Abstract
Background and purpose
Cardiovascular outcome trials demonstrate that glucagonlike peptide‐1 receptor agonists (GLP‐1RAs) reduce the risk of major adverse cardiovascular events in patients with type 2 diabetes (T2D), whereas dipeptidyl peptidase‐4 inhibitors (DPP‐4is) have not shown cardiovascular benefits. We compared acute ischemic stroke (AIS) with T2D treated with either a GLP‐1RA or DPP‐4i prior to the index stroke.
Methods
This national cohort study included AIS patients with T2D from 2017 to 2020 in Denmark who were users of a GLP‐1RA or DPP‐4i. To be categorized as a user, we required at least 12 months of exposure and no concurrent treatment with another newer glucose‐lowering medication during the last 3 months prior to the index stroke. GLP‐1RA users were compared to users of DPP‐4i while adjusting for the calendar year of index stroke, age, sex, comorbidity, and socioeconomic factors.
Results
The study included 1567 AIS events with T2D; 593 were users of GLP‐1RA and 974 of DPP‐4i. The absolute risk of a very severe stroke was 2.4% (95% confidence interval [CI] = 1.2–3.7) in GLP‐1RA users and 6.1% (95% CI = 4.6–7.7) in DPP‐4i users. The corresponding adjusted risk ratio (aRR) of GLP‐1RA versus DPP‐4i was 0.49 (95% CI = 0.24–1.00). The aRRs of 30‐day and 365‐day mortality were 0.55 (95% CI = 0.32–0.94) and 0.72 (95% CI = 0.53–0.98), respectively.
Conclusions
The risk of a very severe stroke as well as the 30‐day and 365‐day poststroke mortality rates were lower among the AIS patients with comorbid T2D receiving GLP‐1RA prior to the index stroke compared to those receiving DPP‐4i. Hence, GLP‐1RA may improve stroke outcomes in comparison with DPP‐4i.
Keywords: cerebrovascular disease/stroke, cohort studies, diabetes, outcome research
INTRODUCTION
Acute ischemic stroke (AIS) patients with comorbid type 2 diabetes (T2D) have an adverse prognosis after stroke including increased risk of recurrent stroke, mortality, readmission, and poorer cognitive outcome [1, 2, 3]. Given the high prevalence of T2D in ischemic stroke patients, 33% in a recent meta‐analysis, this represents an important challenge within stroke care [4]. Despite the significantly increased risk of stroke as well as adverse poststroke outcomes, there is a paucity of available treatments that specifically target the risk of stroke in patient with T2D [5].
A series of large‐scale cardiovascular outcome trials (CVOTs) have created robust clinical evidence of cardiovascular benefits with a reduction in major adverse cardiovascular events with glucagonlike peptide‐1 receptor agonists (GLP‐1RAs) and sodium–glucose cotransporter‐2 inhibitors (SGLT2is) in patients with T2D and high cardiovascular risk or established atherosclerotic cardiovascular disease [6]. Two CVOTs using GLP‐1RA also demonstrated a significant reduction in the risk of nonfatal stroke [7, 8] and a subsequent meta‐analysis of all the CVOTs of GLP‐1RAs found a reduction in the risk of fatal or nonfatal stroke [9, 10]. As SGLT2i did not show a reduction in stroke and dipeptidyl peptidase‐4 inhibitors (DPP‐4is) have not shown cardiovascular benefits, the recent stroke and diabetes guidelines recommend the use of GLP‐1RA in patients with a history of AIS or transient ischemic attack [6, 11, 12].
However, research focusing on real‐world populations of AIS patients with comorbid T2D including comparative analyses of the newer glucose‐lowering medications is sparse [5].
The aim of this study was to compare the clinical outcomes, including stroke severity, mortality, recurrent stroke, readmission, and length of stay in AIS patients with comorbid T2D treated with either a GLP‐1RA or DPP‐4i prior to the index stroke.
METHODS
Study design and study population
This nationwide, population‐based cohort study used information from high‐quality medical registries. The study included AIS patients with T2D from 2017 to 2020 in Denmark who were users of a GLP‐1RA or DPP‐4i prior to the index stroke.
