Introduction: Sodium–glucose cotransporter 2 inhibitors (SGLT2i) have demonstrated significant cardiovascular benefits, and their use has markedly increased in recent years. However, in routine clinical practice, treatment duration is much shorter and more variable than in clinical trials, and it remains uncertain whether such abbreviated use limits the full therapeutic potential of SGLT2i.
Aims: To quantify the effect of different treatment durations of SGLT2i on cardiovascular outcomes in people with type 2 diabetes (T2D).
Methods: We emulated a hypothetical pragmatic trial using nationwide healthcare data from South Korea. Among patients who were aged 18 years or older with T2D, we compared the following treatment strategies: 1) initiation of SGLT2i with treatment durations of <1 year, 2) 1–2 years, 3) 2–3 years, and 4) >3 years. The primary outcomes of interest were major adverse cardiovascular events (MACE) and hospitalization for heart failure (HHF). We employed the cloning, censoring, and weighting method to emulate the randomization, and weighted Kaplan-Meier methods were applied to estimate 5-year absolute risks, calculated risk differences, and risk ratios (RRs). The 95% confidence intervals (CIs) were obtained from 300 non-parametric bootstraps.
Results: 1,174,088 eligible patients were included (mean age, 57.3 years; 40.5% female). Compared to patients who used SGLT2i for less than 1 year, those treated for 1–2 years, 2–3 years, and more than 3 years had progressively lower risks of MACE (RRs: 0.93 [95% CI, 0.90–0.96], 0.82 [0.78–0.85], and 0.69 [0.67–0.72], respectively), as well as HHF (RRs: 0.93 [0.90–0.97], 0.90 [0.80–0.98], and 0.74 [0.67–0.79], respectively) over a 5-year follow-up. The results remained consistent across several demographic and clinical subgroups, and in sensitivity analyses accounting for patients who discontinued treatment due to adverse effects.
Conclusions: These findings underscore the importance of sustained SGLT2i use to maximize their cardiovascular benefits.
Keywords: Target trial emulation, sodium-glucose cotransporter 2 inhibitors, cardiovascular outcomes.

Introduction: The current recommended international normalised ratio (INR) in guidelines is mainly derived from trials in Western populations. The optimal INR target in Asian patients with atrial fibrillation (AF) remains unclear.
Aims: This study aims to determine the optimal INR target for Asian patients receiving warfarin with non-valvular AF (NVAF) with and without mitral stenosis (MS).
Methods: This retrospective cohort study included patients newly diagnosed with AF with and without MS between 1 January 2010 and 31 December 2023 using territory-wide electronic health records in Hong Kong. The INR-specific incidence rates of thromboembolism and bleeding were computed using the Rosendaal method. The incidence rate ratio (IRR) was calculated by conditional Poisson regression. The association between repeated INR measurements and outcomes was further evaluated using the generalised estimating equation (GEE) model.
Results: A total of 11,518 patients with AF without MS and 1,070 patients with AF-MS were identified in this study. The incidence rate of thromboembolism decreased sharply as the INR increased from 1.5-1.99 to 2.0-2.49, both in patients with AF without MS and AF-MS. The lowest bleeding incidence (2.4 events per 100 person-years) in patients with AF without MS was observed at INR of 2.0-2.49, rising to 3.4 events per 100 person-years at INR of 2.5-2.99. In patients with AF-MS, the incidence rate of bleeding remained relatively low when INR was in the range of 1.5-2.99, with a slight increase at 3.0-3.49. Both conditional Poisson regression and GEE models found a significant risk of bleeding among patients with AF without MS when INR≥2.5 but it increased among patients with AF-MS when INR≥3.
Conclusions: INR 1.5-2.5 is non-inferior to INR 2.0-3.0 for Asian patients with NVAF and an INR range of 2.0-2.49 may be an optimal target. Standard-intensity anticoagulation (INR 2.0-3.0) is still preferable for Asian patients with AF-MS.

