Aims/hypothesis: Prolongation of the heart rate corrected QT interval (QTc) is seen during episodes of hypoglycaemia in type 1 diabetes. We studied the relationship between spontaneous hypoglycaemia and the QT interval and hypothesised that the choice of heart rate correction affects the observed change in QTc.
Methods: Twenty-one participants with type 1 diabetes (aged 58 +/- 10 years with duration of diabetes 34 +/- 12 years) had continuous glucose and ECG monitoring for 72 h. QT and RR intervals were measured during hypoglycaemia (blood glucose or continuous glucose measurements <or=3.5 mmol/l) and compared with euglycaemia (5-12 mmol/l). QT intervals were measured using the semi-automated tangent method from signal-averaged ECG and corrected using Bazett's formula, Fridericia's formula, the nomogram method and a linear subject-specific method.
Results: Hypoglycaemia was present in 14 participants. With Bazett's formula, QTc changed significantly from euglycaemia to hypoglycaemia (422 +/- 30 vs 432 +/- 33 ms; p = 0.02). Heart rate, QT intervals and QTc corrected with formulas other than Bazett's were not associated with a significant change (p = 0.07-0.29). During hypoglycaemia, significantly lower values of QTc compared with the subject-specific method were seen for Fridericia's formula (p = 0.02) and the nomogram method (p = 0.04).
Conclusions/interpretation: Spontaneous hypoglycaemia was associated with a modest increase in QTc. Bazett's formula resulted in overcorrection of QTc while both Fridericia's formula and the nomogram method undercorrected the QTc compared with the subject-specific method during hypoglycaemia. The results may indicate that the use of a fixed heart rate correction formula can lead to misleading results in investigations of spontaneous hypoglycaemia.