Diagnostic potential of serum miR-197 and IL-18 levels in predicting bronchopulmonary dysplasia: a prospective cohort study

Abstract

Background

Bronchopulmonary dysplasia (BPD) remains a major complication of prematurity, contributing significantly to infant mortality and long-term morbidity. Current diagnostic methods for BPD rely on clinical criteria assessed at 36 weeks postmenstrual age, underscoring the urgent need for earlier predictive biomarkers to enable timely intervention. MicroRNAs (miRNAs) and inflammatory cytokines, such as interleukin-18 (IL-18), have emerged as promising circulating biomarkers for various diseases. Specifically, miR-197 has been implicated in the regulation of inflammatory pathways and lung disease, whereas IL-18 functions as a key pro-inflammatory cytokine. However, their combined role in the early prediction of BPD remains unexplored. This study aimed to investigate the dynamic changes in serum miR-197 and IL-18 levels, as well as their diagnostic potential as early biomarkers of BPD in preterm infants.

Methods

A total of 129 premature newborns admitted to the Neonatal Intensive Care Unit at the First People’s Hospital of Yinchuan from January 2022 to December 2023 were enrolled in this prospective cohort analysis. The participants were stratified based on BPD status, in accordance with the National Institute of Child Health and Human Development (NICHD) guidelines, yielding 32 cases of BPD and 97 non-BPD controls. The serum IL-18 levels were measured through enzyme-linked immunosorbent assay (ELISA), while quantitative real-time polymerase chain reaction (qRT-PCR) was implemented for determining miR-197 expression at three critical time intervals: postnatal days 1, 7, and 14. The potential relationships between these biomarkers over the study period were determined via bivariate correlation analyses. The diagnostic potential of miR-197 and IL-18, both individually and in combination, was assessed through receiver operating characteristic (ROC) curve analyses.

Results

The BPD group required significantly longer durations of mechanical ventilation and Continuous Positive Airway Pressure (CPAP) assistance, and had a lower gestational age relative to the non-BPD group, and the variations were statistically significant. The serum miR-197 levels decreased significantly in the BPD group compared to those in the non-BPD group on postnatal days 1, 7, and 14, whereas IL-18 levels increased significantly (P < 0.05). However, serum IL-18 levels gradually increased the BPD cohort over time (P < 0.05). The serum levels of miR-197 and IL-18 levels exhibited an inverse correlation in the infants with BPD on postnatal days 1, 7, and 14 (r = -0.697, -0.816, and -0.746, respectively; P < 0.01). ROC curve analysis demonstrated that the combined detection of serum miR-197 and IL-18 levels on postnatal day 7 proved highly effective in estimating the risk of BPD.

Conclusions

Serum levels of miR-197 and IL-18 are likely associated with the onset and progression of BPD. The combined measurement of the serum levels of miR-197 and IL-18 may represent a promising biomarker strategy for the early detection of BPD and the assessment of disease severity.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12887-026-06580-5.

Introduction

Bronchopulmonary dysplasia (BPD) represents a serious and persistent complication observed in premature neonates, marked by injury to immature pulmonary tissues and aberrant repair processes driven by an inflammatory cascade involving pro- and anti-inflammatory factors [1, 2]. BPD arises from an interplay between genetic predisposition and multiple perinatal insults, including inflammation, infection, mechanical ventilation, and hyperoxia, which disrupt normal pulmonary development. According to clinical studies, BPD occurs more frequently in neonates delivered prior to 32 weeks of gestation, and severe cases exhibit a mortality rate as high as 10% within the first year [3]. Infants with severe BPD are vulnerable to a range of complications, such as increased susceptibility to wheezing disorders, particularly bronchiolitis and asthma. Significant advances in prenatal medicine and neonatal care technologies, particularly those developed for addressing the rising prevalence of pregnancies among women of advanced maternal age, have significantly improved the survival rates of preterm infants [4]. Paradoxically, these advancements in neonatal care have coincided with a rising occurrence of BPD. Current evidence suggests that the pathogenesis of BPD is mediated by complex gene-environment interactions, which compromise neonatal survival and impose significant burdens on affected families and healthcare systems [5]. Given these challenges, the early detection and prediction of BPD are critically important, as timely interventions may significantly mitigate disease progression and improve clinical outcomes. Recent studies have explored the biological implications of epigenetic modifications, including altered microRNAs (miRNAs) expression, in the pathogenesis of BPD [6]. Emerging evidence suggests that aberrant intrauterine conditions, such as hypoxia, hyperoxia, and inadequate nutrition, can alter gene expression patterns and modulate miRNA levels. These epigenetic signatures hold promise as candidate biological markers for assessing BPD risk.

