INTRODUCTION
sodium-glucose co-transporter 2 inhibitor (SGLT2i) has brought a paradigm shift in the management of type 2 diabetes mellitus (T2DM). Several large-scale randomized controlled trials of SGLT2i have consistently demonstrated cardio-renal benefits with reduced rates of heart failure (HF) hospitalization and adverse renal outcomes [1-8]. More importantly, in patients with HF and a preserved ejection fraction (HFpEF), SGLT2i is the first class of agents that has been recently shown to reduce their risks of HF hospitalization and cardiovascular death [9].
Increased serum ketone levels has been reported with the use of SGLT2i, although its significance remains to be elucidated. While an alternative fuel hypothesis has been proposed to explain the cardiovascular benefits brought by SGLT2i [10], the risk of ketoacidosis, albeit small, has also raised considerable concern among patients and clinicians, especially when used in insulin-treated patients. It is well known that patients with insulin deficiency or insulin-treated patients are at an increased risk of SGLT2i-related ketoacidosis [11]. However, since most studies that evaluated serum ketone response after SGLT2i were conducted in insulin-naïve patients with short duration of T2DM [12-17], the extent of increase in serum ketone levels and its clinical relevance among insulin-treated patients, who often have longer duration of diabetes and potentially more insulin deficient, have not been clearly defined. Therefore, we conducted this 24-week randomized, open-label, active comparator-controlled study to evaluate the effects of SGLT2i on serum ketone levels among patients with T2DM who are inadequately controlled with insulin, in comparison with a dipeptidyl peptidase-4 inhibitor (DPP4i), to control for the effects of improvement in glycaemia. In addition, secondary outcomes including the differences between SGLT2i and DPP4i on cardiometabolic parameters including high-density lipoprotein cholesterol (HDL-C) cholesterol efflux capacity (CEC), liver fat and fibrosis, as well as cardiac function were also evaluated.
METHODS
Study design
The DISTINCTION (Metabolic Responses of Dapagliflozin versus Sitagliptin in Type 2 Diabetes Patients Inadequately Controlled with Insulin Therapy) study (http://www.clinicaltrials.gov; Unique identifier: NCT03959501) was an investigator initiated, single-centre, randomized, open-label, active comparator-controlled interventional study that was designed and executed independent of the funder. The study design and protocol were reviewed and approved by the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (Ref: UW 17-166). Written informed consent was obtained from all recruited participants prior to any study-related procedures.
Study participants
A total of 60 participants were recruited from the Diabetes Clinic of Queen Mary Hospital, Hong Kong, where over 3,000 patients with T2DM were being followed up. In this study, Chinese Individuals who had T2DM, aged between 21 and 75 years were eligible if their body mass index (BMI) was between 21 and 40 kg/m2, and with their glycosylated hemoglobin (HbA1c) between 8% and 10.5% while on single or two doses of insulin therapy (intermediate-acting human insulin, premixed human insulin or insulin analogues) with or without metformin. Participants also had to be on stable insulin doses, which was defined as less than 10% changes in their total daily insulin dose for at least 3 months prior to enrolment. Key exclusion criteria included type 1 diabetes mellitus, history of ketoacidosis, concurrent use of sulphonylurea or loop diuretics, prior use of SGLT2i, DPP4i or glucagon-like peptide-1 receptor agonists in the preceding 3 months, history of intolerance to SGLT2i or DPP4i, an estimated glomerular filtration rate (eGFR) <45 mL/min/1.73 m2 as calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, history of acute or chronic pancreatitis, benign or malignant pancreatic tumours, bladder cancer, severe liver disease with elevated plasma alanine aminotransferase (ALT) ≥5 times the upper limit of normal, active or history of malignancy in the preceding 5 years, and hospitalization for acute illness in the preceding 3 months before enrolment. Individuals who had severe mental disorder, pregnant or breastfeeding were also excluded.
Randomization and study intervention
Eligible participants were randomly assigned in a 1:1 ratio to receive dapagliflozin 10 mg daily or sitagliptin 100 mg daily for 24 weeks. Randomization was performed in blocks and the sequence was generated using a computer-based allocation method. After randomization, all participants remained on the same insulin dose for 12 weeks unless they had hypoglycaemic events, as defined by the presence of hypoglycaemic symptoms with self-monitored blood glucose concentrations of ≤3.9 mmol/L, or hypoglycaemia judged by the investigators. At 12 weeks, insulin doses were titrated to achieve both fasting and pre-prandial blood glucose concentrations between 4.1 and 7.0 mmol/L according to a titration algorithm (Supplementary Table 1).
Clinical and biochemical assessments
A total of four study visits, including the baseline visit, were arranged over a follow-up period of 6 months. Participants attended all study visits after an overnight fast for at least 8 hours. During the baseline visit, demographic data including age, sex, smoking, and alcohol consumption were obtained. Detailed medical and medication histories were ascertained and verified from the Computer Management System of the Hospital Authority, Hong Kong. In each study visit, anthropometric parameters including body weight (BW), height, BMI, waist circumference (WC), and blood pressure (BP) were measured. Central obesity was defined as WC ≥90 cm in men and ≥80 cm in women [18]. Fasting bloods were drawn for plasma glucose, lipid profile, HbA1c, liver function test, serum creatinine, and C-peptide levels. Chronic kidney disease (CKD) was defined as eGFR <60 mL/min/1.73 m2 or the presence of albuminuria with urine albumin to creatinine ratio ≥30 mg/g. Serum ketone (β-hydroxybutyrate [BHB]) and free fatty acid (FFA) levels were determined at baseline, 12 and 24 weeks. In both baseline and final visits, HDL-C CEC and serum high sensitivity C-reactive protein (hsCRP) levels were measured. Moreover, all participants were invited for cardiac and hepatic assessments using two-dimensional echocardiography (ECHO) and vibration-controlled transient elastography (VCTE), respectively (Please see below for details).
Measurements of circulating BHB, FFA, and hsCRP levels
Serum BHB (Abcam ab83390, Abcam, Cambridge, MA, USA) and FFA levels (Roche Diagnostics GmbH, Mannheim, Germany) were measured by colorimetric assays. Serum hsCRP levels were measured with a high-sensitivity, particle-enhanced immune-turbidimetric assay (Roche Diagnostics).
