Development and Validation of a Machine Learning-Based Risk Prediction Model for PICC-Related Bloodstream Infections in Premature Infants Using SHAP Interpretability

Development and Validation of a Machine Learning-Based Risk Prediction Model for PICC-Related Bloodstream Infections in Premature Infants Using SHAP Interpretability | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development and Validation of a Machine Learning-Based Risk Prediction Model for PICC-Related Bloodstream Infections in Premature Infants Using SHAP Interpretability Yongqin Guo, Yingying Dou, Wenxia Song, Lihong Wang, Li Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8162982/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Objective To develop a risk prediction model for peripherally inserted central catheter-related bloodstream infections (PICC-CRBSI) in preterm infants. Methods We conducted a prospective study of 490 preterm infants with PICC placement in a Chinese tertiary neonatal ICU (June 2023–December 2024). The data were split into training and validation sets at a 7:3. CRBSI was the primary outcome. Feature selection was performed using LASSO regression, the Boruta algorithm, and recursive feature elimination (RFE), and prediction models were constructed using four machine learning algorithms, including logistic regression (LR), decision tree (DT), random forest (RF), and light gradient boosting machine (LightGBM). The optimal model was selected through performance comparison. Results CRBSI occurred in 68 cases (13.9%), predominantly caused by Gram-negative bacteria (e.g., Klebsiella pneumoniae, Escherichia coli). The random forest model performed best among the four machine learning models, with an area under the receiver operating characteristic curve (AUC) of 0.984, precision of 0.857, recall of 0.900, specificity of 0.976, accuracy of 0.966, and F1-score of 0.878. The SHAP summary plot identified the top 7 most important features in order: C-reactive protein (CRP), white blood cell count (WBC), respiratory rate, blood oxygen saturation, number of punctures, age at catheterization, and birth weight. Conclusion Our study developed and interpreted a risk prediction model for PICC- CRBSI in preterm infants. It is expected to assist clinicians in timely risk stratification and targeted interventions, thereby reducing the incidence of CRBSI and improving the prognosis of preterm infants. Clinical Trial Number: Not applicable. Machine learning Preterm infants PICC Bloodstream infection Prediction model Figures Figure 1 Figure 2 Figure 3 Introduction A peripherally inserted central catheter (PICC) is a technique that involves puncturing peripheral superficial veins with a catheter whose tip is placed in the superior or inferior vena cava. PICC is widely used in newborns, especially preterm infants. It avoids the pain of repeated punctures, protects peripheral veins, and serves as a vital route for parenteral nutrition and drug delivery [ 1 , 2 ]. Because PICC is invasive, it can lead to catheter-related bloodstream infections (CRBSI) in preterm infants, who have underdeveloped immune systems. CRBSI raises the risk of death, extends hospital stays, and increases medical expenses. It also worsens both short- and long-term outcomes [ 3 , 4 ]. The incidence of PICC-CRBSI in preterm infants ranges from 7.25% to 13.78% [ 5 ]. Bloodstream infections in preterm infants often present atypically and are hard to distinguish from other neonatal diseases; misdiagnosis rates can reach 20–30% [ 6 ]. Therefore, early identification of high-risk infants for CRBSI is thus crucial in managing PICC in preterm infants. Currently, blood culture is the gold standard for the diagnosis of bloodstream infection, but it is slow and has limited sensitivity. Limited blood volume in preterm infants further lowers positive detection rates, complicating early diagnosis [ 7 ]. Thus, building an objective, high-performance risk prediction model for PICC-CRBSI in preterm infants is clinically important. Machine learning has been increasingly used in medicine [ 8 ], with advantages in disease prediction [ 9 – 11 ]. Most research on machine learning to predict bloodstream infection risk has focused on adults [ 12 , 13 ], with few studies involving neonates. In addition, machine learning models can be challenging to interpret, making it hard to understand how predictions are made [ 14 ]. To address this, we aim to build a risk prediction model for PICC-CRBSI in preterm infants using machine learning and interpret the optimal model with shapley additive explanations (SHAP) to reveal key risk