[Clinical and instrumental elements predictive of left ventricular insufficiency in acute myocardial infarct: multivariate analysis in patients treated with thrombolytic therapy].

BACKGROUNDS: During the course of acute myocardial infarction (AMI), the appearance of signs of left ventricular failure (LVF) (cardiogenic shock, acute pulmonary edema, congestive heart failure) is a prognostically negative event which is still relatively frequent even in patients receiving fibrinolytic therapy. The early identification of patients exposed to such a risk would allow adequate diagnostic and therapeutic preventive measures to be taken.

AIM: To evaluate, in a population of AMI patients undergoing thrombolysis and without any serious complications at the moment of hospitalisation, which anamnestic, clinical and instrumental data obtained within the first 24 hours best identify those who will subsequently develop full-blown LVF. Secondary aim is to evaluate the role that extension of coronary disease plays in determining the occurrence of LVF.

METHODS: The study involved 104 consecutive patients aged < 75 years admitted to hospital for AMI with ST-segment elevation, within 12 hours of the onset of symptoms, in Killip class 1-2 upon entry to the CCU, and treated with thrombolytic therapy. The study design included the collection of anamnestic and clinical data upon admission to the CCU; an enzymatic curve during the first 4 days; the ECG at entry, and 4 and 24 hours after the beginning of fibrinolysis; the chest X-ray, the 2D-echocardiography (2D-echo) and the hemodynamic measurements within the first 24 hours; a coronary angiography on the tenth day (or earlier if clinically necessary).

RESULTS: Seventeen patients (16%) presented signs of LVF; 8 (7.6%) with cardiogenic shock, 9 with congestive heart failure: 3 died (3%), all for shock. Univariate analysis correlated LVF with: 1) the indices of the extension of ischemic/necrotic damage: number of derivations with ST elevation (p < 0.04) and Q waves (p < 0.05) at first ECG, maximum peak of myocardial enzyme (p < 0.02), wall motion score index (p < 0.001), percentage extension of asynergy (p < 0.001), presence of remote asynergy (p < 0.001), left ventricular (LV) end-systolic (p < 0.001) and end-diastolic volume (p < 0.01), and LV ejection fraction (EF) (p < 0.001) at 2D-echo; 2) the indices of hemodynamic involvement: Killip class 2 at entry (p < 0.02), pulmonary venous flow diversion at chest X-ray (p < 0.001), systolic (p < 0.05), diastolic (p < 0.01) and mean (p < 0.01) pulmonary pressure, capillary wedge pressure (p < 0.01), and the LV systolic work index (p < 0.05). Multivariate analysis showed that the only independent variable predictive of LVF was the EF at 2D-echo (p < 0.001): the sensitivity and specificity of EF was respectively 36% and 97% at cut-off value of 0.30, and 93% and 69% at cut-off value of 0.45. Multivessel coronary disease was found more frequently in patients who developed LVF (p < 0.05) and was correlated with 2D-echo LV involvement: presence of remote asynergies (p < 0.0001), lower EF (p < 0.01), higher wall motion score index (p < 0.001) and percentage extension of asynergy (p < 0.01).

CONCLUSIONS: The incidence of LVF in patients with AMI, without serious complications at onset, is still relatively high (16%) even if they are treated with thrombolysis. Of all evaluated clinical and instrumental indices, multivariate analysis showed that EF at 2D-echo was the only independent variable predictive of LVF. Extension of coronary disease correlated with development of LVF. Moreover, worse LV performance and greater regional contractility involvement at 2D-echo correlated with extension of coronary disease. Consequently, echocardiography would appear to be bed-side, simple, reliable and accurate mean of establishing a prognosis from the moment a patient with AMI is admitted to a CCU.