Unveiling MicroRNA-21 as a non-invasive diagnostic biomarker in chronic obstructive pulmonary disease-associated pulmonary hypertension: Comparative insights with NT-proBNP

ABSTRACT

Background:

Pulmonary hypertension (PH) is a fatal complication of chronic obstructive pulmonary disease (COPD) that worsens its prognosis. However, the associated diagnostic markers are uncertain. We assess the diagnostic utility of microRNA-21 (miR-21) and NT-proBNP for PH diagnosis in COPD patients.

Methods:

We selected three groups of participants: Group 1 consisted of 25 patients with COPD who did not have PH (COPD); Group 2 included 35 patients with COPD and PH (COPD-PH); and Group 3 comprised 30 healthy controls. Demographic, spirometric, echocardiographic-Doppler study, and biochemical variables were recorded and compared among these three groups. Serum miR-21 was measured by quantitative real-time PCR.

Results:

Serum miR-21 level in COPD-PH patients was significantly higher compared with that in COPD patients and controls (P < 0.001). Compared with the COPD group, the COPD-PH group had greater miR-21 and brain natriuretic peptide (BNP) levels but significantly reduced forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), tricuspid annular plane systolic excursion (TAPSE), 6-minute walk test, and right ventricular (RV) functional parameters. Receiver Operating Characteristic curve analysis revealed that miRNA-21 had good diagnostic value for PH diagnosis in COPD patients, with an area under the curve (AUC) of 1.000, 100% sensitivity, and 100% specificity at a cutoff >7.94. While NT-proBNP demonstrated good accuracy, with an AUC of 0.936, sensitivity of 88.6%, and specificity of 88% at a cutoff >390.1 pg/mL.

Conclusion:

miR-21 may serve as a useful and reliable biomarker for diagnosing PH in COPD patients, outperforming the traditional biomarker NT-proBNP.

INTRODUCTION

Chronic obstructive pulmonary disease (COPD) has a high global health burden that affects 10% of the world’s population and results in 3 million deaths annually.[1] COPD is characterised by progressive airflow limitation that is not fully reversible, causing chronic respiratory symptoms, abnormal lung structure, and lung function damage.[2] Approximately 10%–30% of COPD patients will develop pulmonary hypertension (PH), which is defined by mean pulmonary artery pressure (PAP) ≥25 mm Hg at rest in the supine position.[3] COPD patients with PH have an increased risk of acute COPD exacerbation and significantly worsened functional parameters. Furthermore, many studies suggest that PH is an independent risk factor for increased mortality in COPD patients.[4,5]

Right heart catheterisation (RHC) is the gold standard for diagnosing PH, but it is a costly, invasive procedure, and unsuitable for all patients. On the other hand, Doppler Echocardiography is the main non-invasive diagnostic tool for predicting PH in patients with COPD. However, it has certain limitations, and it is difficult to implement in primary hospitals.[5] Therefore, there is an urgent need to identify novel, simple, reliable, non-invasive diagnostic tools like biomarkers for the diagnosis of PH-associated COPD.

MicroRNAs (miRNAs) are non–coding small endogenous RNA fragments containing 21–25 nucleotides; they act via binding the 3′-untranslated (3’UTR) region of the target messenger RNAs (mRNAs).[6] Among the different expression patterns of miRNAs, miR-21, which is a key regulator, suppresses apoptosis and promotes proliferation, which is mainly associated with cardiac hypertrophy and heart failure.[1] However, its utility as a marker of the diagnosis of COPD-PH patients has not been investigated thoroughly. In the present study, we assessed the diagnostic utility of circulating miR-21 and NT-proBNP for PH diagnosis in patients with COPD, as well as their capacity to differentiate COPD patients with PH from both non-PH COPD patients and healthy controls.

SUBJECTS AND METHODS

Study design

This is a case–control study.

Ethical approval

This study was approved by the Research Ethics Committee of the Faculty of Medicine for Girls, Al-Azhar University, Cairo, Egypt (approval number 2024092499, September 2024).

Sample size calculation

By applying the formula for proportion in a population using the OpenEpi, Version 3 open-source calculator—SSCC—found at (https://www.openepi.com/SampleSize/SSPropor.htm). According to Luca and Bodrug,[7] the design effect for cluster surveys is 0.22, the hypothesised proportion (frequency) of the outcome factor in the population is 20%, the two-sided confidence level (1-alpha) is set at 95%, the significance limit is set at 5% (P value < 0.05), and the power of study is set at 80%. As a result, the final sample size was 60, which was the projected optimal sample size of 55 COPD patients plus 10% of the non-response rate.