HDL-C CEC measurements
In brief, HDL-CEC was measured in apolipoprotein B (apoB)-depleted serum after removing apoB-containing lipoproteins by polyethylene glycol precipitation. RAW264.7 mouse macrophages (ATCC, Manassas, VA, USA) were seeded at (70,000 cells/well) in 24-multi well plates. Cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) plus 10% fetal bovine serum (FBS) and antibiotics in 5% CO2 for 2 days to reach 70% to 80% confluence. Macrophages were then labelled for 24 hours with 1 µCi/mL of [3H] cholesterol in the presence of 5% FBS. To upregulate ATP-binding cassette transporter A1 (ABCA1) in RAW264.7 cells, 0.3 mM cyclic adenosine monophosphate (Sigma, St. Louis, MO, USA) in 0.2% bovine serum albumin/DMEM medium was added to the cell culture and incubated for another 16 hours. Cells were then washed once with phosphate buffer prior to the addition of 2.5% apoB-depleted serum as HDL fraction for 4 hours incubation. Media were finally removed and cells were lysed in 0.1M NaOH. Samples of both cells and media were counted by liquid scintillation for radioactivity. The efflux of [3H] cholesterol was calculated as the percentage of radiolabel in the media compared with that present in the media plus cells. Background efflux, as measured in the absence of apoB-depleted serum, was subtracted in all experiments.
Echocardiography assessments
Comprehensive transthoracic ECHO examination was performed using commercially available ECHO machines (Vingmed Vivid E9, General Electric Vingmed Ultrasound, Milwaukee, WI, USA) at baseline and follow-up, as previously described [19]. Images were obtained using a 3.5-MHz transducer and digitally stored into three cardiac cycles for analysis by EchoPAC version 112.0 (General Electric Vingmed, Horten, Norway). Inter-ventricular septal dimension and left ventricular (LV) posterior wall thickness at end-diastole (inter-ventricular septal dimension [IVSd] and left ventricular posterior wall thickness at end-diastole [LVPWd], respectively) were measured using a two-dimensional ECHO guided M-mode approach. LV mass was calculated according to the Devereux formula, while LV volumes and ejection fraction (LVEF) were measured using the modified biplane Simpson’s method in both apical four- and two-chamber views. Left atrial volume (LAV) was assessed by single-plane disk summation method in apical four-chamber view. LAV index was determined by LAV divided by body surface area of the participants. Pulsewave and tissue Doppler imaging were applied to assess LV diastolic function in apical four-chamber view. Peak trans-mitral flow velocities in early (E wave) and late diastole (A wave) were measured to calculate the E/A ratio. Deceleration time of the E wave was also determined. Peak velocities of septal and lateral mitral annulus in early diastole (e’) was also measured by tissue Doppler imaging to determine the average E/e’.
Vibration-controlled transient elastography assessments
VCTE assessment was performed using Fibroscan (Echosens, Paris, France) as described previously [20]. Controlled attenuation parameter (CAP) and liver stiffness (LS), which reflect degree of hepatic steatosis and fibrosis, respectively, were measured by two operators with experience in performing over 500 measurements. The inter-observer reliability was satisfactory, as reflected by an intra-class correlation of 0.98 for CAP and 0.97 for LS measurements. Both CAP and LS measurements were represented by the median of 10 reliable measurements, defined when the interquartile range (IQR) was <30%, and with a success rate of >60%. To ensure validity of the results, only CAP values with IQR >40 dB/m were used. All examinations were conducted using the M probe in the first attempt. XL probe was used only when M probe failed to produce valid and reliable measurements, especially in participants with BMI ≥30 kg/m2. Hepatic steatosis was graded by published CAP cut-offs: 248–267, 268–279, and ≥280 dB/m for mild, moderate and severe hepatic steatosis, respectively [21]. Advanced fibrosis (F3) and cirrhosis (F4) were defined by LS cut-offs: F3 9.6 –11.4 kPa and F4 ≥11.5 kPa (M probe); F3 9.3–10.9 kPa and F4 ≥11.0 kPa (XL probe) [22].
Outcomes of interest
The primary outcome was the change in fasting serum BHB levels before and after treatment with either dapagliflozin or sitagliptin for 24 weeks. Secondary outcomes were changes in cardiometabolic measures including BW, BP, HbA1c, lipid, and FFA levels from baseline to week 24. In post hoc analyses, changes in CEC, echocardiographic parameters, CAP and LS measurements from baseline to week 24 were also evaluated.
Statistical analysis
Statistical analysis was performed using IBM SPSS version 26.0 software (IBM Corp., Armonk, New York, USA). All data in the study was analysed based on an intention-to-treat principle, where all participants randomized were included and analysed in the group to which they were originally allocated. Last observation-carried forward method was applied when handling missing data. Data that were not normally distributed as determined by Kolmogorov-Smirnov test, including serum triglyceride, ALT, AST, BHB, and FFA levels, were logarithmically transformed before all analyses. Values were reported as mean±standard deviation (SD), medians with 25th and 75th percentiles (for skewed data), or percentages, as appropriate. Paired t-test was performed to compare changes in continuous variables within each intervention group. Chi-square and independent t-tests were used for comparisons of categorical and continuous variables, respectively. Sex-adjusted P values were presented when comparing changes in variables with significant gender differences, which included serum BHB and FFA levels. Pearson correlation analysis was conducted to determine the associations of changes in BHB levels and echocardiographic parameters with changes in other clinical variables. In all statistical tests, two-sided P values <0.05 were considered significant.
RESULTS
A total of 60 participants were recruited and randomized in a 1:1 ratio to receive either dapagliflozin 10 mg daily (n=30) or sitagliptin 100 mg daily (n=30). All except one participant, who dropped out due to withdrawal of consent, completed the whole study period of 24 weeks. Complete data of serum BHB, biochemical variables and VCTE assessments were available for 29 participants in the dapagliflozin group and 30 participants in the sitagliptin group (Fig. 1). Table 1 summarizes the baseline characteristics of all randomized participants in the study. In both groups, more than 60% of the participants had central obesity and their mean duration of diabetes was more than 17 years. At baseline, over 90% of the participants were on twice daily insulin injections (58% on twice daily premixed human insulin and 42% on twice daily intermediate-acting human insulin). Their glycaemic control however was suboptimal with mean HbA1c levels ≥8.5% (≥69 mmol/mol) in both groups. Importantly, all clinical parameters, including baseline co-morbidities such as hypertension, cardiovascular disease (CVD), and CKD, as well as concomitant medications were well balanced between groups. Fasting C-peptide levels were low in both groups and there were no significant differences. Notably, the prevalence of fatty liver disease at baseline was more than 70%. In both groups, more than 60% and 10% of the study participants had severe hepatic steatosis and advanced liver fibrosis, respectively. Serum BHB and FFA levels were comparable between both groups at baseline.
