The arrhythmogenicity of theophylline. A multivariate analysis of clinical determinants

Abstract

Objective: to determine the relation of serum theophylline concentrations (STC) to the occurrence of cardiac arrhythmias in a population of hospitalized patients.

Study design: A cross-sectional study of 100 patients receiving theophylline who had electrocardiograms at the time STC had been obtained. Categorization was based on STC without knowledge of clinical status: group 1 (25) had STC less than 2.5 mg/L; group 2 (25), STC greater than or equal to 2.5 less than 10 (6.2 +/- 0.4) mg/L; group 3 (25), STC greater than or equal to 10 less than or equal to 20 (15.8 +/- 0.4) mg/L; and group 4 (25), STC greater than 20 (27.8 +/- 1.1) mg/L.

Setting: A university-affiliated teaching hospital.

Results: Two-thirds of the patients were receiving theophylline in an oral form, with a similar distribution found in all groups. Groups were not different with respect to age, clinical diagnoses, electrocardiographic patterns, arterial pH or gases, serum calcium, or treatment with beta-agonists. Although fewer patients in group 1 were receiving steroids than those in the other groups, and serum potassium was lower in group 3 than in group 1 (4.0 +/- 0.6 vs 4.5 +/- 0.5 mEq/L, p less than 0.05), on multivariate analysis, serum potassium was not an independent predictor of arrhythmia, and treatment with steroids or digoxin was, at most, only weakly associated with the presence of arrhythmia. Heart rate was directly related to STC (r = 0.64, p less than 0.01), and STC was the strongest independent predictor of arrhythmia in this population. Although only 20 percent of patients in group 1 or 2 had arrhythmias, 48 percent of group 3 (p less than 0.05) and 56 percent of group 4 (p less than 0.05) had this finding. Moreover, the odds ratio of group 3 having an arrhythmia compared with group 1 was 3.7, 1.5 to 11.7, p less than 0.01. Multifocal atrial tachycardia was found in 8 percent of the patients in group 3 and 16 percent in group 4 but in none of the patients in groups 1 or 2. Two patients with multifocal atrial tachycardia died suddenly within 24 hours of having their STC measurement; none had concomitant ventricular ectopy.

Conclusion: Theophylline causes tachycardia and serious arrhythmias even at STC considered to be therapeutic. Multifocal atrial tachycardia, an arrhythmia associated with use of this drug, may herald sudden cardiac death.