Keywords
INTRODUCTION
Left ventricular systolic function and residual ischemia, evaluated before hospital discharge, are important prognostic factors in patients who have suffered an acute myocardial infarction (AMI).1,2 However, these parameters can change spontaneously over the following months. Experimental studies have shown that myocardial contraction abnormalities following a coronary obstruction are potentially reversible,3,4 and isotopic ventriculography has been used to follow the changes in ventricular volumes and the ejection fraction (EF) of the left ventricle following AMI.5-20 Myocardial perfusion tomography synchronized with ECG (gated-SPECT) allows stress and resting perfusion to be evaluated, and permits estimations of the ventricular volumes and left ventricular systolic function.21-24 However, no papers report the use of this technique to monitor these parameters following AMI. The aim of this work was to determine the changes in these parameters in patients who had suffered their first noncomplicated AMI, between the time prior to hospital discharge and one year later.
PATIENTS AND METHODS
Study subjects
The study subjects were 74 hospitalized patients (55±9 years of age; 9 women) who had suffered their first noncomplicated AMI, from a total of 196 consecutive patients admitted to the Coronary Unit of our hospital over a period of one year. Forty-one of this larger total were excluded because they had suffered a previous AMI, 34 because they had complications during their stay in hospital, 18 because they had been revascularized (12 by angioplasty and six by surgery) before the second examination, 12 because they refused a second examination at one year, 9 because of serious associated noncardiac disease, 3 because of inadequate gated-SPECT examination, 3 because of a further infarction, and 2 because of death during the first year post-AMI.
Gated-SPECT
Before hospital discharge (6-8 days after admission), and again one year after AMI, the study subjects underwent stress-rest myocardial perfusion 99mTc-tetrofosmin gated-SPECT. The stress-rest test was performed using an ergometric bicycle with an initial load of 50 watts, applying 25-watt increments every 3 min until the appearance of symptoms, an inadequate blood pressure response or a fall of >0.2 mV in the ST segment. Between 30 and 60 s before finishing the stress test, patients were given an intravenous dose of 99mTc-tetrofosmin (8 mCi). Stress images were obtained 15-30 min later. A second dose of 20-25 mCi was then administered and rest images obtained 30-60 min later. Acquisitions were made with an Elscint SP4 gamma camera equipped with a low-energy, high-resolution collimator. Sixty projections were obtained in step-and-shoot mode for a 180º semicircular orbit, starting at the oblique anterior right position, angled at 30º, and with detection every 3º for 20 s each. In the resting gated-SPECT study, 8 images were obtained per projection angle during each cardiac cycle. Images were reconstructed (Butterworth filter, order 5, section frequency 0.4 cycles/pixel), obtaining short, long horizontal and long vertical axis sections.
The LV was divided into 20 segments: 6 in the long vertical axis, 6 in the long horizontal axis, 4 in the short basal axis and 4 in the short medial axis. Points were awarded to each (from 1-4) according to the uptake level: 1=normal, 2=slight defect, 3=moderate defect, and 4=important defect. All studies were evaluated by consensus by three expert observers, and the resting perfusion indices (sum of scores at rest/20) and ischemia indices (sum of stress scoressum of rest scores/20).
From the resting gated-SPECT study, the volumes and EF of the LV were calculated using a three-dimensionally-operating automatic quantification algorithm that takes into account the volume of the ventricular cavity in the synchronized images of the short axis.25
Statistical analysis
Comparisons were made between the stress test and scintigraphic values (volumes, EF, resting perfusion index in the infarcted area, and ischemic index) recorded at the first and second examinations for all patients, the subgroups of inferolateral and anterior infarction patients, those who received/did not receive fibrinolytic treatment, and those treated/not treated with angiotensin converting enzyme inhibitors or betablockers at the time of the second study. Comparisons were also made between the number of patients requiring treatment with nitrates, calcium antagonists and betablockers at the times of the first and second examinations.
The Chi-squared test was used to compare associations between categorical variables, whereas the Student's t was used test for continuous variables. To analyze changes between the first and second examination, Student's t test was used for paired data. The differences between means and the 95% confidence intervals for these differences were calculated. Statistical significance was set at 5%. The interobserver variability (differences between two observers/mean value of the two observers) was calculated for the ventricular volumes and the EF of the same examinations (evaluated separately by 2 observers).
