The value of QT interval dispersion for identification of total mortality risk in non‐insulin‐dependent diabetes mellitus

Sawicki PT, Kiwitt S, Bender R, Berger M (Heinrich‐Heine University of Düsseldorf, Düsseldorf; Germany). The value of QT interval dispersion for identification of total mortality risk in non‐insulin‐dependent diabetes mellitus. J Intern Med 1998; 243 : 49–56. Objectives To delineate different risk markers including the difference between the maximum and the minimum length of the QT interval in ECG corrected for heart rate (QTc dispersion) as predictors of total, cardiac and cerebrovascular mortality in NIDDM patients. Design Case‐control, follow‐up study until death or for a period of 15 to 16 years. Setting Tertiary care centre, University Hospital of Düsseldorf, Germany. Subjects 216 unselected consecutive NIDDM patients. Main outcome measures Total, cardiac, and cerebrovascular mortality. Results During the follow‐up 158 (73%) patients died. In the Cox proportional hazards model QTc dispersion was the most important independent predictor of total mortality (risk ratio (RR) 3.3; difference for RR: 0.05 s 1/2 ; P = 0.001). Additional independent risk markers were age, male sex, systolic blood pressure, diabetic retinopathy, micro‐ or macroproteinuria, total serum cholesterol and HDL cholesterol. The QTc dispersion was also an independent predictor of cardiac and cerebrovascular mortality. Conclusions The results of this long‐term follow‐up study indicate that QT dispersion in a routine ECG is a useful marker to identify NIDDM patients with a high mortality risk.