factors and enhance clinical interpretability and practical value. Methods Study population The current study prospectively included preterm infants who received PICCs in the Neonatal Intensive Care Unit of a tertiary hospital in China between June 2023 and December 2024.Inclusion criteria: (1) Gestational age < 37 weeks, birth weight < 2000g; (2) Admitted to the hospital within 12 hours after birth; (3) Underwent PICC placement in this department; (4) No bloodstream infection before catheter placement; (5) Family members had been fully informed and signed the informed consent form. Exclusion criteria: (1) Those who had the PICC removed within 48 hours after insertion; (2) Those with incomplete clinical data or medical records. The study protocol was approved by the Institutional Review Board at the Medical Ethics Committee of Changzhi Maternal and Child Health Hospital (Committee Name:Ethics Committee of Changzhi Maternal and Child Health Hospital.Ethics Number: CZSFYLL2024.026) and complies with the ethical principles of the Declaration of Helsinki..We thoroughly explained the study's purpose, procedures, potential benefits and risks, as well as the right to withdraw at any time to all parents or guardians of eligible children. After ensuring their full understanding, we obtained written informed consent from each guardian. Sample size calculation The sample size was calculated using the formula n = [Z α ² (1-P)×P]/δ², with an allowable error (δ) of 0.05 and Z α of 1.96 [ 15 ]. With a reported PICC-CRBSI incidence rate of 13.78% in preterm infants and a 20% allowance for data loss, the calculation gave a required sample of 220. To strengthen the results, 490 preterm infants were included and then randomly split into a training set (n = 343) and a validation set (n = 147) in a 7:3 ratio. Assessment of variables After conducting a literature review and discussion, the candidate variables were identified as follows. The first category, general information, included gestational age, birth weight, gender, 5-minute Apgar score, mode of delivery, parity, neonatal respiratory distress syndrome, and the need for mechanical ventilation. The second category, catheter-related factors, included age at the time of catheter insertion, insertion site, number of punctures, need for repositioning, duration of catheter placement, frequency of catheter maintenance, episodes of occlusion, use of a connector wrap, and any displacement. The third category, vital signs, consisted of axillary temperature, heart rate, breathing rate, mean blood pressure, and blood oxygen level. The fourth category, laboratory tests, included red blood cell count (RBC), white blood cell count (WBC), platelet count (PLT), neutrophil percentage (NEUT), lymphocyte percentage (LYMPH), and C-reactive protein (CRP). CRBSI was diagnosed based on clinical practice guidelines of the infectious diseases society of America [ 16 ]. Laboratory data collected within 72 hours prior blood culture collection was defined as Specimen 1, while data collected on the day of blood culture collection was defined as Specimen 2. Comparing laboratory indicators collected prior to infection with those obtained at the onset of infection enables quantification of host immune dynamics and facilitates the identification of specific biomarkers. All Vital sign monitoring data are the average date collected within 24 hours on the day when the blood culture specimens were obtained. Data collection methods After systematic training, two researchers independently used standardized data collection forms to review and collect data. Disease-related and laboratory information were extracted from the electronic hospital medical records, while catheter-related factors and vital signs came from nursing electronic documentation. The two researchers checked and verified data together. Variables with over 20% missing values were excluded from analysis. The team discussed and resolved any disagreements during data collection or entry. Laboratory Indicator Monitoring Blood Culture Specimen monitoring: Two blood samples are required, one from a venous catheter and the other obtained via peripheral venipuncture. The volume of each blood draw should be at least 2 milliliters. Blood cultures are performed using a bilateral method, with HP/PYP blood culture bottles, and the blood specimens are incubated. Bacterial species identification is conducted using an automated microbial analysis system manufactured by bioMérieux (France). NICU physicians and hospital infection control department doctors collaborate to exclude contaminated specimens and