Metformin versus sulphonylureas for new onset atrial fibrillation and stroke in type 2 diabetes mellitus: a population-based study

doi:10.21203/rs.3.rs-685526/v1

Download PDF

Research Article

Metformin versus sulphonylureas for new onset atrial fibrillation and stroke in type 2 diabetes mellitus: a population-based study

https://doi.org/10.21203/rs.3.rs-685526/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 03 Feb, 2022

Read the published version in Acta Diabetologica →

Version 1

posted

You are reading this latest preprint version

Background

To compare the rates of incident atrial fibrillation (AF), stroke, cardiovascular mortality, and all-cause mortality between metformin and sulphonylurea users in type 2 diabetes mellitus.

Methods

This was a retrospective population-based cohort study of type 2 diabetes mellitus patients receiving either sulphonylurea or metformin monotherapy between January 1 st , 2000 and December 31 st , 2019. The primary outcome was new-onset AF or stroke. Secondary outcomes were cardiovascular, non-cardiovascular and all-cause mortality. Propensity score matching (1:2 ratio) between sulphonylurea and metformin users was performed, based on demographics, CHA-DS-VASc score, Charlson comorbidity index, past comorbidities, and medication use. Cox regression was used to identify significant risk factors. Competing risk analysis was conducted using cause-specific and subdistribution hazard models. Sensitivity analysis using propensity score stratification, high dimensional propensity score and inverse probability of treatment weighting were conducted.

Results

A total of 36228 sulphonylurea users and 72456 metformin users were included in the propensity score-matched cohort. Multivariable Cox regression showed that sulphonylurea users had higher risks of incident AF (hazard ratio [HR]: 2.89, 95% confidence interval [CI]: 2.75-2.77; P<0.0001), stroke (HR: 3.23, 95% CI: 3.01-3.45; P<0.0001), cardiovascular mortality (HR: 3.60, 95% CI: 2.62-4.81; P<0.0001), and all-cause mortality (HR: 4.35, 95% CI: 3.16-4.75; P<0.0001) compared to metformin users. Similarly significant results were observed using cause-specific and subdistribution hazard models. Sensitvity analysis using other propensity score techniques also yielded higher risks in sulphonylurea users.

Conclusions

Sulphonylurea use was associated with higher risks of incident AF, stroke, cardiovascular mortality and all-cause mortality compared to metformin.

Critical Care & Emergency Medicine

Molecular Genetics

Cardiac & Cardiovascular Systems

Biomedical Engineering

Sulphonylurea

metformin

diabetes

atrial fibrillation

stroke

big data

Type 2 diabetes mellitus (T2DM) is one of the most prevalent chronic disorders around the world which leads to significant morbidity and mortality [1]. It is predicted that by 2040, T2DM would have a global prevalence of 600 million people [2, 3]. T2DM is associated with multisystem complications including atrial fibrillation (AF) and stroke [4]. The manifestations of metabolic syndrome transform the epicardial adipose tissue to cause fibrosis, resulting in atrial remodelling [5]. Furthermore, the sympathetic activity is also increased in T2DM patients [6]. Hence, it has been reported that T2DM patients has a 40% increase in AF risk [7, 8]. Besides, The T2DM insulin resistance state leads to endothelium-dependent vasodilation impairment [9]. The T2DM induced hyperglycemia also damages the vessels by reducing the endothelium-derived NO via inhibiting the phosphatidylinositol 3 kinase pathway. This leads to increase inflammation and formation of foam cells [10]. As such, the T2DM patients also have an increased risk of stroke incidences due to the direct or indirect diabetic-induced changes in the cerebral vessels.

Glycaemic control is essential to reduce the occurrence of AF and stroke in patients with T2DM [8, 11]. Yet, studies investigating the risk of new-onset AF and stroke in diabetic patients on different diabetes medications are scarce. Metformin is the first-line antidiabetic drug treatment recommended by international guidelines, given its high efficacy and safety in ameliorating metabolic controls [12]. Metformin has been reported to reduce stroke severity, cardiovascular caused mortality, and all-cause mortality among T2DM patients [13]. On the other hand, sulphonylurea is also another commonly prescribed antidiabetic agent for patients who are refractory to metformin [14]. Sulphonylurea suspected to be associated with increases the risk of hypoglycaemic events, which raises the risk of adverse cardiovascular events and mortality [15]. However, some studies indicated that sulphonylurea does not increase all-cause mortality or cardiovascular mortality compared to metformin [16]. The following study aims to compare the outcomes of new-onset AF and stroke between metformin and sulphonylurea users amongst T2DM patients in order to provide insights on the cardiovascular effects of antidiabetic medication.

Study design and population

This retrospective population-based cohort study received approval from The Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee. The inclusion criteria were patients with T2DM between January 1st, 2000 to December 31st, 2019. Patients who received both sulphonylureas and metformin on follow-up, did not receive either sulphonylureas or metformin, and those with prior AF or stroke were excluded. The patients were identified from the Clinical Data Analysis and Reporting System (CDARS). This territory-wide database centralizes patient information from affiliated local hospitals to establish comprehensive clinical and medical data, including demographics, past comorbidities, medications, and laboratory data. CHA-DS-VASc score and Charlson comorbidity index were also calculated. The detailed standard ICD-9 codes to identity prior comorbidities and outcomes of new-onset AF or stroke are shown in Supplementary Table 1.

Different variability measures of fasting blood glucose and HbA1c were calculated, including standard deviation (SD), absolute successive variability score, percentage successive variability score, normalized score, normalized absolute successive variability score, normalized percentage successive variability score, coefficient of variation, SD/initial value, and variability independent of mean. The detailed definitions of the calculations of these variability measures were provided in Supplementary Table 2.