We used the Danish Stroke Registry [13] to identify all acute stroke events in the study period. Patients with intracerebral hemorrhage, type 1 diabetes (T1D), and no diabetes were excluded. Furthermore, we restricted the population to patients living in Denmark with a valid unique 10‐digit CPR (personal registration number provided to all Danish citizens by the Danish Civil Registration System) [14] in the year prior and poststroke. Lastly, we identified the AIS events with use of a newer glucose‐lowering medications (GLP‐1RA, SGLT2i, or DPP‐4i) and excluded the patients with mixed use of the drug classes or patients exposed to SGLT2i (flowchart, Figure 1). Users of SGLT2i were excluded from the study due to a low number of outcomes during the study period.
FIGURE 1.
Flowchart of the study cohort. CPR, Danish personal registration number; DPP‐4, dipeptidyl peptidase‐4; GLP‐1, glucagonlike peptide‐1; ICH, intracerebral hemorrhage; IS, ischemic stroke; SGLT‐2, sodium–glucose cotransporter‐2.
Data sources
We used the CPR number to link information from different national registries: the Danish Stroke Registry [13, 15], the Danish National Patient Registry [16], the Danish Prescription Database [17], the Register of Laboratory Results for Research [18], and Statistics Denmark [19]. A detailed description of the data sources and definitions of variables are provided in Table S1.
Identification and classification of T1D and T2D
We used information on the use of any glucose‐lowering medication from the Danish Prescription Database and information on diabetes diagnoses from the Danish Stroke Registry and the Danish National Patient Registry to (i) identify AIS patients with any kind of diabetes, (ii) identify AIS patients who only received glucose‐lowering medications for gestational diabetes or polycystic ovary syndrome, and (iii) decide whether the patient had T1D or T2D. Further information is given in Table S2.
Exposures
To be categorized as a user of a GLP‐1RA or DPP‐4i required at least 12‐month exposure to one of the drug classes prior to the index stroke and for the past 3 months no concurrent treatment with another of the newer glucose‐lowering medications (GLP‐1RA, SGLT2i, DPP‐4i).
Characteristics and outcomes
Baseline characteristics of all the included events were determined at the time of the stroke admission, unless otherwise indicated in the text or in Table S1.
The comorbidity of AIS was characterized using the Charlson Comorbidity Index (CCI) [20]. The CCI was given without diabetes and diabetes‐related complications. Furthermore, the characteristics included socioeconomics, laboratory results, stroke severity, and prestroke medication. The laboratory results (HbA1c, cholesterol, and estimated glomerular filtration rate [eGFR]), if available, represented the measurement closest to the stroke admission date within a 7‐day time window. The stroke severity was assessed with the score on the Scandinavian Stroke Scale (SSS) recorded at the time of hospital admission. A patient was considered a user of a given class of medication (statins, antiplatelet drugs, anticoagulants, and other glucose‐lowering medications) at baseline if the patient had redeemed a prescription within the year prior to the stroke admission. The duration of T2D was defined as time from first glucose‐lowering medications to time of the stroke admission.
The clinical outcomes were as follows: very severe index stroke (SSS score = 0–14); mortality, recurrent stroke, and all‐cause, unplanned readmissions at 30 and 365 days; and the length of stay (acute and total hospital stay). Please see Table S1 for further details and data sources of the clinical outcomes.
Statistical analysis
The study was event‐based. Hence, individuals could be included with more than one stroke episode.
We computed the absolute risks of very severe stroke and death; for cumulative incidences of recurrent stroke and readmission, we used the Aalen–Johansen estimator, treating death as a competing risk. Poisson and Cox regressions were used to estimate risk ratios (RRs) and hazard rate ratios (HRRs).
Adjustments were made by using inverse probability of treatment weighting to estimate average treatment effects in the comparison of GLP‐1RA and DPP‐4i. Logistic regression was used for the propensity model and included calendar year of index stroke, age, sex, living arrangements, prior stroke, prior myocardial infarction, atrial fibrillation, smoking, hypertension, CCI, and immigrant, income, and occupational status (total effect). As stroke severity was considered a mediator, it was not included in the propensity model. However, to explore direct effects, subanalyses including stroke severity were performed. Large weights were handled with trimming and by recalculating weights in the trimmed population. Missing information was handled using multiple imputation by chained equations with predictive mean matching.
