From the Departments of Internal Medicine (Pluennecke) and Cardiology (Stoler) and the Baylor Cardiovascular Institute (Roberts), Baylor University Medical Center, Dallas, Texas.

Corresponding author: William C. Roberts, MD, Baylor Cardiovascular Institute, 4 Roberts, 3500 Gaston Ave., Dallas, Texas 75246.

His medications included lisinopril, 40 mg daily; hydrochlorothiazide, 25 mg daily; potassium chloride, 10 twice daily; terazosin hydrochloride, 2 mg daily; enteric-coated acetylsalicylic acid, 325 mg daily; nifedipine, 30 mg daily; timolol maleate ophthalmic solution, each eye daily; leuprolide acetate injection, every 90 days; and 1 aspirin daily.

On presentation to the emergency department, he was awake, alert, and comfortable. His blood pressure was 92/48 mm Hg; heart rate, 58 beats per minute; respiratory rate, 16 breaths per minute, and he was afebrile. He weighed 84 kg (186 pounds) and was 183 cm (72 inches) tall. Oxygen saturation was 97% on room air. His neck veins were flat, and he had no carotid bruits. There were no precordial rubs or murmurs. A fourth heart sound was present. A few bibasilar rules were heard posteriorly on pulmonary examination. No abnormalities were observed on abdominal examination. There was no peripheral cyanosis, clubbing, or edema. The peripheral pulses were good. Neurologic examination disclosed no abnormalities.

Serum sodium was 135 mEq/L; potassium, 3.7 mEq/L; chloride, 99 mEq/L; bicarbonate, 29 mEq/L; blood urea nitrogen, 37 mg/dL; creatinine, 1.5 mg/dL; glucose, 224 mg/dL; albumin, 3.3 g/dL; total bilirubin, 0.8 mg/dL; hemoglobin, 12.2 g/dL; hematocrit, 35%; mean corpuscular volume, 93 fL; white blood cells, 14.8 x 10³/?L, with 84% neutrophils, 7% lymphocytes, and 9% monocytes; platelets, 216 x 10³/?L; aspartate aminotransferase, 211 U/L; alanine aminotransferase, 66 U/L; alkaline phosphatase, 75 U/L; creatine kinase, 1340 U/L; and troponin, 36 mg/L. Total serum cholesterol was 137 mg/dL. The serum triglycerides were 59 mg/dL. Electrocardiogram was consistent with inferior (posterior) wall AMI, with Q waves and ST segment elevation in leads II, III, and VF. ST segments in lead aVL were depressed. Leads V1 and V2 also had Q waves that appeared to be consistent with a previously healed myocardial infarction in the anterior wall.

The patient was admitted to the coronary care unit in a hemodynamically stable condition. He was treated with heparin, aspirin, oxygen, and nitroglycerin. He was free of chest pain upon arrival to the coronary care unit.

On hospital day 1, left-sided cardiac catheterization was performed: the left ventricular pressure was 90/13 mm Hg; inferior wall akinesis; anterolateral wall hypokinesis; and left ventricular ejection fraction, 45%. Coronary angiography disclosed the following degrees of diameter narrowing: left main, 30%; left anterior descending, 90% (mid); ramus intermedius, 90%; left circumflex, 90%; and right, 100%, with collaterals to the distal right from the left circumflex. A short run of atrial fibrillation occurred and resolved spontaneously. This was his only arrhythmia during hospitalization.

On hospital day 2, he complained of mid back pain that was positional. He remained hemodynamically stable in sinus bradycardia.

On day 3, while with his family in his room feeling well and laughing, he suddenly began “gasping for air.” He rapidly was ventilated with an ambu bag, and telemetry showed sinus tachycardia. His pulse was palpable. He was intubated, and good breath sounds were audible bilaterally. He then became pulseless. The heart rate decreased to about 40 beats per minute. Epinephrine/atropine, dopamine, and normal saline were administered to no avail. Pericardiocentesis yielded 35 mL of blood without functional improvement. Nearly an hour of cardiopulmonary resuscitation was fruitless.

CASE DISCUSSION

ROBERT C. STOLER, MD: This 75-year-old man presented with an inferior wall AMI and did not receive thrombolytic therapy. Cardiac catheterization disclosed multivessel coronary artery disease, and while awaiting coronary artery bypass surgery, the patient died suddenly an estimated 3 to 6 days after onset of the AMI. Pulseless electrical activity was noted during attempts at resuscitation, which was unsuccessful. While assessing the potential causes of the patient's demise, it will be helpful to keep in mind a brief differential diagnosis of possible reversible causes of pulseless electrical activity, which include hypovolemia, hypoxemia, cardiac tamponade, tension pneumothorax, acidosis, hypothermia, hyperkalemia, drug overdose, and pulmonary embolus. Sudden cardiac death occurring in the period immediately following AMI can be divided broadly into 2 categories: electrical or mechanical events. Electrical causes of sudden death include tachy- and bradyarrhythmias, while a mechanical source involves rupture of some segment of the myocardium.

