OBJECTIVE 1) To investigate ECG changes following co-proxamol (paracetamol 325 mg and dextropropoxyphene 32.5 mg) overdose in comparison to co-codamol (paracetamol and codeine) or co-dydramol (paracetamol and dihydrocodeine) in a prospective study. 2) To examine the relationship between estimated dextropropoxyphene dose and ECG changes in a larger patient population. BACKGROUND Co-proxamol is a common cause of drug-induced death and hospital admission in the United Kingdom. ECG changes following dextropropoxyphene have been reported in animals and man, including QRS prolongation. METHODS The prospective study was conducted on 15 patients and controls with overdose. A retrospective study of a cohort of 159 co-proxamol overdoses from a combined data set from Edinburgh and Newcastle, Australia was also conducted. The measured or estimated "four hour" plasma paracetamol level was used as a surrogate of the amount of dextropropoxyphene ingested. RESULTS In the prospective study co-proxamol overdose caused statistically significant QRS prolongation (mean [95% CI] 99.36 [96.19, 102.53] msec), compared to the other combination opioid-paracetamol products (82.84 [80.81, 84.88] msec) but no effect on PR or QTc. QRS duration increase was evident soon after exposure and remained prolonged and stable over the following 24 h. In the retrospective cohort study a dose dependency of effect on QRS was documented, although the correlation coefficient relating paracetamol level to effect was relatively weak (r = 0.338, Sig. [2-tailed] 0.003, n = 74). CONCLUSIONS QRS is significantly prolonged in co-proxamol overdose, and this prolongation is dose dependent. These findings have clinical relevance to the management of patients with co-proxamol poisoning.