determine whether the pathogenic microorganisms in the blood specimen are the cause of CRBSI. Complete blood count (CBC) samples are collected simultaneously with blood cultures. About 2 milliliters of venous blood is collected in EDTA-K2 anticoagulant tubes, which prevent clotting, for testing. CBCs are analyzed using an automated hematology machine that measures blood components through both impedance (electrical resistance) and laser scattering methods. The CRP test was performed using the nephelometric method with the automatic specific protein analyzer BN-11. Data Preprocessing First, records with over 20% missing data are removed. If less than 20% of values are missing, gaps are filled using the mean for the same gestational age. The type of each variable is checked: gender and twin status are set as categories, while vital signs and lab results are treated as continuous variables. To prevent differences in scale, all numerical features are normalized to between 0 and 1. This keeps features on the same scale and stops any single feature from dominating the model. After cleaning, 490 cases remained for analysis. Statistical Methods Normally distributed quantitative data are shown as mean ± standard deviation and compared between groups with the independent samples t-test. Non-normally distributed quantitative data are shown as median (P25, P75) and compared using the rank sum test. Qualitative data are summarized as frequency and percentage, and compared using the chi-square test or Fisher’s exact test. Analyses were done in SPSS 24.0. Results were considered statistically significant at P < 0.05. Machine learning algorithms All samples were randomly shuffled using R 4.2.2 software and then allocated to the training set and validation set in a 7:3 ratio. Three feature selection methods, namely LASSO regression, Boruta algorithm, and recursive feature elimination (RFE), were employed. Four models were developed: logistic regression (LR), decision tree (DT), random forest (RF), and light gradient boosting machine (LightGBM). Each model was trained and tested with ten-fold cross-validation. Hyperparameters were adjusted using grid search. Finally, the performance of the four machine learning models was evaluated and compared using AUC, precision, recall, specificity, accuracy, and F1 score (F1-Score). Results A total of 490 preterm infants meeting the study criteria were enrolled. Of these, 68 developed CRBSI (incidence rate 13.9%). A total of 76 pathogenic bacteria were isolated from 68 positive blood cultures, with Gram-negative bacteria being the majority (56 strains, 73.68%). Klebsiella pneumoniae was found most often (31 strains, 40.79%), followed by Escherichia coli (18 strains, 23.68%), Pseudomonas aeruginosa (4 strains, 5.26%), and three other Gram-negative bacteria (3.95%). The 20 detected Gram-positive bacteria (26.32%) included Staphylococcus epidermidis (10 strains, 13.16%), Staphylococcus aureus (6 strains, 7.89%), Group B Streptococcus (2 strains, 2.63%), and Enterococcus spp. (2 strains, 2.63%). In the data of 490 children, univariate analysis showed that gestational age, weight, 5-minute Apgar score, mechanical ventilation, catheterization age, puncture times, catheter dwell time, catheter maintenance frequency, axillary temperature, respiratory rate, blood oxygen saturation, RBC, WBC, NEUT, and CRP had statistically significant differences (P < 0.05). The results are shown in Table 1 . Table 1 General information of premature infants Variables Total (n = 490) Non-CRBSI (n = 422) CRBSI (n = 68) P Statistic Fetal age,[n (%),(W)] 0.011 9.012 34 ≤ Fetal age<37 93 (18.98) 85 (20.14) 8 (11.76) 28 ≤ Fetal age<34 328 (66.94) 285 (67.54) 43 (63.24) <28 69 (14.08) 52 (12.32) 17 (25.00) Birth weight,[n(%),(g)] < 0.001 18.306 1500 ≤ Birth weight < 2000 99 (20.20) 92 (21.80) 7 (10.29) 1000 ≤ Birth weight < 1500 314 (64.08) 275 (65.17) 39 (57.35) 7 433 (88.37) 379 (89.81) 54 (79.41) ≤ 7 57 (11.63) 43 (10.19) 14 (20.59) Mode of delivery,n(%) 0.55 0.356 Natural childbirth 229 (46.73) 200 (47.39) 29 (42.65) Cesarean section 261 (53.27) 222 (52.61) 39 (57.35) Number of fetuses,n(%) 0.415 0.664 Single fetus 396 (80.82) 344 (81.52) 52 (76.47) Twins 94 (19.18) 78 (18.48) 16 (23.53) NRDS,n(%) 0.392 0.732 No 185 (37.76) 163 (38.63) 22 (32.35) Yes 305 (62.24) 259 (61.37) 46 (67.65) Mechanical ventilation,n(%) 0.003 9.023 No 405 (82.65) 358 (84.83) 47 (69.12) Yes 85 (17.35) 64 (15.17) 21 (30.88) Age in days at catheter placement,[n(%),(d)] 0.003 8.615 ≤ 7 198 (40.41) 159 (37.68) 39 (57.35) >7 292 (59.59) 263 (62.32) 29 (42.65) Catheter insertion site, n(%) 0.081 