ABSTRACT
This study considers the clinical manifestations and risk factors of digoxin toxicity and establishes an appropriate cut-off serum level for the diagnosis of toxicity. A retrospectivestudy of 125 hospitalized patients whose serum digoxin was assayed in 1998 was conducted. Of the 125 subjects, 42 (33.6%) were classified as having definite digoxin toxicity, 9 (7.2%) were classified as having probable digoxin toxicity, and 74 (59.2%) were classified as non-toxicated. Of the patients with definite digoxin toxicity, 24 (57.1%) had cardiac manifestations, seven (16.7%) had non-cardiac manifestations, and 11 had manifestations of both types. The commonest manifestation was atrial fibrillation with block. Average daily doses of digoxin in the patients with definite digoxin toxicity and those without intoxication varied from 0.125 to 0.5 ng/ml. There was no significant statistical difference in digoxin dosage between those with and those without digoxin toxicity. Seven univariate factors of digoxin toxicity were examined: logistic regression analysis showed that, serum BUN and serum chloride were independent associated factors of digoxin toxicity: the finding suggests that renal impairment and volume contraction are strong determinants of digoxin toxicity. Mean (SD) serum digoxin levels among the patients with and without toxicity were 2.28 (1.3) and 1.05 (0.6) ng/ml respectively (p = 0.000). The best cut-off level determined by Receiver Operating Characteristic (ROC) analysis was 1.97 ng/ml. However, a low sensitivity and a high specificity make serum digoxin levels a diagnostic rather than a screening tool. The manifestations of digoxin toxicity among Thai inpatients are no different from those of other populations. The best cut-off level of serum digoxin for the diagnosis of toxicity is 2 ng/ml.