Study groups

Between September 2024 and May 2025, 60 patients with COPD were selected from the internal medicine, cardiology, and chest inpatient departments at AL-Zahraa Hospital, Al-Azhar University, Cairo, Egypt. The Global Initiative for Chronic Obstructive Lung Disease guidelines (forced expiratory volume during the first second of forced breath [FEV1] to forced vital capacity [FVC] ratio after bronchodilation <0.70) were used to guide the transthoracic echocardiography (TTE) that was performed on all patients after they were diagnosed with COPD. According to the Chinese Guidelines for the Diagnosis and Treatment of Pulmonary Hypertension 2018,[8] all patients were categorised into COPD and COPD-PH groups based on their pulmonary artery systolic pressure (PASP). PASP ≤50 mmHg or peak tricuspid regurgitation <3.4 m/s were used to define COPD groups, whereas PASP >50 mmHg or peak tricuspid regurgitation >3.4 m/s were used to establish COPD-PH groups. Additionally, none of the COPD patients had a history of respiratory infections for at least 4 weeks before trial enrolment, and all patients were clinically stable. Furthermore, as controls, Group 3 consisted of 30 healthy, age- and sex-matched non-COPD adults.

Exclusion criteria include the following

(1) Patients with cardiovascular co-morbidities; (2) patients with preexisting mental illness, diabetes mellitus, liver, and renal failure; and (3) PH caused by malignancy, autoimmune diseases, left heart failure, pulmonary thromboembolism, and other unknown causes were excluded.

Clinical analyses

General information with special emphasis on sex, age, body mass index (BMI), smoking index, and disease duration. Pulmonary function tests, arterial blood gas, and TTE results were collected and recorded. Finally, a serum specimen was collected, and the N-terminal pro-brain natriuretic peptide (NT-proBNP) and microRNA-21 were measured.

A pulmonary function test (Spirometry) was performed with the SPIROSIFT SP5000 (Japan). Vital capacity percentage, forced vital capacity (FVC%), forced expiratory volume in the first second (FEV1%), and FEV1/FVC ratio were all measured.

Transthoracic Echocardiography (TTE): was carried out at the echocardiography unit of the cardiology department using a GE-Vivid 9 machine equipped with a phased array cardiac probe and simultaneous electrocardiographic recording to enable flow timing. Patients with compromised systolic function were excluded based on evaluation of left ventricular ejection fraction (LVEF). During diastole, the right ventricle’s (RV) outflow diameters and basal and mid-cavity diameters were measured. (RV end diastolic area − RV end systolic area)/RV end diastolic area × 100% is the formula used to determine the fractional area change (FAC). The PASP was calculated using the peak systolic velocity and the systolic pressure gradient across the tricuspid valve. The tricuspid annular plane systolic excursion, or TAPSE, is a measurement of the RV annular plane’s systolic excursion towards the apex. Using tissue Doppler imaging, the diastolic and systolic velocities of the RV myocardium were determined. 2D speckle tracking echocardiography was utilised for 2D RV strain imaging in order to quantify RV-global longitudinal strain (RV-GLS).[9]

The 6-minute walk test (6MWT) is employed as a bedside test to evaluate exercise capacity. It was carried out in compliance with the ATS Guideline[10] as follows: The patient was told to move at his own pace along a 27-metre straight hallway with two cones at either end on flat, firm ground. Additionally, at the beginning (0 min) and finish (6 min) of the walk test, blood pressure, heart rate, and oxygen saturation were recorded.

Laboratory investigations

Methods

Sample collection and processing

Every patient had seven millilitres of venous blood extracted. After centrifuging 4 ml at 2500 g for 10 minutes, the plasma was carefully extracted and centrifuged for an additional 10 minutes at 18,000 g. To extract and quantify hsa-miR-21-5p using RT-qPCR, the supernatant was moved to 2 ml microcentrifuge tubes and kept at −80°C. In order to quantify the levels of brain natriuretic peptide (BNP) by ELISA, the remaining 3 millilitres of blood were put in a serum gel separator tube, centrifuged for serum separation, and the resulting sera were kept at −20°C.

