Independent influence of type 2 diabetes on reduced cardiopulmonary fitness in patients after percutaneous coronary intervention: a cross-sectional study

Abstract

Previous studies have found a significant association between type 2 diabetes (T2DM) and impaired cardiopulmonary fitness (CRF); however, little evidence was shown in patients after percutaneous coronary intervention (PCI). This study aimed to evaluate the independent effects of T2DM on CRF in patients who have undergone successful percutaneous coronary intervention (PCI) and received guideline-directed medical therapy. Additionally, we explored whether this association is influenced by factors such as demographic features, physical activity level, duration of diabetes, time from index PCI, and history of occlusion myocardial infarction. We retrospectively analyzed data from post-PCI patients who consecutively visited the Cardiac Rehabilitation Center at Beijing Anzhen Hospital between September 2023 and July 2024. To isolate the impact of T2DM on cardiovascular fitness, we implemented strict exclusion criteria for confounding comorbidities, particularly heart failure. Cardiorespiratory fitness was quantified through gold-standard measures: peak oxygen uptake (VO2max) and metabolic equivalents (METs). Baseline characteristics were compared between patients with T2DM and non-diabetic patients (DM group vs. non-DM group). A multivariable regression model was used to evaluate the independent effect of T2DM on CRF, adjusting for confounding factors such as demographic features, physical activity level, duration of diabetes, time since index PCI, and residual comorbidities. Subgroup analyses and interaction tests were performed to assess the impact of T2DM across different subgroups. 201 patients (150 non-DM and 51 DM patients) were included in the final analysis. Hypertension was significantly more prevalent in DM patients (68.6 vs. 42.7%, p = 0.001), while other comorbidities, anthropometric measurements, lifestyle factors, and time from index PCI showed no significant differences between groups (all p > 0.05). Multivariate logistic regression analyses demonstrated significant negative associations between T2DM and both VO2max and METs. After adjusting for basic demographic and lifestyle factors (Model 1), T2DM was inversely associated with VO2max (β=−98.3, 95% CI −193.4 to −3.3, p = 0.044) and METs (β=−0.4, 95% CI −0.8 to −0.0, p = 0.05). These negative associations remained robust and became stronger in Model 2, which further adjusted for physical activity status, hypertension, hyperlipidemia, history of occlusion myocardial infarction, time from index PCI, DM duration, and using beta-blockers, showing more pronounced inverse relationships with both VO2max (β=−212.3, 95% CI −389.4 to −35.3, p = 0.02) and METs (β=−0.9, 95% CI −1.6 to −0.2, p = 0.014). Subgroup analyses indicated consistent inverse associations, with no significant effect modification based on sex, age, body mass index (BMI), time since the index PCI, physical activity status, or a history of occlusion myocardial infarction. Our study demonstrates that T2DM is an independent negative predictor of CRF in post-PCI patients, with consistent findings across various subgroups and robust results after adjusting for confounding factors. These findings underscore the importance of CRF assessment in post-PCI patients and highlight the need for targeted interventions to improve CRF in individuals with T2DM.

Introduction

Cardiovascular disease (CVD) continues to be the predominant cause of mortality worldwide, imposing substantial burdens on healthcare systems through increased costs and reduced health outcomes¹. Despite advances in revascularization techniques and pharmacological therapies, ischemic heart disease remains the primary contributor to cardiovascular mortality, accounting for approximately 9.44 million deaths annually (95% UI: 8.82–9.96 million) and 185 million disability-adjusted life years (DALYs)². Recent evidence from large-scale randomized controlled trials and systematic reviews indicates that percutaneous coronary intervention (PCI) is effective in relieving angina symptoms and enhancing quality of life. However, it does not significantly affect long-term mortality or reduce the incidence of major adverse cardiovascular events (MACE) in patients with stable coronary artery disease (CAD)^3,4. These findings highlight the importance of identifying effective predictors for risk stratification and comprehensive management after PCI.

Cardiorespiratory fitness (CRF), commonly assessed through peak oxygen uptake (VO2peak) or metabolic equivalents (METs), is increasingly recognized as a strong predictor of cardiovascular outcomes and mortality⁵. Meta-analyses have shown that, among healthy men and women, 1-MET increase in CRF is associated with a 13% reduction in overall mortality and a 15% reduction in cardiovascular mortality⁶. In CVD patients with low fitness levels, a decline in CRF of more than 2.0 METs is linked to a 74% increase in mortality risk⁷. Another critical determinant of cardiovascular outcomes is type 2 diabetes mellitus (T2DM), which affects 20–40% of cardiovascular patients and exacerbates atherosclerosis, adverse outcomes, and mortality^8,9. Despite substantial advances in drug-eluting stent (DES) technology that have markedly reduced stent-related complications, large-scale clinical evidence has consistently demonstrated that T2DM independently predicts adverse outcomes following PCI^10,11.

