Bronchopulmonary dysplasia (BPD) is a severe chronic lung disease of extreme preterm infants. Recently, the advances in perinatal care have resulted in reductions in neonatal morbidity and mortality, but the incidence of BPD has not decreased [1, 2]. BPD has been characterized by arrested alveoli and pulmonary vessel development [3]. Infants with BPD may lead to respiratory dysfunction, cardiopulmonary dysfunction, nervous system development retardation and potential death [4, 5]. The therapeutic approaches for BPD, such as non-aggressive ventilator measures, glucocorticoids, diuretics, caffeine and vitamins, are only partly satisfactory [6].

Inflammation is a major contributor to the pathogenesis of BPD [7]. In BPD patients, the increase of pro-inflammatory cytokines, such as TNF-$\alpha$, IL-1$\beta$, IL-6, monocyte chemo-attractant proteins, and the decrease of anti-inflammatory cytokines IL-10 in the lung, tracheal aspirate and peripheral blood induced imbalance of pro-inflammatory and anti-inflammatory cytokines has been demonstrated [8]. Recently, neutrophil-to-lymphocyte ratio (NLR) [9], eosinophil-to-basophil ratio (EBR) [10], platelet-related indicators, such as platelet-to-lymphocyte ratio (PLR) [11] and platelet-to-neutrophil ratio (PNR) [12] calculated from parameters in peripheral blood, can be used to evaluate the prognosis of inflammatory diseases with increased sensitively, without increasing the cost. PLR and PNR can act as novel prognostic markers in various clinical conditions, such as cardiovascular disease [13, 14], rheumatic disease [11], acute ischemic stroke [15], bacterial infection [12], cancer [16] and preterm necrotizing enterocolitis (NEC) [17]. PLR is associated with prognosis in patients with early-onset sepsis (EOS) in neonates [18]. PLR and PNR were associated with a variety of lung diseases. Qu et al. [19] found that PLR can be used as a parameter in predicting coronavirus disease-19, as the patients with significantly elevated platelets during treatment had longer average hospitalization days. Luo et al. [20] found that combination of the PLR and fibrinogen was an independent prognostic factor for 5-year overall survival in patients with non-small cell lung cancer. Serdar Karakaya et al. [21] demonstrated that PLR could be used as an independent predictive biomarker for the use of diffusing capacity of the lungs for carbon monoxide decline which is used to identify bleomycin-induced pulmonary toxicity. PLR also has significative diagnostic value in patients with solitary pulmonary nodules [22]. Meanwhile, PNR have good applied value in the diagnosis of neonatal pneumonia with high sensitivity and specificity [23].

In a previous study, we found that NLR at 72 h after birth can be considered as an important early indicator of BPD in preterm infants [24]. However, the relationship between the platelet-related indicators and BPD remains unknown. In this study, we aimed to assess the potential role of platelet-related indicators in early prediction for BPD.

BPD is a severe clinical syndrome of lung injury that impedes alveolarization and microvascular development in preterm infants. Inflammation, caused by hyperoxia, chorioamnionitis and postnatal infection, is the main factor that initiates lung injury and contributes to BPD in preterm infants. It is clear that inflammation-related signaling pathways, such as NF-kB, Toll-like receptor 4, interferon, and pro-inflammatory cytokines are associated with the development of BPD [25, 26]. Platelets, small anucleate cellular fragments that are released by megakaryocytes, are critical for inflammatory and immune responses. Platelets have been shown to influence leukocyte recruitment and cytokine response [27], to limit bacterial growth and dissemination [28], to influence activation of the vascular endothelium and the coagulation system [29], and to shape immune responses to pathogens and tumor cells [30]. During the development of sepsis, platelets are one of the first cells to respond when the pro-inflammatory and pro-coagulant mechanisms are disrupted [28, 29]. In an inflammatory model, researchers found that many interactions of platelets with the leading edge of adherent neutrophils, and recruited neutrophils searched for activated platelets to initiate inflammation in the early phase [31]. Looney et al. [32] reported that it is vital for the coordinated interactions between neutrophils and platelets in the lung microcirculation of acute lung injury and the elimination of platelets in the blood significantly reduced lung damage in a preclinical study.

Inflammation plays a key role in lung development in extremely preterm infants and is a major contributor to the pathogenesis of BPD. The lung has been confirmed to be a primary site of terminal platelet biogenesis with approximately 10 million platelets per hour [33]. Thrombocytopenia is associated with a more disturbed host response [34]. Therefore, we speculated that the platelet count may be related to the outcomes of BPD. In our study, we found that the significant decrease in P counts at 72 h, at 2 w and at 4 w and the significant increase in P counts at 1 w in infants with BPD, which associations may be due to inflammation in the pulmonary and peripheral blood. However, Chen et al. [35] found that the platelet counts at birth were higher in 115 preterm infants who were delivered at a gestational age $\le$28 weeks or a birth weight $\le$1000 g who developed BPD, but we found no difference in platelet counts at birth in the current study. This discrepancy may be owing to the different study populations including the different gestational age and birth weight, and further clarification is needed in future studies.