Measurement of serum N-terminal proBNP by ELISA

Serum N-terminal proBNP levels were measured using the commercially available human ELISA kits, from Bioassay Technology Laboratory (BT LAB, Zhejiang, China), Lot no. 202203013, Catalog no. E1287Hu. The assay was carried out following the manufacturer’s instructions using an ELISA system that comprised an ELISA washer (ELX 50 Biokit, Italy), a plate reader (AS 1851 from DAS, Italy), and a plate shaker-incubator (Thermo-Shaker from EU for Grant Instruments Ltd, Cambs, England).

miRNA extraction and qRT-PCR

Mature hsa-miR-21-5p was extracted using the miRNeasy kit (Cat. NO. 217004, Qiagen, Germany) following the manufacturer’s instructions using 200 μl of plasma. Thermo Fisher Scientific’s Nanodrop (ND-1000) was used to assess the concentration and purity of RNA. Using the miRCURY LNA RT Kit (Cat. No. 339340, Qiagen, Germany), complementary DNA (cDNA) was created following the manufacturer’s instructions. Using a real-time PCR QuantStudio 5 system (Applied Biosystem, Thermo Fisher Scientific, Singapore), the miRCURY LNA SYBR Green PCR Kit (Cat. No. 339345, Qiagen, Germany) was used to measure the expression of hsa-miR-21-5p. An endogenous control was miR-103. cDNA (at a concentration of 20 ng/μl), 2 μl of miRNA primer, and 10 μl of Syber green master mix were all present in the 20 μl reaction volume.

The qRT-PCR cycling steps were as follows: initial activation step at 95°C for 2 minutes, then 40 cycles, each including a denaturation step at 95°C for 10 seconds, then combined annealing and extension at 56°C for 60 seconds. In each cycle, fluorescence was measured.

After the thermal profile, a melting curve was created by tracking fluorescence as the temperature was gradually raised (from 65 to 95°C) in order to guarantee specific amplification. The target miRNA showed a single, distinct peak in the melting curve analysis.

q-PCR findings Compute: By deducting the Ct value of miR-103 from the Ct values of miR-21 in each sample, the delta cycle threshold (Ct) was determined. The average normalised ΔCT of the control was then subtracted from all other ΔCT values, including the control values, to determine ΔΔCT using normalised ΔCT data. 2−ΔΔCt was used to compute fold changes for relative quantification.

Arterial blood gases (ABG) analysis was done using a Rapid Lab 248 blood gases analyser (Siemens Medical Solutions, Malvern, PA, US); measurements were made of O2 saturation, partial pressure of oxygen (PaO2) mm Hg, partial pressure of carbon dioxide (PaCO2) mm Hg, power of hydrogen (pH) and bicarbonate (HCO3) mmol/L.

Statistical analysis

Data analysis was performed using SPSS Inc.’s (Statistical Package for Social Sciences) program (Chicago, Illinois, USA), version 23.0. The normality of the variables under study was examined using the Shapiro-Wilk test. The mean and standard deviation were used to identify variables that are regularly distributed. In contrast, the median and interquartile range (first to third quartile) were used to identify non-normally distributed variables. We employed the Mann-Whitney U test to compare the quantitative variables of the groups under study. We employ the Chi-square test (displayed as frequency and percentage) to compare the qualitative variables. The degree of correlation between the variables was evaluated using Spearman’s correlation analysis. A receiver operating characteristic (ROC) curve analysis was used to assess the discriminating ability of the studied parameters. The area under the curve (AUC), sensitivity, specificity, and P value were taken for each factor. A P value of less than 0.05 was considered significant.

RESULTS

Results of the demographic features, biochemical variables, pulmonary function tests, ABG, and echocardiographic examinations of the COPD-PH group in comparison to the COPD and control groups were presented in Table 1. The COPD patients and control groups did not differ in terms of age, smoking index, BMI, or sex distribution. FEV1 and FEV1/FVC, two measures of lung function, were significantly lower in COPD groups than in controls. The FEV1 and FEV1/FVC were lower in the COPD-PH group than in the COPD group.

Table 1.

Demographic, clinical characteristics, echo-Doppler, and laboratory parameters among the studied population