Given that both T2DM and CRF independently predict cardiovascular outcomes, their potential interaction merits investigation. However, data regarding the impact of T2DM on CRF in post-PCI populations remain scarce. Therefore, our study aimed to: (1) evaluate the independent effect of T2DM on CRF in post-PCI patients, and (2) investigate potential interaction effects of sex, age, and BMI, diabetes duration, and physical activity status on the association between T2DM and CRF.

Methods

Study design and population

This retrospective cross-sectional study was conducted at a single institution. We retrospectively analyzed data from post-PCI patients who consecutively visited the Cardiac Rehabilitation Center at Beijing Anzhen Hospital for cardiac rehabilitation assessment between September 2023 and July 2024. The inclusion criteria for the study included patients aged 20 to 80 years who had undergone complete PCI and had received guideline-directed medical therapy within the past three years. All participants routinely underwent thorough clinical evaluations within one month prior to enrollment. These evaluations included laboratory tests, echocardiography, and cardiac imaging to rule out any progression of coronary disease. Additionally, three independent cardiovascular specialists assessed each patient to ensure there were no absolute or relative contraindications to exercise testing, as defined by American Heart Association guidelines¹². Exclusion criteria were as follows: (1) heart failure as defined by current guidelines; (2) history of other cardiac surgeries including coronary artery bypass grafting or any open thoracic surgery; (3) any valvular heart disease; (4) uncontrolled cardiac arrhythmias including atrial fibrillation, frequent premature ventricular Contractions, and ventricular tachycardia; (5) sever hepatic/renal dysfunction including diabetic nephropathy; (6) history of chronic obstructive pulmonary disease (COPD) or abnormal pulmonary function found in spirometer test prior to CPET; (7) lower limb dysfunction of any etiology; (8) anemia or thyroid dysfunction; (9) pulmonary hypertension; (10) active infection; (11) tumor; (12) severe obesity (BMI > 40 kg/m²); (13) premature termination of exercise testing or the peak respiratory exchange ratio (RER) < 1.05; and (14) development of significant symptoms during exercise testing, including severe chest pain, malignant arrhythmias, or ST-segment changes indicative of myocardial ischemia.

Ethical considerations

This study was conducted following the “Declaration of Helsinki” by the World Medical Association. Approval and informed consent was waived by the “Clinical Research Ethics Committee of Beijing Anzhen Hospital, Capital Medical University” as the patient identity information has been concealed.

Data collection and definition of variables

All patients underwent cardiac rehabilitation assessment at their initial presentation, with medical data systematically recorded in a standardized electronic database. Patients routinely completed comprehensive questionnaires covering demographic information, comorbidity history, medication usage, and daily physical activity patterns using the International Physical Activity Questionnaire-Long Form (IPAQ-LF)¹³. To minimize self-reporting bias and ensure data accuracy, two trained healthcare professionals independently reviewed medical records to verify and supplement the collected information. All diagnoses were uniformly coded according to the 10th International Code of Diseases (ICD-10-CM). Anthropometric measurements, cardiovascular risk factors (including smoking status and alcohol consumption), and comorbidities (including dyslipidemia, cardiovascular system disease, chronic obstructive pulmonary disease, chronic kidney disease, and hepatic dysfunction) were systematically documented. Laboratory data were collected from clinical records, including liver and renal function tests, N-Terminal pro-brain natriuretic peptide (NT-proBNP) levels, and hemoglobin concentrations, along with detailed medication histories.

T2DM was defined according to current guidelines as meeting any of the following criteria: fasting plasma glucose ≥ 126 mg/dL, glycated hemoglobin A1c (HbA1c) ≥ 6.5%, and/or the requirement for oral hypoglycemic agents or insulin therapy¹⁴. Patients with T2DM were further stratified into three groups based on diabetes mellitus duration (DM duration), estimated from the initiation of glucose-lowering medication: <1 years, 1-3years, and > 3 years. Daily physical activity levels were assessed using the validated International Physical Activity Questionnaire-Long Form (IPAQ-LF). Physical activity (PA) levels were calculated in metabolic equivalent task (MET) minutes per week by multiplying the activity duration by its corresponding MET value¹⁵. Total physical activity was computed as the sum of all domain-specific activities. Based on the IPAQ scoring protocol, PA status was categorized into three activity levels: low, moderate, and high groups¹⁶.

