Effect of Inhaled Corticosteroids on Clinical Efficacy and Blood Glucose In COPD: A Retrospective Cohort Study

Summary

This retrospective cohort study evaluated whether adding inhaled corticosteroids to LABA affects blood glucose and clinical efficacy in 150 non‑diabetic COPD patients. At 12 months, ICS/LABA significantly improved lung function and CAT scores compared with LABA alone, without altering glycemic parameters or increasing adverse events.

Abstract

Inhaled corticosteroids (ICS) are commonly used in chronic obstructive pulmonary disease (COPD), but their potential association with hyperglycemia remains unclear. Therefore, evaluating the glycemic safety of ICS in non‑diabetic COPD patients has direct clinical importance for preventing potential metabolic complications during long‑term management. This retrospective cohort study aimed to evaluate whether adding ICS to long‑acting β₂‑agonist (LABA) therapy affects clinical efficacy and glycemic parameters in non‑diabetic COPD patients. A total of 150 hospitalized COPD patients (January 2022–December 2024) were included. Based on their documented treatment regimens, patients were divided into a control group (formoterol monotherapy, n = 75) and an observation group (formoterol/budesonide combination, n = 75). Primary outcomes were FEV₁, FEV₁/FVC, fasting plasma glucose (FPG), and HbA1c. Secondary outcomes included CAT score, liver/kidney function tests, arterial blood gases, bone mineral density (BMD), and adverse reactions. Baseline characteristics were balanced between groups (all P > 0.05). At 12 months, the observation group showed greater improvements in FEV₁ and FEV₁/FVC (both P < 0.01) and a larger reduction in CAT score (P < 0.001) compared with controls. No significant differences were observed in FPG (P = 0.837), HbA1c (P = 0.134), liver/kidney parameters, arterial blood gases, BMD, or adverse event incidence (P = 0.754) between the two groups. In this retrospective cohort of non‑diabetic COPD patients, adding ICS to LABA therapy improved lung function and symptoms without statistically significant effects on blood glucose or other safety parameters. These findings suggest that ICS/LABA combination may be a safe and effective option in carefully selected patients, but further prospective studies are needed to confirm the absence of long‑term glycemic risk.

Introduction

Chronic obstructive pulmonary disease (COPD) is a common respiratory condition with high morbidity and mortality¹. Prolonged exposure to tobacco smoke or toxic airborne pollutants causes airway obstruction that is poorly reversible, leading to dyspnea and reduced quality of life^2,3,4. Although COPD is prevalent, it is often underdiagnosed until advanced stages.

Inhaled corticosteroids (ICS) and long‑acting β₂‑agonists (LABA) are frequently used to treat COPD. Inhalation delivers medication directly to the lungs, maximizing local efficacy while minimizing systemic effects. ICS reduces airway inflammation, and budesonide is a commonly used ICS with favorable local selectivity^5,6,7. LABA, such as formoterol, relaxes bronchial smooth muscle by activating β₂‑adrenergic receptors, thereby improving lung function and relieving dyspnea⁸. In current practice, LABA and ICS are often combined to enhance therapeutic efficacy.

However, ICS therapy has been associated with adverse events including pneumonia, fractures, and potentially diabetes, although the evidence is conflicting^9,10,11. The relationship between ICS use and hyperglycemia or diabetes risk in COPD patients remains unclear, particularly in patients without pre‑existing diabetes. Most previous studies included mixed populations or did not specifically exclude diabetic individuals, leaving uncertainty about the glycemic safety of ICS in non‑diabetic COPD patients.

Several features of this study strengthen its contribution: exclusive inclusion of non‑diabetic COPD patients to avoid confounding by pre‑existing diabetes; a 12‑month longitudinal follow‑up to assess medium‑term effects; and a comprehensive multi‑index assessment covering pulmonary function, glycemic parameters, liver/kidney function, arterial blood gases, bone mineral density, and adverse events. This retrospective cohort study was therefore designed to compare the clinical efficacy and safety (with a focus on glycemic parameters) of LABA/ICS combination (formoterol/budesonide) versus LABA monotherapy (formoterol) in non‑diabetic COPD patients. By evaluating lung function, symptom scores, blood glucose, liver and kidney function, bone mineral density, and adverse events over 12 months, this study aims to provide preliminary evidence on whether adding ICS offers additional benefits without increasing glycemic or other systemic risks in this specific population.

Protocol

This clinical study was conducted in compliance with the ethical principles outlined in the Declaration of Helsinki¹² and received approval from the Ethics Committee of Bishan Hospital, Chongqing Medical University, prior to study initiation (Approval number: KYLL2022005). The Ethics Committee waived the requirement for written informed consent due to the retrospective nature of the study and the use of anonymized clinical data. The reagents and the equipment used are listed in the Table of Materials.