Table 2 summarizes the changes in clinical and biochemical variables after 24 weeks of study intervention. Both dapagliflozin and sitagliptin led to significant HbA1c lowering (–1.25% vs. –0.98%, or –13.6 mmol/mol vs. –10.7 mmol/mol; P=0.198, respectively) without significant changes in their total insulin doses. With regard to the serum ketone response which is the primary outcome of interest in this study, median fasting serum BHB levels (normal range, 20 to 1,000 µmol/L) increased significantly by 27% from 372 to 472 µmol/L after 24 weeks of dapagliflozin treatment (P<0.05), resulting in significant differences in the change of serum BHB levels between the two groups (90.5 μmol/L vs. –11.9 μmol/L, sex-adjusted P=0.045 for dapagliflozin and sitagliptin, respectively). Although serum BHB levels started to increase after the use of dapagliflozin, the difference comparing with baseline was only significant at 24 weeks but a trend was observed at 12 weeks. In Pearson correlation analysis, change in serum BHB levels positively correlated with change in FFA levels (r=0.433, P=0.017) but not with the other clinical and metabolic parameters (Table 3). However, change in FFA levels were comparable between dapagliflozin and sitagliptin groups from baseline to 24 weeks.
Notably, despite similar glycaemic improvement, dapagliflozin led to significant reductions in BW and WC as compared with sitagliptin (P=0.006 and P=0.028 for BW and WC, respectively). Moreover, these changes were accompanied by significant improvements in hepatic steatosis and fibrosis, as reflected by the reductions in CAP and LS measurements in the dapagliflozin group as compared with sitagliptin (P=0.007 and P=0.022 for CAP and LS, respectively). These improvements seemed to be more readily observed among those with severe hepatic steatosis (CAP >279 dB/m) and significant liver fibrosis (LS ≥5.8 kPa) at baseline (Supplementary Table 2). Moreover, as compared with sitagliptin, both HDL-C levels and CEC significantly improved after the use of dapagliflozin (P=0.041 and P=0.045 for HDL-C and HDL-CEC, respectively).
Among the randomized participants, 50 of them had ECHO performed both at baseline and their final visits, and with all echocardiographic measurements available for analysis (Table 4). At baseline, all echocardiographic parameters were comparable between the dapagliflozin and sitagliptin groups. However, at 24 weeks, average E/e’ was significantly lower in the group treated with dapagliflozin compared with those on sitagliptin (9.68 vs. 11.3, P=0.037, respectively). In Pearson correlation analysis, change in average E/e’ levels inversely correlated with change in HDL-CEC levels (r=–0.405, P=0.045) but not with that of the other clinical variables including serum BHB levels (Supplementary Table 3, Supplementary Fig. 1).
During the whole study period, both dapagliflozin and sitagliptin were well tolerated and none of the participants developed symptoms suggestive of euglycemic diabetic ketoacidosis (DKA). After 24 weeks of treatment, a significantly higher proportion of these insulin-treated participants on dapagliflozin achieved a composite end-point comprising HbA1c reduction ≥1%, weight loss ≥1 kg and absence of hypoglycaemia, than those randomized to sitagliptin (23.3% vs. 3.3%, P=0.023, respectively).
DISCUSSION
The present study demonstrated that among insulin-treated patients with long duration of T2DM, use of dapagliflozin for 24 weeks led to a modest but significant increase in serum ketone levels. Moreover, as compared with sitagliptin, despite similar degree of HbA1c lowering, use of dapagliflozin as addon to insulin therapy provided significant metabolic benefits with weight reduction, improvements in hepatic steatosis and fibrosis, as well as HDL-C and its cholesterol efflux function in patients with T2DM. Indices of LV diastolic function was also significantly better at 24 weeks in patients treated with dapagliflozin. Hence, these cardiometabolic benefits are unlikely mediated by improvement in glycaemia.
Several mechanisms have been proposed to explain the increased serum ketone levels after use of SGLT2i [23]. Although reduced renal clearance of ketone bodies has been reported, studies have suggested that overproduction of ketone bodies played a more major role in causing increased ketone levels during SGLT2 inhibition, especially among those with preserved renal function [24]. SGLT2i disrupts insulin-to-glucagon ratio, and shifts substrate utilization from glucose to lipid oxidation, resulting in accelerated lipolysis, enhanced fat oxidation and ketogenesis [14,25]. Insulin deficiency and brisk reduction of exogenous insulin doses are both risk factors of SGLT2i-related ketoacidosis [11]. Indeed, in a recent study involving 1,278 Japanese insulin-naïve patients with T2DM, it was shown that those with the greatest serum ketone excursion after 24 weeks of canagliflozin treatment had significantly longer duration of diabetes, lower BMI, and baseline serum insulin levels [16]. Therefore, in contrast to previous studies which evaluated ketone responses after short-term use of SGLT2i [13,15,26], or in those insulin-naïve patients with relatively shorter duration of diabetes [12,16,17], our study was the first to evaluate the changes in serum ketone levels after chronic use of SGLT2i, in an exclusively insulin-treated population with a long duration of T2DM (mean duration ≥17 years). Interestingly, the elevation of serum BHB levels after dapagliflozin only became significant at 24 weeks but not at 12 weeks. Moreover, despite the longer duration of diabetes in our participants, their magnitude of elevation in serum ketone levels was modest when compared with the 78% increase reported in a previous study using canagliflozin for a similar duration [16]. These observations are likely due to concomitant insulin treatment in our participants, although intrinsic differences in the pharmacodynamic properties among SGLT2i cannot be ruled out. Further studies are required to validate our findings.
The improvement in HDL-CEC after SGLT2i treatment is another novel finding. ABCA1-related CEC, which has been shown as an independent cardiovascular risk marker in population-based studies [27], is impaired in T2DM [28]. Our findings are in contrast to a previous randomized placebo-controlled trial that failed to demonstrate significant CEC improvement after 12 weeks of dapagliflozin in 31 patients with T2DM [29]. This could be related to differences in the methods of HDL-CEC measurements, as our study utilized the radiolabelled cholesterol method which is more widely used than the fluorescently labelled cholesterol method in their study [30]. Differences in study population could also be another reason. In contrast to their study with only 60% of the participants were on insulin therapy, our study consisted of exclusively insulin-treated patients. Furthermore, it is known that CEC tends to correlate with HDL-C levels [27,30]. However, contrary to ours and most other studies [31], serum HDL-C tended to decrease after dapagliflozin in that study [29]. Therefore, whether this difference in HDL-C responses after SGLT2i could have also contributed to the apparent discordant findings in CEC remains to be confirmed with further studies.
