Antidiabetic drug and chronic respiratory disease in type 2 diabetes: a network meta-analysis and Mendelian randomization analysis

Abstract

Background:

Multimorbidity management is crucial in aging populations, particularly for chronic respiratory diseases (CRDs) and diabetes. Their coexistence increases exacerbation and mortality risks. Antidiabetic drugs may differentially impact respiratory outcomes.

Objectives:

To evaluate the effects of antidiabetic drugs on CRDs in individuals with type 2 diabetes (T2D).

Design:

Systematic review with conventional and network meta-analyses (NMAs).

Data sources and methods:

We conducted a systematic review with conventional meta-analysis, NMA for drug comparisons, and two-sample Mendelian randomization for genetic evidence. We assessed nine antidiabetic drug classes for associations with CRD exacerbations, incidence, and all-cause mortality in individuals with T2D.

Results:

Our analysis of 33 studies (n = 939,064) found that glucagon-like peptide-1 receptor agonists (GLP1RAs) offered the strongest protection against acute exacerbations (hazard ratio vs insulin: 0.40, 95% credible interval 0.29–0.56) and had the highest first-rank probability (72.95%), followed by sodium-glucose cotransporter-2 inhibitors (SGLT2is; 46.19%). Thiazolidinediones reduced incident CRD by 21% (odds ratio (OR) 0.79, 0.69–0.91), while metformin lowered all-cause mortality risk by 16% (OR 0.84, 0.74–0.95). Mendelian randomization confirmed GLP1R expression was associated with reduced asthma risk (OR 0.9987, 0.9979–0.9996).

Conclusion:

GLP1RAs provide the strongest protection against CRD exacerbations in T2D patients, with SGLT2is as the second-most effective option. These findings highlight the potential for personalized antidiabetic therapy in multimorbidity management. Further studies should validate these findings and elucidate underlying mechanisms.

Trial registration:

PROSPERO (ID: CRD42024542379).

Plain language summary

Antidiabetic drug and chronic respiratory disease

Our study revealed GLP1RAs as most protective against respiratory exacerbation. Thiazolidinediones reduced incident chronic respiratory disease in T2D. Antidiabetic drugs decreased mortality in individuals with comorbid T2D and chronic respiratory disease, specially metformin. Genetic evidence linked GLP1R activation to lower asthma risk.

Graphical abstract

Keywords

asthma chronic obstructive pulmonary disease chronic respiratory disease network meta-analysis type 2 diabetes

Introduction

In an aging society, the management of multimorbidity has become a significant public health concern.^1,2 Chronic respiratory diseases (CRDs) and diabetes are leading causes of global mortality, ranking third and fourth, respectively,³ and their frequent coexistence elevates the risk of acute exacerbations and death.^4
–8

A variety of antidiabetic drugs are used clinically, each with distinct mechanisms. These drugs may differentially impact CRD, as they share pathophysiological pathways such as oxidative stress, chronic inflammation, and insulin resistance.^9,10 Recent evidence suggests that glucagon-like peptide-1 receptor agonists (GLP1RAs) and sodium-glucose cotransporter-2 inhibitors (SGLT2is) may reduce acute exacerbation risk in patients with type 2 diabetes (T2D) and chronic obstructive pulmonary disease (COPD).^11
–14 Other agents, including metformin, thiazolidinediones (TZDs), and sulfonylureas, have been associated with fewer hospitalizations and lower mortality.^15–23 Recent meta-analyses also indicate GLP1RAs²⁴ and SGLT2is²⁵ may lower respiratory disease incidence in T2D.

However, a comprehensive comparison of different antidiabetic drug classes and their associations with the key clinical outcomes of acute exacerbation, incidence, and all-cause mortality of CRDs has not been performed. We therefore conducted comprehensive meta-analyses and a network meta-analysis (NMA) to examine and compare these effects. We further employed two-sample Mendelian randomization (TSMR) to explore potential causal links at the genetic level.

Methods

The protocol for this systematic review and NMA was prospectively registered in PROSPERO (ID: CRD42024542379). This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 and the extension statement for network meta-analyses (PRISMA-NMA).^26,27

Search strategy

PubMed, Embase, Web of Science, Cochrane Library, and Scopus were searched for studies published up to February 1, 2024 (Appendix 1). Two reviewers independently screened records, extracted data, and assessed study quality using the Newcastle–Ottawa Scale (NOS); discrepancies were resolved by a third reviewer.

Eligibility criteria

Eligible studies were original articles in English involving adults with diagnosed T2D and CRD (asthma or COPD). They had to compare the use of one or more of nine predefined antidiabetic drug classes (metformin, sulfonylureas, TZDs, alpha-glucosidase inhibitors (AGIs), dipeptidyl peptidase-4 inhibitors (DPP-4is), SGLT2is, GLP-1RAs, non-sulfonylurea insulin secretagogues, and insulin) against non-use, placebo, or another active drug for ⩾3 weeks, providing quantitative effect estimates (odds ratio (OR), hazard ratio (HR), or risk ratio) with 95% confidence intervals (CIs).

Data extraction and quality assessment

Two reviewers independently extracted data using a standardized form, capturing study characteristics, population details, and outcomes. The methodological quality of included observational studies was assessed using the NOS, with scores detailed in Table S2. Any discrepancies in extraction or assessment were resolved through discussion with a third reviewer.

Statistical analysis

To evaluate the effects of antidiabetic drugs on CRDs in individuals with T2D, we conducted a systematic review with traditional meta-analyses to assess outcomes including disease incidence, acute exacerbations, and all-cause mortality. For exacerbation risk, where sufficient direct and indirect comparisons were available across studies, we performed a NMA to enable comparative effect estimates and ranking of interventions.