Statistical analysis and outcomes

The study outcomes were new-onset AF, new-onset stroke, cardiovascular mortality and all-cause mortality after the initiation of sulphonylurea or metformin. Patients were followed up to the endpoints of new-onset AF/stroke, mortality or the study end (December 31st, 2019). Hong Kong Death Registry, an official government registry that registered death records of the Hong Kong population, provided the mortality data. There was no adjudication of the outcomes, as this relied on the ICD-9 coding or record in the death registry. However, the coding was performed by the clinicians or administrative staff, who were not involved in the mode development.

Descriptive statistics were used to summarise the characteristics of the patient cohort. Mean (standard deviation [SD]) are utilised to depict the continuous variables and presented count (%) were applied to illustrate categorical variables. A standardised mean difference (SMD) of no more than 0.2 between the treatment groups post-weighting was deemed insignificant. Propensity score matching with a 1:2 ratio was used to create the control cohort of sulphonylureas vs. metformin drug use based on demographics, CHA-DS-VASc score, Charlson comorbidity index, past comorbidities, and medication history.

Univariate Cox regression was used to determine the significant risk factors and uncover the exposure effects of metformin and sulphonylurea on the adverse outcomes. Multivariate Cox regression models were employed with adjustments based on the significant confounding elements. Hazard ratios (HRs) with corresponding 95% confidence interval (CI) and P value were reported. Cause-specific hazard models and subdistribution hazard models were conducted to further discern possible competing risks.

Other matching approaches are also tried to identify the treatment effects of metformin versus sulphonylureas. This includes propensity score stratification [17], propensity store matching with inverse probability weighting [18], and high-dimensional propensity score approach [19]. PS stratification splits the dataset into several strata on the basis of the individual’s PS alone without reference to the treatment group. The treatment effect is then estimated within each stratum, and an overall estimated treatment effect is calculated by taking a weighted average across each strata. Inverse probability weighting uses the whole dataset but reweights individuals to increase the weights of those who received unexpected exposures. It effectively generates a pseudopopulation with near-perfect covariate balance between treatment groups. IPW applies weights corresponding to 1/PS for patients in the treated cohort and [1/(1 − PS)] for those in the control cohort. The high-dimensional propensity score (HDPS) is an automated data-driven or empirical approach to deriving variables from administrative data for inclusion in propensity score models.

The two-tailed significance tests were regarded as significant if the P values were less than 0.05. There was no imputation performed for missing data. No blinding was performed for the predictor as the values were obtained from the electronic health records automatically. Data were analysed using RStudio software (Version: 1.1.456) and Python (Version: 3.6).

Baseline characteristics

Initially, the cohort comprised of 273876 T2DM patients. Upon excluding patients that match the exclusion criteria (8809 patients with both medication use, 79186 without either medication use, and 10946 patients with prior diagnosis of AF or stroke) yields a study cohort of 174935 patients (Fig. 1). The cohort involved 46.61% males and had initial drug exposure at a mean age of 64.5 years old [SD: 12.2]) after a mean follow-up duration of 3535.1 days (SD: 1250.4). Of these, 36228 (20.7%) patients received sulphonylurea and 138707 (79.3%) received metformin. Amongst this cohort, 50304 patients exhibited 10925 new-onset AF (IR = 0.26%) and 15233 new-onset stroke (9.7%), 450 cardiovascular caused mortality (incidence ratio (IR) = 0.26%), non-cardiovascular mortality and all-cause mortality (IR = 38.75%). The comparisons of the baseline and clinical characteristics between T2DM patients with sulphonylurea and metformin monotherapy before and after propensity score matching are shown in Table 1. The baseline and clinical characteristics of type 2 diabetes patients stratified by new-onset AF or stroke outcomes are detailed in Supplementary Tables 3 and 4, respectively. The two groups showed a similar density of propensity scores after matching (Supplementary Fig. 1). No significant baseline and clinical characteristics differences were observed between the two groups of patients.

Outcome Predictors

Compared to metformin monotherapy users, the sulphonylurea monotherapy users had higher rates of new onset AF (before: 15.54% vs. 3.81%, SMD = 0.40; after: 15.54% vs. 5.44%, SMD = 0.33), stroke (before: 21.33% vs. 5.41%, SMD = 0.48; after: 21.33% vs. 6.8%, SMD = 0.43), and all-cause mortality (before: 71.94% vs. 17.47%, SMD = 1.31; after: 71.94% vs. 24.7%, SMD = 1.07) after 1:2 propensity score matching. Kaplan-Meier survival curves of new-onset AF, new-onset stroke, cardiovascular mortality, and all-cause mortality, after 1:2 propensity score matching, stratified by sulphonylurea and metformin use, shows that metformin users are less likely to experience these adverse outcomes (Fig. 2 and Fig. 3)

Significant univariate predictors for all-cause mortality, cardiovascular mortality, new onset AF and new onset stroke after 1:2 propensity score matching are shown in Table 2. Compared with metformin monotherapy, sulphonylurea monotherapy demonstrated adverse medication effects on cardiovascular mortality (HR: 3.78 [3.05, 4.68], P value < 0.0001), all-cause mortality (HR: 4.25 [4.17, 4.34], P value < 0.0001), new onset AF (HR: 2.84[2.73, 2.96], P value < 0.0001), and new onset stroke (HR: 3.18[3.07, 3.3], P value < 0.0001). The trend remains the same after propensity score matching and adjusting for significant demographics, CHA-DS-VASc score, past comorbidities, and non-sulphonylurea/metformin medications in the multivariate Cox model. The trend that sulphonylurea are associated with higher risks of new-onset AF, new-onset stroke, cardiovascular mortality and all cause mortality are consistent after matching via different approaches (Table 3).

