A meta-analysis of prognostic indicators to predict seizures, arrhythmias or death after tricyclic antidepressant overdose

Objectives: To systematically review and summarize studies on the accuracy of ECG and tricyclic antidepressant (TCA) concentration as prognostic indicators of the risk of seizures, ventricular arrhythmia (VA) or death in patients with TCA overdose. Methods: Articles were identified with MedLine and Cochrane register of controlled clinical trials searches and review of medical toxicology textbooks. Quality of the included studies was assessed. Pooled estimates of sensitivity, specificity, likelihood ratios and Summary Receiver Operating Characteristics (SROC) curves were generated. Results: A total of 18 studies were included in the analysis. The pooled sensitivity (Se) and specificity (Sp) of the QRS for predicting seizures were 0.69 [95% CI 0.57-0.78] and 0.69 [95% CI 0.58-0.78] as compared to 0.75 [95% CI 0.61-0.85] and 0.72 [95% CI 0.61-0.81] for the TCA concentration. The Se and Sp of the QRS to predict VA were 0.79 [95% CI 0.58-0.91] and 0.46 [95% CI 0.35-0.59] compared to 0.78 [95% CI 0.56-0.90] and 0.57 [95% CI 0.46-0.67] for the TCA concentration. The Se and Sp of the QRS to predict death were 0.81 [95% CI 0.54-0.94] and 0.62 [95% CI 0.55-0.68] compared to 0.76 [95% CI 0.49-0.91] and 0.60 [95% CI 0.47-0.72] for the TCA concentration. Very few studies evaluated the accuracy of QTc, T 40 ms axis and the R/S ratio. Conclusions: Overall, the studies suggested that the ECG and TCA concentration have similar but relatively poor performance for predicting complications, such as seizures, VA or death, associated with TCA overdose.