RESULTS
Clinical characteristics
Forty nine of the 74 patients presented with inferolateral infarction, and 25 presented with an anterior infarction (all were assessed by ECG; among the six patients who had no Q wave, SPECT confirmed five inferolateral and one anterior infarction). Fifty three patients were smokers, 34 were hypertensive, and 35 were hypercholesterolemic. Half of the patients presented with pre-infarction angina. Thirteen percent showed a peripheral vasculopathy. Some 40% of the patients received thrombolytic treatment during the first 6 h following the onset of symptoms. None was treated with fibrinolytics after 6 hours. Neither did any receive primary angioplasty. Only 25 patients were catheterized before discharge or during follow-up.
Some 83% of the patients received treatment with beta blockers at the time of the first examination, falling to 61% at the time of the second (P<.0001). No significant differences were seen in the number of patients treated with nitrates (15% compared to 22%), calcium antagonists (14% compared to 20%) or inhibitors of angiotensin converting enzyme (16% compared to 19%) at the times of the two examinations.
Gated-SPECT
Interobserver variability was 2.3% (r=0.997) for the calculation of diastolic volume, 3.8% (r=0.995) for end-systolic volume, and 1.4% (r=0.995) for EF.
During the first year after AMI, the EF of the LV improved significantly (from 46.4±9.9 to 50.9±11.7%; P<.0001; mean difference 4.5; 95% CI, 0.98-8.00) for the 74 patients as a whole. No significant changes were seen for diastolic volume (from 104±34 to 105±39 ml; P=.869; mean difference 0.5; 95% CI, -11.00-12.5), nor for systolic volume (from 57±26 to 54±34 ml; P=.278; mean difference 4.5; 95% CI, -7.2 to -12.6), although the latter did show a tendency towards reduction. The increase in EF was significant in the 18 patients with an EF=40% (from 32.2±5.3 to 39.2±12.5%; P=.035; mean difference 7.0; 95% CI, 0.49-13.5) and in the 56% of patients with an EF>40% (from 50.9±5.8 to 54.6±8.6%; P=.003; mean difference 3.7; 95% CI, 0.95-6.45) at the time of the first examination (Figure 1).
Fig. 1. Changes in EF between the first and second examinations in patients with EF1<=40% and >40% before release from hospital.
No significant differences were seen between patients who had received/not received fibrinolytic treatment with respect to ergometric results (MET, maximum cardiac frequency, maximum systolic arterial pressure) or those of gated-SPECT (FE, volumes and indices of perfusion and ischemia). EF increased in the 35 patients who did not receive thrombolytic treatment (from 46.9±7 to 52.8±6.2%; P=.006; mean difference 4.5; 95% CI, 1.3467.654), in the 46 patients treated with beta blockers (from 45.4±9 to 51.7±11%; P<.003; mean difference 6.3; 95% CI, 2.1410.46), and in the 61 patients that did not receive treatment with inhibitors of angiotensin converting enzyme (from 46.9±9 to 51.6±11%; P=.011; mean difference 4.7; 95% CI, 1.0978.3) at the time of the second investigation.
A significant improvement was seen in the systolic volume and EF in patients who had suffered inferolateral infarction, but not in those with an anterior lesion (Table 1). The EF of the LV increased >5% in 25 of the 49 patients (51%) with inferolateral infarction, and in 7 of the 25 (28%) with an anterior infarction (Figure 2).
Fig. 2. Proportion of patients with changes of >5% in the groups with inferolateral and anterior infarctions.
Significant increases were recorded in maximum O2 consumption (from 6.2±1 to 7.6±2 MET; P<.0001; mean difference 1.4; 95% CI, 0.9-1.9), maximum cardiac frequency (104±19 compared to 125±22 beats/min; P<.0001; mean difference 20.4; 95% CI, 13.627.1), and maximum systolic pressure (162.2±26 compared to 177.1±29.6 mm Hg; P<.0001; mean difference 14.9; 95% CI, 5.823.9) between the first and second examinations, although no significant changes were seen in ischemia indices. Nevertheless, there was a significant improvement in the resting perfusion indices in the infarcted region in those who had suffered an inferolateral infarction (Table 2; Figure 3).
