Association of sulfonylureas with the risk of dementia: A population-based cohort study

Background

Sulfonylureas are oral glucose-lowering medications positioned as a second-line therapy for type 2 diabetes. Evidence relating them to cognitive decline has been mixed. The objective was to determine whether sulfonylurea use was associated with a differential risk of dementia compared with dipeptidyl peptidase-4 (DPP4) inhibitor use.

Methods

Using administrative data from residents in Ontario, Canada, adults aged ≥66 years who were new users of a sulfonylurea or a DPP4 inhibitor from June 14, 2011, to March 31, 2021 entered this population-based retrospective cohort study. Dementia was ascertained using a validated algorithm for Alzheimer's disease and related dementias. Propensity-score weighted Cox proportional hazards models were used to obtain adjusted hazard ratios (aHR) and confidence intervals (CI) for time to incident dementia. The observation window started at 1 year after cohort entry to mitigate protopathic bias due to delayed diagnosis. The primary analysis used an intention-to-treat exposure definition. A separate propensity-score weighted analysis was conducted to explore within-class differences in dementia risk among sulfonylurea new users selected from the primary cohort.

Results

Among 107,806 DPP4 inhibitor new users and 37,030 sulfonylurea new users, sulfonylureas compared with DPP4 inhibitors were associated with a higher risk of dementia (18.4/1000 person-years; aHR [95% CI] = 1.09 [1.04–1.15]) over a mean follow-up of 4.82 years from cohort entry. Glyburide compared to gliclazide exhibited a higher dementia risk (aHR [95% CI] = 1.17 [1.03–1.32]).

Conclusion

New use of a sulfonylurea especially glyburide was associated with a higher dementia risk compared with new use of a DPP4 inhibitor in older adults with diabetes.     

Post-initiation predictors of discontinuation of the sodium-glucose cotransporter-2 inhibitors: A comparative cohort study from the United Kingdom

Aims

To assess post-initiation predictors of discontinuation of sodium-glucose cotransporter-2 (SGLT2) inhibitors compared to dipeptidyl-peptidase-4 (DPP-4) inhibitors in the United Kingdom.

Materials and Methods

We conducted a comparative population-based retrospective cohort study using primary care data from the UK Clinical Practice Research Datalink (CPRD) with linked data to hospital and death records. We included new metformin users who initiated either SGLT2 inhibitors or DPP-4 inhibitors between January 2013 and October 2019. The main outcome was treatment discontinuation, defined as the first 90-day gap after the estimated treatment end date. We used a series of extended Cox models to assess which time-dependent predictors were associated with treatment discontinuation. To test if the hazard ratio of discontinuation for each predictor was statistically different between SGLT2 and DPP-4 inhibitors, an exposure-predictor interaction term was added to each model.

Results

There were 2550 new users of SGLT2 inhibitors and 8195 new users of DPP-4 inhibitors. Approximately 69% of SGLT2 inhibitor and 74% of DPP-4 inhibitor users had discontinued treatment by the end of follow-up. Occurrence of fractures after treatment initiation was a significant predictor of discontinuation of SGLT2 inhibitors (hazard ratio [HR] 4.13, 95% confidence interval [CI] 2.12-8.06) but not DPP-4 inhibitors (HR 0.93, 95% CI 0.79-1.11). The rate of treatment discontinuation was significantly higher for those with low estimated glomerular filtration rate and minimal contact with the healthcare system. Efficacy endpoints, such as heart failure and glycated haemoglobin level, were not associated with treatment discontinuation.

Conclusions

Our findings reflect some discrepancy between the available evidence and prescribing behaviour for SGLT2 inhibitors.     

A New Family of Nonconforming Elements with $H$(curl)-Continuity for the 3D Quad-Curl Problem

We propose and analyze a new family of nonconforming finite elements for the three-dimensional quad-curl problem. The proposed finite element spaces are subspaces of $\boldsymbol{H}$(curl), but not of $\boldsymbol{H}$(grad curl), which are different from the existing nonconforming ones [10,12,13]. The well-posedness of the discrete problem is proved and optimal error estimates in discrete $\boldsymbol{H}$(grad curl) norm, $\boldsymbol{H}$(curl) norm and $L^2$ norm are derived. Numerical experiments are provided to illustrate the good performance of the method and confirm our theoretical predictions.