Nonetheless, our observed significant improvement in HDL-C levels with dapagliflozin, as compared with sitagliptin treatment, was in keeping with the overall improvement in metabolic profile after SGLT2i. Indeed, our study showed that dapagliflozin also led to significant attenuation in the severity of fatty liver disease, which is present in more than 70% of patients with T2DM [22]. Previous studies have already demonstrated that dapagliflozin could delay progression of liver fibrosis in patients with T2DM and significant fibrosis on VCTE at baseline [32]. Recently, a pilot study also showed that empagliflozin could alleviate hepatic steatosis, ballooning and fibrosis in patients with biopsy-proven non-alcoholic steatohepatitis and T2DM [33].
To our knowledge, our study is also the first head-to-head prospective study to compare the cardiometabolic efficacy of SGLT2i versus DPP4i as add-on therapy in exclusively insulin-treated patients with T2DM. DPP4i is an appealing option for glycaemic control, owing to its minimal risk of hypoglycaemia, weight-neutral properties and tolerable safety profile. Although previous prospective and real-world retrospective studies with head-to-head comparison between SGLT2i and DPP4i have reported superior cardiovascular benefits of the former, only few participants were on insulin therapy [34,35]. On the other hand, in a recent randomized study involving 44 Japanese patients with short duration of T2DM and without CVD at baseline, no significant difference in cardiac function was observed between participants assigned to empagliflozin and sitagliptin for 12 weeks [36]. Notably, insulin treatment has been associated with adverse cardiac outcomes including diastolic dysfunction in patients with diabetes [37]. Diastolic dysfunction often precedes clinical HF in T2DM [38]. Previous echocardiographic studies have shown that SGLT2i could augment LV diastolic function in patients with T2DM regardless of the presence of clinical HF [39,40]. Recently, empagliflozin was shown to reduce HF hospitalization among patients with HFpEF [9]. In our study, as compared with sitagliptin, significantly lower average E/e’ was observed at 24 weeks after dapagliflozin treatment. Our findings have therefore provided further clinical support that SGLT2i as a class is beneficial to LV diastolic function in patients with normal LVEF, including those with long duration of T2DM and on insulin therapy. Our study also observed a correlation between changes in HDL-CEC and average E/e’ in the dapagliflozin group. Whether there is a relationship between the improvement in HDL-CEC and LV function warrants investigations in further studies.
Our study has several limitations. First, the sample size is small. Although hyperketonaemia has been suggested as one of the possible mechanisms for improved cardiac function after SGLT2i, our sample size might have inadequate power for further post hoc analyses such as the correlations between change in ketones and improvement in LV diastolic function. Secondly, the study period is relatively short, which could have explained the lack of significant changes in other echocardiographic parameters that might take a longer time to improve than the average E/e’. Indeed, a recent echocardiographic study suggested the improvement in LV filling pressure to occur early in the course of SGLT2i treatment, as reflected by the reduction of E/e’ brought by empagliflozin could happen as early as one day after treatment [41]. Moreover, since all our participants had normal LVEF at baseline; thus, rendering it difficult for more in-depth echocardiographic analysis to evaluate for any differential beneficial effects of SGLT2i in patients with preserved and reduced LVEF. Thirdly, this is an open-label study. However, all technicians involved in serum BHB, FFA, and CEC measurements, VCTE operators and the cardiologist were blinded to study treatment allocation of the participants. Moreover, only serum BHB levels, but not the other circulating ketone bodies, were measured in our study. Furthermore, serum glucagon levels were also not assessed in our study participants. Lastly, since our local labelling for eGFR to start SGLT2i was still 45 mL/min/1.73 m2 at the commencement of our study, whether serum ketone response after SGLT2i differs in patients with worse renal function remains to be addressed in further studies.
In conclusion, our study demonstrated that despite similar HbA1c lowering as sitagliptin, treatment with dapagliflozin for 24 weeks significantly reduced adiposity, attenuated fatty liver disease and improved HDL functionality with slightly better LV diastolic function. More importantly, among these insulin-treated patients with long duration of T2DM, serum ketone levels only rose modestly and none of them had symptoms suggestive of euglycemic DKA. That said, both clinicians and patients should always practice caution with the use of SGLT2i, especially in situations known to precipitate euglycemic DKA such as during the perioperative period and concurrent illness [23]. Nonetheless, SGLT2i represents a safe and effective strategy as add-on therapy to insulin both for HbA1c lowering and optimization of the overall cardiometabolic health in patients with T2DM.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2021.0319.
Supplementary Table 1.
The insulin dosing titration algorithm
dmj-2021-0319-suppl1.pdf
Supplementary Table 2.
Subgroup analysis with dapagliflozin-treated participants stratified by the presence of severe hepatic steatosis and significant liver fibrosis at baseline (n=30)
dmj-2021-0319-suppl2.pdf
Supplementary Table 3.
Correlations between change in average E/e’ and changes in other clinical variables among participants in the dapagliflozin group (n=25)
dmj-2021-0319-suppl3.pdf
Supplementary Fig. 1.
Correlations between change in average E/e’ and change in β-hydroxybutyrate (BHB) levels among participants in the dapagliflozin group (n=25).
dmj-2021-0319-suppl4.pdf
NOTES
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: C.H.L., K.C.B.T.
Acquisition, analysis, or interpretation of data: C.H.L., M.Z.W., D.T.W.L., D.S.H.C., C.H.Y.F., S.W.M.S., W.Y., A.C.H.L., J.K.Y.L., Y.C.W., K.K.H.Y.
Drafting the work or revising: C.H.L., K.S.L.L., K.C.B.T.
Final approval of the manuscript: K.C.B.T.
FUNDING
This study was supported in part by funding from Astra Zeneca, and from an Endowment Fund established for the “Sir David Todd Professorship in Medicine” awarded to Kathryn Choon-Beng Tan. Astra Zeneca had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Acknowledgements
We thank all the participants who contributed to this study. We thank Ms. Rachel Wong for her technical assistance in the measurements of serum BHB, FFA, and hsCRP levels.
Table 1.
