In the present study we evaluated the change of LV diastolic function in patients with type 2 diabetes who started SGLT2i. Despite the fact that most patients had potential risk factors for cardiovascular complication, there were no significant changes of LV diastolic parameters after SGLT2i administration. Even in the high-risk group for diabetes-related cardiovascular events, the diastolic profile did not improve after additional SGLT2i administration. In patients younger than 60 years of age, E/e' was significantly decreased after administration of SGLT2i compared to patients aged 60 years or older.
Currently, SGLT2i is strongly recommended for T2DM patients with established atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease 10)11). In addition, newly updated HF guidelines also recommend SGLT2i for T2DM patients with either established cardiovascular disease or high risk for it in order to prevent hospitalization for HF 15)16). However, the evidences are weak as to whether SGLT2i should be considered as a primary antidiabetic agent in patients who have only indicators of high risk for cardiovascular complication without established HF or other types of cardiovascular disease. Recent clinical trials consistently showed that SGLT2i improved the rates of HF hospitalization in patients with type 2 diabetes, but most of enrolled patients already had atherosclerotic cardiovascular disease 5)6)17). In patients without documented cardiovascular disease, benefit of SGLT2i over other types of antidiabetic agents in terms of glucocentric and cardiovascular outcome is unclear.
Since LV diastolic dysfunction is considered as the early clinical manifestation of diabetic cardiomyopathy including HF with reduced ejection fraction as well as HF with preserved ejection fraction, several trials have assessed the improvement of diastolic parameters after the administration of SGLT2i in patients with T2DM. Soga et al. prospectively showed that the E/e' significantly decreased 6 months after the administration of dapagliflozin in 57 diabetic patients 12). However, all patients enrolled in the study were under stable heart failure. Matsutani et al. also reported improvement of E/e’ after 3 months treatment with canagliflozin in 37 diabetic patients, but about a third of enrolled patients had pre-existing cardiovascular disease 13). Meanwhile, a recent randomized controlled trial in Korea demonstrated that SGLT2i did not significantly affect resting e’ velocity, E/e’, LA volume index despite it improved diastolic parameter after exercise 14). Taken together, the effect of SGLT2i on diastolic function is inconclusive for the following reasons; 1) enrolled patients mostly had documented cardiovascular disease, 2) the treatment duration was heterogeneous, and 3) sample sizes were too small to draw clear conclusion.
In our data, initial LV diastolic function was mostly within the normal range despite potential risk factors of enrolled patients for cardiovascular disease. Nevertheless, since E/e’ value linearly correlates with LV filling pressure, its decrement is meaningful for indicating diastolic function improvement 18)19). However, there was no significant decrease in E/e’ value 3 months after administration of SGLT2i in our study. A recent cohort study demonstrated that first-line diabetic treatment with SGLT2i had similar effect for cardiovascular outcome and safety profiles compared to metformin 20). These results might support that most apply a glucocentric approach and recommend other types initial medication such as metformin for most diabetics without cardiovascular disease, leaving SGLT2i as an alternative option.
Generally, LV diastolic dysfunction is highly prevalent in elderly patients whose relaxation of myocardium is impaired with increasing age. Our data showed that the change of E/e’ was significantly different between patients younger and older than 60 years. Considering that SGLT2i induces volume depletion, blood pressure reduction and weight reduction especially in elderly patients, a patient-centered choice of antidiabetic drugs for balancing of benefits and harms across patients with T2DM with different cardiovascular and/or kidney risk is needed.
Study limitations
There are some limitations of the current study. First, the study was designated as a single-arm prospective observation, so we could not compare the effectiveness of SGLT2i on cardiac function with patients who took other types of glucose-lowing agents. Second, this study comprised a small number of patients, but our data enrolled relatively the largest number of patients compared with previous studies dealing with the benefit of SGLT2i on diastolic function, and which met the sample size calculation based on the significance of E/e’ change. Third, the present study was focused on the effect of SGT2i on changes of diastolic parameter as early markers of diabetic cardiomyopathy in patients without cardiovascular disease despite high potential risk factors. Therefore, baseline E/e’ value was not elevated, so it could be limited to detect changes of diastolic parameter by administration of SGLT2i. However, considering that the E/e’ value closely correlates with LV filling pressure and pulmonary capillary wedge pressure in HF with or without decrement of LVEF, we thought that the change of E/e’ within normal range could be meaningful.