Competing risk analysis

Cause-specific and subdistribution hazard models demonstrates that sulphonulurea users have an increased risk for cardiovascular mortality, all-cause mortality, new-onset AF and new-onset strokein comparison to metformin users (HR > 1, P value < 0.0001) (Table 3). Correspondingly, the unbiased estimates of the cumulative incidence functions with competing risk analysis for cardiovascular mortality, all-cause mortality, new-onset AF and new-onset stroke stratified by sulphonylurea and. metformin use before and after 1:2 propensity score matching were presented in Fig. 4.

The main finding for the present study is that in comparison to metformin users, sulphonylurea users have a higher risk of new-onset AF, new-onset stroke, cardiovascular mortality, and all-cause mortality.

Metformin and sulphonylurea are two of the most prescribed drugs for initial diabetic control. However, it was uncertain if the two drugs would have different clinical outcomes. In accord with the previous data, our data revealed that sulphonylurea would be more perilous to heart failure and cardiovascular mortality [20]. This is partly explained by the fact that metformin enhances insulin sensitivity while sulphonylurea acts by increasing endogenous insulin levels. As such, metformin can help with weight reduction, and causes fewer hypoglycemic events compared to sulphonylurea [21, 22]. Besides increased cardiovascular mortality, the raise in all-cause mortality amongst sulphonylurea users may be due to drug-induced hyperinsulinemia. This is consistent with the findings in several studies from different part of the world [23–25]. It has been proposed that hyperinsulinaemia would raise the risk of cancers risks by increasing mitogenesis [26]. There are reports demonstrating an increased risk of cancer in patients on sulphonylurea in comparison to metformin users [27]. Besides, sulphonylurea was also shown to increase risks of serious pneumonia, renal function decline and dementia [28–30].

An abundance of evidence suggested strong associations between T2DM and AF occurrences [31]. This is not only because AF and T2DM have mutual risk factors such as atherosclerosis and hypertension; T2DM is also an independent risk factor for AF occurrence, as shown in the Framingham study [32]. Reasonable glycemic control would, therefore, reduce the AF onset as well as lessen the mortality rate among the patients with AF. As with our investigations, metformin has been demonstrated to bring down the risk of AF and ventricular arrhythmias compared to sulphonylureas [33]. Metformin has antioxidant effects and anti-inflammatory effects that reduce atrial fibrosis and remodeling [34, 35]. Meanwhile, the impact of sulphonylurea on the AF onset was less well investigated. The sulphonylurea-induced potassium channel blockade is hypothesized to inhibit the ischemic preconditioning, shorten the action potential duration and increase the cardiac excitability, resulting in AF [36].

Metformin was also exemplified to have a lower risk of stroke onset compared to sulphonylurea. In fact, metformin can reduce stroke onset independent of its glycemic control [37]. Metformin activation of the glial cells AMPK have neuroprotective effects [38]. Similar to its effect on AF, metformin also demonstrates extra anti-inflammatory by activating the nuclear factor erythroid 2-related factor (Nrf2) pathways in neurons. It also has antioxidant effects by increasing glutathione and catalase levels [39]. In contrast, sulphonylurea-induced hypoglycemia contributes to stroke occurrence. Hypoglycemic events have been reported to increase stroke risk via pathways such as increased platelet activation, thrombotic and atherosclerotic cytokine release, and inflammation [40, 41]. Furthermore, sulphonylurea stimulates the release of atherogenic C-peptide [42]. Therefore, it is expectable that metformin has a more protective effect than sulphonylurea in terms of AF and stroke.

There are some limitations in this study that should be acknowledged. Firstly, the data is from a single locality, hence external validation is needed to explore the applicability of findings onto other populations. Secondly, important clinical factors, such as smoking history and physical activity are not routinely coded into the local health database, which are significant predictors for diabetes and the primary outcomes. Finally, this study only shows the association between metformin and reducing AF and stroke outcomes but cannot elucidate causal relationships. There is a need for further studies to investigate the mechanism of how metformin exerts those additional protective effects compared to sulphonylurea.

Compared to metformin users, sulphonylurea use was significantly associated with higher risks of new onset AF, new onset stroke, cardiovascular mortality, and all-cause mortality.

Ethics approval and consent to participate: This retrospective population-based cohort study received approval from The Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee

Consent for publication: All authors have read and approved submission of the manuscript which has not been published and is not being considered for publication elsewhere in whole or part in any language.

Availability of data and materials: The data will be available upon requests to the corresponding authors

Competing interests: The authors have no conflicts of interest to declare.

Funding: There is no funding for this project to declare.

Authors' contributions:
JD, GZ, CC, OHIC: data analysis, data interpretation, statistical analysis, manuscript drafting, critical revision of manuscript

SL, KSKL, WTW, TL, AKCW, SHC: project planning, data acquisition, data interpretation, critical revision of manuscript

QZ, GT: study conception, study supervision, project planning, data interpretation, statistical analysis, manuscript drafting, critical revision of manuscript

Acknowledgements: Not applicable.