Baseline characteristics of the study participants
| Baseline variable | Dapagliflozin (n=30) | Sitagliptin (n=30) | P value | ||
|---|---|---|---|---|---|
| Clinical parameters | |||||
| Men, % | 53.3 | 66.7 | 0.292 | ||
| Age, yr | 56.9±10.7 | 60.6±7.03 | 0.120 | ||
| Ever smoker, % | 40.0 | 40.0 | 1.000 | ||
| Duration of diabetes, yr | 17.1±9.56 | 19.3±8.50 | 0.357 | ||
| BMI, kg/m2 | 26.4±3.87 | 26.9±2.90 | 0.570 | ||
| WC, cm | 92.7±8.28 | 92.9±8.19 | 0.949 | ||
| Men | 94.8±8.96 | 95.0±6.85 | 0.952 | ||
| Women | 90.3±6.98 | 88.6±9.34 | 0.615 | ||
| Central obesity, % | 83.3 | 66.7 | 0.136 | ||
| Systolic BP, mm Hg | 132±13.1 | 137±14.5 | 0.143 | ||
| Diastolic BP, mm Hg | 73.1±7.27 | 75.6±7.99 | 0.207 | ||
| Biochemistry parameters | |||||
| Fasting glucose, mmol/L | 9.35±3.32 | 9.82±2.89 | 0.564 | ||
| HbA1c, % | 8.94±0.60 | 8.79±0.56 | 0.321 | ||
| HbA1c, mmol/mol | 74.20±6.57 | 72.60±6.11 | 0.321 | ||
| C-peptide, nmol/L | 0.29 (0.19–0.39) | 0.28 (0.17–0.46) | 0.444 | ||
| HDL-C, mmol/L | 1.25±0.34 | 1.24±0.41 | 0.945 | ||
| HDL-CEC, % | 22.2±5.24 | 21.8±6.32 | 0.792 | ||
| LDL-C, mmol/L | 2.08±0.64 | 1.86±0.77 | 0.235 | ||
| Triglyceridesa, mmol/L | 1.15 (0.90–1.63) | 1.35 (0.88–2.18) | 0.328 | ||
| ALTa, U/L | 22 (16–37) | 23 (19–31) | 0.813 | ||
| ASTa, U/L | 20 (18–29) | 22 (17–24) | 0.447 | ||
| eGFR, mL/min/1.73 m2 | 90.4±19.6 | 81.2±16.1 | 0.050 | ||
| ≥60 mL/min/1.73 m2, % | 90.0 | 93.3 | 1.000 | ||
| Albuminuria status | 0.865 | ||||
| A1, % | 53.3 | 56.7 | |||
| A2, % | 33.33 | 23.3 | |||
| A3, % | 13.3 | 20.0 | |||
| Fasting FFA, µmol/L | 91.5 (63.2–165) | 110 (86.6–160) | 0.278 | ||
| Fasting BHB, µmol/L | 372 (315–521) | 401 (325–514) | 0.389 | ||
| hsCRPa, mg/L | 0.96 (0.51–2.62) | 0.89 (0.42–2.49) | 0.341 | ||
| VCTE | |||||
| CAP, dB/m | 285±48.2 | 292±56.6 | 0.625 | ||
| Minimal <248 dB/m | 23.3 | 20.0 | |||
| Mild 248–267 dB/m | 16.7 | 23.3 | |||
| Moderate 268–279 dB/m | 0.0 | 3.3 | |||
| Severe >279 dB/m | 60.0 | 63.3 | |||
| LSa, kPa | 5.75 (4.33–8.95) | 6.10 (4.98–8.15) | 0.675 | ||
| F0/1 <5.8 kPa | 50.0 | 46.7 | |||
| F2 5.8–9.5 kPa | 36.7 | 43.3 | |||
| F3 9.6–11.4 kPa | 6.6 | 3.3 | |||
| F4 >11.4 kPa | 6.7 | 6.7 | |||
| Medical diseases, % | |||||
| Hypertension | 80.0 | 83.3 | 0.739 | ||
| Coronary artery disease | 13.3 | 13.3 | 1.000 | ||
| Stroke | 6.7 | 6.7 | 1.000 | ||
| STDR | 6.7 | 10.0 | 1.000 | ||
| Concomitant medications, % | |||||
| Metformin | 100 | 100 | 1.000 | ||
| Pioglitazone | 13.3 | 16.7 | 0.718 | ||
| ACEI | 50.0 | 53.3 | 0.796 | ||
| ARB | 27.7 | 26.7 | 1.000 | ||
| Statin | 63.3 | 80.0 | 0.152 | ||
| Fibrate | 0.0 | 10.0 | 0.237 | ||
| Aspirin | 26.7 | 26.7 | 1.000 | ||
| Total daily insulin dosage, units | 42.0±15.9 | 45.0±14.1 | 0.447 | ||
Values are presented as percent, mean±standard deviation, or median (interquartile range). Albuminuria status was assessed with a random urine sample, and categorized according to urine albumin to creatinine ratio (A1: <30 mg/g; A2: ≥30–<300 mg/g; A3 ≥300 mg/g).
BMI, body mass index; WC, waist circumference; BP, blood pressure; HbA1c, glycosylated hemoglobin; HDL-C, high-density lipoprotein cholesterol; CEC, cholesterol efflux capacity; LDL-C, low-density lipoprotein cholesterol; ALT, alanine aminotransferase; AST, aspartate aminotransferase; eGFR, estimated glomerular filtration rate; FFA, free fatty acid; BHB, β-hydroxybutyrate; hsCRP, high sensitivity C-reactive protein; VCTE, vibration-controlled transient elastography; CAP, controlled attenuation parameter; LS, liver stiffness; STDR, sight threatening diabetic retinopathy; ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin II receptor blocker.
a Log-transformed before analysis.
Table 2.
