restrictive Left Ventricular Filling Pattern After Myocardial Infarction: Significance of Concomitant Preserved Systolic Function

Restrictive left ventricular (LV) filling identifies a high risk subgroup, following myocardial infarction (MI). The extent and significance of systolic dysfunction in this group is not clear. The aim of our study was to determine the incidence and extent of systolic dysfunction in patients with restrictive filling and nonrestrictive filling examining the prognostic implications. Doppler parameters of LV diastolic function were measured in 102 post‐MI subjects within 4 days. Restrictive filling was defined as the presence of E:A ratio > 2 or E:A ratio 1–2 with MDT ≤ 140 msec. Follow‐up was to a median of 11 months. Restrictive filling (group A) was found in 19 (19%) of 102 patients. Patients with this pattern were more likely to have systolic dysfunction than those without (group B); 63% and 35%, respectively, P = 0.024. Eight (42%) of 19 patients in group A had relatively preserved systolic function. At 11 months 14 patients had developed heart failure (HF), 6 in group A (32%) and 8 in group B (10%), P = 0.012. There were two deaths (11%) in group A and 7 (8%) in group B, P = ns. Seven (88%) of 8 patients in group A with relatively preserved systolic function were alive and free of heart failure at follow‐up compared to 4 of 11 patients (36%) in group A with systolic dysfunction (P = 0.026). Restrictive filling can be associated with relatively preserved systolic function after MI and these patients have a relatively good outcome. Patients with restrictive filling post‐MI are a heterogenous group emphasizing the evaluation of both systolic and diastolic function.

Left ventricular diastolic dysfunction is common after myocardial infarction (MI) and the nding of a restrictive type of left ventricular (LV) lling carries important prognostic implications. Previous studies based on pulsed-wave Doppler evaluation of transmitral ow velocity patterns have demonstrated that patients with this pattern have a particularly poor short-and long-term outcome that is independent of measures of systolic function. [1][2][3] Restrictive lling is commonly associated with signi cant impairment of systolic function 1,[4][5] where it implies the presence of elevated left ventricular lling pressures 5,6 and a high likelihood of the presence or subsequent development of symptomatic heart failure. 2,7 However, in our experience this pattern may occur in some patients in the presence of relatively preserved systolic function, and the clinical implications for this group have not previously been de ned. The purpose of this study was to compare the clinical characteristics of patients with and without a restrictive LV lling pattern after MI and determine the incidence and extent of systolic dysfunction in each group. We also examined the prognostic implications of relative preservation of systolic function.

Study Population
We prospectively studied 102 patients (70 male, aged 66 6 11 years) with a diagnosis of acute MI who were recruited from the coronary care unit at our hospital. The diagnosis of MI was based on the presence of chest pain characteristic of myocardial ischemia/infarction or conventional electrocardiographic changes together with enzymatic evidence of myocardial necrosis (elevation of serum creatine kinase to at least twice the upper limit of the normal range). Patients with atrial brillation were excluded from the study as were patients with severe mitral regurgitation. Patients were also excluded if the mitral ow velocity pro le could not be characterized for technical reasons. Follow-up data was collected at a median interval of 11 months. At this point patient symptoms were established from individual case note review and telephone contact. Signi cant heart failure was de ned as New York Heart Association (NYHA) functional class III or IV and assessed without knowledge of the echocardiographic parameters. Cause of death was determined by reviewing hospital records and, when necessary, death certi cates. The study was approved by the institutional committee on medical research and full informed consent was given by all study subjects.