Introduction: Proprotein convertase subtilisin-kexin type 9 inhibitors (PCSK9i) are novel lipid-lowering medications (LLMs) but their real-world prescribing and response in Asian populations remain unclear.
Aims: To describe patient characteristics, concomitant treatments after PCSK9i initiation, and lipid changes in Hong Kong patients prescribed monoclonal PCSK9i.
Methods: Cohort study using electronic health records from the Hong Kong Hospital Authority of patients newly prescribed alirocumab or evolocumab between 1 August 2016 and 12 December 2024. Descriptive statistics were used to analyse patient characteristics, LLM treatment trajectories, and lipid changes at one-year of follow-up.
Results: Among 3,381 PCSK9i initiators, 3,296 met eligibility criteria (mean age 62.8 years; 70.5% male; median baseline low-density lipoprotein cholesterol [LDL-C] 2.9 mmol/L). Before PCSK9i initiation, 90.3% of patients received a LLM, 82.9% with a statin and 64.5% with ezetimibe. For longitudinal analysis, 2,195 patients had ≥1 year of follow-up (median 2.6 years; interquartile range [IQR] 1.6–4.0). 84.0% of patients initiated a PCSK9i in combination with another LLM, primarily statins and ezetimibe. Between baseline and one year after PCSK9i initiation, the use of any statin or ezetimibe declined by 5.8% and 12.5%, respectively, while high-intensity statin use declined by 11.8%. Mean relative LDL-C decreased by 44.5% (95% confidence interval [CI]: 41.8%–47.1%), and 66.9% of patients achieved an LDL-C < 1.8 mmol/L. Non-high-density lipoprotein cholesterol (non-HDL-C) decreased by a mean of 40.2% (95% CI: 38.6%-41.7%). Greater reductions in LDL-C and non-HDL-C were observed among patients receiving PCSK9i plus a statin or ezetimibe as compared with PCSK9i alone.
Conclusions: The absolute number of patients using PCSK9i in Hong Kong is low, but has increased rapidly since becoming funded in 2021. One-year follow-up in our cohort demonstrates that PCSK9i were highly effective in reducing LDL-C and non-HDL-C in clinical practice, particularly when combined with a statin or ezetimibe.

Aims: Drug-induced liver injury (DILI) is a significant concern during anti-tuberculosis (anti-TB) treatment. The associations between diabetes mellitus (DM) comorbidity and anti-TB DILI have been reported but with inconsistent results. Based on the anti-TB treatment cohort in Jiangsu Province, the aim of this study is to answer whether DM influences anti-TB DILI occurrence in Chinese patients.

Methods: A prospective cohort study was conducted, enrolling patients receiving anti-TB treatment from four designated anti-TB hospitals from June 2017 to December 2024. Patients were grouped by DM status and followed up until completion of treatment. Propensity score matching (PSM) was performed at a 1:1 ratio to ensure comparability between groups of interest. Multivariate Cox regression analysis was employed to identify the factors influencing DILI occurrence, with results presented as hazard ratios (HR) and 95% confidence intervals (95%CI).

Results: A total of 3,952 patients with TB were included, among whom 759 had DM. During the treatment period, 522 cases of DILI occurred. The incidence rates of DILI were 14.62% in DM patients and 12.87% in non-DM patients (c2=2.200, P=0.333). The median times to DILI development were 45 days and 41 days, respectively (Z=-0.664, P=0.507). After PSM, 350 cases were included in each group, and multivariate Cox regression analysis showed that DM comorbidity did not increase the risk of DILI during TB treatment (HR=0.921, 95%CI: 0.705-1.202, P=0.544). The preventive use of hepatoprotective drugs significantly reduced the risk of DILI (HR=0.622, 95%CI: 0.470-0.823, P<0.001), while elevated baseline aspartate aminotransferase levels increased DILI risk (HR=1.013, 95%CI: 1.006-1.020, P<0.001).

Conclusions: Based on the prospective cohort study, the occurrence of DILI during anti-TB treatment in TB patients is not associated with the presence of comorbid DM, but is related to the preventive use of hepatoprotective drugs and baseline aspartate aminotransferase levels.

