Pathophysiology and Natural History Pericardial Disease

Incidence and significance of pericardial effusion in patients with acute myocardial infarction (AMI) have not been established. To evaluate these issues, we studied prospectively 138 consecutive patients with AMI. An echocardiogram was obtained in each 1, 3, and 10 days and 3 and 6 months after admission. Fifty four patients with unstable angina and 57 without heart disease were studied as controls. Echocardiographic diagnostic criteria of pericardial effusion were established from 33 additional patients undergoing surgery. Pericardial effusion was found in 28% of patients with AMI. Twenty-five percent of patients with AMI had pericardial effusion on the third day, vs 8% of patients with unstable angina (p < .02) and 5% of patients without heart disease (p < .01). At 1, 3, and 10 days and 3 and 6 months prevalence of pericardial effusion was 17%, 25%, 21%, 1 1%, and 8%, respectively. There was no case of tamponade. Pericardial effusion was more common in anterior AMI (p < .02) and in patients with heart failure (p < .05) but it was not significantly associated with early pericarditis, peak creatine kinase-MB, the level of anticoagulation, or mortality. Thus, pericardial effusion is a common event in patients with AMI (incidence of 28%), but does not result in specific complications. The reabsorption rate of pericardial effusion is slow and, in our experience, mild or moderate pericardial effusion does not preclude heparin therapy. Circulation 73, No. 2, 294-299, 1986. THE INCIDENCE of pericardial effusion (PE) during acute myocardial infarction (AMI) is not well established: It was formerly considered to be an uncommon occurrence,` but there have been marked discrepancies between recently reported series4 5 and so questions regarding PE and AMI have not yet been answered. In addition, although individual cases of massive hemopericardium associated with free wall ruptures6 7 or with anticoagulant therapy have been described,8' 9 the causes and the clinical significance of PE in most patients have not yet been confirmed. Therefore, it is not known whether the finding of PE favors the diagnosis of subclinical early postinfarction pericarditis, or whether it is predictive of the subsequent development of the postinfarction syndrome. It is also not established whether the finding of PE in patients with AMI is an indication to modify prophylactic anticoagulant therapy. Thus, the present echoFrom Servicio de Cardiologia, Departamento de Medicina, Ciudad Sanitaria "Vall d'Hebron," Barcelona, Spain. Address for correspondence: Dr. Enrique Galve, Servicio de Cardiologia, Ciudad Sanitaria "Vall d'Hebron," P Valle Hebron s/n, Barcelona 08035. Spain. Received June 7. 1985; revision accepted Oct. 10, 1985. cardiographic study was designed to determine prospectively the incidence, natural history, clinical relevance, and related factors of PE in the course of AMI. Material and methods Patients. A total of 282 patients were distributed in four groups. Group I included 138 consecutive patients admitted to our coronary unit with AMI. AMI was diagnosed by the presence of at least two of the three following diagnostic criteria: prolonged suggestive chest pain, characteristic electrocardiographic (ECG) changes (new Q waves), and elevation of serum creatine kinase (CK)-MB level. Seven out of these 138 patients died within a few hours after admission before the first echocardiographic study could be recorded. The remaining 131 patients (1 15 men and 16 women, with ages ranging from 35 to 81 years, mean age 58 + 12 years) were prospectively studied with sequential echocardiographic recordings on days 1, 3, and 10 after admission, and 3 and 6 months after discharge. Ten patients were excluded because their echocardiographic recordings were considered of inadequate quality. Accordingly, group I consisted of 121 analyzable patients. Group II was a control group including 54 patients admitted for unstable angina. Ages and sex distribution were comparable to those of patients in group I. Diagnostic criteria of unstable angina were effort angina with a clearly progressive pattern or angina at rest with repeated attacks or with attacks lasting longer than 15 min; and absence of new Q waves on the echocardioCIRCULATION 294 by gest on July 6, 2017 http://ciajournals.org/ D ow nladed from PATHOPHYSIOLOGY AND NATURAL HISTORY-PERICARDIAL DISEASE gram and of elevated serum CK-MB levels. None of these patients had a previous history of myocardial infarction or had undergone cardiac surgical procedures. Echocardiographic study was performed in this group on the third day after admission. Two patients were excluded because their echocardiograms were of inadequate quality. Group III was another control group consisting of 57 subjects without any clinical, radiologic, or ECG evidence of cardiovascular disease and in whom any previous or present condition that might have been associated with PE was ruled out. Ages and sex distribution of these subjects were comparable to those of patients in group I. Two patients were excluded because their echocardiograms were of inadequate quality. Group IV consisted of 33 patients (with ages and sex distributions comparable to those in Group I) with any of several cardiac conditions who underwent cardiac surgery and in whom an