Recently, a number of studies have shown that platelet-related indicators can be used as prognostic biomarkers for a variety of diseases [11, 12, 13, 14, 15, 16, 17, 18]. Several studies demonstrated the significance of PLR and PNR in evaluating and predicting the severity of systemic inflammation, infections, immune and neoplastic diseases [19, 20, 21, 22, 23]. Studies also demonstrated that the PLR as an inflammatory marker along with NLR, which provides information about the disease activity and severity in early onset sepsis [36] and cancer [37]. Yun et al. [38] demonstrated that PNR and PLR as novel inflammatory biomarkers could predict the clinical outcome after aneurysmal subarachnoid hemorrhage. Liao et al. [23] clarified that peripheral blood parameter of NLR and PNR have good applied value in the diagnosis of neonatal pneumonia with high sensitivity and specificity. Similarly, inflammation and immune response depend on N, P and L counts in BPD patients. In our previous study, we confirmed that the higher NLR at 72 h after birth can be considered as an early indicator of BPD [24]. Accordingly, we speculated that the platelet-related indicators may be associated with BPD. In this study, we confirmed that the value of PLR at birth and PNR at birth were higher in preterm infants with BPD, while PLR at birth and PNR at birth were independently associated with the risk for BPD. When combining gestational age with PLR at birth and PNR at birth, the sensitivity and specificity for predicting BPD were improved. At the same time, we found that P at 2 w in infants with BPD was lower than that in infants without BPD, suggesting that it could be used as an independent predictor of BPD. When combined with PLR at birth, PNR at birth, P at 2 w, and gestational age, the sensitivity of the diagnosis of BPD was increased.

Severe complications such as NEC, pulmonary hypertension, IVH and ROP are often associated with premature infants, which can lead to poor prognosis. In this study, we observed a correlation between PLR and severe adverse complications. The occurrences were higher for adverse outcomes such as NEC, pulmonary hypertension, IVH grade 3 or 4, ROP requiring intervention, and death of infants with PLR $\ge$135.33 at birth, PNR $\ge$129.12 at birth and P at 2 w 285.8. Go et al. [39] found mean platelet volume $\ge$10.2 fL within 12 h of birth correlates with mortality among infants born 32 weeks’ gestation. However, there are few studies on the prediction of adverse outcomes in premature infants by platelet-related indicators. We speculated that platelet-related indicators can be a meaningful indicator for complications of preterm infants and further relevant prospective studies are needed.

The main advantage of our research is that platelet-related indicators are easily accessible with no extra expense in the clinic practice. However, the present study has some limitations. The specificity and sensitivity of the indicators are not the strongest, and it is necessary to combine gestational age and other indicators to predict BPD. Meanwhile, the subjects were relatively few and the potential residual confounding could not be eliminated in the data from the regional two-centers. A larger number of subjects and multicenter studies are required to confirm the findings in the present study.

We confirmed that the platelet-related indicators PLR at birth, PNR at birth, and P at 2 w combined with gestational age, might be considered as a significant indicator of BPD. Furthermore, the occurrence rates were higher for adverse outcomes with PLR at birth $\ge$135.33, PNR at birth $\ge$129.12 and P at 2 w 285.8 in preterm infants.

BPD, bronchopulmonary dysplasia; PLR, platelet-to-lymphocyte ratio; PNR, platelet-to-neutrophil ratio; NICU, neonatal intensive care units; AUC, area under the curve; NLR, neutrophil-to-lymphocyte ratio; EBR, eosinophil-to-basophil ratio; NEC, necrotizing enterocolitis; PMA, post-menstrual age; EOS, early-onset sepsis; NRDS, newborn respiratory distress syndrome; ROP, retinopathy of prematurity; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis; EDTA, ethylene dinitro tetraacetic acid; CBC, complete blood count; N, neutrophil; L, lymphocyte; P, platelet; ROC, receiver operating characteristic; W, week; SE, standard error; 95% CI, 95% confidence interval; OR, odds ratio; SD, standard deviation.

The research data was uploaded as a file of supplementary materials. All data reported in this paper will also be shared by the corresponding author upon request.

XZ and YS designed the research study. CuiC and JZ performed the research. ChaC analyzed the data. SC and NL collected data and drafted the work. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The research protocol was approved by the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University (Ethic Approval Number: No.2020-064) and Yiwu Maternity and Children Hospital (Ethic Approval Number: No.000021), and all of the participants provided signed informed consent.

Not applicable.

This research was funded by Taizhou Social Development Science and Technology Project (21ywb136), Zhejiang Medical Health Science and Technology Project, China (2023KY1293), Zhejiang Medical Association clinical research fund project, China (2024ZYC-B51) and Wenzhou Science and Technology Research Project, China (Y2023003).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/j.ceog5110216.