Parameters (unit)	Groups, n			Significance
	COPD patients without PH (Group 1) n=25	COPD patients with PH (Group 2) n=35	Healthy groups n=30 (Group 3)	P 1	P 2	P 3
Gender (M/F)	22/3	31/4	19/11
Age (years)	55.0 (46.0–66.0)	57.0 (50.0–64.0)	57.0 (50.0–65.0)	0.952	0.747	0.848
Fold change of miRNA-21	5.57 (4.2–6.8)	14.7 (10.7–19.2)	1.09 (1.07–1.18)	<0.001*	<0.001*	<0.001*
BNP (pg/mL)	180.7 (119.9–325.7)	957.5 (515.3–1455.0)	91.9 (80.2–102.6)	<0.001*	<0.001*	<0.001*
Disease duration (years)	14.0 (10.0–20.0)	12.0 (10.0–18.0)	0.00 (0.000–0.000)	0.645	<0.001*	<0.001*
FEV1	64.0 (57.5–68.0)	56.0 (48.0–64.0)	83.5 (80.0–86.2)	0.007*	<0.001*	<0.001*
FEV1.FVC	63.0 (55.0–65.0)	55.0 (45.0–60.0)	72.5 (70.0–76.0)	0.002*	<0.001*	<0.001*
PaO2	79.0 (73.5–83.5)	60.0 (55.0–75.0)	93.0 (90.0–95.0)	<0.001*	<0.001*	<0.001*
RVOT (mm)	28.2 (23.3–31.3)	34.2 (32.4–36.4)	20.2 (19.05–21.5)	<0.001	<0.001*	<0.001*
RV basal (mm)	30.2 (26.4–33.4)	43.7 (38.3–47.3)	23.6 (21.05–28.2)	<0.001	<0.001*	<0.001*
RV mid (mm)	28.5 (26.2–33.1)	40.5 (36.8–45.6)	22.6 (20.2–25.7)	<0.001	<0.001*	<0.001*
PASP (mmHg)	25.6 (21.8–31.5)	49.6 (42.3–56.4)	18.6 (13.9–21.5)	<0.001	<0.001*	<0.001*
RV.S’ (cm/s)	10.9 (9.8–11.9)	9.1 (7.8–9.5)	13.7 (12.3–16.3)	<0.001	<0.001*	<0.001*
RV-GLS (%)	20.9 (20.1–21.9)	18.3 (17.1–19.8)	24.7 (22.7–26.02)	<0.001	<0.001*	<0.001*
TAPSE (mm)	20.0 (18.5–20.5)	16.2 (12.3–17.5)	21.8 (20.5–23.4)	<0.001	<0.001*	<0.001*
LVEF (%)	61.4±1.9	58.9±2.7	63.8±2.6	<0.001	<0.001	<0.001
FAC (%)	42.4 (41.1–45.9)	32.5 (29.8–37.4)	51.4 (49.8–53.5)	<0.001*	<0.001*	<0.001*
6MWD (m)	411.7 (389.5–420.5)	364.7 (263.2–385.3)	491.7 (489.1–520.8)	<0.001*	<0.001*	<0.001*

FEV1=, FEV1.FVC =, PAO2 =, RVOT=right ventricular outflow, RV S’=RV systolic annular velocity assessed by tissue Doppler imaging, RV-GLS=RV-global longitudinal strain, TAPSE=tricuspid annular plane systolic excursion, LVEF=left ventricular ejection fraction, FAC=fractional area change, 6MWD=6-minute walk distance. P¹ indicates A comparison between Group 1 and Group 2. P² indicates A comparison between Group 1 and Group 3. P³ indicated A comparison between Group 2 and Group 3

On echocardiographic examinations, COPD-PH patients had significantly higher RV dimensions compared to COPD patients and controls. Additionally, the COPD-PH group’s RV systolic function was significantly lower than that of the COPD and control groups, as indicated by lower values of TAPSE, RV S’, FAC, and RV-GLS. Additionally, even though LVEF readings were within the normal range for all groups, COPD patients exhibited lower LVEFs than controls.

Furthermore, the COPD-PH group experienced considerable hypoxia in comparison to the COPD group, as demonstrated by lower PaO2 and a shorter walking distance as measured by a 6-minute walk test.

The expressions of miR-21 and serum BNP were significantly greater in COPD groups compared to healthy controls, with the greatest elevation observed in COPD-PH patients.

Univariate analysis revealed that many non-invasive clinical parameters were associated with COPD-PH, such as pulmonary function test, TAPSE values, 6-min walk test, and serum miRNA-21 (P < 0.001), as indicated in Table 2.

Table 2.