All confidence intervals (CIs) were calculated with a 95% confidence level and were based on cluster robust standard errors taking into account the multiplicity of records per person.
We used the Stata 16.1 (StataCorp, College Station, TX, USA) for all analyses.
Ethical approval
Under Danish law, registry‐based studies require no ethical approval or patient consent. The study was approved by the Danish Data Protection Agency (2019–899/10–0033) and the Danish Clinical Registries.
RESULTS
Of the 7130 AIS events in patients with T2D, 593 were in users of GLP‐1RA and 974 in users of DPP‐4i. Among the excluded events, 353 were in users of SGLT‐2. Figure 1 shows a flowchart of the study cohort and details of the exclusions. The number of users of DPP‐4i was higher within the first years of the study period compared to GLP‐1RA users, but was similar by 2020, 226 versus 198. At all times, the number of the excluded SGLT2i users was considerably lower. Figure 2 shows yearly counts of AIS with use of GLP‐1RA, SGLT2i, and DPP‐4i from 2017 to 2020.
FIGURE 2.
Ischemic strokes with use of glucagonlike peptide‐1 (GLP‐1) receptor agonist, sodium–glucose cotransporter‐2 (SGLT‐2) inhibitor, or dipeptidyl peptidase‐4 (DPP‐4) inhibitor.
Baseline characteristics of users of GLP‐1RA and DPP‐4i (Table 1) showed that patients with AIS and T2D who were users of GLP‐1RA were younger, were more frequently men, had less severe strokes, had higher income, were more often employed, and less frequently were immigrants/descendants of immigrants compared to users of DPP‐4i. Furthermore, the users of GLP‐1RA had less comorbidity with lower rates of prior stroke, atrial fibrillation, and CCI ≥ 2. There were also differences in HbA1c and eGFR between the groups as well as in the use of statins, antiplatelet drugs, and oral anticoagulants. Details of baseline characteristics are shown in Table 1.
TABLE 1.
Baseline characteristics of study cohort.
| IS with TD2 with prior use of | GLP‐1RA | DPP‐4i |
|---|---|---|
| Users, n | 593 | 974 |
| Age, years, median (IQR) | 70.1 (61.6–75.2) | 76.1 (69.3–82.3) |
| Sex, male, % (n) | 65.8 (390) | 61.3 (597) |
| Severity, SSS score, median (IQR) | 51.0 (43.0–56.0) | 49.0 (39.0–55.0) |
| Very severe stroke [SSS = 0–14], % (n) | 2.4 (14) | 6.1 (59) |
| Severe stroke [SSS = 15–29], % (n) | 4.6 (27) | 8.0 (77) |
| Moderate stroke [SSS = 30–44], % (n) | 20.4 (119) | 23.2 (223) |
| Mild stroke [SSS = 45–58], % (n) | 72.5 (422) | 62.6 (601) |
| Same‐day admission, % (n) | 55.1 (322) | 60.1 (582) |
| Days to admission, median (IQR) | 0.5 (0.5–1.0) | 0.5 (0.5–1.0) |
| Above average income, % (n) | 22.2 (122) | 15.1 (139) |
| Living with someone, % (n) | 60.2 (350) | 50.9 (488) |
| Living alone, % (n) | 38.2 (222) | 44.4 (425) |
| Living–other arrangement, % (n) | 1.5 (9) | 4.7 (45) |
| Immigrant or descendent, % (n) | 6.9 (41) | 9.0 (88) |
| Educational level, median (IQR) | 30.0 (20.0–30.0) | 30.0 (20.0–30.0) |