Pulmonary function testing protocol and definition of abnormalities

All participants routinely underwent standardized pulmonary function testing (PFT) using a computerized spirometer (Madecare Medical Systems Limited, Hebei, China) in accordance with American Thoracic Society/European Respiratory Society (ATS/ERS) guidelines¹⁷. Measured parameters included forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and FEV1/FVC ratio. Pulmonary function abnormalities were defined according to current guidelines as follows: (1) Obstructive pattern: FEV1/FVC < lower limit of normal (LLN) or < 0.70, with reduced FEV1 (< 80% predicted); (2) Restrictive pattern: FVC < 80% predicted with normal or increased FEV1/FVC ratio; (3) Mixed pattern: both obstructive and restrictive abnormalities present¹⁸.

Cardiopulmonary exercise testing

Cardiopulmonary exercise testing (CPET) was performed on an electromagnetically braked cycle ergometer using the CPX-600 system (Madecare Medical Systems Limited, Hebei, China), integrated with a metabolic cart for breath-by-breath gas exchange analysis. Before each test, the gas analyzers and flow sensors were calibrated according to manufacturer specifications. The exercise protocol consisted of three phases: (1) a 3-minute rest period in the seated position for baseline measurements; (2) a 3-minute warm-up period of unloaded cycling at 60 rpm; and (3) a progressive incremental phase with workload increasing by 10–20 watts/minute, individually adjusted based on the patient’s estimated exercise capacity to achieve test duration between 8 and 12 min. Throughout the test, 12-lead ECG, blood pressure, and oxygen saturation were continuously monitored. The test was terminated when participants reached volitional exhaustion or met predetermined stopping criteria as per current guidelines¹⁹. To ensure a true maximal effort, at least two of the following criteria were required: (1) respiratory exchange ratio (RER) > 1.10; (2) achievement of ≥ 85% age-predicted maximum heart rate; (3) plateau in VO2 despite increasing workload; and (4) Rating of Perceived Exertion (RPE) ≥ 17 on the Borg 6–20 scale.

Maximum oxygen uptake (VO2max) was determined as the highest 30-second average of oxygen consumption achieved during the last exercise stage. Metabolic equivalents (METs) were calculated by dividing the measured VO2max (mL/kg/min) by 3.5²⁰. The reliability of CPET measurements was ensured through strict adherence to quality control procedures, including regular equipment maintenance and standardized testing protocols.

Statistics

Continuous variables are presented as mean ± standard deviation or median (interquartile range) based on their distribution, and categorical variables as numbers (percentages). Between-group comparisons were performed using Student’s t-test or Mann-Whitney U test for continuous variables and chi-square or Fisher’s exact test for categorical variables. Multivariate logistic regression analysis was conducted to evaluate the independent association between T2DM and METs. Confounders included age, sex, BMI, smoking status, cardiovascular risk factors, medicines and commodities. The results are presented as β with 95% confidence intervals (CI). Subgroup analyses were performed according to sex, age (≤ 60 vs.> 60 years), BMI categories (< 24 vs.≥24 kg/m²), time from index PCI, PA status, and history of occlusion myocardial infarction (OMI). Interaction tests were conducted to assess the consistency of the association across these subgroups. Statistical analyses were performed using R version 4.3.3 and IBM SPSS Statistics for Windows, version 24 (IBM Corp., Armonk, N.Y., USA), with p < 0.05 considered statistically significant.

Results

Final analyzed population

From September 1, 2023, to July 30, 2024, a total of 399 asymptomatic patients who had undergone prior PCI and had no contraindications for CPET were initially enrolled. After applying exclusion criteria and further elimination of those who met exclusion criteria, 201 patients were ultimately included in the final analysis. A flowchart of the study is provided in (Fig. 1).

Fig. 1.

Flowchart of patients enrolled in the study. PCI indicates percutaneous coronary intervention; CPET, cardiopulmonary exercise testing; COPD, chronic obstructive pulmonary disease.

Baseline characteristics

Table 1 shows the baseline characteristics of 201 patients stratified by T2DM status (150 non-DM vs. 51 DM patients). The subgroup distribution by sex, age, BMI, history of OMI, time from index PCI, and PA status is shown in (Fig. 2). Mean age was comparable (57.3 ± 10.5 vs. 56.4 ± 10.7 years, p = 0.604), with predominantly male participants in both groups. Anthropometric measurements, lifestyle factors (smoking, drinking, PA status), and time from index PCI showed no significant differences between groups (all p > 0.05). Among DM patients, 49.0% had diabetes duration > 3 years. Notably, hypertension was significantly more prevalent in DM patients (68.6 vs. 42.7%, p = 0.001), while other comorbidities (hyperlipidemia, history of OMI) were similarly distributed. No significant differences were observed in medication use between the two groups. The characteristics of the participants excluded in the study were shown in (Supplementary Table 1).