Spearman’s correlation of variables correlated with pulmonary hypertension among COPD patients (n=60)

	PASP=Pulmonary Artery Systolic Pressure
	Correlation coefficient (r)	P
Age/years	0.103	0.436
Fold change of miRNA-21	0.619	<0.001*
BNP (pg/mL)	0.883	<0.001*
Disease duration	0.073	0.580
FEV1.	−0.624	<0.001*
FEV1/FVC	−0.732	<0.001*
PaO2	−0.895	<0.001*
RVOTD (mm)	0.843	<0.001*
RV.BASAL (mm)	0.848	<0.001*
RV.MID (mm)	0.951	<0.001*
RV.S (cm/s)	−0.923	<0.001*
RV-GLS (%)	−0.93	<0.001*
TAPSE (mm)	−0.962	<0.001*
FAC (%)	−0.943	<0.001*
6MWD (m)	−0.939	<0.001*

The ROC analysis showed that serum miR-21 levels were valuable in the diagnosis of PH in COPD patients, which had an AUC of 1.000, while sensitivity was 100%, and specificity was 100% at a cutoff value of >7.94, with a P value of <0.001. Also, BNP could predict PH with a cutoff predictive value of >390.1 pg/mL, while sensitivity was 88.6%, and specificity was 88%, with an AUC of 0.936 [Table 3, Figure 1].

Table 3.

ROC curve for the discriminative ability of the studied parameters to differentiate COPD cases without elevated PAP (n=25) from COPD cases with elevated PAP (n=35)

Variable(s)	Cutoff point	AUC	Sensitivity%	Specificity %	P	95% Confidence interval
Fold change of miRNA 21	>7.94	1.000	100.0	100.0	<0.001*	1.000	1.000
BNP	>390.1	0.936	88.6	88.0	<0.001*	0.877	0.995

AUC: area under the curve

Figure 1.

ROC curve for miRNA-21 and serum BNP as a diagnostic for COPD-associated pulmonary hypertension

DISCUSSION

The most frequent causes of PH are lung and left-sided cardiac disorders. Since PH development has no distinct clinical manifestations and is typically ignored until the illness worsens, it is detected far too late. Until recently, RHC has been the most effective method for diagnosing PH. However, due to its intrusive nature, it may be essential to develop new, simple, and trustworthy non-invasive diagnostic tools like biomarkers.[11] Recently, several biomarkers have been discovered and demonstrated to be promising for the identification of PH, particularly miRNAs.[12] Furthermore, they are quite stable in blood and could be detected in blood samples and analysed using qRT-PCR techniques, which offer a simple diagnostic method.[13]

In our work, we investigated the diagnostic utility of circulating miR-21 in distinguishing patients with COPD-PH, a condition classified as Group 3 PH according to the WHO classification, from patients with COPD alone and healthy controls. We demonstrated that serum miRNA-21 was significantly elevated in COPD groups compared to healthy controls, with the greatest elevation observed in COPD-PH patients, indicating a clear association between miR-21 expression and PH development in the COPD patients. Furthermore, the correlation analysis revealed a positive correlation between miRNA-21 levels and PH severity, which further supports its involvement in the pathophysiology of PH secondary to COPD. Notably, ROC curve analysis revealed that miR-21 has 100% sensitivity, 100% specificity and 1.000 AUC at a cutoff value of >7.94 for diagnosing PH in COPD patients. This suggests that miR-21 could be used as a good promising non-invasive diagnostic biomarker to identify PH in COPD patients, where invasive RHC is often impractical as a screening tool.

Our findings are consistent with previous studies. According to He et al.,[14] they employed a mouse model that combined exposure to cigarette smoke with injections of cigarette smoke extract to stimulate COPD. It was discovered that miR-21 levels in lung tissue were significantly higher and inversely linked with lung function, and knocking down miR-21 considerably decreased inflammation and lung damage. These findings are consistent with our study and provide evidence of miR-21’s involvement in the development of COPD.

On the other hand, Chang et al.[2] observed that patients with hypoxia-related PH, including those with COPD, had significantly higher levels of miR-21 in their serum and pulmonary tissue, which may be linked to endothelial dysfunction and pulmonary vascular remodelling, two important mechanisms in the development of COPD-associated PH.

In parallel with our study, Dimitry et al.[11] found that serum miR-21 had high diagnostic value (AUC of 0.742, sensitivity of 60%, and specificity of 95%) in differentiating patients with pulmonary arterial hypertension (PAH) (Group 1 PH) from other subtypes of PH and healthy individuals. Remarkably, our findings revealed superior diagnostic accuracy, with miR-21 demonstrating 100% sensitivity and specificity (AUC = 1.000) for distinguishing patients with COPD-PH from those with COPD-only and healthy controls. Although these values are slightly lower than those observed in our study, they still support the potential of miR-21 as a promising biomarker for PH. Notably, their study focused specifically on patients with Group 1 PAH, a subtype that is mainly defined by vascular remodelling in the absence of underlying lung disease. In contrast, our study specifically targeted patients with PH-associated COPD (Group 3 PH), where hypoxia and chronic inflammation play a central role in vascular pathology. This targeted focus on COPD-PH may explain the higher diagnostic accuracy observed in our findings compared to the broader Group 1-focused cohort in Dimitry et al.’s work. Furthermore, the work by Dimitry et al.[11] indicates that the type of PH influences the expression levels of miR-21, as Group 1 PH patients had the highest levels of miR-21, Group 3 PH patients had moderately elevated levels, while Group 2 PH patients and healthy controls had the lowest levels. This “graded expression pattern” suggests miR-21 could be used to distinguish different PH subtypes.