Bertoni AG, et al. Diabetes-related morbidity and mortality in a national sample of U.S. elders. Diabetes Care. 2002;25(3):471–5.
Osei K. Global epidemic of type 2 diabetes: implications for developing countries. Ethn Dis. 2003;13(2 Suppl 2):S102-6.
Ogurtsova K, et al. IDF Diabetes Atlas: Global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Res Clin Pract. 2017;128:40–50.
Lin PJ, et al. Multiple chronic conditions in type 2 diabetes mellitus: prevalence and consequences. Am J Manag Care. 2015;21(1):e23–34.
Watanabe H, et al. Metabolic syndrome and risk of development of atrial fibrillation: the Niigata preventive medicine study. Circulation. 2008;117(10):1255–60.
Huggett RJ, et al. Impact of type 2 diabetes mellitus on sympathetic neural mechanisms in hypertension. Circulation. 2003;108(25):3097–101.
Huxley RR, et al. Meta-analysis of cohort and case-control studies of type 2 diabetes mellitus and risk of atrial fibrillation. Am J Cardiol. 2011;108(1):56–62.
Seyed Ahmadi S, et al. Risk of atrial fibrillation in persons with type 2 diabetes and the excess risk in relation to glycaemic control and renal function: a Swedish cohort study. Cardiovascular Diabetology. 2020;19(1):9.
Paneni F, et al. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Eur Heart J. 2013;34(31):2436–43.
Tesfamariam B, Brown ML, Cohen RA. Elevated glucose impairs endothelium-dependent relaxation by activating protein kinase C. J Clin Invest. 1991;87(5):1643–8.
Zabala A, et al. Risk of first stroke in people with type 2 diabetes and its relation to glycaemic control: A nationwide observational study. Diabetes Obes Metab. 2020;22(2):182–90.
American Diabetes A. (7) Approaches to glycemic treatment. Diabetes Care. 2015;38 Suppl:S41-8.
Whitlock RH, et al. A Safety Comparison of Metformin vs Sulfonylurea Initiation in Patients With Type 2 Diabetes and Chronic Kidney Disease: A Retrospective Cohort Study. Mayo Clin Proc. 2020;95(1):90–100.
Christensen DH, Rungby J, Thomsen RW. Nationwide trends in glucose-lowering drug use, Denmark, 1999–2014. Clin Epidemiol. 2016;8:381–7.
Douros A, et al. Sulfonylureas as second line drugs in type 2 diabetes and the risk of cardiovascular and hypoglycaemic events: population based cohort study. BMJ. 2018;362:k2693.
Hemmingsen B, et al. Sulfonylurea versus metformin monotherapy in patients with type 2 diabetes: a Cochrane systematic review and meta-analysis of randomized clinical trials and trial sequential analysis. CMAJ Open. 2014;2(3):E162-75.
Austin PC. An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies. Multivariate Behav Res. 2011;46(3):399–424.
Desai RJ, Franklin JM. Alternative approaches for confounding adjustment in observational studies using weighting based on the propensity score: a primer for practitioners. BMJ. 2019;367:l5657.
Schneeweiss S, et al. High-dimensional propensity score adjustment in studies of treatment effects using health care claims data. Epidemiology. 2009;20(4):512–22.
Roumie CL, et al., Comparative Safety of Sulfonylurea and Metformin Monotherapy on the Risk of Heart Failure: A Cohort Study. J Am Heart Assoc, 2017. 6(4).
Palmer SC, et al. Comparison of Clinical Outcomes and Adverse Events Associated With Glucose-Lowering Drugs in Patients With Type 2 Diabetes: A Meta-analysis. JAMA. 2016;316(3):313–24.
Bonds DE, et al. The association between symptomatic, severe hypoglycaemia and mortality in type 2 diabetes: retrospective epidemiological analysis of the ACCORD study. BMJ. 2010;340:b4909.
Pyorala M, et al. Plasma insulin and all-cause, cardiovascular, and noncardiovascular mortality: the 22-year follow-up results of the Helsinki Policemen Study. Diabetes Care. 2000;23(8):1097–102.
Reiff S, Fava S. All-cause mortality in patients on sulphonylurea monotherapy compared to metformin monotherapy in a nation-wide cohort. Diabetes Res Clin Pract. 2019;147:62–6.
Azoulay L, Suissa S. Sulfonylureas and the Risks of Cardiovascular Events and Death: A Methodological Meta-Regression Analysis of the Observational Studies. Diabetes Care. 2017;40(5):706–14.
Belfiore A, Malaguarnera R. Insulin receptor and cancer. Endocr Relat Cancer. 2011;18(4):R125-47.
Soranna D, et al. Cancer risk associated with use of metformin and sulfonylurea in type 2 diabetes: a meta-analysis. Oncologist. 2012;17(6):813–22.
Singh S, Loke YK, Furberg CD. Long-term use of thiazolidinediones and the associated risk of pneumonia or lower respiratory tract infection: systematic review and meta-analysis. Thorax. 2011;66(5):383–8.
Hung AM, et al. Kidney function decline in metformin versus sulfonylurea initiators: assessment of time-dependent contribution of weight, blood pressure, and glycemic control. Pharmacoepidemiol Drug Saf. 2013;22(6):623–31.
Orkaby AR, et al. Metformin vs sulfonylurea use and risk of dementia in US veterans aged >/=65 years with diabetes. Neurology. 2017;89(18):1877–85.
Sun Y, Hu D. The link between diabetes and atrial fibrillation: cause or correlation? J Cardiovasc Dis Res. 2010;1(1):10–1.
Benjamin EJ, et al. Independent risk factors for atrial fibrillation in a population-based cohort. The Framingham Heart Study. JAMA. 1994;271(11):840–4.
Ostropolets A, et al. Metformin Is Associated With a Lower Risk of Atrial Fibrillation and Ventricular Arrhythmias Compared With Sulfonylureas: An Observational Study. Circ Arrhythm Electrophysiol. 2021;14(3):e009115.
Isoda K, et al. Metformin inhibits proinflammatory responses and nuclear factor-kappaB in human vascular wall cells. Arterioscler Thromb Vasc Biol. 2006;26(3):611–7.
Ren H, et al. Metformin alleviates oxidative stress and enhances autophagy in diabetic kidney disease via AMPK/SIRT1-FoxO1 pathway. Mol Cell Endocrinol. 2020;500:110628.
Leonard CE, et al. Pro- and Antiarrhythmic Actions of Sulfonylureas: Mechanistic and Clinical Evidence. Trends Endocrinol Metab. 2017;28(8):561–86.
Jia J, et al. Neuropharmacological Actions of Metformin in Stroke. Curr Neuropharmacol. 2015;13(3):389–94.
Jin Q, et al. Improvement of functional recovery by chronic metformin treatment is associated with enhanced alternative activation of microglia/macrophages and increased angiogenesis and neurogenesis following experimental stroke. Brain Behav Immun. 2014;40:131–42.
Ashabi G, et al. Pre-treatment with metformin activates Nrf2 antioxidant pathways and inhibits inflammatory responses through induction of AMPK after transient global cerebral ischemia. Metab Brain Dis. 2015;30(3):747–54.
Joy NG, et al. Effects of Acute and Antecedent Hypoglycemia on Endothelial Function and Markers of Atherothrombotic Balance in Healthy Humans. Diabetes. 2015;64(7):2571–80.
Razavi Nematollahi L, et al. Proinflammatory cytokines in response to insulin-induced hypoglycemic stress in healthy subjects. Metabolism. 2009;58(4):443–8.
Marx N, et al. C-peptide colocalizes with macrophages in early arteriosclerotic lesions of diabetic subjects and induces monocyte chemotaxis in vitro. Arterioscler Thromb Vasc Biol. 2004;24(3):540–5.