Fig. 3. Gated-SPECT of a patient with inferolateral infarction before release from hospital (A) and one year later (B). The upper part of the figure shows the nonsynchronized resting perfusion images of the long vertical axis (ELV), short axis (EC), and long horizontal axis (ELH), showing an important defect of inferolateral perfusion in A (arrows) but a clear improvement in B. The lower part of the figure shows resting perfusion images synchronized in end-diastole (D) and end-systole (S), and the volume curve with FE that increased from 45 to 51% between the two examinations.
DISCUSSION
In publications reporting the use of isotopic ventriculography between 1 and 10 days post-AMI, the EF of the LV is described to improve in 24%-41% of patients.7,9,12 In addition, the majority of studies that have compared left ventricular systolic function before hospital discharge with that seen 6 months later, also report significant improvements in EF.14,15,17,18,20 The fact that other authors8,11,16 have not found significant differences between EF values in the third and sixth months following AMI seems to support the idea that the majority of these improvements in the first three months following AMI.
The present study used gated-SPECT to evaluate the systolic function of the LV and myocardial perfusion before hospital discharge and then again one year after a first AMI. This is a very appropriate technique since evaluations of ventricular volumes and EF are highly reproducible, as reflected in the present results. A little more than half the patients who suffered an inferolateral infarction experienced an increase in EF>5% between the first and second examinations. However, this improvement was seen in only 28% of patients with anterior infarctions; in 20%, the EF actually fell by >5% (a decrease seen in only 6% of the patients with inferolateral infarction). The increase in EF was significant for patients both with EF=40% or >40% (at the time of the first investigation, and in those who had suffered an inferolateral, but not an anterior infarction. The smaller number of patients with infarctions in the latter area might explain, at least in part, why, unlike in other studies, significance was not reached. Although changes in segmentary contractility between the first and second examinations were not measured, the improvement in global EF in some patients is probably due to improvement in the contractility of the infarcted area since, in all patients, this was normal in all other regions, and none had suffered an infarction at any other site.
The improvement of systolic function has mainly been described in subgroups of patients that have received thrombolytic treatment26-29 or primary angioplasty,30 and especially in those with an anterior AMI and low EF. However, in the present study, and in agreement with Warren et al.,31 EF was also seen to improve in those who had not received thrombolytic treatment. Other authors32,33 indicate that this improvement could be related to treatment with inhibitors of angiotensin converting enzyme, but in the present study improvements were seen in the EF of patients not treated with these drugs.
An improvement in the systolic function of the LV was also seen in the subgroup of patients on beta blocker treatment one year after AMI.
It has been suggested that the improved perfusion seen in scintigraphs between the acute or subacute phase and the chronic phase in nonrevascularized patients could be due to the spontaneous repermeabilization of the causative artery and/or a recovery of the microcirculation or cell function after the ischemia and reperfusion stage.34-36 Since only 34% of the present patients were catheterized, the results cannot provide support either way.
Wackers et al.37 were the first to describe that the proportion of perfusion defects observed with 201Tl diminished over time in patients, following AMI. Pellikka et al.35 showed that the distribution of perfusion defects diminished between 18-48 h and 6-14 days in patients with re-perfused AMI. Several authors have speculated that this recovery might also be attributed to the development of collateral circulation and the recovery of the microcirculation or transitory alterations to cellular function that contribute to the capture of the radiotracer.34,37 Toba et al.38 suggest that the scintigraphic improvement seen in nonsynchronized perfusion images during the follow-up of AMI patients might be due not only to an improvement in perfusion, but also to the recovery of regional contractility in the context of stunned myocardium. The present results appear to support the first of these hypotheses since stunned myocardium corresponds, by definition, to a hypocontractile myocardium with conserved perfusion. The present patients basically showed improvements in the resting perfusion indices of the infarcted area between the first and second examinations. In any event, the small number of patients with serious systolic dysfunction allowed no definitive conclusion to be drawn with respect to the possible spontaneous recovery of the viable myocardium.
In conclusion, left ventricular systolic function, as measured by gated SPECT during the first week following an AMI, can change over the following year. In 51% of patients with inferolateral infarction, and in 28% of those with anterior infarction, EF increased by >5% between the first and second examinations. This improvement was accompanied by increased resting myocardial perfusion.
Correspondence: Dr. J. Candell-Riera.
Servicio de Cardiología, Hospital Universitari Vall d'Hebron,
P.o Vall d'Hebron, 119-129, 08035 Barcelona, Spain.
E-mail: jcandell@hg.vhebron.es