MiRNAs represent an evolutionarily conserved class of small non-coding RNAs, typically 18–25 nucleotides long, and they play critical roles in regulating gene expression post-transcriptionally through specific interactions with target mRNAs, thereby significantly affecting disease etiology [7]. Additionally, previous research has revealed that miRNAs exhibit remarkable stability in vitro following isolation from the bloodstream, making them promising diagnostic biomarkers for various diseases, including lung cancer [8], heart failure [9], and acute myocardial infarction [10]. Their variable expression patterns during pulmonary morphogenesis [11] and their association with various respiratory disorders, including asthma, idiopathic pulmonary fibrosis, and lung cancer [12–14], provide compelling evidence in support of their essential regulatory roles in lung development and the pathophysiology of pulmonary diseases. Previous studies have demonstrated that miRNA levels undergo progressive alterations in BPD, indicating their involvement in its pathogenesis and highlighting their potential as novel diagnostic markers. The distinct miRNAs expression patterns in the tracheal aspirates of premature neonates with severe disease further validate their prospective application as biological markers for determining the severity of BPD [15]. Lal et al. [16] observed that the reduction in miR-876-3p expression at birth serves as a reliable indicator of severe BPD in neonates born with extremely low birth weight. Using mouse models of hyperoxia-induced BPD, previous studies have detected 201 differentially expressed miRNAs, such as the significantly downregulated miR-150, miR-126, and miR-151, as well as the markedly upregulated miR-21 and miR-34a [17, 18].

MiR-197 is encoded by a genetic locus at chromosomal region 1p13.3, has been shown to undergo persistent dysregulation across multiple disease states. Mechanistic investigations have demonstrated that miR-197 modulates key cellular processes, including growth, apoptosis, development, migratory potential, and inflammatory and immune responses, through interactions with specific targets [19]. It has been demonstrated that the differential expression of miR-197 likely serves as a promising non-invasive biomarker for pneumonia [20]. Additionally, miR-197 regulates key endothelial cell functions, including viability, angiogenesis, and inflammatory responses [21]. A previous study identified altered miRNAs expression in the pulmonary artery tissues of individuals with chronic obstructive pulmonary disease (COPD), such as reduced miR-197 expression [22]. MiR-197 expression was found to correlate with the extent of vascular remodeling in the lungs and impaired airflow. Although both miR-197 and miR-145 demonstrate favorable individual diagnostic sensitivity and specificity for non-small cell lung cancer (NSCLC), their combined detection further improves diagnostic sensitivity [23]. It has been demonstrated that miR-197 exacerbates LPS-induced cardiomyocyte injury by inhibiting Silent Information Regulator 2 Homolog 1 (SIRT1) expression, thereby indicating its potential as a promising molecular biomarker for treating sepsis [24].

Interleukin (IL)-18, a member the IL-1 family, functions as a pro-inflammatory cytokine, and is primarily released by activated mononuclear macrophages. IL-18 represents a pivotal immunoregulatory cytokine that serves critical roles in modulating helper T cell responses and orchestrating numerous immunological processes [25]. Accumulating evidence suggests that IL-18 can disrupt the intricate equilibrium between pro- and anti-inflammatory pathways. The excessive release of IL-18 has been shown to exacerbate damage to inflammatory tissues and accelerate the progression of pulmonary fibrosis. These findings highlight its essential role in the etiology of respiratory illnesses, especially adult acute respiratory distress syndrome (ARDS) and COPD [26]. Interestingly, clinical studies have demonstrated that miR-197 expression exhibits an inverse relationship with the severity of hepatitis B infections. By conducting gain- and loss-of-function analyses, mechanistic investigations have demonstrated that miR-197 overexpression significantly downregulates IL-18 expression during gene expression and protein synthesis, whereas miR-197 knockdown induces a contrary effect [27].

The changes in the serum levels of miR-197 and IL-18 in infants with BPD, as well as their clinical significance, remain poorly investigated to date. The present study aimed to characterize the dynamic expression profiles of miR-197 and IL-18 in serum samples collected from preterm neonates with BPD at different postnatal time points. We further evaluated the diagnostic utility of serum miR-197 and IL-18 levels in BPD and explored their potential as promising biomarkers for early risk stratification and clinical assessment.