Table 1. Baseline and clinical characteristics of type-2 diabetes patients on sulphonylurea vs. metformin medications before and after 1:2 propensity score matching.

* for SMD 0.2

Characteristics	Before matching			After 1:2 matching
Characteristics	Sulphonylurea (N=36228) Mean(SD);Max;N or Count(%)	Metformin (N=138707) Mean(SD);Max;N or Count(%)	SMD	Sulphonylurea (N=36228) Mean(SD);Max;N or Count(%)	Metformin (N=72456) Mean(SD);Max;N or Count(%)	SMD
*Outcomes*
Mortality	26065(71.94%)	24239(17.47%)	1.31*	26065(71.94%)	17899(24.70%)	1.07*
Cardiovascular mortality	251(0.69%)	208(0.14%)	0.08	251(0.69%)	126(0.17%)	0.08
AF	5633(15.54%)	5292(3.81%)	0.40*	5633(15.54%)	3947(5.44%)	0.33*
Stroke	7728(21.33%)	7505(5.41%)	0.48*	7728(21.33%)	4932(6.80%)	0.43*
*Demographics*
Male gender	17522(48.36%)	64026(46.15%)	0.04	17522(48.36%)	34983(48.28%)	<0.01
Female gender	18706(51.63%)	74681(53.84%)	0.04	18706(51.63%)	37473(51.71%)	<0.01
Baseline age, years	71.3(10.9);105.1;n=36228	63.0(11.9);104.1;n=138707	0.72*	71.3(10.9);105.1;n=36228	71.1(10.7);99.5;n=72456	0.01
<50	1350(3.72%)	19379(13.97%)	0.37*	1350(3.72%)	2699(3.72%)	<0.01
[50-60]	4738(13.07%)	38069(27.44%)	0.36*	4738(13.07%)	9462(13.05%)	<0.01
[60-70]	8094(22.34%)	37957(27.36%)	0.12	8094(22.34%)	16212(22.37%)	<0.01
[70-80]	13724(37.88%)	32263(23.25%)	0.32*	13724(37.88%)	27557(38.03%)	<0.01
>80	8325(22.97%)	11052(7.96%)	0.42*	8325(22.97%)	16526(22.80%)	<0.01
*Past comorbidities*
CHA-DS-VASc score	2.1(1.3);7.0;n=36228	1.6(1.3);7.0;n=138707	0.44*	2.1(1.3);7.0;n=36228	2.1(1.3);7.0;n=72456	0.01
Charlson score	2.9(1.3);15.0;n=36228	2.2(1.6);16.0;n=138707	0.45*	2.9(1.3);15.0;n=36228	2.9(1.4);15.0;n=72456	<0.01
Systemic embolism	9(0.02%)	46(0.03%)	0	9(0.02%)	18(0.02%)	<0.01
Hypertension	9804(27.06%)	25562(18.42%)	0.21*	9804(27.06%)	19429(26.81%)	0.01
Heart failure	869(2.39%)	2657(1.91%)	0.03	869(2.39%)	1723(2.37%)	<0.01
Coronary heart disease	4015(11.08%)	10055(7.24%)	0.13	4015(11.08%)	7847(10.83%)	0.01
Hemiplegia or paraplegia	63(0.17%)	110(0.07%)	0.03	63(0.17%)	126(0.17%)	<0.01
Renal diseases	148(0.40%)	1798(1.29%)	0.1	148(0.40%)	290(0.40%)	<0.01
Neurological diabetic complication	66(0.18%)	402(0.28%)	0.02	66(0.18%)	131(0.18%)	<0.01
Osteoporosis	15(0.04%)	53(0.03%)	<0.01	15(0.04%)	30(0.04%)	<0.01
Ophthalmical diabetic complication	189(0.52%)	1466(1.05%)	0.06	189(0.52%)	376(0.51%)	<0.01
Liver diseases	56(0.15%)	280(0.20%)	0.01	56(0.15%)	112(0.15%)	<0.01
Ventricular tachycardia/fibrillation	127(0.35%)	455(0.32%)	<0.01	127(0.35%)	250(0.34%)	<0.01
Dementia	180(0.49%)	575(0.41%)	0.01	180(0.49%)	357(0.49%)	<0.01
Anemia	1534(4.23%)	3925(2.82%)	0.08	1534(4.23%)	2985(4.11%)	0.01
AMI	499(1.37%)	1946(1.40%)	<0.01	499(1.37%)	995(1.37%)	<0.01
COPD	587(1.62%)	1650(1.18%)	0.04	587(1.62%)	1161(1.60%)	<0.01
IHD	2224(6.13%)	7533(5.43%)	0.03	2224(6.13%)	4428(6.11%)	<0.01
PVD	156(0.43%)	640(0.46%)	<0.01	156(0.43%)	310(0.42%)	<0.01
Gastrointestinal bleeding	625(1.72%)	2157(1.55%)	0.01	625(1.72%)	1234(1.70%)	<0.01
Malignancy	218(0.60%)	755(0.54%)	0.01	218(0.60%)	436(0.60%)	<0.01
Obesity	113(0.31%)	1046(0.75%)	0.06	113(0.31%)	226(0.31%)	<0.01
*Medications*
Number of diabetes mellitus drugs	1.8(0.6);6.0;n=36228	1.8(0.6);6.0;n=138707	0.03	1.8(0.6);6.0;n=36228	1.8(0.5);5.0;n=72456	0.05