Echocardiography
Examination. Within four days of admission transthoracic two-dimensional and Doppler echocardiography were performed. Patients were studied in the left lateral decubitus position with a Hewlett-Packard Sonos 2000 or 2500 machine (Agilent Technologies, Andover, MA, USA) equipped with a 2.5-MHz transducer. All studies were performed by a single experienced operator. Heart rate and blood pressure were recorded at the termination of the procedure. Color ow Doppler imaging was used to quantify mitral regurgitation. M-mode and Doppler recordings were made at a sweep speed of 50 or 100 mm/sec and studies were recorded on 0.75 ins SVHS video tape. In all study subjects echocardiographic parameters were measured of ine from video tape recordings with at least three and ve consecutive beats averaged each two-dimensional and Doppler parameter, respectively. Ectopic and postectopic beats were not analyzed.
Evaluation of Left Ventricular Systolic Function. Left ventricular end-systolic and end-diastolic dimensions were recorded in the parasternal long-axis view at a level immediately distal to the mitral valve tips from the two-dimensional-guided M-mode image. Left ventricular ejection fraction was calculated from the apical four-chamber and two-chamber views using the apical biplane method of discs. 8 A wall-motion score was calculated for each study based on a nine-segment model that has been extensively validated. 9 Visual assessment of left ventricular function was based on a grading scheme for systolic thickening. The score 3 was used for hyperkinesis, 2 for normokinesis, 1 for hypokinesis, 0 for akinesis, and -1 for dyskinesis. The overall wall-motion score was calculated by dividing the sum of the scores in each segment by nine. An ejection fraction of , 40% was used to represent significant systolic dysfunction. 1 0 Evaluation of Left Ventricular Diastolic Function. The transmitral ow velocity prole was recorded from the apical four-chamber view with the pulsed-wave sample volume located at the tips of the mitral valve lea ets in diastole. The maximum E and A wave velocities were recorded. Mitral deceleration time was measured as the time from peak E velocity to the point of intercept of the deceleration slope with the baseline. The mitral deceleration time was not recorded in patients with summation of E and A waves. The isovolumic relaxation time (IVRT) was recorded by pulsed-wave Doppler evaluation with the sample volume midway between the mitral lea ets and the aortic annulus. Using this technique the timing of aortic closure and mitral opening could be identi ed on the same Doppler trace. Restrictive left ventricular lling was de ned as an E:A ratio of . 2 or a ratio of 1-2 and a mitral deceleration time of , 140 msec. This de nition has been used previously. 1 Statistical Analysis. Continuous variables are expressed as mean 1 SD. Comparisons between continuous data in patients with and without restrictive lling were made using the paired or unpaired t-test and Mann-Whitney test for parametric and nonparametric data, respectively. Categorical data was compared by the chi-square test.

Baseline Demographic Data
Of 102 patients studied, 19 (19%) had evidence of restrictive lling (group A). Table I shows the age and sex of the two groups in addition to the clinical characteristics. Patients in group A were of a comparable age to those in group B but were more likely to be male. Peak creatine kinase was higher in group A. Pulmonary edema at presentation was more common in group A, but the difference was not statistically signi cant. There was no difference between the two groups in the proportion presenting with anterior infarction or number with a previous MI. Both systolic and diastolic blood pressure were similar in the two groups. Duration of hospital stay for group A patients was 11.8 6 10.8 days and for group B, 8.3 6 4.5 days (P 5 ns). Within group A there was no signi cant difference in age between patients with preserved and nonpreserved systolic function (age 65 6 10 and 62 6 7 years, respectively; P 5 ns) and no signi cant difference in heart rate between these two groups (77 6 15 and 75 6 12 beats/min respectively; P 5 ns). Table II shows the two-dimensional and Doppler echocardiographic data for groups A and B. In addition to MDT, differences in left ventricular lling between the two groups were re ected in signi cant differences in the E:A velocity ratio and IVRT between the two groups (P , 0.0001 and , 0.001, respectively).

Echocardiographic Evaluation of Left Ventricular Function
The ejection fraction and wall-motion score were lower in group A than group B (both P 5 0.01). Figure 1 shows the ejection fraction in each group A patient compared to group B. Eight (42%) patients in group A had preserved systolic function. In four of these patients the ejection fraction was between 40% and 50%. In two patients it was . 55%. In group B, 58 (70%) patients had preserved systolic function. Patients in group A were less likely to have evi-  dence of relatively preserved systolic function (P 5 0.01).

Clinical Status at Follow-up
Follow-up was complete to 11 months. Nine patients died and 14 had developed heart failure (Fig. 2). There had been two deaths (11%) in group A and seven (8%) in group B (P 5 ns). Signi cantly more patients had developed heart failure in group A (6 [32%]) compared to group B (8 [10%]); P 5 0.012. Among the eight group A patients with preserved systolic function at presentation, one had developed heart failure and none had died. Of the 11 patients in this group who presented with concomitant systolic dysfunction, ve developed heart failure and two had died (Fig. 3). Group A patients with signi cant systolic dysfunction were more likely to be in heart failure or have died at follow-up compared to those with preserved systolic function (P 5 0.026).