Materials and methods

Patient cohort

A prospective cohort analysis was carried out, involving 129 premature neonates hospitalized in the Neonatal Intensive Care Unit (NICU) at the First People’s Hospital of Yinchuan from January 2022 to December 2023. The inclusion criterion was premature birth (gestational age < 32 weeks) with admission to the NICU within 24 h of birth. Among the neonates enrolled in this study, 32 were diagnosed with BPD, while 97 were not classified as having BPD based on the diagnostic guidelines. The BPD cohort was further subdivided into a severe BPD group (n = 10) and a mild BPD group (n = 22) based on disease severity. The severe group included infants who required more than 30% supplemental oxygen or respiratory assistance via Continuous Positive Airway Pressure (CPAP) or mechanical ventilation at 36 weeks of corrected gestational age [28]. A diagnosis of BPD was made based on the 2018 National Institute of Child Health and Human Development (NICHD) guidelines [28], which define it as the requirement for oxygen supplementation via mechanical ventilation, non-invasive positive pressure ventilation, high-flow nasal cannula, or low-flow oxygen, at a postmenstrual age of 36 weeks. The exclusion criteria included the presence of (1) major congenital malformations (2), chromosomal aberrations, and (3) complex congenital cardiac disease (excluding patent ductus arteriosus and minor septal defects). Ethical clearance for the study was granted by the Institutional Review Board of the First People’s Hospital of Yinchuan on September 5, 2022 (Ethical Approval ID: 20220811), and written informed consent was provided by the legal guardians of all enrolled participants.

Sample acquisition

Samples of venous blood were collected from all the study participants at three postnatal intervals (days 1, 7, and 14) using sterile venipuncture procedures. The samples were subsequently centrifuged at 3000 × g for 10 min, and the obtained serum samples were promptly aliquoted into RNase-free cryovials and stored at -80 °C to minimize freeze-thaw cycles prior to batch analyses.

Enzyme-linked immunosorbent assay (ELISA)

The serum IL-18 levels were quantified using a commercially sourced ELISA kit (Wuhan Yipu Biotechnology Co., Ltd., Cat. No. 20240012#, Wuhan, China) in accordance with the manufacturer’s instructions.

Quantitative real time polymerase chain reaction (qRT-PCR)

Total RNA was extracted from the serum samples using an RNeasy Mini Kit (Haigene, Harbin, China). The isolated total RNA was reverse transcribed for synthesizing complementary DNA (cDNA) using the Maxima First Strand cDNA Synthesis Kit for qRT-PCR (Thermo Fisher Scientific, Cat. No. D1801, USA). qRT-PCR was performed using the MasterCycler epgradient RealPlex system (Eppendorf, Hamburg, Germany) with the Maxima SYBR Green/ROX qPCR Master Mix (Eppendorf, Hamburg, Germany). The primers for miR-197 and U6 were purchased from Qiagen (Duesseldorf, Germany). The sequences of the forward and reverse primers for miR-197 were: 5ʹ-GCGTTCACCACCTTCTCCA-3ʹ and 5ʹ-AGTGCAGGGTCCGAGGTATT-3ʹ, respectively, whereas those for U6 were 5ʹ-CTCGCTTCGGCAGCACA-3ʹ and 5ʹ-AACGCTTCACGAATTTGCGT-3ʹ, respectively. The expression levels were normalized to that of the U6 endogenous control and quantified using the comparative Ct (ΔΔCq) method.

Statistical analyses

The statistical analyses were carried out using SPSS software (v25.0). The results were expressed as the mean ± standard deviation (x̄ ± s), and comparisons between groups were conducted using independent t-tests. Frequencies or percentages were used to present categorical data, which were analyzed with the chi-square (χ²) test. The data collected across different time intervals were analyzed through one-way analysis of variance (ANOVA), and pairwise group differences were examined using the Student-Newman-Keuls (SNK-q) post hoc test. Correlation analyses were conducted with Pearson’s and Spearman’s correlation methods. The predictive potential of miR-197 was assessed through receiver operating characteristic (ROC) curve analysis, and P < 0.05 indicated statistical significance. Given the exploratory nature of this study and the relatively small sample size of the BPD cohort (n = 32), multivariable adjustment was not performed to minimize the risk of model overfitting and preserve the robustness of the statistical findings. The potential confounding factors, including gestational age and birth weight, were evaluated as baseline characteristics, and their association with BPD are presented in Table 1.

Table 1.

