Background A shortening of electrocardiographic QT interval has been observed in obese subjects after weight loss, but previous results may have been biased by inappropriate heart rate (HR) correction. Methods Electrocardiography (ECG) recordings of 49 (35 females) severely obese patients before and 12 months after Roux-en-Y gastric bypass (RYGB) surgery were analysed. QT interval (QTc) was calculated by using four different equations, i.e. Bazett, Fridericia, Framingham and Hodges. Results Irrespectively of the used correction formula, QTc interval length was reduced after the surgery (QTc Bazett −31 ± 18 ms; QTc Fridericia −12 ± 15 ms; QTc Framingham −14 ± 15 ms; QTc Hodges −9 ± 15 ms; all Ps < 0.001), but QTc Bazett reduction was significantly greater than the reduction in QTc calculated upon the other three equations (all Ps < 0.001). Moreover, changes in QTc Bazett (P < 0.001) but not in QTc Fridericia , QTc Framingham and QTc Hodges (all Ps > 0.05) were significantly correlated with concurrent changes in HR. Multivariate regression analyses revealed a significant independent association of serum insulin levels with QTc Fridericia , QTc Framingham and QTc Hodges values (all Ps < 0.05) preoperatively, whilst changes in QTc interval length after the surgery were not consistently associated to concurrent changes in metabolic traits. Conclusions Our data show that the extent of weight lossassociated QTc interval shortening largely depends on the applied HR correction equation and appears to be overestimated when the most popular Bazett's equation is used.