| Not working, % (n) | 76.1 (420) | 89.1 (835) |
| Smoker, % (n) | 24.2 (129) | 22.8 (187) |
| Former smoker, % (n) | 39.3 (209) | 39.0 (320) |
| Never smoker, % (n) | 36.5 (194) | 38.2 (313) |
| Prior stroke, any stroke, % (n) | 27.5 (163) | 30.2 (294) |
| Prior ischemic stroke, % (n) | 26.8 (159) | 29.3 (285) |
| Prior intracerebral hemorrhage, % (n) | 1.3 (8) | 2.7 (26) |
| CCI, median (IQR) | 1.0 (0.0–2.0) | 1.0 (0.0–3.0) |
| CCI = 0, % (n) | 32.7 (194) | 28.6 (279) |
| CCI = 1, % (n) | 27.7 (164) | 21.7 (211) |
| CCI ≥ 2, % (n) | 39.6 (235) | 49.7 (484) |
| Atrial fibrillation, % (n) | 16.6 (98) | 27.2 (264) |
| Prior myocardial infarction, % (n) | 13.2 (77) | 13.7 (131) |
| Hypertension, % (n) | 82.9 (491) | 80.5 (779) |
| Peripheral artery disease, % (n) | 10.1 (58) | 10.1 (95) |
| HbA1c, mmol/mol, median (IQR) | 58.5 (49.0–70.0) | 55.0 (48.0–65.0) |
| Missing HbA1c value, % (n) | 30.5 (181) | 35.1 (342) |
| Cholesterol: total, mmol/L, median (IQR) | 3.9 (3.2–4.7) | 3.9 (3.3–4.8) |
| Missing cholesterol value, % (n) | 23.7 (141) | 22.9 (223) |
| LDL cholesterol, mmol/L, median (IQR) | 1.8 (1.2–2.6) | 1.9 (1.4–2.6) |
| Missing LDL cholesterol value, % (n) | 27.6 (164) | 25.6 (250) |
| eGFR, mL/min/1.73 m2, median (IQR) | 74.0 (53.0–90.0) | 61.0 (40.0–85.0) |
| <45, % (n) | 14.3 (85) | 26.9 (262) |
| 45–60, % (n) | 12.5 (74) | 16.6 (162) |
| 60–90, % (n) | 36.4 (216) | 30.7 (299) |
| >90, % (n) | 23.4 (139) | 15.1 (147) |
| Missing eGFR value, % (n) | 13.3 (79) | 15.1 (104) |
| Duration of T2D, years, median (IQR) | 13.7 (9.0–18.6) | 11.1 (7.1–15.3) |
| Duration > 5 years, (n) | 90.7 (538) | 83.8 (816) |
| Statins, % (n) | 76.2 (452) | 70.7 (689) |
| Antiplatelet drugs, all, % (n) | 59.4 (352) | 50.1 (488) |
| Acetylsalicylic acid, % (n) | 39.1 (232) | 32.2 (314) |
| Vitamin K antagonist, % (n) | 4.4 (26) | 8.0 (78) |
| DOAC, % (n) | 12.0 (71) | 16.4 (160) |
| Biguanides, % (n) | 74.9 (444) | 73.3 (714) |
| Sulfonylureas, % (n) | 9.4 (56) | 15.8 (154) |
| Insulin, % (n) | 59.2 (351) | 25.8 (251) |
Abbreviations: CCI, Charlson Comorbidity Index; DOAC, direct oral anticoagulant; DPP‐4i, dipeptidyl peptidase‐4 inhibitor; eGFR, estimated glomerular filtration rate; GLP‐1RA, glucagonlike peptide‐1 receptor agonist; IQR, interquartile range; IS, ischemic stroke; LDL, low‐density lipoprotein; SSS, Scandinavian Stroke Scale; TD2, type 2 diabetes.
The absolute and relative risks of a very severe stroke and mortality at 30 and 365 days are shown in Table 2. The absolute risk of a very severe stroke among the GLP‐1RA users was 2.4% (95% CI = 1.2–3.7), whereas the risk among the DPP‐4i users was 6.1% (95% CI = 4.6–7.7) and the adjusted RR GLP‐1RA/DPP‐4i was 0.49 (95% CI = 0.24–1.00). The 30‐ and 365‐day mortality risk was lower for GLP‐1RA users compared to DPP‐4i users and the adjusted RR was 0.55 (95% CI = 0.32–0.94) and 0.72 (95% CI = 0.53–0.98), respectively (Table 2).
TABLE 2.