1. Study design and participants

This retrospective cohort study included 150 patients diagnosed with COPD at Bishan Hospital of Chongqing Medical University between January 2022 and December 2024. Data were extracted from the hospital’s electronic medical record system. Patients were divided into two groups according to the treatment they actually received: the control group (formoterol monotherapy, n = 75) and the observation group (formoterol combined with budesonide, n = 75). The study flow is illustrated in Figure 1. Data acquisition and analysis were performed by researchers blinded to the clinical treatment assignments.

From the initial pool of eligible patients, 75 patients were randomly selected from each group to achieve balanced sample sizes for statistical comparison. This selection was performed after data extraction, not prospectively. All selected patients had complete follow-up records at all time points (10 days, 3 months, 6 months, 12 months) as documented in routine clinical notes and laboratory databases; therefore, no loss to follow-up occurred within the analyzed cohort.

1. Inclusion criteria
   Patients were included if they met all of the following criteria: (1) aged 50–80 years; (2) confirmed COPD diagnosis based on symptoms, risk exposure history, physical examination, and post-bronchodilator FEV₁/FVC <70%¹³; (3) prescribed treatment with either LABA alone (formoterol) or LABA/ICS combination (formoterol/budesonide). (4) COPD severity was classified according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2023 criteria, based on post-bronchodilator forced expiratory volume in one second (FEV₁) as a percentage of predicted value (% predicted). All enrolled patients had moderate to very severe COPD (GOLD stages 2-4).
2. Exclusion criteria
   Exclusion criteria were: (1) prior diagnosis of type 1 or type 2 diabetes mellitus; (2) fasting plasma glucose ≥7.0 mmol/L or HbA1c ≥6.5% at baseline¹⁴; (3) other conditions affecting glucose metabolism (Cushing’s syndrome, thyroid dysfunction, pancreatic disease); (4) concomitant non-COPD respiratory or systemic diseases (bronchial asthma, bronchiectasis, interstitial lung disease, lung cancer, myocardial infarction, osteoporosis); (5) recent use of systemic glucocorticoids, or long-term use of diuretics, immunosuppressants, or biologic agents; (6) end-stage COPD, acute exacerbation at enrollment, severe hepatic or renal impairment, or inability to cooperate with follow-up.

2. Data source and treatment groups

Patients were divided into groups based on their documented treatment regimens retrieved from electronic medical records. The control group received formoterol monotherapy (4.5 µg twice daily via dry powder inhaler). The observation group received a fixed-dose combination of formoterol (4.5 µg) and budesonide (160 µg) per inhalation, twice daily. The total daily dose was 320 µg budesonide and 9 µg formoterol. Treatment adherence was determined by reviewing telephone follow-up records (which were documented as part of routine clinical care every 3–6 months), and patients who changed their medication regimen during the follow-up period were excluded from the final analysis (none met this exclusion criterion).

1. Outcome measures
   Baseline data collected included age, sex, body mass index (BMI), smoking history, disease duration, FEV₁, FEV₁/FVC, CAT score, FPG, HbA1c, ALT, AST, SCr, arterial blood gases (PaO₂, PaCO₂, pH), and BMD.
   Primary outcomes were obtained as follows:  FEV₁ and FEV₁/FVC measured at admission (baseline), 10 days, 3 months, and 12 months were retrieved from the pulmonary function test database. FPG and HbA1c measured at baseline and 12 months were obtained from the laboratory information system. FPG was measured by the hexokinase method using commercial kits and an automatic biochemical analyzer.
   Secondary outcomes were extracted from routine clinical records: (1) CAT score¹⁵at baseline, 10 days, 3 months, and 12 months; (2) Serum creatinine (SCr)¹⁶ at baseline, 6 months, and 12 months. Blood samples (5 mL fasting venous blood) were centrifuged at 1500 × g for 10 min; SCr was measured at 546 nm using an automatic biochemical analyzer; (3) ALT and AST¹⁷ at baseline, 6 months, and 12 months. Serum was separated by centrifugation at 1500 × g for 10 min; ALT and AST were measured at 340 nm on the same analyzer. The ALT/AST ratio was calculated; (4) Arterial blood gases (PaO2, PaCO2, pH)¹⁸at baseline and 12 months. Radial artery blood (1.5 mL) was collected and analyzed immediately using a portable blood gas analyzer; (5) Bone mineral density (BMD)¹⁹at baseline and 12 months, measured by dual-energy X-ray absorptiometry (DXA); (6) Adverse reactions (palpitation, headache, skeletal muscle spasm, tremor)²⁰ recorded throughout the treatment period. Incidence (%) was calculated as (number of patients with adverse events / total patients in group) × 100. All measurements were extracted from routine clinical records; no scheduled study-specific visits were added.