We assessed heterogeneity using the I² metric and employed random-effects models to analyze the association between various antidiabetic drugs and the risks of acute exacerbation, incidence, and all-cause mortality of CRDs. HRs, ORs, and relative risks were pooled as ORs, with data directly obtained from studies and preference given to covariate-adjusted estimates when available.²⁸ Funnel plots and Egger’s test were used to evaluate small-study bias and symmetry. Subgroup analyses were conducted by age (<65 vs >65 years), geographic region (Asia vs North America), outcome (COPD vs asthma), effect estimate type (OR vs HR), NOS quality (>8 vs <8), and sample size (>10 vs <10 k). Sensitivity analyses were performed using a one-by-one exclusion method, considering only metformin due to the limited sample size for other drugs.^29,30

We conducted an NMA of studies comparing two antidiabetic drugs for acute exacerbation of CRDs. For instance, if a study compared drug A + drug B with drug A + drug C, only the differing drugs (B vs C) were evaluated. Inconsistency was assessed via the node-splitting method with Bayesian p-values comparing direct and indirect estimates, and rank probabilities were calculated and visually displayed as bar plots. Bayesian NMA was performed using a random-effects model with JAGS and R (v4.3.1, R Core Team, 2023), utilizing the gemtc (v1.0–2) and rjags (v4–15) packages. Additionally, five sensitivity networks were run: monotherapy only, COPD only, asthma only, HR-only, and OR-only studies.

All analyses were performed using STATA version 16.0 (Stata Corporation, College Station, TX, USA) and R software (version 4.3.1) with the gemtc and TwoSampleMR packages. Statistical significance was set at a two-tailed p-value <0.05.

Two-sample Mendelian randomization

For the TSMR analysis, genetic instruments (single-nucleotide polymorphisms, SNPs) for antidiabetic drug target genes were selected from the Genotype-Tissue Expression project (Table S3). For each gene, the SNP with the strongest association to its expression was chosen, validated via its established association with glycated hemoglobin (HbA1c) levels, and subjected to linkage disequilibrium clumping (distance = 1000 kb, r² < 0.01). Where necessary, proxy SNPs were used (r² > 0.8; palindromic SNPs with Minor Allele Frequency >0.3). Outcome data for asthma and COPD were obtained from published genome-wide association studies via the OpenGWAS database (Table S4). The primary TSMR analysis, performed using the TwoSampleMR package (v.0.6.0), applied the inverse-variance weighted or Wald ratio method to estimate ORs (95% CIs) per unit increase in genetically predicted gene expression. We subsequently assessed potential mediation by inflammatory biomarkers. First, we estimated the effect of each gene expression instrument on 91 inflammatory biomarkers (EBI-GWAS Catalog). Biomarkers with significant associations were then analyzed for their effect on asthma and COPD. The mediation effect was calculated as the product of these two coefficients (β1 * β2).³¹

Results

Literature selection, study characteristics, and quality assessment

Of 10,078 identified records, 33 studies involving 939,064 individuals with T2D and CRD met the inclusion criteria (Figure 1).^{17,20
–22,23,32
–36} Study and participant characteristics are detailed in Table S1. Included studies were conducted in the USA, UK, Taiwan, Australia, China, Japan, and Hong Kong. Sample sizes ranged from 94 to 285,992 participants, with a mean age of 62.3 years and 54.1% men. All nine predefined antidiabetic drug classes—metformin, sulfonylureas, TZDs, AGIs, DPP4is, SGLT2is, GLP1RAs, non-sulfonylurea insulin secretagogues, and insulin—were represented. Based on the NOS, the methodological quality of included studies was high (scores 7–9).

Figure 1.

PRISMA flow diagram.

The effect of antidiabetic drugs on acute exacerbation in individuals with combined T2D and CRD

Conventional meta-analysis

Thirteen studies (201,473 individuals) compared antidiabetic drug use versus non-use for acute exacerbation. Overall, antidiabetic drug use was associated with a 14% lower risk of exacerbation (OR 0.86, 95% CI 0.76–0.96; Figure 2(c)). Specifically, SGLT2is (OR 0.20, 95% CI 0.13–0.32) and TZDs (OR 0.89, 95% CI 0.81–0.99) were associated with significant risk reduction. TZDs reduced the risk of asthma exacerbation (OR 0.79, 95% CI 0.63–1.00; Figure 2(b)) but not COPD exacerbation (OR 0.91, 95% CI 0.81–1.02; Figure 2(a)).

Figure 2.

A meta-analysis of the impact of using antidiabetic drugs on the risk of acute exacerbation of CRD. (a) Risk of exacerbation of chronic obstructive pulmonary disease. (b) Risk of asthma exacerbation. (c) Risk of exacerbation of CRD.

Subgroup analyses indicated metformin reduced exacerbation risk in individuals aged >65 years (OR 0.78, 95% CI 0.70–0.87), in studies with NOS score <8 (OR 0.74, 95% CI 0.60–0.91), and in studies with sample size <10,000 (OR 0.36, 95% CI 0.20–0.67; Figure S1). A sensitivity analysis for metformin found no significant change when individual studies were omitted (Figure S2). No significant publication bias was detected (Egger’s test p = 0.15; Figure S3). A sensitivity analysis was only conducted for metformin use versus non-use due to insufficient prior evidence.

The results from the one-by-one exclusion method showed that the association of metformin with the risk of CRD exacerbation remained insignificant (Figure S2). The Egger test results in this analysis showed no significant bias (p-value = 0.15) in the relationship between antidiabetic drugs and acute exacerbation of CRD, and the funnel plot was consistent with the Egger test results (Figure S3).

The network meta-analysis

The NMA included studies directly comparing two distinct antidiabetic drugs, involving eight classes: metformin, sulfonylureas, TZDs, AGIs, DPP4is, SGLT2is, GLP1RAs, and insulin (Figure 3).

Figure 3.