Table 1.

Baseline characteristics of patients according to T2DM status.

Variables	Non-DM patients (n = 150)	DM patients (n = 51)	p-value
Demographic characteristic
Age(years), mean ± SD	57.3 ± 10.5	56.4 ± 10.7	0.604
Age group (years), n (%)
<=60	84 (56.0%)	33 (64.7%)	0.276
> 60	66 (44.0%)	18 (35.3%)
Gender (female)	28 (18.7%)	9 (17.6%)	0.871
Height (cm)	169.9 ± 7.3	168.4 ± 6.3	0.191
Weight (Kg)	74.4 ± 12.2	72.7 ± 10.0	0.35
BMI, kg/m2	25.7 ± 3.3	25.6 ± 3.0	0.83
BMI group (kg/m2), n (%)			0.436
< 24	50 (33.3%)	14 (27.5%)
>=24	100 (66.7%)	37 (72.5%)
Lifestyle indicators
Smoking			0.297
Never smokers	64 (42.7%)	16 (31.4%)
Current smokers	27 (18.0%)	13 (25.5%)
Former smokers	59 (39.3%)	22 (43.1%)
Drinking, n (%)			0.701
Never drinkers	86 (57.3%)	31 (60.8%)
Current drinkers	26 (17.3%)	10 (19.6%)
Former drinkers	38 (25.3%)	10 (19.6%)
PA status, n (%)			0.881
Low	77 (51.3%)	28 (54.9%)
Moderate	52 (34.7%)	17 (33.3%)
High	21 (14.0%)	6 (11.8%)
Time from index PCI, n (%)			0.632
1 month to 1 year	91 (60.7%)	29 (56.9%)
1 year to 3 years	59 (39.3%)	22 (43.1%)
DM duration, n (%)			< 0.001
<1 year	0 (0.0%)	5 (9.8%)
1 year to 3 years	0 (0.0%)	10 (19.6%)
> 3 years	0 (0.0%)	25 (49.0%)
Comorbidities, n (%)
Hypertension	64 (42.7%)	35 (68.6%)	0.001
Hyperlipidemia	73 (48.7%)	27 (52.9%)	0.598
History of OMI	43 (28.7%)	13 (25.5%)	0.662
Echocardiographic findings
LVEDD (mm)	47.1 ± 3.9	46.0 ± 4.3	0.121
LVESD (mm)	30.6 ± 4.1	30.2 ± 4.5	0.531
LVEF (%)	61.9 ± 6.1	62.0 ± 5.7	0.924
CPET
VO2max (mL/min)	1512.1 ± 400.4	1396.0 ± 350.0	0.067
METs (mL/kg/min)	5.8 ± 1.3	5.5 ± 1.3	0.139
Medications
Beta blockers	73 (48.7%)	32 (62.7%)	0.082
ARNI	20 (13.3%)	6 (11.8%)	0.773
ARB	23 (15.3%)	12 (23.5%)	0.182
ACEI	11 (7.3%)	2 (3.9%)	0.392
Statins	138 (92.0%)	47 (92.2%)	0.971
PCSK-9 inhibitor	12 (8.0%)	3 (5.9%)	0.619

Data are presented as mean (standard deviation), median (interquartile range), or proportion (%) as appropriate.

BMI body mass index, PA physical activity, PCI percutaneous coronary intervention, DM type 2 diabetes mellitus, OMI occlusion myocardial infarction, LVEDD left ventricular end-diastolic diameter, LVESD left ventricular end-systolic diameter, LVEF left ventricular ejection fraction, VO2max maximum oxygen uptake, METs metabolic equivalents, CPET cardiopulmonary exercise testing, ARNI angiotensin receptor-neprilysin Inhibitor, ARB angiotensin receptor blocker, ACEI angiotensin-converting enzyme inhibitor, PCSK-9 proprotein convertase subtilisin/kexin type 9 inhibitor.

Fig. 2.

The subgroup distribution of METs by sex, age, BMI, history of OMI, time from index PCI, and PA status. BMI body mass index, DM type 2 diabetes mellitus, OMI occlusion myocardial infarction, PCI percutaneous coronary intervention, PA physical activity.