Comparative analyses of clinical data between BPD and non-BPD cohorts

Patient demographics		BPD group (n = 32)	Non-BPD group (n = 97)	t/χ²	P-value
Infant characteristics	Gestational age (weeks), mean ± SD	30.1 ± 0.7	30.4 ± 0.5	2.26	0.026
	Birth weight (gm), mean ± SD	1550 ± 122	1625 ± 109	3.261	0.001
	Gender (male), n (%)	15 (46.8)	46 (47.4)	0.003	0.957
	Duration of CPAP (days)	24.81 ± 3.1	8.47 ± 2.9	27.19	< 0.01
	Duration of mechanical ventilation (days)	12.12 ± 0.97	2.7 ± 1.12	42.28	< 0.01
	NRDS [n (%)]	6 (18.7)	9 (9.2)	2.1	0.147
	Necrotizing enterocolitis, n (%)	3 (9.3)	6 (6.2)	0.377	0.539
	Sepsis, n (%)	4 (12.5)	13 (13.4)	0.017	0.895
Maternal characteristics	Maternal smoking (yes), n (%)	10 (31.2)	17 (17.5)	2.738	0.097
	Chorioamnionitis, n (%)	7 (21.9)	12 (12.4)	1.73	0.188
	Premature rupture of membranes, n (%)	6 (18.7)	10 (10.3)	1.264	0.261
	Antenatal steroids, n (%)	27(84.3)	77 (79.4)	0.384	0.535
	Gestational Diabetes, n (%)	6 (18.7)	16 (19.7)	0.086	0.769

NRDS Neonatal respiratory distress syndrome

Results

Patient cohort

A total of 138 newborns were selected between January 2022 and December 2023, of whom 5 were not included due to brief hospital stays, and 2 were excluded due to surgical procedures during hospitalization. Among the remaining 131 infants, 32 received a diagnosis of BPD. Of the remaining 99 newborns without BPD, 2 were excluded owing to incomplete clinical data or hemolyzed or insufficient blood samples that prevented reliable analysis. The remaining 97 newborns were assigned to the non-BPD group. The BPD cohort consisted of 22 newborns with mild BPD and 10 with severe BPD. Among the infants, 85 were classified as early preterm (gestational age 28–31⁺⁶ weeks), and 44 as moderate preterm (gestational age 32–33⁺⁶ weeks) (Table S1).

Comparative evaluation of clinical data between BPD and non-BPD cohorts

The BPD cohort required significantly longer durations of CPAP and mechanical ventilation, and exhibited significantly lower birth weights and gestational ages, relative to those of the non-BPD group (P < 0.05; Table 1). There were no statistically significant differences in maternal factors between the BPD and non-BPD cohorts (P > 0.05).

Comparative analyses of serum miR-197 and IL-18 levels between BPD and non-BPD cohorts at different time points

The relative miR-197 expression levels exhibited significant downregulation in the preterm neonates with BPD (P < 0.05) relative to those in the non-BPD cohort at postnatal days 1, 7, and 14, whereas IL-18 expression was significantly upregulated (P < 0.05). Notably, IL-18 expression progressively increased over time in the BPD group, and the variations in expression between the BPD and non-BPD groups were statistically significant across all three time points (Table 2; Figs. 1 and 2).

Table 2.

Comparative analysis of serum miR-197 and IL-18 levels in the BPD and non-BPD cohorts at different postnatal time intervals

Biomarkers	Postnatal day	BPD group(n = 32)	Non-BPD group (n = 97)	t/χ²	P-value
miR-197	Day 1	0.845 ± 0.13	1.07 ± 0.20	5.95	< 0.01
	Day 7	0.769 ± 0.10	0.94 ± 0.11*	8.14	< 0.01
	Day 14	0.636 ± 0.13	0.9 ± 0.12*	10.78	< 0.01
IL-18	Day 1	140.7 ± 49.5	93.3 ± 29.8	5.10	< 0.01
	Day 7	225.3 ± 46.8*	155.9 ± 39.2*	8.14	< 0.01
	Day 14	240.1 ± 63.9*	159.4 ± 16.5*	11.41	< 0.01

*Statistically significant difference in expression, relative to that on postnatal day 1 (P< 0.05)

Fig. 1.

Comparative evaluation of serum miR-197 levels between BPD and non-BPD cohorts at different time intervals. *Statistically significant difference in expression, relative to that on postnatal day 1 (P < 0.05). The Y-axis represents serum miR-197 levels, and the X-axis represents different days. Each dot represents the mean value, with error bars indicating SD

Fig. 2.

Comparative evaluation of serum IL-18 levels between the BPD and non-BPD cohorts at different time intervals. *Statistically significant differences in Il-18 expression, relative to that on postnatal day 1 (P < 0.05). The Y-axis represents serum IL-18 levels, and the X-axis represents different days. Each dot represents the mean value, with error bars indicating SD

Comparison of serum miR-197 levels across three neonate groups

Detailed subgroup analyses were further conducted to assess the clinical value of predicting the severity of BPD. As indicated in Table 3, serum miR-197 levels were lower in neonates with severe BPD compared to those in the mild BPD group on the first day after birth, and the same pattern was observed on days 7 and 14 (Table 3).