Network of available comparisons of antidiabetic drugs for the risk acute exacerbation of chronic respiratory disease. The circle size in each network corresponds to the number of participants randomly assigned to the treatment comparison. The thickness of each line corresponded to the number of trials comparing the two connected treatments.

GLP1RAs were associated with the greatest reduction in exacerbation risk, with ORs (95% credible interval) of 0.61 (0.45–0.83) versus metformin, 0.64 (0.52–0.79) versus sulfonylureas, 0.73 (0.54–0.96) versus TZDs, 0.67 (0.50–0.91) versus AGIs, 0.66 (0.53–0.82) versus DPP4is, and 0.40 (0.29–0.56) versus insulin. SGLT2is also significantly lowered risk compared to most other drugs, though not versus GLP1RAs (OR 1.10, 0.81–1.40) or TZDs (OR 0.82, 0.58–1.20). Insulin was consistently associated with the highest relative risk (Table 1).

Table 1.

Network meta-analysis of the association between antidiabetic drugs and the risk of acute exacerbation of chronic respiratory disease.

	Metformin	Sulfonylurea	TZD	AGI	DDP4i	SGLT2i	GLP1RA	Insulin
Metformin		1.00(0.82,1.30)	1.30(0.95,1.70)	1.10(0.86,1.40)	1.10(0.84,1.40)	1.50(1.10,2.10)	1.60(1.20,2.20)	0.66(0.51,0.85)
Sulfonylurea	0.96(0.77,1.20)		1.20(0.92,1.60)	1.10(0.84,1.30)	1.00(0.88,1.20)	1.50(1.20,1.90)	1.60(1.30,1.90)	0.63(0.49,0.82)
TZD	0.79(0.59,1.10)	0.82(0.62,1.10)		0.87(0.67,1.10)	0.85(0.64,1.10)	1.20(0.86,1.70)	1.30(0.92,1.80)	0.52(0.37,0.73)
AGI	0.91(0.71,1.20)	0.95(0.75,1.20)	1.20(0.88,1.50)		0.98(0.77,1.20)	1.40(1.00,1.90)	1.50(1.10,2.00)	0.60(0.44,0.81)
DDP4i	0.92(0.73,1.20)	0.96(0.82,1.10)	1.20(0.88,1.60)	1.00(0.80,1.30)		1.40(1.10,1.80)	1.50(1.20,1.90)	0.61(0.46,0.81)
SGLT2i	0.65(0.47,0.89)	0.68(0.54,0.84)	0.82(0.58,1.20)	0.72(0.52,0.97)	0.70(0.55,0.88)		1.10(0.81,1.40)	0.43(0.30,0.59)
GLP1RA	0.61(0.45,0.83)	0.64(0.52,0.79)	0.78(0.56,1.10)	0.67(0.50,0.91)	0.66(0.53,0.82)	0.94(0.73,1.20)		0.40(0.29,0.56)
Insulin	1.50(1.20,2.00)	1.60(1.20,2.10)	1.90(1.40,2.70)	1.70(1.20,2.30)	1.60(1.20,2.20)	2.30(1.70,3.30)	2.50(1.80,3.40)

Estimates are presented as hazard ratios (95% credible interval).

Bold values represent statistically significant result.

AGI, alpha-glucosidase inhibitors; DPP4i, dipeptidyl peptidase-4 inhibitors; GLP1RA, glucagon-like peptide-1 receptor agonists; SGLT2i, sodium-glucose cotransporter-2 inhibitors; TZD, thiazolidinediones.

Treatment rankings, presented as bar plots in Figure 4, identified GLP1RAs as the most likely best treatment (first-rank probability 72.95%), followed by SGLT2is (second-rank probability 46.19%). TZDs, metformin, and AGIs had the highest probabilities for the third (33.37%), fourth (40.81%), and fifth (32.84%) ranks, respectively, while insulin had the highest probability of ranking last (99.18%; Table S3).

Figure 4.

Bar plots showing the probability of each antidiabetic drug being at first to eight ranks. The intensity of the color distinguishes the rank probabilities of the effect of each drug from first to eighth.

The node-splitting method indicated no significant inconsistency between direct and indirect evidence across all treatment loops (p > 0.05; Figure S4). Sensitivity analyses, including those restricted to COPD exacerbations, yielded consistent findings, with GLP1RAs remaining the most effective option (Figures S7 and S8, Tables S5 and S6; Supplemental Appendix 3.1).

Evidence of combination therapy on the risk of acute exacerbation of CRD

Evidence regarding specific combination therapies was limited to two articles. One study reported that dual therapy with metformin and a sulfonylurea was associated with a lower risk of COPD exacerbation compared to other dual regimens (OR 0.78, 95% CI 0.64–0.95).¹⁷ In contrast, another study found the combination of a sulfonylurea and a TZD was superior to sulfonylurea plus metformin (OR 0.69, 95% CI 0.51–0.94).³² Additionally, triple therapy with metformin, sulfonylurea, and TZD reduced exacerbation risk compared to metformin, sulfonylurea, and an AGI (OR 0.81, 95% CI 0.68–0.96). These contrasting findings highlight the need for further investigation to clarify the effects of specific antidiabetic drug combinations.

The effect of antidiabetic drugs on incident CRD

Seven studies (182,224 individuals) assessed the association between antidiabetic drug use and new‑onset CRD, including metformin, TZDs, sulfonylureas, AGIs, insulin, non‑sulfonylurea secretagogues, and DPP4is.

Pooled results indicated that TZD use was associated with a 21% lower risk of incident disease (OR 0.79, 95% CI 0.69–0.91; Figure S30). No other drug class showed a significant association. Subgroup analyses did not reveal differential effects for asthma versus COPD incidence (Figure S31). One study not meeting our outcome definition³⁷ reported that SGLT2is reduced incident respiratory disease risk compared to DPP4is (HR 0.65, 95% CI 0.54–0.79).⁴⁶ No significant publication bias was detected (Egger’s test p = 0.07; Figure S32).