echocardiographic recording performed on the day previous to surgery was available. During operation, after opening the pericardium, pericardial fluid was aspirated through a catheter with special care to remove fluid from all pericardial sinuses and recesses. The volume of pericardial fluid was measured and the results were correlated to the echocardiographic findings. Clinical findings. The following parameters were studied: (1) site of the infarction. Infarctions were classified as "Q wave infarction" (when new Q waves present) and "non-Q wave infarction" (when only changes in the ST-T segment were present). Q wave infarctions were subdivided into anterior (Q wave from V, to V3-V4) and inferior (Q wave in leads II, III, and aVF). True posterior infarctions (R/S ratio higher than 1 in V1) were included in the inferior subgroup, and lateral infarctions (Q waves in V5-V6 and/or lead I and aVL) were included in the anterior subgroup, except when associated with inferior infarction. (2) Myocardial enzyme curve. Measurements of SGOT, total CK, and CK-MB were carried out every 6 hr through the first 2 days after admission. Correlations were made with the peak value of CK-MB. (3) Heart failure. Heart failure was considered to be present when the patient had the characteristics of Killip's classes II, III, or IV. (4) Early postinfarction pericarditis. Since there are not well-established diagnostic criteria for this type of pericarditis, the two following different conventional definitions based on clinical parameters were established: (A) pericardial friction rub regardless of associated findings; and (B) pericardial friction rub and/or two of the following criteria: fever, pericardial pain, suggestive ECG changes. (5) X-ray changes. For all patients at least three x-ray chest films were available, two of which were obtained within the 4 first days after admission and the third of which was obtained on day 10. All were examined looking for changes in heart size, findings suggesting heart failure, and signs of pleural effusion. (6) Resuscitation maneuvers, electroversion, and electrocatheter insertion. The presence and time sequence of these events were tabulated along with the incidence of with PE. (7) Anticoagulant therapy. All the patients with AMI and those with unstable angina received subcutaneous calcium heparin at an initial dosage of 2500 IU/10 kg of body weight every 12 hr (full-dose heparin regimen) to maintain a ratio patient/control of partial thromboplastin time (PTT) in the 1.8 to 2.5 range. In those patients with past or present gastroduodenal ulcer, significant arterial hypertension, or old age, a dosage of 7500 IU/12 hr was administered (low-dose heparin regimen), with a subsequent lengthening of the patient's PTT to not more than 10 sec over his or her control PTT. Both the two dose regimens were adjusted over two successive control periods to keep the mentioned values in the preestablished ranges. Echocardiographic study. Initially, all recordings were carried out with an M mode Ekoline 20A ultrasonoscope. In all cases in which images of persisting effusions (echo-free space in the posterior pericardial sac present in the first three recordings without changes) were found (five patients) or in which doubts regarding interpretation of findings arose among observers (16 patients), an additional two-dimensional study with an ATL Mark 300 mechanical sector scanner echocardiograph was carried out. Two-dimensional examinations were obtained in conventional views (longand short-axis parasternal, four-chamber apical, and four-chamber and short-axis subcostal). The M mode criteria for the diagnosis of PE were drawn from our correlation study of echocardiographic and surgical findings: The amount of fluid found at operation in each patient was correlated with the echocardiographic patterns described by Horowitz et al.10 (figure 1). Types A, B, and C2 (no case of pattern C1 was found) represent patterns in which epicardium and posterior pericardium are not separated from each other throughout the cardiac cycle; these patterns were consistently found in patients in whom less than 50 ml of pericardial fluid was recovered at operation. There was complete overlap among values, precluding any estimation of the amount of fluid from the echocardiographic pattern. By contrast, all patients with D pattern had more than 50 ml fluid at operation, larger amounts than those associated with the other echocardiographic patterns. Type E represents pericardial thickening and was not associated with effusion. On the basis of these results, we accepted as a diagnostic finding of PE only the separation between pericardial sheets throughout the cardiac cycle, as in Horowitz's D pattern, and we assumed that this implied an amount of fluid greater than 50 ml. For two-dimensional examinations, it was accepted that the posterior echo-free space detected by M mode was caused by PE if the following criteria were met: (1) the echo-free space persisted throughout the cardiac cycle, without end-diastolic obliteration; (2) it ranged from atrioventricular junction to at least the vicinity of papillary muscle; (3) it was anterior to the descending aorta, and (4) it could be differentiated from other structures such as coronary sinus, descending aorta, and occasional pleural effusion or ventricular pseudoaneurysm.