Table 3.

Comparison of serum miR-197 and IL-18 levels across three infant groups

	Severe BPD group (n = 10)	Mild BPD group (n = 22)	Control group (n = 97)	t/χ²	P-value
miR-197
1st day	0.794 ± 0.11	0.876 ± 0.13	1.07 ± 0.20	17.64	< 0.001
7th day	0.742 ± 0.11	0.782 ± 0.12	0.94 ± 0.11	28.46	< 0.001
14th day	0.606 ± 0.12	0.652 ± 0.11	0.9 ± 0.12	59.61	< 0.001
IL-18
1st day	152.6 ± 44.4	121.5 ± 34.5	93.3 ± 29.8	30.18	< 0.001
7th day	242.8 ± 44.8	212.6 ± 52.2	155.9 ± 39.2	44.31	< 0.001
14th day	242.6 ± 46.8	230.6 ± 60.2	159.4 ± 16.5	91.95	< 0.001

Comparison of serum IL-18 levels across three neonate groups

The serum levels of IL-18 were highest in neonates with severe BPD, significantly exceeding those in the mild and non-BPD groups at all three time points on days 1, 7, and 14 (Table 3).

Correlation between serum miR-197 and IL-18 levels in infants with BPD

Pearson and Spearman correlation analyses demonstrated that the serum levels of miR-197 and IL-18 were significantly inversely correlated in preterm infants with BPD at postnatal days 1, 7, and 14 (Table 4; Fig. 3).

Table 4.

Correlation between serum miR-197 and IL-18 levels in BPD

Serum levels of miR-197 Postnatal day	Serum levels of IL-18
	R	P-value
1	-0.697	< 0.001
7	-0.816	< 0.001
14	-0.746	< 0.001

Fig. 3.

Correlation between serum miR-197 and IL-18 levels in infants with BPD on postnatal days 1, 7, and 14. The Y-axis represents serum IL-18 levels, and the X-axis represents serum miR-197 levels

Diagnostic potential of miR-197 in BPD

The diagnostic potential of serum miR-197 levels in BPD at various time points was assessed by ROC curve analysis, and the findings are provided in Table 4; Fig. 4A. On postnatal day 14, the serum levels of miR-197 exhibited a high diagnostic efficacy in predicting BPD, with a cut-off value of 0.721 (area under the curve (AUC): 0.926, sensitivity: 75.26%, specificity: 92.78%, and P < 0.001). On postnatal day 7, the serum levels of miR-197 demonstrated a strong predictive potential for BPD, with an AUC, sensitivity, and specificity of 0.878, 75.26%, and 84.37%, respectively (P < 0.001) (Table 5; Fig. 4A).

Fig. 4.

ROC curve analysis of the diagnostic potential of the serum levels of miR-197 and IL-18 in predicting BPD. Diagnostic performance of the serum levels of (A) miR-197, (B) IL-18, (C) miR-197 combined with IL-18 in predicting BPD

Table 5.

Predictive potential of miR-197 in diagnosis BPD at various postnatal intervals

miR-197 levels on postnatal day	AUC	Sensitivity (%)	Specificity (%)	Cut-off value	P -value
1	0.827	75	83.5	0.585	< 0.001
7	0.878	75.26	84.37	0.596	< 0.001
14	0.926	75.26	92.78	0.721	< 0.001

Diagnostic potential of IL-18 in BPD

The diagnostic utility of serum IL-18 levels for BPD at various postnatal time points was assessed through ROC curve analysis, and the findings are provided in Table 6; Fig. 4B. On postnatal day 7, the serum levels of miR-197 exhibited superior diagnostic efficacy in predicting BPD, with a cut-off value of 0.68 (AUC: 0.885, sensitivity: 71.1%, specificity: 92.87%, P < 0.001) (Table 5; Fig. 4B).

Table 6.

Diagnostic potential of serum IL-18 levels in predicting BPD at different postnatal intervals

IL-18 levels on postnatal day	AUC	Sensitivity (%)	Specificity (%)	Cut-off value	P -value
1	0.793	84.5	65.6	0.501	< 0.001
7	0.885	71.1	92.87	0.680	< 0.001
14	0.899	91.75	81.25	0.791	< 0.001

Combined diagnostic potential of the serum levels of miR-197 and IL-18 in predicting BPD

To improve the predictive accuracy, the combined diagnostic potential of the serum miR-197 and IL-18 levels on postnatal day 7 was evaluated by ROC curve analysis (Table 7; Fig. 5C). The combined levels demonstrated a sensitivity, specificity, and AUC were 94.8%, 93.7%, and 0.976, respectively, indicating a substantially superior predictive potential than that of either marker alone (Table 6; Fig. 4C).

