Aims We propose a study design with controlled hypoglycaemia induced by subcutaneous injection of insulin and matched control episodes to bridge the gap between clamp studies and studies of spontaneous hypoglycaemia. The observed prolongation of the heart rate corrected QT interval (QTc) during hypoglycaemia varies greatly between studies. Methods We studied ten adults with type 1 diabetes (age 41 ± 15 years) without cardiovascular disease or neuropathy. Single-blinded hypoglycaemia was induced by a subcutaneous insulin bolus followed by a control episode on two occasions separated by 4 weeks. QT intervals were measured using the semi-automatic tangent approach, and QTc was derived by Bazett's (QTcB) and Fridericia's (QTcF) formulas. Results QTcB increased from baseline to hypoglycaemia (403 ± 20 vs. 433 ± 39 ms, p < 0.001). On the euglycaemia day, QTcB also increased (398 ± 20 vs. 410 ± 27 ms, p < 0.01), but the increase was less than during hypoglycaemia (p < 0.001). The same pattern was seen for QTcF. Plasma adrenaline levels increased significantly during hypoglycaemia compared to euglycaemia (p < 0.01). Serum potassium levels decreased similarly after insulin injection during both hypoglycaemia and euglycaemia. Conclusions Hypoglycaemia as experienced after a subcutaneous injection of insulin may cause QTc prolongation in type 1 diabetes. However, the magnitude of prolongation is less than typically reported during glucose clamp studies, possible because of the study design with focus on minimizing unwanted study effects.