BACKGROUND: Blunted heart rate variability (HRV) and presence of ventricular late potentials (VLPs) are known to correlate with an increased risk of ventricular tachycardia and sudden cardiac death in acute myocardial infarction (AMI). In the present study, we investigated the effect of glucose-insulin-potassium (GIK) solution on the VLPs and HRV in AMI. METHODS: Seventy-two consecutive patients with first Q wave AMI were randomized to GIK solution and placebo. HRV analysis and ambulatory electrocardiographic recordings were taken in all patients between 24 and 48 h. Sub-maximal exercise testing and echocardiography were performed and signal-averaged electrocardiography (SAECG) was recorded before discharge. RESULTS: Total filtered QRS duration (FQRS: 102 +/- 7 versus 108 +/- 11 ms; P < 0.05), low-amplitude signal (LAS: 25 +/- 8 versus 32 +/- 11 ms; P < 0.01) and frequency of VLPs (21 versus 45%; P < 0.05) were found to be significantly lower while root-mean-square voltage of the terminal 40 ms of QRS (RMS-40: 45 +/- 18 versus 36 +/- 20 microV; P < 0.05), and left ventricular ejection fraction (EF: 55 +/- 6 versus 48 +/- 7; P < 0.05) were significantly higher in the GIK group when compared to placebo. During the hospital period, the presence and frequency of post-myocardial infarction angina were significantly lower in the GIK group (15 versus 29%, P < 0.05), whereas an insignificant decrease in frequency of ventricular arrhythmias was observed in these patients. On HRV analysis, there was no significant difference between two groups in either time domain (SD, SDNN, RMS-SD) or frequency domain (HF, LF, LF/HF ratio) parameters. CONCLUSION: GIK solution may be beneficial to VLPs, ischaemic events, and left ventricular systolic performance in the early period of AMI. This therapy has no significant effect on HRV in AMI patients.