Number of cardiovascular drugs	1.9(1.2);5.0;n=36228	1.6(1.2);5.0;n=138707	0.25*	1.9(1.2);5.0;n=36228	1.9(1.2);5.0;n=72456	<0.01
ACEI/ARB	18529(51.14%)	65870(47.48%)	0.07	18529(51.14%)	37451(51.68%)	0.01
Beta blockers	14309(39.49%)	45613(32.88%)	0.14	14309(39.49%)	28559(39.41%)	<0.01
Calcium channel blockers	18428(50.86%)	53427(38.51%)	0.25*	18428(50.86%)	36821(50.81%)	<0.01
Diuretics	8083(22.31%)	22413(16.15%)	0.16	8083(22.31%)	15739(21.72%)	0.01
Stains	8605(23.75%)	30530(22.01%)	0.04	8605(23.75%)	17046(23.52%)	0.01
Antidiabetic drugs	8288(22.87%)	13678(9.86%)	0.36*	8288(22.87%)	14755(20.36%)	0.06
Lipid-lowering drugs	40(0.11%)	527(0.37%)	0.05	40(0.11%)	80(0.11%)	<0.01
Gliclazide	1402(3.86%)	9344(6.73%)	0.13	1402(3.86%)	2761(3.81%)	<0.01
Nitrates	15(0.04%)	240(0.17%)	0.04	15(0.04%)	30(0.04%)	<0.01
Antihypertensive drugs	17(0.04%)	249(0.17%)	0.04	17(0.04%)	34(0.04%)	<0.01
Anticoagulants	1(0.00%)	45(0.03%)	0.02	1(0.00%)	2(0.00%)	<0.01
Antiplatelets	22(0.06%)	423(0.30%)	0.06	22(0.06%)	44(0.06%)	<0.01
*Laboratory tests*
Lymphocyte, x10^9/L	1.9(0.9);14.7;n=2772	2.0(0.9);47.2;n=8884	0.09	1.9(0.9);14.7;n=2772	1.8(0.8);14.7;n=4924	0.04
Neutrophil, x10^9/L	5.6(2.9);37.2;n=2772	5.3(2.7);41.1;n=8884	0.07	5.6(2.9);37.2;n=2772	5.4(2.7);41.1;n=4924	0.06
Platelet, x10^9/L	254.3(88.3);1560.0;n=2912	263.1(82.1);1098.0;n=8525	0.1	254.3(88.3);1560.0;n=2912	261.0(88.4);1080.0;n=5262	0.08
Potassium, mmol/L	4.2(0.5);8.0;n=20729	4.2(0.5);14.6;n=72334	0.04	4.2(0.5);8.0;n=20729	4.2(0.5);8.0;n=40019	0.04
Albumin, g/L	38.4(5.2);53.9;n=5916	39.5(4.8);55.0;n=18556	0.2*	38.4(5.2);53.9;n=5916	39.1(5.0);52.0;n=10279	0.14
Sodium, mmol/L	139.4(3.3);153.0;n=20740	139.5(3.0);160.3;n=72377	0.04	139.4(3.3);153.0;n=20740	139.5(3.2);153.0;n=40022	0.06
Urea, mmol/L	6.5(2.8);56.3;n=17896	6.0(2.4);61.3;n=63368	0.21*	6.5(2.8);56.3;n=17896	6.3(2.6);61.3;n=35235	0.06
Creatinine, umol/L	92.4(35.3);783.0;n=18011	84.6(30.2);1019.0;n=63689	0.24*	92.4(35.3);783.0;n=18011	90.3(32.3);1014.0;n=35432	0.06
Total protein, g/L	74.3(7.0);104.6;n=5900	74.7(6.4);109.2;n=18517	0.06	74.3(7.0);104.6;n=5900	74.5(6.7);109.2;n=10241	0.04
Alkaline phosphatase, U/L	78.9(41.8);2503.0;n=16786	75.0(31.9);3273.0;n=57204	0.1	78.9(41.8);2503.0;n=16786	75.6(35.4);1721.0;n=31235	0.08
Aspartate transaminase, U/L	32.6(53.2);777.0;n=1540	32.1(93.7);4994.0;n=4895	0.01	32.6(53.2);777.0;n=1540	33.5(151.9);4994.0;n=2702	0.01
Alanine transamise, U/L	24.5(24.6);874.0;n=13870	26.7(23.1);870.0;n=45855	0.09	24.5(24.6);874.0;n=13870	23.9(21.2);870.0;n=25311	0.02
Bilirubin, umol/L	11.2(10.6);145.0;n=831	12.7(13.6);346.7;n=2142	0.13	11.2(10.6);145.0;n=831	12.7(11.0);157.0;n=1208	0.14
Triglyceride, mmol/L	1.7(1.1);46.4;n=23778	1.8(1.4);90.8;n=89917	0.06	1.7(1.1);46.4;n=23778	1.7(1.1);41.3;n=47366	<0.01
LDL, mmol/L	2.9(0.9);10.5;n=14474	2.9(0.9);10.5;n=53250	0.06	2.9(0.9);10.5;n=14474	2.9(0.8);9.7;n=28923	<0.01
Total cholesterol, mmol/L	4.8(1.0);12.9;n=23835	4.8(1.0);20.1;n=90076	0.09	4.8(1.0);12.9;n=23835	4.8(1.0);15.3;n=47480	<0.01
HDL, mmol/L	1.2(0.3);4.2;n=21130	1.2(0.3);4.7;n=81686	0.01	1.2(0.3);4.2;n=21130	1.2(0.3);3.9;n=42479	0.01
HbA1c, g/dL	12.4(1.9);18.3;n=3088	12.8(1.8);18.6;n=10127	0.21*	12.4(1.9);18.3;n=3088	12.3(1.8);18.4;n=5635	0.02
Glucose, mmol/L	9.5(4.2);50.4;n=7970	9.1(4.4);64.3;n=14151	0.09	9.5(4.2);50.4;n=7970	8.9(4.1);64.1;n=12652	0.14
