We describe a case of a 28-year-old woman who was admitted to our hospital with severe diabetic ketoacidosis. She was known to have had type 1 diabetes for 10 years. During the previous 2 days, she had gone to a party, drank a considerable amount of alcohol, and did not administer her regular dose of insulin.
On admission, she was semicomatose and tachypnoic, her blood pressure was 90/70 mmHg, and her heart rate 80 bpm. Laboratory tests showed severe metabolic acidosis (pH 6.92, bicarbonate 2.2 mmol/l, pCO2 1.49 kPa), very high blood glucose (75 mmol/l), hyponatremia (104.3 mmol/l), hypochloremia (70 mmol/l), severe hyperkalemia (8.5 mmol/l), and elevated blood urea (20.3 mmol/l) and creatinine (317 μmol/l). Blood ethanol level was 0.2 g/l. Screening for possible intoxication, including cocaine, opiates, and amphetamines, was negative. Electrocardiogram (ECG) showed sinus rhythm with wide QRS complexes and diffuse nonspecific ST changes.
The patient was treated with continuous intravenous saline and insulin infusion. After 12 h, her blood glucose decreased to 17.5 mmol/l (pH 7.23, bicarbonate 12.0 mmol/l, potassium 5.12 mmol/l, and sodium 127.8 mmol/l). Blood urea decreased to 14.6 mmol/l and creatinine to 154 μmol/l. ECG was also normalized. After 36 h, the patient experienced transient stabbing chest pain, which was partially relieved by the change of body position. Complex ventricular arrhythmias, including short runs of ventricular tachycardia, were noticed. Repeat ECG revealed mild ST elevations in leads II, III, and aVF with negative T-waves in leads V2–V4. Echocardiography revealed somewhat depressed left ventricular systolic function (LVEF 45%) with hypokinesis of the posterior and inferior walls. Serum troponin I increased to 343 ng/ml (normal value ≤0.4 ng/ml). On day 3 she was pain free but still had frequent premature ventricular beats. Troponin gradually decreased to 178 ng/ml. ECG showed ST segment normalization with flattening of T-waves in leads II, III, and aVF. Repeat echocardiography on day 5 showed reversal of posterior/inferior wall hypokinesis and normalization of left ventricular systolic function. The patient had an uneventful recovery. Coronary angiography on day 13 revealed normal coronary arteries with no evidence of coronary artery disease.
Different electrocardiographic patterns, including acute pseudoinfarction, have already been described in patients with ketoacidosis and hyperkalemia (1,2). None of these patients, however, had evidence of myocardial necrosis, as seen in our case. Despite very high levels of cardiac specific troponin I, echocardiography demonstrated rapid reversibility of wall motion abnormalities that corresponded to ECG changes. This is in contrast to previous observations showing no compromise but even increased myocardial contractility during diabetic ketoacidosis (3). The mechanism of myocardial necrosis in our patient is unclear. It might have been a late consequence of severe acid-base and electrolyte disturbances that might have triggered coronary spasms leading to ischemic myocardial necrosis. The coincidence of infectious myocarditis is less likely, but cannot be excluded.
In conclusion, severe diabetic ketoacidosis might be associated with myocardial necrosis of unknown mechanism leading to transient wall motion abnormalities and ventricular arrhythmias.