Moreover, several studies have highlighted the functional role of miR-21 in the pathogenesis of PH, particularly in hypoxia-driven settings such as COPD-associated PH (Group 3 PH). Karnati et al.[15] and Yang et al.[16] demonstrated that in experimental models, miR-21 was upregulated in response to chronic hypoxia and contributes to pulmonary vascular remodelling by promoting smooth muscle proliferation and inflammatory signalling, and its sequestration in these models attenuated hypoxia-induced vascular remodelling, suggesting its active involvement in disease progression in hypoxia-driven PH conditions like COPD-PH. Similarly, Sarkar et al.[17] found a direct correlation between miR-21 levels and elevated pulmonary artery pressures in hypoxic settings, reinforcing its link with Group 3 PH pathogenesis. Collectively, these findings support the use of miR-21 not only as a diagnostic biomarker for identifying PH within the COPD patients (Group 3 PH) but also to help differentiate them from individuals with COPD alone or other PH subtypes. Moreover, miR-21 may act as a pathogenic factor in hypoxia-induced PH, contributing to disease progression through vascular remodelling and inflammatory pathways.

In contrast, other studies have reported decreased miR-21 levels in certain PH subtypes. Gruning et al.[18] observed reduced miR-21 expression in PAH patients following exercise intervention, and Düzgün et al.[19] found significantly lower circulating miR-21 levels in patients with Group 1 PAH and CTEPH compared to healthy controls. Caruso et al.[20] similarly reported reduced miR-21 in human lung tissue and serum from PAH patients. These discrepancies may reflect subtype-specific miR-21 expression or stage-dependent variations in miR-21 regulation. Importantly, although some previous studies reported decreased miR-21 levels in Group 1 PH and CTEPH (Group 4), our study is the first to demonstrate a marked upregulation of miR-21 specifically in Group 3 PH (COPD-PH), highlighting its diagnostic specificity within the COPD-PH patient group. Notably, we observed excellent diagnostic performance with an AUC of 1.0, and 100% sensitivity and specificity in our study, indicating that miR-21 may serve as a highly specific and non-invasive tool for diagnosing PH in COPD patients, enabling earlier intervention and possibly improving outcomes in this high-risk group.

In addition to miR-21, NT-proBNP, a recognised cardiac biomarker, demonstrated diagnostic value in detecting PH-associated COPD. Our results are consistent with earlier meta-analyses, such as those conducted by Tian et al.[21] and Su et al.,[22] which found that patients with COPD-PH had NT-proBNP levels that were considerably greater than those with COPD alone. Su et al.[22] in a meta-analysis of 18 trials including 8,534 participants, confirmed the diagnostic significance of NT-proBNP in this cohort with a standardised mean difference of 0.82 (P < 0.0001).

Moreover, Tian et al.[21] demonstrated that an NT-proBNP threshold of 175 pg/mL produced an AUC of 0.651, indicating moderate diagnostic accuracy, in patients hospitalised with acute exacerbations of COPD (AECOPD). Collectively, these findings support the diagnostic utility of miR-21 and NT-proBNP, either alone or in combination, as promising non-invasive biomarkers for diagnosing PH in COPD patients (WHO Group 3 PH).

CONCLUSION

miRNA-21 demonstrates strong potential as a diagnostic biomarker for diagnosing PH in COPD patients, outperforming the conventional biomarker NT-proBNP in terms of diagnostic accuracy. Its expression also facilitates discrimination between COPD patients with and without PH, supporting its role in personalised patient assessment and management.

The limitations of the present study are its single-centre design and limited sample size. Future studies with larger patient cohorts and longitudinal follow-up are needed to validate these findings and assess how miR-21 and NT-proBNP can be integrated into clinical practice, potentially guiding earlier intervention and improving outcomes in high-risk PH populations. Additionally, more experimental research is required to explore the mechanisms underlying miRNA-21’s overexpression.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.