Risk of “very severe stroke” and 30‐ and 365‐day mortality and comparison of ischemic stroke between users of GLP‐1RA and users of DPP‐4i.
| Absolute risk, % (95% CI) | Risk ratio (95% CI) | |
|---|---|---|
| Very severe stroke | ||
| GLP‐1RA | 2.4 (1.2–3.7) | |
| DPP‐4i | 6.1 (4.6–7.7) | |
| GLP‐1RA/DPP‐4i, crude | 0.39 (0.22–0.70) | |
| GLP‐1RA/DPP‐4i, adjusted | 0.49 (0.24–1.00) | |
| 30‐day mortality | ||
| GLP‐1RA | 3.9 (2.3–5.4) | |
| DPP‐4i | 10.0 (8.1–11.8) | |
| GLP‐1RA/DPP‐4i, crude | 0.39 (0.25–0.61) | |
| GLP‐1RA/DPP‐4i, adjusted | 0.55 (0.32–0.94) | |
| 365‐day mortality | ||
| GLP‐1RA | 11.3 (8.7–13.9) | |
| DPP‐4i | 24.3 (21.5–27.2) | |
| GLP‐1RA/DPP‐4i, crude | 0.46 (0.36–0.60) | |
| GLP‐1RA/DPP‐4i, adjusted | 0.72 (0.53–0.98) | |
Abbreviations: CI, confidence interval; DPP‐4i, dipeptidyl peptidase‐4 inhibitor; GLP‐1RA, glucagonlike peptide‐1 receptor agonist.
Table 3 shows cumulative incidences and the HRRs of recurrent stroke and readmission. The incidence of recurrent stroke at 30 and 365 days was lower among GLP‐1RA users compared to DDP‐4i users; however, the overall adjusted HRR showed no significant difference (0.78, 95% CI = 0.48–1.29).
TABLE 3.
Thirty‐ and 365‐day cumulative incidences of recurrent stroke and readmission and comparison of GLP‐1RA users to users of DPP‐4i.
| 30‐day incidence (95% CI) | 365‐day incidence (95% CI) | HRR (95% CI) | |
|---|---|---|---|
| Recurrent stroke | |||
| GLP‐1RA | 1.4 (0.7–2.6) | 6.4 (4.6–8.6) | |
| DPP‐4i | 1.7 (1.0–2.7) | 7.2 (5.7–9.0) | |
| GLP‐1RA/DPP‐4i, crude | 0.82 (0.53–1.28) | ||
| GLP‐1RA/DPP‐4i, adjusted | 0.78 (0.48–1.29) | ||
| Readmission | |||
| GLP‐1RA | 15.3 (12.4–18.3) | 47.2 (43.0–51.2) | |
| DPP‐4i | 16.7 (14.4–19.2) | 49.3 (46.0–52.5) | |
| GLP‐1RA/DPP‐4i, crude | 0.88 (0.76–1.02) | ||
| GLP‐1RA/DPP‐4i, adjusted | 1.06 (0.88–1.26) |
Abbreviations: CI, confidence interval; DPP‐4i, dipeptidyl peptidase‐4 inhibitor; GLP‐1RA, glucagonlike peptide‐1 receptor agonist; HRR, hazard rate ratio.
Readmissions were comparable between the groups (Table 3). Length of stay was also found comparable. The acute hospital stay in GLP‐1RA was median 3 (interquartile range [IQR] = 2–8) days and in DPP‐4i it was 3 (IQR = 2–7) days, whereas the total hospital stay was 4 (IQR = 2–10) and 4 (IQR = 2–11) days, respectably.
Subanalysis including adjustments of stroke severity underlined that stroke severity was a mediator of mortality (Table S3).
DISCUSSION
In a national population‐based cohort of patients with AIS with T2D who were users of either GLP‐1RA or DPP‐4i, the risk of a very severe stroke was lower among the users of GLP‐1RA compared to those receiving DPP‐4i prior to the index stroke. Moreover, the risk of poststroke mortality was lower in the patients using GLP‐1RAs. No clear differences in risk of recurrent stroke, all‐cause unplanned readmission, or length of stay were observed.