The effect of antidiabetic drugs on all-cause mortality in individuals with combined T2D and CRD

Eight studies (141,103 individuals) evaluated all-cause mortality, assessing metformin, TZDs, GLP1RAs, sulfonylureas, and insulin. Meta-analysis showed that metformin use was associated with a 16% lower risk of all-cause mortality compared with non-use (OR 0.84, 95% CI 0.74–0.95). This protective association remained significant in a sensitivity analysis restricted to patients with COPD (OR 0.86, 95% CI 0.78–0.94; Figure S33). No significant publication bias was detected (Egger’s test p = 0.65; Figure S34).

Mendelian randomization of the relationship between antidiabetic drugs target gene with asthma and COPD

TSMR assessed potential causal relationships between genetically predicted expression of antidiabetic drug target genes and CRDs (Table S15). Higher expression of GLP1R (the target for GLP1RAs) was associated with reduced asthma risk (OR 0.998, 95% CI 0.997–0.999). In contrast, higher expression of SLC5A2 (encoding SGLT2) was weakly associated with increased COPD risk (OR 1.002, 95% CI 1.000–1.004). Expression of PRKAB1 (related to metformin action) was associated with increased risk of both COPD (OR 1.099, 95% CI 1.001–1.207) and asthma (OR 1.006, 95% CI 1.002–1.011). Expression of sulfonylurea-related genes ABCC8 and KCNJ11 was associated with increased COPD risk (OR 1.174, 95% CI 1.087–1.267 and OR 1.108, 95% CI 1.052–1.167, respectively), but not with asthma. No significant causal relationships were detected for TZD- or insulin-related genes.

In a mediation analysis, no inflammatory biomarkers showed statistically significant mediation effects. However, a suggestive inverse relationship was observed between SLC5A2 expression and levels of the chemokine CXCL10 (β = −0.043, SE 0.025, p = 0.093; Table S17).

Discussion

This comprehensive meta-analysis and NMA, comprising 33 observational studies, evaluated the effects of antidiabetic drugs on CRDs in individuals with T2D. Our principal findings indicate that GLP1RAs offer the strongest protection against acute exacerbations, while insulin is associated with the highest relative risk. Mendelian randomization analysis provided genetic evidence supporting a causal role for GLP1R activation in reducing asthma risk. TZDs were associated with a reduction in both exacerbations and incident disease, and metformin was linked to lower all-cause mortality in patients with T2D and COPD.

Comparison with previous evidence

Prior meta-analyses have not comprehensively assessed antidiabetic drugs’ effects on acute exacerbations and all-cause mortality in T2D patients with asthma or COPD. We address this gap with the first NMA directly comparing multiple agents for exacerbation risk. Earlier work, including an NMA of cardiorenal trials, found SGLT2is reduced asthma incidence (unlike GLP1RAs and DPP4is), but was limited to placebo comparisons of four drug classes and focused solely on disease onset.³⁸ Other meta-analyses reported reduced respiratory risk with GLP1RAs and SGLT2is, but were restricted to drug use versus non-use analyses.^13,24,25,39 Our study advances beyond these limitations by systematically ranking all major drug classes, demonstrating GLP1RAs as most protective, followed by SGLT2is.

Potential mechanisms

GLP1RAs lower blood glucose through multiple mechanisms including insulin secretion, glucagon suppression, and delayed gastric emptying.⁴⁰ Beyond glycemic control, they exert anti-inflammatory and antioxidant effects, restore protease balance, and may modulate immune function in COPD.⁴¹ Their protective role in respiratory diseases may also involve reducing airway hyperresponsiveness, as observed in animal models,^42,43 isolated human airways,⁴⁴ and patients with T2D and COPD.⁴⁵

Pulmonary dysfunction, potentially driven by insulin resistance and microangiopathy, may be an underrecognized complication of T2D.^46,47 GLP1RAs have been shown to improve lung function parameters, including FVC, FEV1, and PEF, particularly in patients with substantial HbA1c reduction.^14,48 This effect may involve GLP-1 receptor expression in alveolar cells.⁴⁹ Supporting this, the LIRALUNG trial reported that liraglutide reduced serum surfactant protein D levels, suggesting a role in alveolar stabilization.¹⁴

DPP-4is may attenuate bronchial hyperresponsiveness by increasing circulating GLP-1 levels.⁵⁰ This is particularly relevant given the elevated DPP-4 expression observed in pulmonary macrophages from COPD patients. However, the relationship between DPP-4 and COPD is complex, with studies reporting both reduced^51,52 and increased⁵³ DPP-4 levels in different COPD populations. These discrepancies in expression may explain the differential respiratory effects observed between GLP1RAs and DPP-4is.

SGLT2is lower serum glucose by blocking renal reabsorption, potentially reducing substrate for pulmonary glucose metabolism and CO₂ production.⁵⁴ Beyond metabolic effects, SGLT2is may mitigate COPD exacerbations through NLRP3 inflammasome inhibition, indicating an anti-inflammatory mechanism.⁵⁵ Supporting this, our Mendelian randomization analysis found that genetically predicted SLC5A2 (encoding SGLT2) expression was inversely correlated with levels of CXCL10, a chemokine involved in inflammation.

TZDs, as PPARγ agonists, enhance insulin sensitivity while directly suppressing NF-κB-mediated proinflammatory signaling.^56,57 This dual metabolic and anti-inflammatory action likely underlies their observed benefit in COPD, where they appear to attenuate both airway and systemic inflammation during exacerbations.⁵⁸

Metformin primarily acts through AMP-activated protein kinase (AMPK) activation to suppress hepatic gluconeogenesis. In the airways, it restores AMPK-α activity, attenuating inflammation and remodeling in preclinical models,^59
–61 reduces oxidative stress and mitochondrial dysfunction,⁶² and modulates the GLP-1/DPP-4 axis.^63
–65 These mechanisms suggest metformin’s benefits in respiratory disease extend beyond glycemic control, involving both metabolic and anti-inflammatory pathways. Furthermore, the observed reduction in all-cause mortality associated with metformin may reflect broader systemic benefits. Metformin has been linked to a lower risk of several cancers in prior studies,^66
–69 and neutral or adverse oncological associations have been reported for other agents such as insulin.⁷⁰ Although our study did not assess cause-specific mortality, this systemic dimension may contribute to the survival advantage seen with metformin in this multimorbid population.