Multivariate logistic regression analyses

As shown in Table 2, multivariate logistic regression analyses demonstrated significant negative associations between T2DM and both VO2max and METs. After adjusting for basic demographic and lifestyle factors (Model 1), T2DM was inversely associated with VO2max (β=−98.3, 95% CI −193.4 to −3.3, p = 0.044) and METs (β=−0.4, 95%CI −0.8 to −0.0, p = 0.05). These negative associations remained robust and became stronger in Model 2, which further adjusted for PA status, time from index PCI, DM duration and comorbidities (hypertension, hyperlipidemia, history of OMI), showing more pronounced inverse relationships with both VO2max (β=−212.3, 95% CI −389.4 to −35.3, p = 0.02) and METs (β=−0.9, 95% CI −1.6 to −0.2, p = 0.014).

Table 2.

Multivariate logistic regression analysis for assessing the association between T2DM and CRF.

	Model 1		Model 2
	β (95% CI)	p value	β (95% CI)	p value
VO2max (mL/min)	−98.3 (−193.4, −3.3)	0.044	−212.3 (−389.4, −35.3)	0.02
METs (mL/kg/min)	−0.4 (−0.8, −0.0)	0.05	−0.9 (−1.6, −0.2)	0.014

VO2max maximum oxygen uptake, METs metabolic equivalents, PA physical activity, OMI occlusion myocardial infarction, DM type 2 diabetes mellitus.

Model 1: adjusted for sex, age, height, weight, BMI, smoking, drinking.

Model 2: adjusted for sex, age, height, weight, BMI, smoking, drinking, PA status, hypertension, hyperlipidemia, history of OMI, time from index PCI, DM duration, and beta-blockers.

Subgroup analysis and interaction analysis

We performed subgroup analyses and interaction analyses to evaluate the consistency of the association between T2DM and METs across various subgroups (Fig. 3). In subgroup analyses, we observed consistent negative associations across all subgroups. Notably, significant associations were found in the following groups: participants aged 60 years or younger (β: −0.8, 95% CI −1.6 to −0.1, P = 0.03) and those older than 60 years (β −1.1, 95% CI −2.0 to −0.2, P = 0.01); individuals with a BMI less than 24 kg/m² (β: −1.6, 95% CI −3.1 to −0.2, P = 0.03); participants who were between 1 month and 1 year post-PCI (β: −1.4, 95% CI −2.4 to −0.4, P = 0.01); those with high levels of physical activity (β: −3.4, 95% CI −6.1 to −0.7, P = 0.03); and individuals without OMI (β: −1.0, 95% CI −1.9 to −0.2, P = 0.02). No significant effect modification was found by sex, age, BMI, time since index PCI, PA status, or history of OMI (all p-interaction > 0.05).

Fig. 3.

Forest plots of subgroup analysis and interaction analysis to evaluate the association between T2DM and METs. Sex, age, height, weight, BMI, smoking, drinking, PA status, hypertension, hyperlipidemia, OMI, time from index PCI, and DM duration were adjusted except the variable itself. DM type 2 diabetes mellitus, METs metabolic equivalents, BMI body mass index, PCI percutaneous coronary intervention, PA physical activity, OMI occlusion myocardial infarction.

Discussion

In our cross-sectional study involving 201 post-PCI patients, multivariate analysis showed that T2DM was independently associated with lower CRF. This association remained significant even after adjusting for potential confounding factors, including sex, age, BMI, time since the index PCI, PA status, and history of OMI. The negative impact of T2DM on CRF was consistent across all subgroup analyses, with no significant interactions identified (all p-interaction values > 0.05). This indicates a robust and independent negative effect of T2DM on CRF.

The importance of assessing CRF in the post-PCI population

According to previously published reference values in the Chinese population, CRF in healthy males aged 50–59 years should be 7.0 ± 1.5 METs (mean ± SD), while in healthy females of the same age group should be 6.5 ± 1.3 METs²¹. In our study, the post-PCI population exhibited lower cardiorespiratory fitness (CRF) levels, measured at 5.8 ± 1.3 METs in the non-diabetic (non-DM) group, compared to previously reported references. Our findings are consistent with the FRIEND (Fitness Registry and the Importance of Exercise National Database) study, which demonstrated that patients with CVD exhibited lower CRF reference standards compared to their age- and sex-matched apparently healthy counterparts²². This indicates that our post-PCI populations demonstrate significantly reduced CRF even after excluding those with with heart failure, cardiac structural abnormalities, arrhythmias, pulmonary dysfunction, and possible comorbidities. This reduction is primarily attributed to impaired myocardial contractility, elevated left ventricular filling pressures, and reduced cardiac output during exercise, which collectively limit oxygen delivery to peripheral tissues^23,24. Although PCI can alleviate patient symptoms, its long-term benefits are transient. The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial showed that while PCI enhances the quality of life for patients with stable coronary artery disease—such as providing relief from angina and improving physical function—these benefits tend to lessen and become less significant after 36 months post-procedure²⁵. Previous meta-analyses have demonstrated a dose-response relationship, revealing that each 1-MET reduction in cardiorespiratory fitness (CRF) is associated with an 11–17% increase in the risk of all-cause mortality²⁶. These findings all suggest that post-PCI patients with guideline-directed medical therapy remain in a low CRF state and are still at high risk for adverse cardiovascular events and mortality.