3. Statistical analysis

Statistical analyses were performed using a standard statistical software package. All tests were two-tailed, and a P-value <0.05 was considered statistically significant.

Normality of continuous data was assessed using the Shapiro–Wilk test. Normally distributed data were expressed as mean ± standard deviation (SD); non-normally distributed data as median with interquartile range (P25–P75).

Between-group comparisons at baseline and each follow-up time point were performed using independent-samples t‑tests for normally distributed variables (FEV₁, FEV₁/FVC, FPG, HbA1c, CAT scores) and Mann–Whitney U tests for non-normally distributed variables (ALT, ALT/AST ratio, SCr, arterial blood gas parameters, BMD). Within-group changes over time were analyzed using paired t‑tests (for normally distributed variables) or Wilcoxon signed-rank tests (for non-normally distributed variables).

To account for repeated measurements (FEV₁, FEV₁/FVC, CAT scores assessed at four time points; ALT/AST ratio, SCr at three time points), a linear mixed-effects model with unstructured covariance was additionally applied, with treatment group, time, and group‑by‑time interaction as fixed effects and patient as a random intercept. This approach addresses the correlation of repeated observations within the same patient.

All 150 patients had complete records for the scheduled follow-up time points (10 days, 3 months, 6 months, and 12 months) as extracted from routine clinical notes and laboratory databases. Patients with incomplete follow-up records were excluded prior to analysis; no patient met this exclusion criterion. Thus, no missing data were present for any of the reported outcomes.

Effect sizes for between-group comparisons are reported as Cohen’s d for t‑tests and as rank-biserial correlation for Mann–Whitney U tests.

Results

Baseline characteristics

Table 1 presents baseline characteristics for the control group (n = 75) and observation group (n = 75). No statistically significant differences were observed between the two groups in any of the measured baseline variables (all P > 0.05). The distribution of GOLD stages (moderate to very severe) was similar between the two groups (P > 0.05). The two groups were derived from the electronic medical records of al...

Discussion

The hallmark of COPD is sustained airway inflammation, which induces airway stenosis and parenchymal damage. Pharmacological management aims to alleviate symptoms, enhance exercise tolerance, and mitigate exacerbation risks. Current evidence regarding ICS and hyperglycemia risk in COPD patients is inconclusive. Therefore, this retrospective cohort study aimed to evaluate the clinical efficacy of ICS/LABA combination therapy versus LABA monotherapy in non‑diabetic COPD patients and its association with glycemic para...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Alanine aminotransferase (ALT) assay kit	Abbott Laboratories	Not applicable	Used with ARCHITECT C8000 analyzer
Arterial blood gas analyzer, portable	Abbott Point of Care Inc.	i-STAT 300	For PaO2, PaCO2, pH measurement; uses single-use cartridges
Aspartate aminotransferase (AST) assay kit	Abbott Laboratories	Not applicable	Used with ARCHITECT C8000 analyzer
Automatic biochemical analyzer	Abbott Laboratories	ARCHITECT C8000	For SCr, ALT, AST, and HbA1c analysis
Centrifuge	Beckman Coulter	Microfuge 20R	For serum separation (1500×g, 10 min)
COPD Assessment Test (CAT) questionnaire	GlaxoSmithKline (published instrument)	Available at www.catestonline.org	Paper-based; no commercial kit required
Dual-energy X-ray absorptiometry (DXA) scanner	Hologic Inc.	Discovery Wi	For bone mineral density (BMD) measurement
Formoterol dry powder inhaler (Oxis Turbuhaler)	AstraZeneca	National Drug Approval No. HJ20130278	4.5 µg per inhalation, 60 doses
Formoterol/budesonide dry powder inhaler (Symbicort Turbuhaler)	AstraZeneca	National Drug Approval No. HJ20140458	160 µg budesonide / 4.5 µg formoterol per inhalation
G*Power software	Heinrich Heine University Düsseldorf	Version 3.1	For sample size calculation
Glucose assay kit (hexokinase method)	Olympus Corporation	OSR 6221	For fasting plasma glucose (FPG) measurement
HbA1c analyzer (turbidimetric immunoassay)	Bio-Rad Laboratories or equivalent	Not applicable	Performed on ARCHITECT C8000 using commercial kit
Portable blood gas analyzer	See “Arterial blood gas analyzer”	Same as above	–
Serum creatinine (SCr) assay kit	Abbott Laboratories	Not applicable	Used with ARCHITECT C8000
Spirometer	CareFusion or equivalent	MasterScreen Pneumo	For FEV1 and FVC measurement according to ATS/ERS standards
Statistical software	IBM Corp.	SPSS Statistics 25.0	For all statistical analyses