Changes in clinical variables before and after study intervention
| Baseline variable | Dapagliflozin (n=30) | Sitagliptin (n=30) | Change from baseline | P value (Δchange) | ||||
|---|---|---|---|---|---|---|---|---|
| At baseline | At week 24 | At baseline | At week 24 | Dapagliflozin | Sitagliptin | |||
| Primary outcome | ||||||||
| BHBa, μmol/L | 372 (315 to 521) | 472 (349 to 587)c | 401 (325 to 514) | 403 (326 to 524) | 90.5 (–8.10 to 193) | –11.9 (–123 to 79.3) | 0.045b | |
| Secondary outcomes | ||||||||
| Body weight, kg | 70.3±12.8 | 69.3±12.5c | 71.5±11.6 | 72.2±12.8 | –0.94±1.89 | 0.65±2.46 | 0.006 | |
| WC, cm | 92.7±8.28 | 91.6±8.19 | 92.9±8.19 | 94.0±9.02c | –1.09±4.47 | 1.11±2.90 | 0.028 | |
| Systolic BP, mm Hg | 132±13.1 | 129±12.9 | 137±14.5 | 136±16.4 | –2.20±13.2 | –1.18±15.8 | 0.787 | |
| Diastolic BP, mm Hg | 73.1±7.27 | 71.7±8.66 | 75.6±7.99 | 74.9±9.26 | –1.40±6.46 | –0.75±6.61 | 0.701 | |
| Total daily insulin dose, units | 42.0±15.9 | 41.4±15.8 | 45.0±14.1 | 45.8±14.9 | –0.60±3.33 | 0.80±2.66 | 0.077 | |
| FG, mmol/L | 9.35±3.32 | 7.21±2.11d | 9.82±2.89 | 6.79±2.02e | –2.14±3.28 | –3.03±3.32 | 0.298 | |
| HbA1c, % | 8.94±0.60 | 7.69±0.86e | 8.79±0.56 | 7.81±0.93e | –1.25±0.78 | –0.98±0.80 | 0.198 | |
| HbA1c, mmol/mol | 74.20±6.57 | 60.60±9.38e | 72.60±6.11 | 61.80±10.10e | –13.60±0.56 | –10.70±8.78 | 0.198 | |
| eGFR, mL/min/1.73 m2 | 91.9±18.1 | 90.2±19.8 | 81.2±16.1 | 77.7±15.6c | –0.21±5.97 | –3.43±8.50 | 0.092 | |
| HDL-C, mmol/L | 1.25±0.34 | 1.30±0.37 | 1.24±0.41 | 1.20±0.41 | 0.05±0.15 | –0.04±0.18 | 0.041 | |
| LDL-C, mmol/L | 2.08±0.64 | 2.06±0.74 | 1.86± 0.77 | 1.76±0.62 | –0.02±0.47 | –0.11±0.62 | 0.545 | |
| TGa, mmol/L | 1.15 (0.90 to 1.63) | 1.15 (0.80 to 1.53) | 1.35 (0.88 to 2.18) | 1.25 (0.80 to 1.70)c | –0.15 (–0.40 to 0.10) | –0.25 (–0.50 to 0.05) | 0.611 | |
| ALTa, U/L | 21.5 (15.8 to 37.3) | 20.0 (15.8 to 27.5) | 22.5 (19.0 to 30.5) | 27.0 (19.8 to 32.0) | –1.00 (3.00 to –5.50) | 2.00 (–2.00 to 5.25) | 0.284 | |
| ASTa, U/L | 20.0 (17.8 to 29.3) | 19.0 (16.8 to 25.0) | 21.5 (16.8–24.0) | 23.0 (20.8 to 27.3)c | 0.00 (–3.50 to 2.00) | 3.00 (–1.00 to 5.25) | 0.160 | |
| CAP, dB/m | 285±48.2 | 268±50.9c | 292±56.6 | 302 ± 58.0 | –17.8±40.6 | 10.0±37.0 | 0.007 | |
| LS, kPa | 5.75 (4.33 to 8.95) | 5.15 (4.58 to 6.75)c | 6.10 (4.98 to 8.15) | 7.00 (5.10 to 8.58) | –0.50 (–1.25 to 0.15) | 0.25 (–0.50 to 1.20) | 0.022 | |
| HDL-CEC, % | 22.2±5.32 | 24.8±7.60c | 21.8± 6.32 | 21.3±6.43 | 2.66±5.14 | –0.50±6.71 | 0.045 | |
| hsCRPa, mg/L | 0.96 (0.51 to 2.62) | 0.63 (0.42 to 2.05) | 0.89 (0.42 to 2.49) | 0.97 (0.42 to 2.22) | –0.19 (–0.95 to 0.002) | –0.10 (–0.60 to 0.63) | 0.177 | |
| FFAa, µmol/L | 91.6 (64.1 to 164) | 119 (69.0 to 164) | 110 (86.6 to 160) | 106 (56.6 to 174) | 9.58 (–41.8 to 45.0) | –6.72 (–43.6 to 26.5) | 0.191b | |
Values are presented as median (interquartile range) or mean±standard deviation.
BHB, β-hydroxybutyrate; WC, waist circumference; BP, blood pressure; FG, fasting glucose; HbA1c, glycosylated hemoglobin; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglyceride; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAP, controlled attenuation parameter; LS, liver stiffness; CEC, cholesterol efflux capacity; hsCRP, high sensitivity C-reactive protein; FFA, free fatty acid.
a Log-transformed before analysis,
b Sex-adjusted P value,
c <0.05,
d <0.01,
e <0.001 vs. baseline.
Table 3.
Correlations between change in serum BHB levels and changes in other clinical and metabolic variables among participants in the dapagliflozin group (n=30)
| Change in clinical and metabolic variable | Change in BHB levels, μmol/L | |
|---|---|---|
| Crude r | P value | |
| Body weight, kg | 0.111 | 0.560 |
| Waist circumference, cm | –0.022 | 0.910 |
| Systolic BP, mm Hg | 0.017 | 0.929 |
| Diastolic BP, mm Hg | 0.239 | 0.204 |
| Total daily insulin dose, units | 0.221 | 0.241 |
| Fasting glucose, mmol/L | 0.068 | 0.723 |
| HbA1c, % | 0.140 | 0.461 |
| Triglyceridea, mmol/L | 0.188 | 0.319 |
| LDL-C, mmol/L | –0.344 | 0.063 |
| HDL-C, mmol/L | –0.052 | 0.783 |
| HDL-CEC, % | 0.107 | 0.575 |
| ALTa, U/L | 0.088 | 0.644 |
| ASTa, U/L | 0.047 | 0.804 |
| eGFR, mL/min/1.73 m2 | 0.011 | 0.956 |
| CAP, dB/m | 0.239 | 0.204 |
| LSa, kPa | –0.208 | 0.270 |
| FFAa, µmol/L | 0.433 | 0.017 |
| hsCRPa | –0.206 | 0.274 |
BHB, β-hydroxybutyrate; BP, blood pressure; HbA1c, glycosylated hemoglobin; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; CEC, cholesterol efflux capacity; ALT, alanine aminotransferase; AST, aspartate aminotransferase; eGFR, estimated glomerular filtration rate; CAP, controlled attenuation parameter; LS, liver stiffness; FFA, free fatty acid; hsCRP, high sensitivity C-reactive protein.
a Log-transformed before analysis.
Table 4.