Although it is well established that people with T2D are at a significantly greater risk of a stroke, and CVOTs of GLP‐1RA suggest a reduction in stroke recurrence, stroke is often overlooked when examining new treatment strategies, whereas information on cardiac outcomes, chronic kidney disease, and heart failure is more frequently available. A recent review underlines the unmet need to reduce the burden of stroke in T2D and highlights the importance of further research in this field [5].
One of the main findings of this study was the association with a lower risk of a very severe stroke in the AIS patients who were users of GLP‐1RAs compared to users of DPP‐4i. Evidence suggests that GLP‐1RAs may have neuroprotective effects, and this may explain the reduction in stroke severity when compared to users of DPP‐4i [21]. Other more complex mechanism may also be in play, for example, this study found that T2D treated with GLP‐1RAs in addition to metformin compared to metformin was associated with better cognitive function and higher circulating endothelial progenitor cells [22].
Another main finding was the lower mortality rate at 30 days and 365 days in the users of GLP‐1RA. As stroke severity is one of the most important prognostic factors of poststroke mortality [23], we did subanalyses of total versus direct effect to explore this further (direct effect with adjustments of stroke severity) and found that stroke severity was a strong mediator of mortality.
Given the number of included events and the low cumulative incidence of recurrent stroke, the precision of comparisons between the groups regarding recurrent stroke was not impressive. Nevertheless, despite there being no statistical difference, it is worth noting that the point estimate for 365‐day recurrent stroke was in the same direction and largely at the same level as the estimate for 365‐day mortality. Hence, the results for recurrent stroke are in line with the evidence from the CVOTs and systematic reviews, suggesting that GLP‐1RA reduces the risk of recurrent strokes [7, 8, 9, 10]. Furthermore, it corresponds to a recent real‐world observational study that found that T2D treated with semaglutide (a GLP‐1RA) was associated with lower risk of stroke in comparison with T2D treated with DDP‐4i and the difference was greater in patients with established atherosclerotic cardiovascular disease [24].
Despite the use of a contemporary cohort of 7130 AIS patients with T2D with a high frequency of cardiovascular risk factors and prior stroke and other cardiovascular diseases, the number of users of newer glucose‐lowering medications GLP‐1RA, SGLT2i, and DPP4i was relatively low. This corresponds well with previous findings of a slow implementation process of newer glucose‐lowering medications, even after the CVOTs showed cardiovascular benefits [25, 26].
Notable strengths of this study were that we were able to compare several clinical outcomes in a complete contemporary national cohort of AIS patients with T2D who were users of either a GLP‐1RA or DPP‐4i prior to the index stroke and adjust for potential confounders as we had detailed information on comorbidity and socioeconomic factors for each stroke event. Of note, all medications in this cohort were subject to reimbursement. Baseline characteristics showed a difference in eGFR as a measure of nephropathy/kidney function. This was taken into account by adjusting for CCI scores that include moderate and severe renal disease.
This being an observational study, it is inherently impossible to verify the complete absence of residual confounding; however, we did control for a wide range of potential confounding variables and obtained a satisfactory balance (please see balance diagnostics in the Supplementary Material). This is reassuring and indicates that our adjusted effect measures were not likely to be substantially influenced by confounding. Ultimately, randomized trials comparing newer glucose‐lowering medication restricted to populations with T2D would be needed to draw firm conclusions about GLP‐1RA versus DPP‐4i effects on clinical outcomes in AIS patients with T2D.
CONCLUSIONS
We found an association of reduced risk of a “very severe stroke” and lower 30‐ and 365‐day mortality in AIS patients with T2D treated with a GLP‐1RA prior to the AIS compared to patients treated with a DPP‐4i. These results together with the results of the CVOTs underline the importance of considering GLP‐1RA in patients at high risk of a stroke or after a stroke [5, 27].
AUTHOR CONTRIBUTIONS
Sidsel Hastrup: Writing – original draft; conceptualization; methodology; project administration. Jakob Nebeling Hedegaard: Conceptualization; methodology; formal analysis; writing – review and editing. Grethe Andersen: Conceptualization; methodology; supervision; funding acquisition; writing – review and editing. Jørgen Rungby: Conceptualization; methodology; supervision; writing – review and editing. Soren Paaske Johnsen: Conceptualization; methodology; project administration; supervision; funding acquisition; writing – review and editing.