Table 7.

Combined diagnostic potential of miR-197 and IL-18 levels in predicting BPD on postnatal day 7

Biomarker	AUC	Sensitivity (%)	Specificity (%)	P-value
miR-197	0.878	75.26	84.37	< 0.001
IL-18	0.885	71.1	92.87	< 0.001
Combined	0.976	94.8	93.7	< 0.001

Fig. 5.

ROC curve analysis showing the combined diagnostic potential of miR-197 and IL-18 levels in predicting severe BPD

Combined diagnostic potential of serum miR-197 and IL-18 levels in predicting severe BPD

ROC curve analysis was performed to evaluate the combined diagnostic efficacy of serum miR-197 and IL-18 levels on day 7 for identifying severe BPD (Fig. 5). The combination yielded a sensitivity, specificity, and AUC of 90%, 87.4%, and 0.979, respectively (95% CI: 0.890–0.992).

Discussion

BPD is a persistent pulmonary disorder associated with prematurity that arises from disrupted lung development due to preterm birth, coupled with persistent inflammatory processes. The prevalence of BPD is escalating as the survival rates of extremely preterm infants improve. BPD imposes a substantial healthcare burden, characterized by prolonged hospitalizations, increased susceptibility to respiratory infections, and a higher risk of neurodevelopmental impairments. BPD stems from an intricate interplay among genetic, epigenetic, developmental, and environmental factors. The central mechanisms underlying the pathogenesis of BPD include (1) genetic predisposition and epigenetic modifications; (2) disrupted lung development, characterized by alveolar simplification and impaired vascularization; (3) postnatal insults such as mechanical ventilation and oxygen toxicity; and (4) sustained inflammatory responses [29, 30]. Furthermore, maternal conditions, including pre-eclampsia, obesity, gestational diabetes, and systemic inflammation, may increase the predisposition to preterm birth and indirectly influence the risk of BPD by altering the intrauterine environment [31]. The present study demonstrates that the factors associated with mechanical ventilation collectively represent significant risk factors for the pathogenesis of BPD. Prolonged exposure to hyperoxia can induce oxidative stress responses, potentially leading to structural and functional abnormalities in pulmonary tissue. The early prediction of BPD is likely a critical prerequisite for improving pulmonary development outcomes in preterm infants.

The lack of definitive treatments for BPD is attributable to its complex pathophysiology and poorly defined clinical classification. It is therefore essential to identify biomarkers that are relevant to its early diagnosis, disease progression, and prediction of therapeutic responses. Owing to their stability, lack of known post-processing modifications, and ease of detection, miRNAs represent promising novel non-invasive biomarkers for clinical application. The extracellular miRNAs present in serum and plasma have attracted considerable interest as reliable indicators for analyzing miRNA expression, serving as potential non-invasive diagnostic markers for various disorders. Emerging research has established miRNAs as key regulators of various cellular processes, including proliferation, migration, apoptosis, angiogenesis, and inflammation, with their dysregulation serving as a diagnostic biomarker for a range of human diseases [21]. MiR-197 is encoded by a genetic locus at chromosomal region 1p13.3, and studies have demonstrated that it is significantly dysregulated in various diseases, including lung, breast, ovarian, and colorectal cancers, as well as hepatocellular carcinoma [32, 33]. Zampetaki et al. [34] revealed that miR-197 expression exhibits a negative correlation with the risk of myocardial infarction, whereas Abd-El-Fattah et al. reported its association with the severity of pneumonia [35]. MiR-197 demonstrates context-dependent functionality, exhibiting significantly divergent expression patterns across studies. Although miR-197 was initially identified as downregulated in primary biliary cirrhosis [36], subsequent studies have shown it to be overexpressed [37], completely absent [38], or downregulated [39], depending on the disease context. To date, the predictive potential of serum miR-197 levels in diagnosing BPD in infants remains to be investigated. In this study, we analyzed the serum levels of miR-197 in infants with and without BPD to evaluate its diagnostic potential. The findings demonstrated that miR-197 expression was significantly downregulated in infants with BPD on postnatal days 1, 7, and 14, compared to that in the control group (P < 0.001). The findings further revealed that the downregulation of miR-197 expression was more pronounced in severe BPD than in mild BPD. These observations collectively indicate that miR-197 likely exhibits a dynamic response to disease severity, indicating that it can serve as a promising novel diagnostic biological marker for BPD.