NLR	4.0(5.3);108.0;n=2772	3.5(3.9);68.5;n=8884	0.11	4.0(5.3);108.0;n=2772	3.8(4.0);49.0;n=4924	0.05
Fasting blood glucose, mmol/L	7.8(2.6);36.3;n=15751	7.7(2.6);67.2;n=58145	0.01	7.8(2.6);36.3;n=15751	7.8(2.6);33.1;n=31839	<0.01
HbA1c, %	7.4(1.4);20.8;n=16681	7.4(1.5);21.9;n=59591	<0.01	7.4(1.4);20.8;n=16681	7.4(1.4);18.5;n=33126	<0.01
*Variability measurements*
Glucose
Absolute successive variability score (glucose)	57.9(20.6);96.8;n=15752	56.8(20.9);96.6;n=58148	0.06	57.9(20.6);96.8;n=15752	56.3(21.2);96.6;n=31839	0.08
Percentage successive variability score (glucose)	50.2(21.6);96.0;n=15752	48.9(21.6);96.2;n=58148	0.06	50.2(21.6);96.0;n=15752	48.8(21.9);96.2;n=31839	0.07
SD of glucose	1.7(1.2);13.1;n=15752	1.7(1.2);30.4;n=58148	0.02	1.7(1.2);13.1;n=15752	1.6(1.2);24.9;n=31839	0.06
Mean glucose, mmol/L	8.1(1.8);24.1;n=15752	8.3(1.9);29.0;n=58148	0.07	8.1(1.8);24.1;n=15752	8.0(1.7);22.7;n=31839	0.07
Normalized absolute successive variability score (glucose)	7.3(2.8);20.1;n=15752	7.0(2.7);20.7;n=58148	0.09	7.3(2.8);20.1;n=15752	7.2(2.8);20.7;n=31839	0.04
Normalized percentage successive variability score (glucose)	6.4(3.0);21.2;n=15752	6.1(3.0);22.2;n=58148	0.09	6.4(3.0);21.2;n=15752	6.3(3.1);22.2;n=31839	0.03
SD/Initial (glucose)	23.4(20.4);418.2;n=15751	23.2(19.9);475.4;n=58148	0.01	23.4(20.4);418.2;n=15751	22.4(19.9);378.8;n=31839	0.05
Coefficient of variation (glucose)	20.2(12.2);110.9;n=15752	20.2(12.4);174.8;n=58148	0.01	20.2(12.2);110.9;n=15752	19.5(12.3);135.0;n=31839	0.06
Variability independent of mean (glucose)	31.8(19.6);188.5;n=15752	31.8(19.9);323.1;n=58148	<0.01	31.8(19.6);188.5;n=15752	30.6(19.7);227.8;n=31839	0.06
HbA1c
Absolute successive variability score (HbA1c)	37.6(21.9);92.3;n=16935	37.0(21.5);93.3;n=60447	0.03	37.6(21.9);92.3;n=16935	36.1(21.9);90.9;n=33644	0.07
Percentage successive variability score (HbA1c)	28.4(20.4);90.9;n=16935	27.5(19.9);90.9;n=60447	0.04	28.4(20.4);90.9;n=16935	26.8(20.0);90.0;n=33644	0.08
SD of HbA1c	0.9(0.7);8.0;n=16934	0.9(0.7);19.3;n=60440	0.02	0.9(0.7);8.0;n=16934	0.9(0.7);18.5;n=33637	0.06
Mean HbA1c, %	7.7(1.1);19.0;n=16935	7.8(1.2);20.3;n=60447	0.08	7.7(1.1);19.0;n=16935	7.6(1.1);20.3;n=33644	0.04
Normalized absolute successive variability score (HbA1c)	4.8(2.7);15.4;n=16935	4.7(2.7);14.6;n=60447	0.05	4.8(2.7);15.4;n=16935	4.6(2.7);14.3;n=33644	0.07
Normalized percentage successive variability score (HbA1c)	3.6(2.6);15.4;n=16935	3.5(2.5);16.2;n=60447	0.06	3.6(2.6);15.4;n=16935	3.5(2.6);13.4;n=33644	0.07
SD/Initial (HbA1c)	12.2(9.9);121.4;n=16686	12.4(10.6);296.4;n=59678	0.01	12.2(9.9);121.4;n=16686	11.7(9.9);253.2;n=33191	0.05
Coefficient of variation (HbA1c)	11.3(7.7);74.5;n=16934	11.4(8.1);166.2;n=60440	0.01	11.3(7.7);74.5;n=16934	10.8(7.7);137.8;n=33637	0.06
Variability independent of mean (HbA1c)	16.0(11.1);109.8;n=16934	16.2(11.6);252.1;n=60440	0.01	16.0(11.1);109.8;n=16934	15.3(11.1);214.2;n=33637	0.06
Lipids
SD of LDL	0.5(0.4);3.4;n=6357	0.5(0.3);5.1;n=23852	0.04	0.5(0.4);3.4;n=6357	0.5(0.3);3.0;n=12390	0.07
SD of HDL	0.1(0.1);1.2;n=13898	0.1(0.1);1.6;n=54920	0.07	0.1(0.1);1.2;n=13898	0.1(0.1);1.2;n=28779	0.06
SD of TC	0.6(0.4);8.3;n=16259	0.6(0.4);11.8;n=62537	0.02	0.6(0.4);8.3;n=16259	0.6(0.4);6.3;n=32886	0.06
SD of TG	0.6(0.9);28.0;n=16167	0.6(1.1);35.1;n=62344	0.06	0.6(0.9);28.0;n=16167	0.6(0.8);32.3;n=32815	0.02