The impact of insulin on airway diseases remains complex. While inhaled insulin has been associated with reduced forced expiratory volume,⁷¹ systemic hyperinsulinemia is linked to increased airway hyperresponsiveness and smooth muscle proliferation.^72,73 Our results are consistent with this mechanistic profile, as insulin was associated with the highest relative risk of exacerbations among all antidiabetic drugs studied.

Public health impact

The rising burden of multimorbidity in aging populations represents a major public health challenge.^2,74 With approximately 539 million individuals living with T2D and an estimated 742 million affected by CRDs worldwide, the co-occurrence of these conditions presents a pressing clinical dilemma.^3,75,76 Patients with both T2D and CRD face substantially elevated risks of exacerbations and all-cause mortality. Our findings indicate that targeted antidiabetic therapy—particularly GLP1RAs, SGLT2is, and TZDs—can meaningfully reduce acute respiratory exacerbation risks. These insights offer evidence-based guidance for optimizing treatment in this high-risk group and underscore the importance of integrated, patient-centered approaches to multimorbidity management.

Strengths and limitations

Our NMA, meta-analysis, and TSMR offer valuable evidence linking antidiabetic drugs to CRDs. The NMA combined direct and indirect evidence to estimate the comparative effects on exacerbation risk, using defined inclusion criteria, a comprehensive literature search, all study designs, quality assessment, standardized data extraction, and careful analysis. We also performed a TSMR to provide genetic evidence of causality and reduce confounding.

Several limitations must be acknowledged. First, the available evidence base constrained the analytical methods feasible for each outcome; while NMA was possible for exacerbations, limited head-to-head data restricted analyses of incidence and mortality to conventional meta-analysis. Thus, the differing results are complementary, each addressing a specific research question within the constraints of the available data. Second, variation in outcome definitions and in the degree of adjustment for confounders across observational studies may introduce bias, despite our use of sensitivity and subgroup analyses. Third, our class-level aggregation of drugs (e.g., all sulfonylureas) does not account for potential intra-class heterogeneity, and the analysis of combination therapies was limited. Fourth, data gaps precluded NMA for mortality and incidence, and specific subgroup analyses (e.g., for asthma exacerbations or non-sulfonylurea secretagogues) were underpowered. Varying degrees of adjustment for potential confounders across the included observational studies is an important limitation. While we performed subgroup analyses by study quality and utilized adjusted estimates where available, residual confounding remains a possibility. Finally, Mendelian randomization estimates reflect lifelong genetic exposure, differing from clinical trial effect sizes.^77
–79

Our findings require validation in large-scale, well-designed studies and randomized controlled trials (RCTs) to confirm associations and elucidate mechanisms.

Conclusion

This comprehensive analysis demonstrates that, among antidiabetic drugs, GLP-1RAs are most effective for preventing acute exacerbations of CRD in patients with T2D, followed by SGLT2is. TZDs are associated with reduced disease incidence, while metformin is linked to lower all-cause mortality in patients with coexisting COPD. These findings provide evidence-based guidance for selecting antidiabetic therapy in this high-risk multimorbid population. Further longitudinal studies and RCTs are required to validate these associations and elucidate the underlying mechanisms.

Supplemental Material

sj-docx-1-tae-10.1177_20420188261437346 – Supplemental material for Antidiabetic drug and chronic respiratory disease in type 2 diabetes: a network meta-analysis and Mendelian randomization analysis

Supplemental material, sj-docx-1-tae-10.1177_20420188261437346 for Antidiabetic drug and chronic respiratory disease in type 2 diabetes: a network meta-analysis and Mendelian randomization analysis by Ikramulhaq Patel, Yin-He Chai, Zi-Qi Wang, Hui Xu, Jin-Yan Zhang, Rafael Simo, Xing-Yao Tang and Jian-Bo Zhou in Therapeutic Advances in Endocrinology and Metabolism

Supplemental Material

sj-pdf-2-tae-10.1177_20420188261437346 – Supplemental material for Antidiabetic drug and chronic respiratory disease in type 2 diabetes: a network meta-analysis and Mendelian randomization analysis

Supplemental material, sj-pdf-2-tae-10.1177_20420188261437346 for Antidiabetic drug and chronic respiratory disease in type 2 diabetes: a network meta-analysis and Mendelian randomization analysis by Ikramulhaq Patel, Yin-He Chai, Zi-Qi Wang, Hui Xu, Jin-Yan Zhang, Rafael Simo, Xing-Yao Tang and Jian-Bo Zhou in Therapeutic Advances in Endocrinology and Metabolism

Footnotes

Acknowledgements

Developed at the MRC Integrative Epidemiology Unit (IEU) at the University of Bristol, OpenGWAS resource is a manually curated collection of complete genome-wide association study (GWAS) summary datasets made available as open source files for download, or by querying a database of the complete data. We thank all investigators of the OpenGWAS project for making GWAS summary datasets publicly available. The UK Biobank was founded by the Medical Charity Welcome Trust, Medical Research Council, Department of Health, Scottish Government, and Northwest Regional Development Agency. We are grateful to all UK Biobank participants. FinnGen is a large public–private partnership that collects and analyzes genome and health data from 500,000 Finnish biobank participants. FinnGen aims to provide novel medically and therapeutically relevant insights and to construct a world-class resource that can be applied in future studies. We acknowledge the participants and investigators of the FinnGen Study. We are grateful for the resources provided by MAGIC Consortia (https://www.magicinvestigators.org/). We thank the NHGRI-EBI Catalog team () for providing up-to-date GWAS summary data for use in this research.