Measuring CRF is a non-invasive, cost-effective method for monitoring CVD outcomes and identifying patients at risk of restenosis²⁷. Several large cohort studies have consistently shown that higher CRF is associated with a stepwise reduction in CVD risk^28,29. For example, the Aerobics Center Longitudinal Study demonstrated that decreased CRF is linked to an increased risk of cardiac events. At the same time, higher CRF remains protective even after adjusting for 10-year CVD risk. CRF assessment is also useful for predicting restenosis. Adachi et al.³⁰ found that successful PCI improves oxygen uptake kinetics and exercise capacity in patients with CVD, while no such improvements are observed in those with restenosis, highlighting the utility of cardiopulmonary exercise testing as a non-invasive tool to assess post-PCI restenosis.

Independent negative impact of T2DM on CRF

Our study specifically examined the impact of T2DM on CRF in post-PCI patients and found a significant negative association after adjusting for potential confounders such as demographic variables, comorbidities, time since the index PCI, diabetes duration, and physical activity status. Our findings align with the previous research conducted on other populations^31,32. Research shows that individuals with T2DM have lower VO2max than those without diabetes across age groups, from adolescents to adults, indicating impaired exercise capacity^33,34. Furthermore, T2DM has been identified as an independent predictor of reduced peak aerobic capacity in patients with chronic heart failure, regardless of left ventricular ejection fraction³⁵. Although studies focusing on CRF in post-PCI populations remain limited, a previous investigation examining factors affecting CRF after PCI demonstrated that T2DM was significantly associated with reduced CRF (OR = 2.138, p = 0.027)³⁶. However, this study did not investigate the independent effect of T2DM on the decline of cardiorespiratory fitness (CRF), nor did it account for the impact of pulmonary function and physical activity on CRF.

CVD patients with or without diabetes often have multiple comorbidities, complicating the interpretation of reduced CRF. The mechanisms underlying the decline in CRF are complex. Abnormal CRF may be associated with multiple factors, including maximal cardiac output, arterial oxygen content, the proportion of cardiac output directed to exercising muscles, and the muscles’ capacity to utilize oxygen³⁷. Impairment in any of these processes can lead to a reduction in CRF. To isolate the specific impact of T2DM on CRF in post-PCI patients, our study applied strict inclusion and exclusion criteria, eliminating confounding conditions like COPD/abnormal lung function, renal dysfunction (including diabetic nephropathy and lower limb dysfunction (including diabetic lower extremity dysfunction), heart failure (including diabetic cardiomyopathy), arrhythmias and structural heart diseases. Our cohort included patients treated with PCI for stable angina (n = 145) and a history of occlusion myocardial infarction (OMI, n = 56). To account for potential confounding, we analyzed the impact of the history of OMI as both a covariate and a subgroup. The results remained robust, indicating that T2DM is an independent risk factor for reduced CRF regardless of history of OMI. Additionally, we considered the duration of diabetes (DM duration) as a potential confounder since previous study have shown that longer DM duration is significantly associated with increased mortality risk in AMI patients³⁸. However, in our study, DM duration did not alter the core results, suggesting that T2DM itself, rather than its duration or complications, exerts a direct negative effect on CRF.

Our findings align with several previous studies that were conducted in patients without cardiovascular complications. Gurdal et al.³⁹found that exercise capacity was found to be significantly decreased in normotensive patients with T2DM without CVD, and this decrease was independent of diastolic dysfunction. Gulsin et al.⁴⁰ demonstrated that T2DM is associated with reduced CRF due to microvascular ischemia and diastolic dysfunction, even in asymptomatic individuals. Jlali et al.⁴¹ found that T2DM patients with normal pulmonary function exhibit impaired skeletal muscle oxygenation during exercise, which leads to reduced CRF, suggesting tailored exercise interventions focusing on improving muscle oxygenation and blood flow for this population. These findings suggest that, beyond the effects of complications, T2DM itself is an independent risk factor contributing to impaired CRF through complex ways.