IL-18 is a pro-inflammatory cytokine that belongs to the IL-1 family, and is constitutively expressed in the cytoplasm of various cell types, including macrophages and keratinocytes. IL-18 exhibits multiple immunomodulatory functions by enhancing interferon-γ (IFN-γ) secretion and modulating helper T cell immune responses [40]. Recent studies suggest that IL-18 plays key roles in the advancement of pulmonary inflammation by activating Th1-type helper T cells and promoting the proliferation of fibroblasts and deposition of collagen. The involvement of IL-18 in these molecular pathways implicates it in the pathophysiology of ARDS and COPD, suggesting its promise as a predictive biological marker for BPD [41]. This study demonstrated that miR-197 was downregulated, whereas IL-18 expression was upregulated in infants with BPD. The findings confirmed the significant upregulation of IL-18 expression in neonates with BPD compared to that in the control group, suggesting a potential association between elevated IL-18 levels and the development of BPD (P < 0.001), consistent with the findings of previous studies [42]. The present study demonstrated that serum IL-18 levels exhibited a significant positive correlation with the clinical severity of BPD in neonates, with IL-18 expression increasing with disease severity (P < 0.001). The observed dose-dependent association suggests that elevated serum levels of IL-18 likely contributes significantly to the onset and advancement of BPD. Notably, the serum levels of miR-197 and IL-18 differed significantly between the BPD and non-BPD groups on postnatal day 1 (Table 2). This observation is particularly noteworthy, as it suggests that these biomarkers likely reflect antenatal inflammatory priming or intrauterine environmental influences, such as subclinical maternal infection, placental dysfunction, or oxidative stress, rather than postnatal lung injury alone. In preterm infants, the intrauterine environment plays a critical role in shaping pulmonary development and immune responses, in that antenatal insults may ‘prime’ the fetal immune system, leading to exaggerated postnatal inflammatory responses and heightened susceptibility to BPD. The early differences in the serum levels of miR-197 and IL-18 may therefore serve as indicators of intrauterine programming, distinguishing antenatal influences from postnatal insults—including mechanical ventilation and hyperoxia—that subsequently exacerbate lung injury. This distinction is clinically relevant, as it highlights the potential of these biomarkers in identifying high-risk infants at birth, thereby enabling the implementation of targeted interventions to mitigate both antenatal and postnatal risks.

To the best of our knowledge, this study is the first to report that the serum levels of miR-197 exhibit a strong negative correlation with serum IL-18 levels on postnatal days 1, 7, and 14 in infants who develop BPD (r = -0.697, -0.816, and − 0.746, respectively; P < 0.01). IL-18 is a pro-inflammatory cytokine that contributes to the onset and progression of BPD. The present study suggests an association between the expression levels of miR-197 and IL-18 in children with BPD. The findings further revealed that miR-197 may function as a pro-inflammatory mediator by modulating the expression of inflammatory genes, contributing to the increased production of pro-inflammatory cytokines and amplification of inflammatory responses, thereby exacerbating lung tissue damage. The findings obtained herein may serve as a foundation for further validation in subsequent studies.

ROC analysis indicated that serum miR-197 and IL-18 levels on postnatal day 7 demonstrated high sensitivity and specificity in estimating BPD risk, thereby highlighting their potential utility as valuable diagnostic biomarkers. Analysis of the combined diagnostic value of miR-197 and IL-18 on postnatal day 7 revealed superior performance, with the highest sensitivity (94.8%), specificity (93.7%), and AUC (0.976), which surpassed the diagnostic accuracy of either miR-197 or IL-18 alone. Compared to previous studies [43], it demonstrated improved specificity and sensitivity. Therefore, the simultaneous measurement of the serum levels of miR-197 and IL-18 could potentially enhance the predictive accuracy of BPD. We further evaluated the combined diagnostic potential of serum miR-197 and IL-18 levels on day 7 for predicting severe BPD. Although the AUC value was markedly high (0.979), the limited sample size of the severe BPD group (n = 10) may have caused model overfitting, potentially leading to an overestimation of diagnostic accuracy. This study contributes to the expanding literature on biomarkers for predicting BPD. Previous studies have primarily investigated individual biomarkers, including inflammatory cytokines, such as IL-33 [43], N-terminal pro-brain natriuretic peptide (NT-proBNP) [44], mediator complex subunit 1 (MED1) [45], and systemic inflammatory markers [46]. However, these studies exhibit significant variability in predictive accuracy (AUC: 0.70–0.90), and face numerous limitations, including poor sample stability, lack of dynamic monitoring, and insufficient specificity in clinical correlations. The novelty of this study lies in exploring the combined application of miRNAs and cytokines for the timely prediction of BPD. This combination achieved a high AUC of 0.976 on day 7, surpassing the predictive performance of various previously reported single biomarkers. Our study highlights that miR-197 and IL-18 can serve as promising biomarkers for BPD, offering distinct advantages over previously reported markers. Owing to its regulatory role in gene expression, miR-197 is particularly relevant to the pathways of lung development and injury, exhibiting significant changes in expression in patients with BPD [47]. IL-18 is a pro-inflammatory cytokine that is indicative of immune activation and may play a crucial role in the early diagnosis of BPD [48]. Elevated levels of IL-18 may indicate early-stage inflammation, whereas alterations in miR-197 expression are likely associated with the processes of lung repair. Therefore, the combined assessment of these markers may provide a comprehensive strategy for the early diagnosis and monitoring of BPD. Given the ease of sample collection from infants in clinical settings and the straightforward detection of the serum levels of IL-18 and miR-197, these indicators can serve as viable biomarkers for clinical application.