Declarations

Ethics approval and consent to participate

The study is a meta-analysis. No new individual-level data involving human participants or animals were collected or analyzed. Therefore, ethical approval and informed consent were not required.

Consent for publication

Not applicable.

Author contributions

Ikramulhaq Patel: Writing – original draft; Writing – review & editing.

Yin-He Chai: Formal analysis; Software; Validation; Writing – review & editing.

Zi-Qi Wang: Methodology; Validation; Writing – review & editing.

Hui Xu: Writing – original draft; Writing – review & editing.

Jin-Yan Zhang: Formal analysis; Methodology; Software; Validation.

Rafael Simo: Methodology; Writing – review & editing.

Xing-Yao Tang: Conceptualization; Writing – review & editing.

Jian-Bo Zhou: Conceptualization; Funding acquisition; Project administration; Supervision; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The National Natural Science Foundation of China (No. 82270866) supported this work.

Competing interests

The authors declare that there is no conflict of interest.

Availability of data and materials

All data used in this study are publicly available.

ORCID iDs

Ikramulhaq Patel

Jian-Bo Zhou

Supplemental material

Supplemental material for this article is available online.

References

Skou

Mair

Fortin

, et al. Multimorbidity. Nat Rev Dis Primers 2022; 8(1): 48.

Aramrat

Choksomngam

Jiraporncharoen

, et al. Advancing multimorbidity management in primary care: a narrative review. Prim Health Care Res Dev 2022; 23: e36.

GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet (London, England) 2020; 396(10258): 1204–1222.

Hashemzadeh

Movahed

MR.

The occurrence of asthma in hospitalized patients with type 2 diabetes mellitus. Intern Med J 2009; 39(10): 699–701.

Cavaillès

Brinchault-Rabin

Dixmier

, et al. Comorbidities of COPD. Eur Respir Rev 2013; 22(130): 454–475.

Anecchino

Rossi

Fanizza

, et al. Prevalence of chronic obstructive pulmonary disease and pattern of comorbidities in a general population. Int J Chron Obstruct Pulmon Dis 2007; 2(4): 567–574.

Castañ-Abad

Montserrat-Capdevila

Godoy

, et al. Diabetes as a risk factor for severe exacerbation and death in patients with COPD: a prospective cohort study. Eur J Public Health 2020; 30(4): 822–827.

Huang

Ruan

, et al. Diabetes mellitus in patients with chronic obstructive pulmonary disease—the impact on mortality. PLoS One 2017; 12(4): e0175794.

Park

Perez

Perez Gandara

, et al. Mechanisms linking COPD to type 1 and 2 diabetes mellitus: is there a relationship between diabetes and COPD? Medicina (Kaunas, Lithuania) 2022; 58(8): 1030.

10.

Uppal

Mohammed

Rajashekar

, et al. Type 2 diabetes mellitus and asthma: pathomechanisms of their association and clinical implications. Cureus 2023; 15(3): e36047.

11.

Pradhan

Yin

, et al. Novel antihyperglycaemic drugs and prevention of chronic obstructive pulmonary disease exacerbations among patients with type 2 diabetes: population based cohort study. BMJ (Clin Res ed) 2022; 379: e071380.

12.

Yen

Wei

Huang

, et al. SGLT-2 inhibitors and the risk of COPD exacerbations and mortality in COPD patients. Ann Am Thorac Soc 2025; 22(6): 846–854.

13.

Karakasis

Patoulias

Fragakis

, et al. Effect of newer glucose-lowering medications on chronic obstructive pulmonary disease exacerbations among individuals with type 2 diabetes. Eur J Intern Med 2025; 136: 153–155.

14.

López-Cano

Ciudin

Sánchez

, et al. Liraglutide improves forced vital capacity in individuals with type 2 diabetes: data from the randomized crossover LIRALUNG study. Diabetes 2021; 71(2): 315–320.

15.

Fawzy

Akenroye

, et al. Metformin use and risk of asthma exacerbation among asthma patients with glycemic dysfunction. J Allergy Clin Immunol Pract 2021; 9(11): 4014–4020.e4.

16.

Rinne

Feemster

Collins

, et al. Thiazolidinediones and the risk of asthma exacerbation among patients with diabetes: a cohort study. Allergy Asthma Clin Immunol 2014; 10(1): 34.

17.

Wang

Lai

Huang

, et al. Use of antidiabetic medications and risk of chronic obstructive pulmonary disease exacerbation requiring hospitalization: a disease risk score-matched nested case-control study. Respir Res 2020; 21(1): 319.

18.

Bishwakarma

Zhang

Lin

, et al. Metformin use and health care utilization in patients with coexisting chronic obstructive pulmonary disease and diabetes mellitus. Int J Chron Obstruct Pulmon Dis 2018; 13: 793–800.

19.

Huang

Tsai

, et al. Metformin use mitigates the adverse prognostic effect of diabetes mellitus in chronic obstructive pulmonary disease. Respir Res 2019; 20(1): 69.

20.

Yen

F-S

Chen

Wei

JC-C

, et al. Effects of metformin use on total mortality in patients with type 2 diabetes and chronic obstructive pulmonary disease: a matched-subject design. PLoS One 2018; 13(10): e0204859.

21.

Yen

Wei

, et al. Sulfonylurea use in patients with type 2 diabetes and COPD: a nationwide population-based cohort study. Int J Environ Res Public Health 2022; 19(22): 15013.

22.

Yen

Wei

Yang

, et al. Thiazolidinedione use in individuals with type 2 diabetes and chronic obstructive pulmonary disease. Front Med 2021; 8: 729518.

23.