Several mechanisms may link T2DM to reduced CRF, with diabetes-induced cardiac impairment playing a central role. Notably, diabetes-induced cardiac dysfunction may precede echocardiography-detectable abnormalities. Liu et al.³⁹ found that even in newly diagnosed diabetic patients without microvascular or macrovascular complications and with normal pulmonary function, CRF remained significantly lower than in the control group, indicating subclinical diabetes-related cardiac dysfunction. For those normotensive patients with T2DM and without significant coronary artery disease, a previous study demonstrated that left ventricular diastolic dysfunction was detected in up to 75% of those patients⁴². Subclinical cardiac dysfunction is prevalent in T2DM, with evidence suggesting that even asymptomatic individuals frequently exhibit subtle left ventricular systolic and diastolic dysfunction, potentially representing the initial manifestation of stage B heart failure^43,44. Similar studies found that myocardial performance index (MPI, also known as Tei index), which serves as a comprehensive assessment parameter, reveals subclinical LV dysfunction in both T2DM and prediabetes before overt clinical manifestations^45,46. These findings suggest that subclinical alterations in myocardial perfusion may be crucial mechanisms underlying cardiac dysfunction development in T2DM. Multiple pathways are involved, including impaired cardiac output adjustment, endothelial dysfunction, reduced muscle blood flow, impaired oxygen diffusion, and mitochondrial dysfunction⁴⁷. T2DM also affects muscle metabolic reflexes, autonomic function, and exercise-dependent pathways regulating PGC-1alpha expression^48,49. Oxidative stress and inflammation also play a role in vascular dysfunction, while glycated hemoglobin demonstrates increased oxygen affinity, further hindering oxygen delivery to tissues⁵⁰. The heightened risk of cardiovascular events in individuals with T2DM cannot be fully explained by traditional cardiovascular risk factors alone. Furthermore, the combination of diabetes mellitus with other risk factors appears to create a synergistic effect, leading to a greater overall risk than would be expected from merely adding the individual risks together⁵¹.

Negative effects of T2DM in subgroups

T2DM consistently negatively affects CRF across all subgroups, particularly among older adults, individuals with a lower BMI, those within one month to one year after PCI, and people with high levels of physical activity. However, no evidence of significant effect modification related to any of the examined characteristics.

Elderly patients with T2DM often exhibit unique clinical characteristics that distinguish them from younger populations⁵². Elderly individuals may present with a higher prevalence of comorbidities, including cardiovascular and renal complications, as well as an increased vulnerability to hypoglycemia due to age-related physiological changes and polypharmacy⁵². Epidemiological data indicate that approximately 40% of older adults with diabetes mellitus present with four or more comorbidities, with predominant disease clusters including diabetes mellitus-hypertension, diabetes mellitus-arthritis-hypertension, and diabetes mellitus-arthritis-hypertension-cardiovascular disease⁵³. These geriatric syndromes may further potentiate the deleterious impact of T2DM on impaired CRF in this patient group. However, additional experimental investigation is warranted to substantiate these findings. The negative impact of T2DM on CRF may be attenuated in individuals with a BMI greater than 24 kg/m². This can be attributed to the fact that obesity itself is a significant contributor to impaired exercise capacity, potentially overshadowing the additional negative effects of diabetes-related pathways in this higher BMI population⁵⁴. The underlying mechanisms may involve factors such as greater metabolic inflexibility and cardiovascular alterations associated with the obese population^55,56. These findings highlight the need for further research to elucidate the complex interplay between obesity, T2DM, and cardiopulmonary fitness.

Additionally, T2DM in the elderly population is more often complicated by factors such as frailty and sarcopenia^57,58. Beyond diabetic-related cardiac dysfunction, T2DM significantly influences age-related muscle health through multiple pathways. This metabolic disorder accelerates sarcopenia development in older adults through complex mechanisms involving nutritional, endocrine, inflammatory, and neurological pathways^59–61. Notably, the bidirectional relationship between T2DM and sarcopenia creates a detrimental cycle, where each condition exacerbates the other, ultimately leading to accelerated functional decline, increased frailty risk, and reduced quality of life in the elderly population⁶².

Based on these considerations, our analysis incorporated daily PA as a potential confounding factor. Studies using the IPAQ questionnaire have demonstrated a strong correlation between PA levels and CRF, with higher PA levels associated with improved CRF and reduced cardiovascular risk¹³. To eliminate the confounding effect of PA, we adjusted PA levels in multivariate regression models and conducted subgroup analyses based on low, moderate, and high PA levels. The negative impact of T2DM on CRF remained significant and consistent across all PA subgroups. No significant interaction was observed between PA levels and the negative impact of T2DM on CRF. Interestingly, we noted that in the high PA subgroup, the effect size of T2DM on METs appeared to be more strongly negative. This suggests that while high PA levels may improve CRF, the presence of T2DM can still exert a substantial adverse effect on CRF, even in individuals with higher PA. Although the interaction was not statistically significant, this may be attributed to the relatively small sample size in the high PA subgroup, which could have limited statistical power. Further research is needed to explore this observation in greater detail.