Ventricular tachyarrhythmias occur frequently in patients following AMI.Ventricular tachycardia is most concerning when it is sustained. Monomorphic ventricular tachycardia usually arises from a preexisting scar of the myocardium, whereas polymorphic ventricular tachycardia is most commonly a manifestation of myocardial ischemia. The prognosis of patients with ventricular tachycardia that occurs within 48 hours of an AMI is generally the same as that of patients without ventricular arrhythmias (1). The occurrence of sustained ventricular tachycardia >48 hours post AMI is associated with increased mortality at 1 year (2). The incidence of ventricular tachycardia increases as the serum potassium falls below 4.5 mEq/L (3).

The incidence of ventricular fibrillation is now decreasing and occurs in <5% of patients with AMI (4). Sixty percent of primary ventricular fibrillations (occurring within 48 hours of AMI) occurs within the first 4 hours and 80% within 12 hours. Mortality in hospital survivors of primary ventricular fibrillation is not increased (1). The prognosis of patients with late ventricular fibrillation (>48 hours after the onset of AMI) is considerably worse, with clearly increased mortality. Late ventricular fibrillation is associated with anterior location of the AMI, and sinus tachycardia or atrial fibrillation at presentation (2).

Treatment of ventricular tachycardia and ventricular fibrillation starts with maintaining the potassium level >4.5 mEq/L and the magnesium level >2.0 mg/dL (5). Since the advent of the coronary care unit, lidocaine prophylaxis has not been shown to improve mortality (6). Prompt defibrillation/cardioversion and pharmacologic maintenance with lidocaine most commonly or intravenous amiodarone are cornerstones of therapy. Electrophysiologic study and implantable cardioverter defibrillator implantation are often warranted, particularly in patients with late ventricular tachycardia or ventricular defibrillation.

Bradyarrhythmias are not infrequently seen in patients with AMI. Mobitz type I block (Wenckebach) is seen in 10% of cases, most commonly in the inferior location (7). This block usually lasts <72 hours, is self-limited, and rarely degenerates to complete heart block. Mobitz type II block occurs in <1% of AMI, usually in the anterior location (8). This is associated with a higher risk of progression to complete heart block and often requires temporary and/or permanent pacemaker placement. Complete heart block nearly always requires pacing, and the prognosis is worse in the setting of an anterior wall AMI. Conversely, complete heart block in the setting of the inferior wall AMI is transient in 90% of cases (7).

Ventricular septal defect occurs in 1% to 3% of AMIs, most commonly between days 1 and 7 (9). Ventricular septal defects are seen more commonly in the elderly, in the setting of systemic hypertension or anterolateral infarct, and possibly in patients who receive thrombolytic therapy (10). A new holosystolic murmur and thrill are seen in at least 50% of those patients who progress to biventricular failure and cardiogenic shock in hours to days.

Papillary muscle rupture results in acute, severe mitral regurgitation, most commonly 2 to 7 days post AMI (9). This complication is most common in the setting of an inferior wall AMI, due to an anatomic difference in blood supply (11). The anterolateral papillary muscle receives a dual blood supply from the left anterior descending and circumflex coronary arteries. The posteromedial papillary muscle receives a single blood supply from the posterior descending artery, making it more likely to become ischemic or rupture during an inferior wall AMI. Patients with this condition develop rapid, severe pulmonary edema and cardiogenic shock. The murmur of mitral regurgitation may not be heard or may disappear as shock progresses.

Ventricular free wall rupture resulting in hemopericardium and tamponade is the source of 15% of deaths from AMI (9). This occurs most commonly within the first 4 days post AMI. Systemic steroids and nonsteroidal anti-inflammatory drugs may increase the risk (12, 13). Generally, free wall rupture occurs in infarcts involving >20% of the myocardium; the incidence increases with advanced age and hypertension, and is higher in women and in patients without prior myocardial infarction (14). Administration of thrombolytics changes the time course of rupture, increasing the incidence early (<24 hours) in the course of AMI but decreasing the incidence late in the course (15).

A high index of suspicion and early recognition are fundamental in the treatment of mechanical complications of AMI. Confirmation is most commonly obtained with echocardiography. Stabilization of the hemodynamics through the use of an intra-aortic balloon pump, vasopressors, inotropes, and vasodilators (if the blood pressure permits) is the first step in treatment. Ultimately, surgical correction of the defect is necessary; generally, mortality rates are better with early repair. Perioperative mortality is higher in patients with ventricular septal defects than in those with papillary muscle rupture (10).

My diagnosis in this case rests on 3 facts: the patient was on telemetry during the arrest, pulseless electrical activity was noted as the primary rhythm rather than a tachy- or bradyarrhythmia, and the patient rapidly progressed to death without evidence of pulmonary edema. This combination of events overwhelmingly makes free wall rupture with tamponade the most likely cause of death.