Mendy

Gopal

Alcorn

, et al. Reduced mortality from lower respiratory tract disease in adult diabetic patients treated with metformin. Respirology (Carlton, Vic) 2019; 24(7): 646–651.

24.

Wang

Pei

, et al. The relationship between the use of GLP-1 receptor agonists and the incidence of respiratory illness: a meta-analysis of randomized controlled trials. Diabetol Metab Syndr 2023; 15(1): 164.

25.

Wei

Yang

Leng

, et al. Use of GLP1RAs and occurrence of respiratory disorders: a meta-analysis of large randomized trials of GLP1RAs. Clin Respir J 2021; 15(7): 847–850.

26.

Hutton

Salanti

Caldwell

, et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations. Ann Intern Med 2015; 162(11): 777–784.

27.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Syst Rev 2021; 10(1): 89.

28.

McBride

Yun

, et al. BRCA1 and BRCA2 gene mutations and colorectal cancer risk: systematic review and meta-analysis. J Natl Cancer Inst 2018; 110(11): 1178–1189.

29.

Lehtonen

Soveri

Laine

, et al. Is bilingualism associated with enhanced executive functioning in adults? A meta-analytic review. Psychol Bull 2018; 144(4): 394–425.

30.

Gunnerud

ten Braak

Reikerås

EKL

, et al. Is bilingualism related to a cognitive advantage in children? A systematic review and meta-analysis. Psychol Bull 2020; 146(12): 1059–1083.

31.

Guo

Zhao

Huang

, et al. Sodium-glucose cotransporter 2 inhibitors, inflammation, and heart failure: a two-sample Mendelian randomization study. Cardiovasc Diabetol 2024; 23(1): 118.

32.

Chen

Tseng

, et al. Combination therapies with thiazolidinediones are associated with a lower risk of acute exacerbations in new-onset COPD patients with advanced diabetic mellitus: a cohort-based case-control study. BMC Pulm Med 2021; 21(1): 141.

33.

Rinne

Liu

Feemster

, et al. Thiazolidinediones are associated with a reduced risk of COPD exacerbations. Int J Chron Obstruct Pulmon Dis 2015; 10: 1591–1597.

34.

Yen

Wei

Chiu

, et al. Cardiovascular outcomes of metformin use in patients with type 2 diabetes and chronic obstructive pulmonary disease. Front Pharmacol 2022; 13: 919881.

35.

Yen

Wei

Yang

, et al. Respiratory outcomes of metformin use in patients with type 2 diabetes and chronic obstructive pulmonary disease. Sci Rep 2020; 10(1): 10298.

36.

Yen

Chang

Wei

, et al. Respiratory outcomes of insulin use in patients with COPD: a nationwide population-based cohort study. Pharmaceuticals (Basel, Switzerland) 2023; 16(5): 643.

37.

PCM

Tan

KCB

Lam

DCL

, et al. Association of sodium-glucose cotransporter 2 inhibitor vs dipeptidyl peptidase-4 inhibitor use with risk of incident obstructive airway disease and exacerbation events among patients with type 2 diabetes in Hong Kong. JAMA Netw Open 2023; 6(1): e2251177.

38.

Wang

Tang

Zhang

, et al. Association between novel glucose-lowering drugs and risk of asthma: a network meta-analysis of cardiorenal outcome trials. Diabetes Res Clin Pract 2022; 183: 109080.

39.

Qiu

Ding

Zhan

, et al. Use of SGLT2 inhibitors and occurrence of noninfectious respiratory disorders: a meta-analysis of large randomized trials of SGLT2 inhibitors. Endocrine 2021; 73(1): 31–36.

40.

Collins

Costello

RA.

Glucagon-like peptide-1 receptor agonists. (Disclosure: Ryan Costello declares no relevant financial relationships with ineligible companies.) Treasure Island (FL): StatPearls Publishing LLC, 2024.

41.

Wang

Mei

Qian

, et al. The role of glucagon-like peptide-1 receptor agonists in chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis 2023; 18: 129–137.

42.

Viby

Isidor

Buggeskov

, et al. Glucagon-like peptide-1 (GLP-1) reduces mortality and improves lung function in a model of experimental obstructive lung disease in female mice. Endocrinology 2013; 154(12): 4503–4511.

43.

Zhu

Zhang

, et al. Glucagon like peptide-1 (GLP-1) modulates OVA-induced airway inflammation and mucus secretion involving a protein kinase A (PKA)-dependent nuclear factor-κB (NF-κB) signaling pathway in mice. Int J Mol Sci 2015; 16(9): 20195–20211.

44.

Rogliani

Calzetta

Capuani

, et al. Glucagon-like peptide 1 receptor: a novel pharmacological target for treating human bronchial hyperresponsiveness. Am J Respir Cell Mol Biol 2016; 55(6): 804–814.

45.

Rogliani

Matera

Calzetta

, et al. Long-term observational study on the impact of GLP-1R agonists on lung function in diabetic patients. Respir Med 2019; 154: 86–92.

46.

Lecube

Simó

Pallayova

, et al. Pulmonary function and sleep breathing: two new targets for type 2 diabetes care. Endocr Rev 2017; 38(6): 550–573.

47.

Simó

Lecube

Looking for solutions to lung dysfunction in type 2 diabetes. Ann Transl Med 2020; 8(8): 521.

48.

Gutiérrez-Carrasquilla

Sánchez

Barbé

, et al. Effect of glucose improvement on spirometric maneuvers in patients with type 2 diabetes: the Sweet Breath study. Diabetes Care 2019; 42(4): 617–624.

49.

Körner

Stöckli

Waser

, et al. GLP-1 receptor expression in human tumors and human normal tissues: potential for in vivo targeting. J Nucl Med 2007; 48(5): 736–743.

50.