FUNDING INFORMATION
Funded by a research grant from Novo Nordisk Denmark A/S.
CONFLICT OF INTEREST STATEMENT
The authors declare the funding as a potential conflict of interest. However, Novo Nordisk Denmark A/S had no influence on data collection, no data access, and no influence on the interpretation of the results.
Supporting information
TABLE S1.
TABLE S2.
TABLE S3.
TABLE S4.
TABLE S5.
Hastrup S, Hedegaard JN, Andersen G, Rungby J, Johnsen SP. Glucagonlike peptide‐1 receptor agonists versus dipeptidyl peptidase‐4 inhibitors in ischemic strokes with diabetes 2. Eur J Neurol. 2024;31:e16329. doi: 10.1111/ene.16329
DATA AVAILABILITY STATEMENT
Research data are not shared. According to Danish law, it is not possible to provide public access to a dataset that is based on linkage of data from nationwide public registries. Access to Danish registry data can be granted to individual researchers only upon seeking approval from the Danish Data Protection Agency.
REFERENCES
- 1. Shou J, Zhou L, Zhu S, Zhang X. Diabetes is an independent risk factor for stroke recurrence in stroke patients: a meta‐analysis. J Stroke Cerebrovasc Dis. 2015;24(9):1961‐1968. [DOI] [PubMed] [Google Scholar]
- 2. Echouffo‐Tcheugui JB, Xu H, Matsouaka RA, et al. Diabetes and long‐term outcomes of ischaemic stroke: findings from get with the guidelines‐stroke. Eur Heart J. 2018;39(25):2376‐2386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lo JW, Crawford JD, Samaras K, et al. Association of prediabetes and type 2 diabetes with cognitive function after stroke: a STROKOG collaboration study. Stroke. 2020;51(6):1640‐1646. [DOI] [PubMed] [Google Scholar]
- 4. Lau LH, Lew J, Borschmann K, Thijs V, Ekinci EI. Prevalence of diabetes and its effects on stroke outcomes: a meta‐analysis and literature review. J Diabetes Investig. 2019;10(3):780‐792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Williams DM, Atkinson M, Evans M. Stroke prevention and treatment in people with type 2 diabetes: is there a role for GLP‐1 (glucagon‐like Peptide‐1) analogues? Stroke. 2023;54(5):1441‐1451. [DOI] [PubMed] [Google Scholar]
- 6. 10. Cardiovascular Disease and Risk Management: Standards of medical Care in Diabetes‐2021. Diabetes Care. 2021;44(Suppl 1):S125‐s150. [DOI] [PubMed] [Google Scholar]
- 7. Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375(19):1834‐1844. [DOI] [PubMed] [Google Scholar]
- 8. Gerstein HC, Colhoun HM, Dagenais GR, et al. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double‐blind, randomised placebo‐controlled trial. Lancet. 2019;394(10193):121‐130. [DOI] [PubMed] [Google Scholar]
- 9. Sattar N, Lee MMY, Kristensen SL, et al. Cardiovascular, mortality, and kidney outcomes with GLP‐1 receptor agonists in patients with type 2 diabetes: a systematic review and meta‐analysis of randomised trials. Lancet Diabetes Endocrinol. 2021;9(10):653‐662. [DOI] [PubMed] [Google Scholar]
- 10. Bellastella G, Maiorino MI, Longo M, et al. Glucagon‐like Peptide‐1 receptor agonists and prevention of stroke systematic review of cardiovascular outcome trials with meta‐analysis. Stroke. 2020;51(2):666‐669. [DOI] [PubMed] [Google Scholar]
- 11. Kleindorfer DO, Towfighi A, Chaturvedi S, et al. 2021 guideline for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline from the American Heart Association/American Stroke Association. Stroke. 2021;52(7):e364‐e467. [DOI] [PubMed] [Google Scholar]
- 12. Davies MJ, Aroda VR, Collins BS, et al. Management of hyperglycaemia in type 2 diabetes, 2022. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of diabetes (EASD). Diabetologia. 2022;65(12):1925‐1966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Johnsen SP, Ingeman A, Hundborg HH, Schaarup SZ, Gyllenborg J. The Danish stroke registry. Clin Epidemiol. 2016;8:697‐702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Mainz J, Hess MH, Johnsen SP. The Danish unique personal identifier and the Danish civil registration system as a tool for research and quality improvement. Int J Qual Health Care. 2019;31(9):717‐720. [DOI] [PubMed] [Google Scholar]