AUTOPSY FINDINGS

WILLIAM C. ROBERTS, MD: At autopsy, the amount of subepicardial fat was so increased that the heart floated in water (Figure). A radiograph of the heart at necropsy disclosed calcium in the left circumflex and right coronary arteries. The 4 major epicardial coronary arteries (right, left main, left anterior descending, and left circumflex) were excised from the heart, fixed in formaldehyde, decalcified, and divided into 5-mm segments, and a histologic section stained by the Movat were prepared from each 5-mm segment. The results of these studies are shown in the Table. Of the 54 5-mm segments examined, 11 (20%) were narrowed between 76% and 95% in cross-sectional area; 27 (50%) were narrowed 51% to 75%, and 16 (30%) were narrowed 26% to 50% in cross-sectional area by plaque alone. No segments were narrowed 25%, and none were narrowed >95% in cross-sectional area. In addition, in 2 segments in the right coronary artery a ruptured plaque was found, and there was hemorrhage into the pultaceous debris of the underlying plaque with superimposed occluding thrombus. The ventricles were cut in bread-loaf fashion parallel to the posterior atrioventricular sulcus. An acute infarct was present in the posterior left ventricular free wall, also involving the ventricular septum and a portion of right ventricular free wall (Figure). The size of the infarct was relatively small. The infarct had ruptured, and the rupture site was at the junction of the ventricular septum and left ventricular free wall posteriorly (Figure). No scars were present in the myocardial wall, a finding indicating no previous infarct. The left ventricular cavity was not dilated.

Why does rupture occur? The answer remains unclear. The subepicardial adipose tissue is increased in ruptured cases compared with nonruptured fatal AMI cases (16). Rupture appears to be infrequent in lean persons with AMI. What the subepicardial adipose tissue has to do with rupture, however, is unclear because most of the fat is over the right ventricle and atria and in the atrioventricular sulci and the fat rarely infiltrates the left ventricular wall. Ventricular scars are absent or if present they are small (17, 18). Left ventricular function usually remains good after AMI in cases that go on to rupture. Systemic hypertension was believed to be of higher frequency in patients with fatal AMI with rupture compared with patients with fatal AMI without rupture, but recent studies have shown that not to be the case (17–20). The acute infarct is usually relatively small, meaning that ventricular function is pretty well preserved (21). That less coronary arterial narrowing by plaque occurs in fatal AMI with rupture compared with fatal AMI without rupture probably signifies few coronary arterial collaterals. Finally, not all rupture cases with AMI extend through the entire left ventricular free wall (22). Sometimes the myocardial wall ruptures and the blood infiltrates the subepicardial adipose tissue without actually breaking through the subepicardial fat. Thus, in these cases hemopericardium does not result.

FOLLOW-UP DISCUSSION

ANNE PLUENNEKE, MD: Rupture of the left ventricular free wall or ventricular septum or papillary muscle is responsible for 15% of hospital deaths from AMI (18). Rupture of the free wall of the infarcted ventricle is the most common mechanical complication of AMI (17). Free wall rupture is second only to pump failure as a cause of death in AMI in the coronary care unit (18).

Patients with cardiac rupture secondary to AMI usually have no history of previous myocardial infarction (17–20). Autopsy studies have shown that ruptured hearts have considerably less coronary narrowing than nonruptured acute infarcts (21). In ruptured hearts, the ventricle typically is no larger in mass than in nonruptured infarcted hearts (17, 23). The infarct is usually smaller in ruptured than in nonruptured cases (17). Rupture is more likely to affect older patients than younger patients (17–20). Thrombolytic therapy does not increase the risk of rupture. A meta-analysis of 4 studies in 1990 confirmed that cardiac rupture is prevented by early thrombolytic therapy but is promoted by late treatment (24). Lastly, ST-segment elevation and Q-wave development on the initial electrocardiogram, along with peak creatine kinase–myocardial band enzyme levels >150 U/L, also correlate with increased risk of cardiac rupture (14).

Being female does not appear to be an independent risk factor for free wall rupture. In total numbers, men rupture more, but the percentage of women who have an AMI and then rupture is higher than that of those who have an AMI without rupture (17–20, 23, 25). Systemic hypertension has not proved to be a risk factor for rupture (17–20, 25, 26). In fact, long-standing hypertension can lead to hypertrophy, which can be protective against rupture (14).

The time course from onset of AMI to free wall rupture ranges from 1 to 21 days. It occurs most commonly between days 2 and 6, with a peak at 50 hours. Nearly a third of ruptures, however, occur within the first 24 hours (27).

The key to diagnosing rupture is having a high index of suspicion. The definitive diagnosis is largely dependent on pericardiocentesis and echocardiography. The finding of gross blood in the pericardial sac strongly suggests rupture and the need for urgent surgery (28).

References