Key words: Diabetes mellitus; Tuberculosis; Drug-induced liver injury

Introduction
Recent studies suggest that the use of glucagon-like peptide-1 receptor agonists (GLP-1 RAs), sodium-glucose cotransporter-2 (SGLT-2) inhibitors or dipeptidyl peptidase-4 (DPP-4) inhibitors, may reduce the risk of asthma exacerbation. However, evidence was limited when comparing these antidiabetic agents.
Aims
To compare the risk of asthma exacerbations among users of SGLT-2 inhibitors, GLP-1RAs, and DPP-4 inhibitors.
Methods
Using data from the TriNetX Global Collaborative Network from January 2016 through March 2025, we emulated three target trials to compare the risk of asthma exacerbation among patients with comorbid asthma and type 2 diabetes. The active-comparator-new-user design was implemented to minimize potential bias. We assigned patients based on their first prescription of study drugs, and propensity score matching was implemented to mimic the randomization. We defined the date of first prescription as the index date. We further examined the heterogeneity of treatment effects based on baseline BMI categories (obese: >29.9; overweight: 24.9–29.9; normal and underweight: <24.9). The study outcome was the composite asthma endpoint, consisting of exacerbation diagnoses, use of oral or injectable steroids, and use of rescue medications.
Results
There were 163,875 patients who received GLP-1 RAs, DPP-4 inhibitors, or SGLT-2 inhibitors between 2016 and the end of March 2025. Compared with DPP-4 inhibitors, the use of GLP-1 RAs and SGLT-2 inhibitors was associated with a lower risk of composite asthma endpoint, with hazard ratios (HRs) of 0.87 (95% CI: 0.85–0.90) and 0.92 (95% CI: 0.89–0.95), respectively. Effect was similar among obese patients and overweight patients, but not among those with normal or underweight.
Conclusions
Our findings indicate that the use of GLP-1 RAs and SGLT-2 inhibitors was associated with a reduced risk of asthma exacerbation. Furthermore, the effect modification by patients’ baseline BMI suggests the need to individualize glucose-lowering therapy.

Keywords: Asthma exacerbation, Glucagon-like peptide-1 receptor agonists, Sodium-glucose cotransporter-2 inhibitors

Introduction:
Pioglitazone, approved for type 2 diabetes, showed potential anticonvulsant effects in animal studies, but its effects in real-world populations remain uncertain.
Objective:
To examine anticonvulsant effects of pioglitazone among patients with type 2 diabetes.
Method:
We conducted a retrospective cohort study under a target trial emulation framework using data from the US Collaborative Network. We identified patients initiating pioglitazone or DPP4 inhibitors (2004-2024), defining index date as initiation date. We applied a one-year washout period to exclude prevalent users. We also excluded patients with prior outcomes of interest or those who received both drugs on the index date. We used ICD-10-CM codes to identify outcomes of interest, including convulsions (R56), epilepsy (G40), and a composite outcome of both. We used intention-to treat analysis and 1:1 propensity score matching (PSM) to emulate treatment assignment, calculating scores based on age, sex, HbA1c, and comorbidities. We applied Cox models to estimate hazard ratios (HR) and 95% confidence interval (CI), using DPP4 inhibitors as reference. We conducted subgroup analyses by age, sex, HbA1c, and body mass index.

Results :
We included 698,270 patients (185,595 pioglitazone, 526,924 DPP4 inhibitors). The mean age was 60.7 years (SD 12.0) and 61.9 years (SD 12.4), with 54.4% and 48.9% male, respectively. After PSM, 185,440 remained in each group with balanced baseline characteristics. We observed an association between a reduced risk of composite outcome (HR=0.95, 95% CI: 0.91-0.99) and individual outcomes, convulsions (HR=0.96, 95% CI: 0.91-1.02) and epilepsy (HR=0.88, 95% CI: 0.82-0.94). Results from subgroup analyses were consistent, except in patients under 65.

Conclusions:
We observed an association between a reduced risk of convulsion or epilepsy and pioglitazone use, compared with DPP4 inhibitors, suggesting the potential anticonvulsant effect of pioglitazone. Further investigation is warranted to explore its anticonvulsant effect.

Keyword: pioglitazone, anticonvulsant, US Collaborative Network, type 2 diabetes.