AF: atrial fibrillation, HF: heart failure, MI: myocardial infarction, COPD: chronic obstructive pulmonary disease, IHD: ischemic heart disease, PVD: peripheral vascular disease, TIA: transient ischemic attack, ACEI: angiotensin-converting-enzyme inhibitors, ARB: angiotensin II receptor blockers, LDL: low density lipoprotein cholesterol, HDL: high density lipoprotein cholesterol, TG: triglycerides, SD: standard deviation.

Table 2. HRs (and 95% CIs) of sulphonylurea vs. metformin from univariate Cox models, multivariate Cox models, cause-specific and subdistribution hazard models for cardiovascular mortality, non-cardiovascular mortality, new onset HF and new onset MI before and after 1:2 propensity score matching.

* for p≤ 0.05, ** for p ≤ 0.01, *** for p ≤ 0.001

Model	Outcome	Sulphonylurea vs. Metformin (Before matching) HR [95% CI]	P value	Sulphonylurea vs. Metformin (After 1:2 matching) HR [95% CI]	P value
Univariate Cox model	New onset AF	3.81[3.67, 3.96]	<0.0001***	2.84[2.73, 2.96]	<0.0001***
	New onset stroke	3.73[3.62, 3.85]	<0.0001***	3.18[3.07, 3.3]	<0.0001***
	Cardiovascular mortality	4.23[3.52, 5.09]	<0.0001***	3.78[3.05, 4.68]	<0.0001***
	All-cause mortality	6.44[6.33-6.55]	<0.0001***	4.25[4.17-4.34]	<0.0001***
Multivariate Cox model^#	New onset AF	2.60[2.51, 2.79]	<0.0001***	2.89[2.75, 2.77]	<0.0001***
	New onset stroke	3.1[2.5, 3.4]	<0.0001***	3.23[3.01, 3.45]	<0.0001***
	Cardiovascular mortality	2.42[1.93, 2.98]	<0.0001***	3.60[2.62, 4.81]	<0.0001***
	All-cause mortality	4.10[3.88, 4.35]	<0.0001***	4.35[3.16, 4.75]	<0.0001***
Cause-specific hazard model	New onset AF	5.58[5.34, 5.84]	<0.0001***	4.04[3.85, 4.24]	<0.0001***
	New onset stroke	5.93[5.73, 6.13]	<0.0001***	4.79[4.62, 4.97]	<0.0001***
	Cardiovascular mortality	6.73[5.58, 8.12]	<0.0001***	5.71[4.6, 7.09]	<0.0001***
	All-cause mortality	7.09[6.94, 7.23]	<0.0001***	4.77[4.67, 4.88]	<0.0001***
Subdistribution hazard model	New onset AF	4.98[4.1, 6.02]	<0.0001***	4.21[3.02, 5.11]	<0.0001***
	New onset stroke	4.85[3.25, 5.43]	<0.0001***	3.55[2.87, 4.53]	<0.0001***
	Cardiovascular mortality	5.43[4.44, 7.09]	<0.0001***	5.09[3.65, 6.88]	<0.0001***
	All-cause mortality	6.54[5.23, 7.1]	<0.0001***	4.54[3.55, 5.11]	<0.0001***

AF: atrial fibrillation, HR: hazard ratio, CI: confidence interval.

^#Adjusted for significant demographics, past comorbidities, and non- sulphonylurea/metformin medications.

Table 3. Risk of incident AF, stroke, cardiovascular mortality and all-cause mortality associated with treatment of sulphonylurea vs. metformin using different matching approaches.

* for p≤ 0.05, ** for p ≤ 0.01, *** for p ≤ 0.001

Outcome	HR after PS stratification [95% CI];P value	HR after HDPS matching [95% CI];P value	HR after PS IPTW [95% CI]; P value
New onset AF	3.85[3.45, 3.99]; <0.0001***	3.56[2.78, 4.02]; <0.0001***	2.45[1.45, 3.38]; <0.0001***
New onset stroke	3.45[3.11, 3.69]; <0.0001***	3.67[3.02, 3.99]; <0.0001***	3.45 [3.02, 3.79]; <0.0001***
Cardiovascular mortality	3.98[3.26, 4.77]; <0.0001***	4.01[3.34, 5.12]; <0.0001***	3.54[3.23, 4.09]; <0.0001***
All-cause mortality	5.57[4.76-6.62]; <0.0001***	6.01[5.56-6.88]; <0.0001***	5.45[4.43-5.69]; <0.0001***

AF: atrial fibrillation, HR: hazard ratio, CI: confidence interval, PS: propensity score, HDPS: high dimensional propensity score, IPTW: inverse probability of treatment weighting.

SupplementaryAppendix20210630.docx

Download PDF

Journal Publication

published 03 Feb, 2022

Read the published version in Acta Diabetologica →

Version 1

posted

You are reading this latest preprint version

Metformin versus sulphonylureas for new onset atrial fibrillation and stroke in type 2 diabetes mellitus: a population-based study

Status:

Journal Publication

Version 1

Abstract

Figures

Background

Methods

Study design and population

Statistical analysis and outcomes

Results

Baseline characteristics

Outcome Predictors

Competing risk analysis

Discussion

Limitations

Conclusions

Declarations

References

Tables

Supplementary Files

Status:

Journal Publication

Version 1