Kotnala

Kim

Rajput

, et al. Contribution of dipeptidyl peptidase 4 to non-typeable Haemophilus influenzae-induced lung inflammation in COPD. Clin Sci (London, England: 1979) 2021; 135(17): 2067–2083.

51.

Somborac-Bačura

Buljević

Rumora

, et al. Decreased soluble dipeptidyl peptidase IV activity as a potential serum biomarker for COPD. Clin Biochem 2012; 45(15): 1245–1250.

52.

Chang

Yang

Jia

, et al. Expression and clinical significance of serum dipeptidyl peptidase IV chronic obstructive pulmonary disease. Am J Med Sci 2016; 351(3): 244–252.

53.

Seys

LJM

Widagdo

Verhamme

, et al. DPP4, the middle east respiratory syndrome coronavirus receptor, is upregulated in lungs of smokers and chronic obstructive pulmonary disease patients. Clin Infect Dis 2018; 66(1): 45–53.

54.

Brikman

Dori

Sodium glucose cotransporter2 inhibitor—possible treatment for patients with diabetes, pulmonary disease and CO₂ retention. Med Hypotheses 2020; 139: 109631.

55.

Wang

, et al. NLRP3 inflammasome involves in the acute exacerbation of patients with chronic obstructive pulmonary disease. Inflammation 2018; 41(4): 1321–1333.

56.

Van Bilsen

van Nieuwenhoven

. PPARs as therapeutic targets in cardiovascular disease. Expert Opin Ther Targets 2010; 14(10): 1029–1045.

57.

Belvisi

Mitchell

JA.

Targeting PPAR receptors in the airway for the treatment of inflammatory lung disease. Br J Pharmacol 2009; 158(4): 994–1003.

58.

Donaldson

Seemungal

Patel

, et al. Airway and systemic inflammation and decline in lung function in patients with COPD. Chest 2005; 128(4): 1995–2004.

59.

Jin

Guo

, et al. Corrigendum: metformin ameliorates inflammation and airway remodeling of experimental allergic asthma in mice by restoring AMPKα activity. Front Pharmacol 2022; 13: 900127.

60.

Calixto

Lintomen

André

, et al. Metformin attenuates the exacerbation of the allergic eosinophilic inflammation in high fat-diet-induced obesity in mice. PLoS One 2013; 8(10): e76786.

61.

Park

Bang

Kwon

, et al. Metformin reduces airway inflammation and remodeling via activation of AMP-activated protein kinase. Biochem Pharmacol 2012; 84(12): 1660–1670.

62.

Wang

Yang

, et al. Metformin prevents methylglyoxal-induced apoptosis by suppressing oxidative stress in vitro and in vivo. Cell Death Dis 2022; 13(1): 29.

63.

Maida

Lamont

Cao

, et al. Metformin regulates the incretin receptor axis via a pathway dependent on peroxisome proliferator-activated receptor-α in mice. Diabetologia 2011; 54(2): 339–349.

64.

Lindsay

Duffy

McKillop

, et al. Inhibition of dipeptidyl peptidase IV activity by oral metformin in Type 2 diabetes. Diabet Med 2005; 22(5): 654–657.

65.

Lenhard

Croom

Minnick

DT.

Reduced serum dipeptidyl peptidase-IV after metformin and pioglitazone treatments. Biochem Biophys Res Commun 2004; 324(1): 92–97.

66.

Tseng

CH.

Metformin and lung cancer risk in patients with type 2 diabetes mellitus. Oncotarget 2017; 8(25): 41132–41142.

67.

Tseng

CH.

Metformin and pancreatic cancer risk in patients with type 2 diabetes. Pancreas 2018; 47(9): e57–e59.

68.

Tseng

CH.

Metformin and risk of hepatocellular carcinoma in patients with type 2 diabetes. Liver Int 2018; 38(11): 2018–2027.

69.

Tseng

CH.

Metformin is associated with a lower risk of colorectal cancer in Taiwanese patients with type 2 diabetes: a retrospective cohort analysis. Diabetes Metab 2017; 43(5): 438–445.

70.

Tseng

CH.

Human insulin therapy is associated with an increased risk of lung cancer: a population-based retrospective cohort study. Front Endocrinol (Lausanne) 2019; 10: 443.

71.

Alabraba

Farnsworth

Leigh

, et al. Exubera inhaled insulin in patients with type 1 and type 2 diabetes: the first 12 months. Diabetes Technol Ther 2009; 11(7): 427–430.

72.

Singh

Bodas

Bhatraju

, et al. Hyperinsulinemia adversely affects lung structure and function. Am J Physiol Lung Cell Mol Physiol 2016; 310(9): L837–L845.

73.

Nie

Jacoby

Fryer

AD.

Hyperinsulinemia potentiates airway responsiveness to parasympathetic nerve stimulation in obese rats. Am J Respir Cell Mol Biol 2014; 51(2): 251–261.

74.

Chowdhury

Chandra Das

Sunna

, et al. Global and regional prevalence of multimorbidity in the adult population in community settings: a systematic review and meta-analysis. EClinicalMedicine 2023; 57: 101860.

75.

Boers

Barrett

, et al. Global burden of chronic obstructive pulmonary disease through 2050. JAMA Netw Open 2023; 6(12): e2346598.

76.

GBD 2021 Diabetes Collaborators. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet (London, England) 2023; 402(10397): 203–234.

77.

Holmes

Ala-Korpela

Smith

GD.

Mendelian randomization in cardiometabolic disease: challenges in evaluating causality. Nat Rev Cardiol 2017; 14(10): 577–590.

78.

Elsworth

Lyon

Alexander

, et al. The MRC IEU OpenGWAS data infrastructure. BioRxiv 2020.08.10.244293, 2020.

79.

Sollis

Mosaku

Abid

, et al. The NHGRI-EBI GWAS catalog: knowledgebase and deposition resource. Nucleic Acids Res 2023; 51(D1): D977–D985.