Dyslipidaemia, a factor worthy of adjustment: reply

We read with great interest the recent paper from Shen et al.1 which documented detrimental effects of type 2 diabetes mellitus (T2DM) and insulin on mortality and hospitalizations due to heart failure in patients with heart failure with preserved ejection fraction (HFpEF). Authors have made multiple adjustments when performing Cox regression analysis, including factors like heart rate where differences between patients with HFpEF and no T2DM, T2DM not on insulin and T2DM on insulin were relatively small (70.5 ± 11.4 vs. 71.6 ± 11.0 vs. 72.0 ± 11.7 bpm). On the contrary, authors did not make adjustments for dyslipidaemia whose regulation is a cornerstone of treatment in secondary prevention of cardiovascular disease due to its beneficial effects, including mortality.2 Authors reported that ‘There was a similar difference in the prevalence of dyslipidaemia (75%, 62%, and 45% of each group, respectively)’ but they probably overlooked that baseline prevalence of dyslipidaemia was almost double in T2DM patients on insulin than in patients without T2DM, associated with a P-value statistically significant at 0.0001 level. In addition, prevalence of coronary artery bypass graft/percutaneous coronary intervention (CABG/PCI) was almost double in the group with T2DM and insulin compared to patients without T2DM (38.5% vs. 19.2%, P < 0.0001). Higher prevalence of CABG/PCI indicates that patients with higher coronary artery disease burden were more common in the less favourable study groups. Considering substantial number of patients in this study that would benefit from the regulation of dyslipidaemia, we believe that additional adjustments of Cox regression models for baseline dyslipidaemia prevalence and history of CABG/PCI would enhance robustness of the study results.