A prospective evaluation of the electrocardiographic manifestations of hypothermia.

OBJECTIVE: To determine the effects of body temperature, ethanol use, electrolyte status, and acid-base status on the electrocardiograms (ECGs) of hypothermic patients.

METHODS: Prospective, two-year, observational study of patients presenting to an urban ED with temperature < or =95 degrees F (< or =35 degrees C). All patients had at least one ECG obtained. Electrocardiograms were interpreted by a cardiologist blinded to the patient's temperature. J-point elevations known as Osborn waves were defined as present if they were at least 1 mm in height in two consecutive complexes.

RESULTS: 100 ECGs were obtained in 43 patients. Presenting temperatures ranged between 74 degrees F and 95 degrees F (23.3 degrees C-35 degrees C). Initial rhythms included normal sinus (n = 34), atrial fibrillation (n = 8), and junctional (n = 1). Osborn waves were present in 37 of 43 initial ECGs. Of the six initial ECGs that did not have Osborn waves present, all were obtained in patients whose temperatures were > or =90 degrees F > or =32.2 degrees C). For the entire group, the Osborn wave was significantly larger as temperature decreased (p = 0.0001, r = -0.441). The correlation between temperature and size of the Osborn wave was strongest in six patients with four or more ECGs (range r = -0.644 to r = -0.956, p = 0.001). No correlation could be demonstrated between the height of the Osborn waves and the serum electrolytes, including sodium, chloride, potassium, bicarbonate, BUN, creatinine, glucose, anion gap, and blood ethanol levels.

CONCLUSIONS: The presence and size of the Osborn waves in hypothermic patients appear to be a function of temperature. The magnitude of the Osborn waves is inversely correlated with the temperature.