Strengths and limitations

The present investigation offers important strengths that enhance the validity and clinical relevance of the findings. First, selecting a post-PCI population provides additional evidence on the impact of T2DM on CRF in a clinically relevant high-risk group. As a real-world study, our research provides valuable insights into actual clinical practice patterns and outcomes. Secondly, our study reduced confounding by excluding participants with comorbidities that could influence CRF. Our comprehensive statistical adjustments also address potential confounding factors, enabling a clearer evaluation of the independent effects of T2DM. Third, These strengths enhance the validity of the findings and their contribution to understanding the relationship between T2DM and CRF. Our findings may provide valuable preliminary evidence to inform future randomized controlled trials.

Several limitations of our study should be acknowledged. First, due to the observational nature of this study, we could only demonstrate associations between T2DM and CRF rather than establish causal relationships. Second, as a single-center retrospective study with a relatively homogeneous patient population, our findings may not be generalizable to other populations or healthcare settings. Given our exclusion of multiple comorbidities that could affect CRF, our findings may not be generalizable to post-PCI patients with comorbidities other than hypertension, hyperlipidemia, and OMI. While acknowledging the retrospective nature of our study, we emphasize key methodological strengths: (1) As a real-world study, our research provides valuable insights into actual clinical practice considering post-PCI patients in our cardiac rehabilitation center; (2) Our stringent inclusion and exclusion criteria minimize selection bia; (3) We employed comprehensive statistical adjustments for potential confounding factors; and (4)Our data are stored in a standardized medical database, resulting in minimal missing data. These considerations help mitigate the limitations of retrospective analyses. Third, physical activity levels were assessed through self-reported questionnaires, which may be subject to potential misclassification. Fourth, we could not account for certain potential confounding factors that might influence CRF, such as muscle mass and visceral fat content, since these were not routinely measured in our clinical setting. Finally, although we excluded patients with established diabetic complications such as diabetic nephropathy and cardiomyopathy, subclinical complications in T2DM patients might have contributed to reduced CRF. Given the limitations of current clinical diagnostic methods, further prospective studies with more comprehensive evaluations of diabetic complications are needed to validate our findings.

Conclusions

In conclusion, our study shows that T2DM is an independent negative predictor of CRF in post-PCI patients. This finding is consistent across various subgroups and remains robust even after adjusting for confounding factors, including demographic features, physical activity level, duration of diabetes, time since index PCI, and other comorbidities. These results highlight the importance of assessing CRF in post-PCI patients and emphasize the need for targeted interventions to improve CRF in individuals with T2DM. Future research should investigate the mechanisms behind the negative impact of T2DM on CRF and explore strategies to mitigate this effect.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

CVD

Cardiovascular disease

CRF

Cardiorespiratory fitness

PCI

Percutaneous coronary intervention

T2DM

Type 2 diabetes mellitus

OMI

Occlusion myocardial infarction

BMI

Body mass index

CPET

Cardiopulmonary exercise testing

PA

Physical activity

VO2max

Maximum oxygen uptake

METs

Metabolic equivalents

COPD

Chronic obstructive pulmonary disease

LVEDD

Left ventricular end-diastolic diameter

LVESD

Left ventricular end-systolic diameter

LVEF

Left ventricular ejection fraction

ARNI

Angiotensin receptor-neprilysin Inhibitor

ARB

Angiotensin receptor blocker

ACEI

Angiotensin-converting enzyme inhibitor

PCSK-9

Proprotein convertase subtilisin/kexin type 9 inhibitor

Author contributions

JH Wu and YT Liu conceptualized the study. JH Wu and Y Feng screened patients in the Cardiac Rehabilitation Center. N Li, Y Shao, and Y Zhang performed the screening and collecting of medical records from the questionnaires and the medical database. YT Liu and SH Zhang performed pulmonary function testing and cardiopulmonary exercise tests. YT Liu collected data, performed the statistical analysis, wrote the manuscript, and ideated and produced the tables and figures. JH Wu reviewed the edited the manuscript.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to them containing information that could compromise the privacy of patients but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This study was conducted following the “Declaration of Helsinki” by the World Medical Association. Approval and informed consent was waived by the “Clinical Research Ethics Committee of Beijing Anzhen Hospital, Capital Medical University” as the patient identity information has been concealed.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-90281-z.