Restrictive Left Ventricular Filling After Myocardial Infarction
Left ventricular diastolic dysfunction is frequently observed after MI. 1-7,1 1 Pulsed-Doppler echocardiography is a convenient, noninvasive tool in this setting. Despite well-recognized limitations, evaluation of the transmitral ow velocity pro le has been used extensively and provides a means of characterizing left ventricular diastolic properties. A wide spectrum of abnormalities have been described previously.
In patients with relatively small infarcts, a pattern of abnormal relaxation is the most frequent nding. 4 A restrictive lling pattern is more common in patients with greater myocardial damage, and in patients with depressed systolic function, it identi es a subgroup at particularly high risk of an adverse outcome. [1][2][3] Previous reports suggest that restrictive lling in the early postinfarction period is not uncommon (13% of all post-MI patients in the study of Nijland et al. 1 and 14% in that by Sakata et al. 2 ). In Nijland's study no patient with restrictive lling had evidence of preserved systolic function de ned as an ejection fraction of $ 45% and a wall-motion score of $ 1.76. Although other studies indicate a low ejection fraction in the group of patients with restrictive lling, these studies do not state how many individual patients had evidence of relative preservation of systolic function. In addition, although it would seem plausible that an  isolated restrictive lling pattern in the absence of systolic dysfunction would carry different prognostic implications, the outcome for this patient subgroup has not been de ned.

Current Study
The prevalence of restrictive lling in the post-MI population of our study is comparable to that found by previous investigators. 1 ,2 Enzymatic infarct size was signi cantly greater in patients with restrictive lling in accordance with the study of Nijland et al. 1 and is consistent with the fact that more extensive systolic dysfunction is related to the presence of increased myocardial stiffness. There was a trend toward a higher prevalence of pulmonary edema in patients with restrictive lling, but this failed to reach statistical signi cance. This is also consistent with increased myocardial stiffness.
We have documented the range of abnormalities of systolic function in patients with restrictive lling after MI. Relative preservation of systolic function was found in 8 (42%) of 19 patients. There is increasing evidence that abnormalities of diastolic function are important contributors to symptoms in patients with congestive cardiac failure, but this is predominantly applicable to those patients with significant systolic dysfunction. 12 ,13 This would be consistent with the relatively low incidence of heart failure at follow-up in the subgroup of our patients with isolated restrictive lling. Our ndings support the assessment of both systolic and diastolic function in a composite manner after MI. This is likely to provide optimal prognostic information.

Limitations
The limitations of transmitral ow velocity patterns in the assessment of diastolic function are widely acknowledged. In addition to the effect of systolic dysfunction, mitral deceleration time can be affected by other factors such as loading conditions and pericardial restraint. However, this method of evaluating ventricular diastolic function is widely available and commonly used in clinical practice. We felt it appropriate to use a technique that was applicable to the routine clinical setting. The mitral deceleration time may appear normal (pseudonormal) in circumstances where abnormal relaxation and restrictive lling co-exist as they have opposing effects on the transmitral ow velocity pro le. This is likely to lead to an un-derestimation of the prevalence of true diastolic function abnormalities when assessed by this method. However, in the present study we were concerned only with the presence of restrictive lling as it is this pattern alone that has been of prognostic use. The short isovolumic ventricular relaxation time in the patients with restrictive lling suggests that these patients had signi cantly reduced myocardial compliance compared to the remaining patients. We did not consider the potential effect of drugs such as intravenous nitrates on the parameters of diastolic function. A single Doppler study within 4 days of infarction may not be considered optimal for the reliable evaluation of diastolic function in these patients. It may have been preferable to perform serial studies during the hospital course such as in the study of Nijland. 1 However, in that study it was clearly demonstrated that the presence of restrictive lling at any point during the early post-MI period conferred an adverse outcome. In view of this, we did not repeat the Doppler evaluation.

Conclusion
Restrictive lling is frequently associated with relatively preserved systolic function after MI. Although patients with systolic dysfunction and restrictive lling have a poor prognosis, those with isolated restrictive lling have a relatively good outcome. This study emphasizes the composite evaluation of both systolic and diastolic function after MI.