The present investigation has several key limitations, which are discussed hereafter. First, the study is subject to two interrelated constraints affecting the generalizability of the findings. Although the cohort represents the NICU population at our institution, the relatively small number of BPD cases (n = 32) limits data variability. Another limitation of this study is the lack of statistical adjustment for prenatal and perinatal confounders, which may have influenced the results. Similarly, the reported ROC curves and AUC values are based on unadjusted biomarker levels, representing another limitation. Additionally, gestational immaturity may have affected the outcomes of this study. Future large-scale confirmatory studies should incorporate these covariates into adjusted ROC models to validate the disease-specific predictive performance of the biomarkers. On the other hand, the single-center design further restricts the applicability of the findings to other clinical settings, as differences in NICU protocols, patient demographics, and regional clinical practices across institutions may lead to inconsistent outcomes during the replication of the study. Second, the small sample size also limits the reliability of the observed miRNA expression patterns. As a limited sample size is more susceptible to individual variabilities, including unique genetic backgrounds or subtle physiological differences among participants, the detected miRNA patterns may not fully reflect true associations with the pathogenesis of BPD. Instead, they may be partially attributed to random individual differences, thereby reducing the credibility of the observed correlations. Although several key maternal factors, including gestational diabetes and chorioamnionitis, were controlled for in the analysis (Table 1), the potential influence of other factors cannot be fully excluded. These factors can be categorized into two categories: (1) well-established antenatal factors, including pre-eclampsia, maternal obesity, and prolonged premature rupture of membranes (PPROM), which are known to increase the risk of preterm birth and BPD by altering the intrauterine environment. These variables were included in the analysis of baseline characteristics, and no significant differences were observed between the BPD and non-BPD groups (P > 0.05); and (2) unmeasured potential contributors, including subtle factors not quantified in this study—including placental pathology (such as mild villous dysplasia), fetal growth restriction, subclinical maternal infection, maternal nutritional status, and intrauterine oxidative stress—may have influenced biomarker levels on postnatal day 1 [6, 31]. These unmeasured variables represent potential confounders that warrant further investigation in future large-scale studies. Furthermore, the small sample size severely limited the ability to perform robust stratified analyses. Specifically, it was not possible to stratify patients based on key variables, including gestational age or growth status, for identifying patient populations that would benefit most from this biomarker assay. The identification of these target subgroups is critical for the clinical translation of the combined biomarkers identified herein and represents a key focus for future research.

Conclusions

In summary, the findings revealed that preterm infants with BPD exhibit lower serum miR-197 levels and higher IL-18 expression, both of which are associated with disease severity and serve as predisposing factors for its development in premature neonates. The integrated assessment of serum miR-197 and IL-18 levels exhibits significant diagnostic potential for estimating the onset of BPD in premature neonates, offering novel insights into its early diagnosis.

Authors’ contributions

Conceptualization: Lie Huang and Meile Cheng; Data curation: Xiaomei Fu and Aipeng Liu; Formal analysis: Xi Xu; Investigation: Lie Huang; Writing—original draft: Lie Huang and Xi Xu; Writing—review & editing: Lie Huang. All the authors have reviewed and approved the final version of the manuscript for publication.

Funding

This research was supported by the Natural Science Foundation, Department of Science and Technology of Ningxia (grant number: 2023AAC03867).

Data availability

Data supporting the findings of the study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The research was performed in accordance with the Declaration of Helsinki and received approval from the Institutional Review Board of the First People’s Hospital of Yinchuan (protocol code: 20220811, approved on September 5, 2022).

Written informed consent was obtained from the parents of all enrolled subjects.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.