- 15. Wildenschild C, Mehnert F, Thomsen RW, et al. Registration of acute stroke: validity in the Danish stroke registry and the Danish National Registry of patients. Clin Epidemiol. 2013;6:27‐36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Schmidt M, Schmidt SA, Sandegaard JL, Ehrenstein V, Pedersen L, Sorensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol. 2015;7:449‐490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Pottegård A, Schmidt SAJ, Wallach‐Kildemoes H, Sørensen HT, Hallas J, Schmidt M. Data resource profile: the Danish National Prescription Registry. Int J Epidemiol. 2017;46(3):798‐798f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Arendt JFH, Hansen AT, Ladefoged SA, Sørensen HT, Pedersen L, Adelborg K. Existing data sources in clinical epidemiology: laboratory information system databases in Denmark. Clin Epidemiol. 2020;12:469‐475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Gregersen R, Wiingreen R, Rosenberg J. Health‐related register‐based research in Denmark. Ugeskr Laeger. 2018;180(43):V02180115‐V02180115. [PubMed] [Google Scholar]
- 20. Thygesen SK, Christiansen CF, Christensen S, Lash TL, Sørensen HT. The predictive value of ICD‐10 diagnostic coding used to assess Charlson comorbidity index conditions in the population‐based Danish National Registry of patients. BMC Med Res Methodol. 2011;11:83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Vergès B, Aboyans V, Angoulvant D, et al. Protection against stroke with glucagon‐like peptide‐1 receptor agonists: a comprehensive review of potential mechanisms. Cardiovasc Diabetol. 2022;21(1):242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Longo M, Di Meo I, Caruso P, et al. Circulating levels of endothelial progenitor cells are associated with better cognitive function in older adults with glucagon‐like peptide 1 receptor agonist‐treated type 2 diabetes. Diabetes Res Clin Pract. 2023;200:110688. [DOI] [PubMed] [Google Scholar]
- 23. Andersen KK, Andersen ZJ, Olsen TS. Predictors of early and late case‐fatality in a nationwide Danish study of 26,818 patients with first‐ever ischemic stroke. Stroke. 2011;42(10):2806‐2812. [DOI] [PubMed] [Google Scholar]
- 24. Evans M, Husain M, Srivastava A, Mangla KK, Kuhlman AB, Lingvay I. Risk of stroke in real‐world US individuals with type 2 diabetes receiving Semaglutide or a dipeptidyl peptidase 4 inhibitor. Adv Ther. 2024;41(5):1843‐1859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Funck KL, Knudsen JS, Hansen TK, Thomsen RW, Grove EL. Real‐world use of cardioprotective glucose‐lowering drugs in patients with type 2 diabetes and cardiovascular disease: a Danish nationwide cohort study, 2012 to 2019. Diabetes Obes Metab. 2021;23(2):520‐529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Knudsen JS, Baggesen LM, Lajer M, et al. Changes in SGLT2i and GLP‐1RA real‐world initiator profiles following cardiovascular outcome trials: a Danish nationwide population‐based study. PLoS One. 2020;15(3):e0229621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Goldenberg RM, Cheng AYY, Fitzpatrick T, Gilbert JD, Verma S, Hopyan JJ. Benefits of GLP‐1 (glucagon‐like peptide 1) receptor agonists for stroke reduction in type 2 diabetes: a call to action for neurologists. Stroke. 2022;53(5):1813‐1822. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
TABLE S1.
TABLE S2.
TABLE S3.
TABLE S4.
TABLE S5.
Data Availability Statement
Research data are not shared. According to Danish law, it is not possible to provide public access to a dataset that is based on linkage of data from nationwide public registries. Access to Danish registry data can be granted to individual researchers only upon seeking approval from the Danish Data Protection Agency.