Changes in echocardiographic parameters before and after treatment with study intervention
| Echocardiographic parameter | Dapagliflozin (n=25) | Sitagliptin (n=25) | Change from baseline | P value (Δchange) | |||
|---|---|---|---|---|---|---|---|
| At baseline | At week 24 | At baseline | At week 24 | Dapagliflozin | Sitagliptin | ||
| IVSd, mm | 11.00±1.42 | 11.30±1.55 | 11.50±1.55 | 11.40±1.76 | 0.18±1.20 | –0.47±1.07 | 0.374 |
| LVPWd, mm | 9.40±1.58 | 9.26±1.34 | 9.60±1.79 | 9.29±1.21 | –0.08±1.26 | –0.41±1.59 | 0.680 |
| LV mass, g | 92.00±19.60 | 90.80±16.60 | 94.70±20.20 | 95.30±16.70 | –1.03±13.50 | 0.27±11.50 | 0.615 |
| LVEF, % | 62.20±9.90 | 62.20±6.56 | 64.30±5.10 | 64.30±5.37 | 0.33±9.42 | –0.15±5.19 | 0.970 |
| E/A | 0.86±0.20 | 0.87±0.20 | 0.98±0.32 | 0.96±0.31 | –0.02±0.18 | –0.03±0.21 | 0.562 |
| DT, ms | 208.2±43.3 | 238.0±28.9a | 213.6±41.4 | 238.0±47.5b | 29.9±45.7 | 24.4±49.0 | 0.687 |
| Septal e’, m/s | 0.0744±0.02 | 0.0744±0.02 | 0.0680±0.02 | 0.0676±0.02 | 0.0004±0.02 | –0.0007±0.01 | 0.928 |
| Lateral e’, m/s | 0.094±0.02 | 0.099±0.02 | 0.097±0.02 | 0.094±0.03 | 0.007±0.01 | –0.005±0.02 | 0.093 |
| Average E/e’ | 10.20±2.64 | 9.68±2.77c | 11.60±2.52 | 11.30±2.67c | –0.55±1.81 | –0.22±2.36 | 0.583 |
| LAVi, mL/m2 | 31.50±7.50 | 30.60±8.27 | 32.20±8.75 | 33.60±12.40 | –0.46±7.24 | 1.07±11.50 | 0.394 |
Values are presented as mean±standard deviation.
IVSd, inter-ventricular septal dimension; LVPWd, left ventricular posterior wall thickness at end-diastole; LV, left ventricular; LVEF, left ventricular ejection fraction; E wave, peak trans-mitral flow velocities in early diastole; A wave, peak trans-mitral flow velocities in late diastole; DT, deceleration time; e’, peak velocities of septal and lateral mitral annulus in early diastole; LAVi, left atrial volume divided by body surface area of the participants.
Paired t-test: aP<0.01, bP<0.05; cP=0.037 for the difference of average E/e’ between dapagliflozin and sitagliptin at 24 weeks.
1. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 2015;373:2117-28.
2. Neal B, Perkovic V, Mahaffey KW, de Zeeuw D, Fulcher G, Erondu N, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017;377:644-57.
3. Wiviott SD, Raz I, Bonaca MP, Mosenzon O, Kato ET, Cahn A, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med 2019;380:347-57.
4. Perkovic V, Jardine MJ, Neal B, Bompoint S, Heerspink HJL, Charytan DM, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med 2019;380:2295-306.
5. McMurray JJ, Solomon SD, Inzucchi SE, Kober L, Kosiborod MN, Martinez FA, et al. Dapagliflozin in patients with heart failure and reduced ejection fraction. N Engl J Med 2019;381:1995-2008.
6. Packer M, Anker SD, Butler J, Filippatos G, Pocock SJ, Carson P, et al. Cardiovascular and renal outcomes with empagliflozin in heart failure. N Engl J Med 2020;383:1413-24.
7. Bhatt DL, Szarek M, Steg PG, Cannon CP, Leiter LA, McGuire DK, et al. Sotagliflozin in patients with diabetes and recent worsening heart failure. N Engl J Med 2021;384:117-28.
8. Bhatt DL, Szarek M, Pitt B, Cannon CP, Leiter LA, McGuire DK, et al. Sotagliflozin in patients with diabetes and chronic kidney disease. N Engl J Med 2021;384:129-39.
9. Anker SD, Butler J, Filippatos G, Ferreira JP, Bocchi E, Bohm M, et al. Empagliflozin in heart failure with a preserved ejection fraction. N Engl J Med 2021;385:1451-61.
10. Heerspink HJ, Perkins BA, Fitchett DH, Husain M, Cherney DZ. Sodium glucose cotransporter 2 inhibitors in the treatment of diabetes mellitus: cardiovascular and kidney effects, potential mechanisms, and clinical applications. Circulation 2016;134:752-72.
11. Goldenberg RM, Berard LD, Cheng AY, Gilbert JD, Verma S, Woo VC, et al. SGLT2 inhibitor-associated diabetic ketoacidosis: clinical review and recommendations for prevention and diagnosis. Clin Ther 2016;38:2654-64.
12. Inagaki N, Kondo K, Yoshinari T, Kuki H. Efficacy and safety of canagliflozin alone or as add-on to other oral antihyperglycemic drugs in Japanese patients with type 2 diabetes: a 52-week open-label study. J Diabetes Investig 2015;6:210-8.
13. Ferrannini E, Baldi S, Frascerra S, Astiarraga B, Heise T, Bizzotto R, et al. Shift to fatty substrate utilization in response to sodium-glucose cotransporter 2 inhibition in subjects without diabetes and patients with type 2 diabetes. Diabetes 2016;65:1190-5.
14. Daniele G, Xiong J, Solis-Herrera C, Merovci A, Eldor R, Tripathy D, et al. Dapagliflozin enhances fat oxidation and ketone production in patients with type 2 diabetes. Diabetes Care 2016;39:2036-41.
15. Yabe D, Iwasaki M, Kuwata H, Haraguchi T, Hamamoto Y, Kurose T, et al. Sodium-glucose co-transporter-2 inhibitor use and dietary carbohydrate intake in Japanese individuals with type 2 diabetes: a randomized, open-label, 3-arm parallel comparative, exploratory study. Diabetes Obes Metab 2017;19:739-43.
16. Polidori D, Iijima H, Goda M, Maruyama N, Inagaki N, Crawford PA. Intra- and inter-subject variability for increases in serum ketone bodies in patients with type 2 diabetes treated with the sodium glucose co-transporter 2 inhibitor canagliflozin. Diabetes Obes Metab 2018;20:1321-6.
17. Kim SR, Lee SG, Kim SH, Kim JH, Choi E, Cho W, et al. SGLT2 inhibition modulates NLRP3 inflammasome activity via ketones and insulin in diabetes with cardiovascular disease. Nat Commun 2020;11:2127.
18. Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009;120:1640-5.
19. Wu MZ, Lee CH, Chen Y, Yu SY, Yu YJ, Ren QW, et al. Association between adipocyte fatty acid-binding protein with left ventricular remodelling and diastolic function in type 2 diabetes: a prospective echocardiography study. Cardiovasc Diabetol 2020;19:197.
20. Lee CH, Seto WK, Lui DT, Fong CH, Wan HY, Cheung CY, et al. Circulating thrombospondin-2 as a novel fibrosis biomarker of nonalcoholic fatty liver disease in type 2 diabetes. Diabetes Care 2021;44:2089-97.
21. Karlas T, Petroff D, Sasso M, Fan JG, Mi YQ, de Ledinghen V, et al. Individual patient data meta-analysis of controlled attenuation parameter (CAP) technology for assessing steatosis. J Hepatol 2017;66:1022-30.
22. Kwok R, Choi KC, Wong GL, Zhang Y, Chan HL, Luk AO, et al. Screening diabetic patients for non-alcoholic fatty liver disease with controlled attenuation parameter and liver stiffness measurements: a prospective cohort study. Gut 2016;65:1359-68.
23. Fleming N, Hamblin PS, Story D, Ekinci EI. Evolving evidence of diabetic ketoacidosis in patients taking sodium-glucose cotransporter 2 inhibitors. J Clin Endocrinol Metab 2020;105:dgaa200.
24. Ferrannini E, Baldi S, Frascerra S, Astiarraga B, Barsotti E, Clerico A, et al. Renal handling of ketones in response to sodium-glucose cotransporter 2 inhibition in patients with type 2 diabetes. Diabetes Care 2017;40:771-6.
25. Ferrannini E, Muscelli E, Frascerra S, Baldi S, Mari A, Heise T, et al. Metabolic response to sodium-glucose cotransporter 2 inhibition in type 2 diabetic patients. J Clin Invest 2014;124:499-508.
26. Al Jobori H, Daniele G, Adams J, Cersosimo E, Triplitt C, DeFronzo RA, et al. Determinants of the increase in ketone concentration during SGLT2 inhibition in NGT, IFG and T2DM patients. Diabetes Obes Metab 2017;19:809-13.
27. Rohatgi A, Khera A, Berry JD, Givens EG, Ayers CR, Wedin KE, et al. HDL cholesterol efflux capacity and incident cardiovascular events. N Engl J Med 2014;371:2383-93.
28. He Y, Ronsein GE, Tang C, Jarvik GP, Davidson WS, Kothari V, et al. Diabetes impairs cellular cholesterol efflux from ABCA1 to small HDL particles. Circ Res 2020;127:1198-210.
29. Fadini GP, Bonora BM, Zatti G, Vitturi N, Iori E, Marescotti MC, et al. Effects of the SGLT2 inhibitor dapagliflozin on HDL cholesterol, particle size, and cholesterol efflux capacity in patients with type 2 diabetes: a randomized placebo-controlled trial. Cardiovasc Diabetol 2017;16:42.
30. Ouimet M, Barrett TJ, Fisher EA. HDL and reverse cholesterol transport. Circ Res 2019;124:1505-18.
31. Zaccardi F, Webb DR, Htike ZZ, Youssef D, Khunti K, Davies MJ. Efficacy and safety of sodium-glucose co-transporter-2 inhibitors in type 2 diabetes mellitus: systematic review and network meta-analysis. Diabetes Obes Metab 2016;18:783-94.
32. Shimizu M, Suzuki K, Kato K, Jojima T, Iijima T, Murohisa T, et al. Evaluation of the effects of dapagliflozin, a sodium-glucose co-transporter-2 inhibitor, on hepatic steatosis and fibrosis using transient elastography in patients with type 2 diabetes and non-alcoholic fatty liver disease. Diabetes Obes Metab 2019;21:285-92.
33. Lai LL, Vethakkan SR, Nik Mustapha NR, Mahadeva S, Chan WK. Empagliflozin for the treatment of nonalcoholic steatohepatitis in patients with type 2 diabetes mellitus. Dig Dis Sci 2020;65:623-31.
34. Pasternak B, Ueda P, Eliasson B, Svensson AM, Franzen S, Gudbjornsdottir S, et al. Use of sodium glucose cotransporter 2 inhibitors and risk of major cardiovascular events and heart failure: Scandinavian register based cohort study. BMJ 2019;366:l4772.
35. Fuchigami A, Shigiyama F, Kitazawa T, Okada Y, Ichijo T, Higa M, et al. Efficacy of dapagliflozin versus sitagliptin on cardiometabolic risk factors in Japanese patients with type 2 diabetes: a prospective, randomized study (DIVERSITY-CVR). Cardiovasc Diabetol 2020;19:1.
36. Hiruma S, Shigiyama F, Hisatake S, Mizumura S, Shiraga N, Hori M, et al. A prospective randomized study comparing effects of empagliflozin to sitagliptin on cardiac fat accumulation, cardiac function, and cardiac metabolism in patients with early-stage type 2 diabetes: the ASSET study. Cardiovasc Diabetol 2021;20:32.
37. Shen L, Rorth R, Cosmi D, Kristensen SL, Petrie MC, Cosmi F, et al. Insulin treatment and clinical outcomes in patients with diabetes and heart failure with preserved ejection fraction. Eur J Heart Fail 2019;21:974-84.
38. From AM, Scott CG, Chen HH. The development of heart failure in patients with diabetes mellitus and pre-clinical diastolic dysfunction a population-based study. J Am Coll Cardiol 2010;55:300-5.
39. Matsutani D, Sakamoto M, Kayama Y, Takeda N, Horiuchi R, Utsunomiya K. Effect of canagliflozin on left ventricular diastolic function in patients with type 2 diabetes. Cardiovasc Diabetol 2018;17:73.
40. Soga F, Tanaka H, Tatsumi K, Mochizuki Y, Sano H, Toki H, et al. Impact of dapagliflozin on left ventricular diastolic function of patients with type 2 diabetic mellitus with chronic heart failure. Cardiovasc Diabetol 2018;17:132.
41. Rau M, Thiele K, Hartmann NK, Schuh A, Altiok E, Mollmann J, et al. Empagliflozin does not change cardiac index nor systemic vascular resistance but rapidly improves left ventricular filling pressure in patients with type 2 diabetes: a randomized controlled study. Cardiovasc Diabetol 2021;20:6.
© 2022. This work is licensed under http://creativecommons.org/licenses/by-nc/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Mei-Zhen Wu; Lui, David Tak-Wai; Chan, Darren Shing-Hei; Carol Ho-Yi Fong; Shiu, Sammy Wing-Ming; Wong, Ying; Lee, Alan Chun-Hong; Joanne King-Yan Lam; Yu-Cho, Woo; Karen Siu-Ling Lam; Yiu, Kelvin Kai-Hang; Tan, Kathryn Choon-Beng