5.037 Head 76 (15.51) 68 (16.11) 8 (11.76) Lower limbs 133 (27.14) 107 (25.36) 26 (38.24) upper limbs 281 (57.35) 247 (58.53) 34 (50.00) Number of punctures,[n(%),(time)] 2 116 (23.67) 82 (19.43) 34 (50.00) Readjust the catheter, n(%) 0.428 0.627 No 256 (52.24) 224 (53.08) 32 (47.06) Yes 234 (47.76) 198 (46.92) 36 (52.94) The indwelling time of PICC,[n(%),(d)] 0.034 4.502 ≤ 14 235 (47.96) 211 (50.00) 24 (35.29) >14 255 (52.04) 211 (50.00) 44 (64.71) Frequency of dressing change, n(%) 0.02 5.371 times per seven days 428 (87.35) 375 (88.86) 53 (77.94) <times per seven days 62 (12.65) 47 (11.14) 15 (22.06) Catheter blockage, n(%) 0.617 0.25 No 444 (90.61) 384 (91.00) 60 (88.24) Yes 46 (9.39) 38 (9.00) 8 (11.76) Catheter connector wrapping, n(%) 0.291 1.114 No 57 (11.63) 46 (10.90) 11 (16.18) Yes 433 (88.37) 376 (89.10) 57 (83.82) Catheter displacement, n(%) 0.873 0.026 No 332 (67.76) 287 (68.01) 45 (66.18) Yes 158 (32.24) 135 (31.99) 23 (33.82) Axillary Temperature, [ M ( P 25 , P 75 ),℃] 36.8 (36.7, 36.9) 36.8 (36.7, 36.9) 36.8 (36.7, 37.12) 0.016 11768.5 Heart Rate,[ M ( P 25 , P 75 ),bpm] 146 (142, 148) 146 (142, 148) 145 (139.5, 148) 0.176 15806 Respiratory Rate ,[ M ( P 25 , P 75 ),bpm] 44 (42, 44) 44 (42, 44) 42 (40, 44) 0.011 17000.5 Mean Blood Pressure,[ M ( P 25 , P 75 ),mmHg] 39 (36, 41) 39 (36, 41) 38 (36, 40.25) 0.052 16440.5 Blood oxygen saturation, M ( P 25 , P 75 ) 93 (92, 94) 93 (92, 94) 92 (91, 94) < 0.001 17941 RBC,[ M ( P 25 , P 75 ),10 12 /L] 1 4.2 (4.1, 4.3) 4.2 (4.1, 4.3) 4.3 (4.1, 4.6) 0.084 12485 2 4.3 (4.18, 4.6) 4.3 (4.2, 4.6) 4.25 (4.1, 4.36) 0.002 17744 WBC,[n(%),10 9 /L] 1 0.074* — 25 4 (0.82) 2 (0.47) 2 (2.94) 2 < 0.001* — 25 21 (4.29) 9 (2.13) 12 (17.65) PLT,[ M ( P 25 , P 75 ),10 9 /L] 1 289 (246, 316) 289 (246, 316) 284 (246.75, 316) 0.881 14186 2 258 (214, 306) 256 (214, 306) 265 (216, 308) 0.179 12894 NEUT0,[ M ( P 25 , P 75 ),%] 1 47.4 (44.82, 49.5) 47.4 (44.97, 49.5) 48.1 (40.5, 52.3) 0.995 14355.5 2 46.8 (41.3, 48.6) 46.5 (41.2, 48.6) 47.6 (44.77, 50.2) 10mg/L),n(%) 1 0.212 1.555 No 448 (91.43) 389 (92.18) 59 (86.76) Yes 42 (8.57) 33 (7.82) 9 (13.24) 2 < 0.001 151.553 No 384 (78.37) 370 (87.68) 14 (20.59) Yes 106 (21.63) 52 (12.32) 54 (79.41) Note: *Fisher's exact test was used. Three methods—Lasso regression, Boruta algorithm, and RFE—were used to screen baseline variables. The intersection of their results identified seven predictive variables, including birth weight, age at catheterization, number of punctures, respiratory rate, oxygen saturation, WBC, and CRP. Four machine learning algorithms, including LR, DT, RF, and LightGBM, were used to construct models and perform internal validation. The results are shown in Fig. 1 . All four models were evaluated. The random forest (RF) model performed best, with an area under the ROC curve of 0.984 (0.965–1.000), precision of 0.857, recall of 0.900, specificity of 0.976, accuracy of 0.966, and an F1 score of 0.878. The detailed information was shown in Table 2 . Table 2 Performance evaluation of four machine learning algorithms Model AUC(95%Cl) Precision Recall Specificity Accuracy F1 LR 0.902(0.802-1.000) 0.552 0.800 0.898 0.884 0.653 DT 0.843(0.738–0.948) 0.550 0.733 0.932 0.912 0.629 RF 0.984(0.965-1.000) 0.857 0.900 0.976 0.966 0.878 LightGBM 0.614(0.493–0.734) 0.231 0.300 0.843 0.769 0.261 Figure 2 illustrates the sequence of variable importance ranking in the RF model, from highest to lowest: CRP, WBC, respiratory rate, blood oxygen saturation, number of punctures, age at catheterization, and birth weight. Figure 3 illustrates the SHAP summary plot, which first displays the feature importance of each variable. In this plot, each point represents a sample. The horizontal axis shows the magnitude of each feature's influence on the prediction. The greater a point's distance from the central line (zero point), the more significantly that feature influences the model's prediction. Discussion PICC catheters provide long-term intravenous access in preterm infants but are associated with substantial CRBSI risks due to immature immune systems. Globally, neonatal CRBSI rates range from 2% to 30%, with higher rates in developing countries[ 17 , 18 ]. In our study, the PICC-CRBSI rate was 13.9% (68 out of 490), which is similar to the rate reported by Ren et al. [ 5 ]. Notably, most infections in this study were caused by Gram-negative bacteria, primarily Klebsiella pneumoniae and Escherichia coli, whereas data from developed countries indicate that Gram-positive bacteria are the primary cause[ 19 , 20 ]. These differences may reflect variations in infection control and clinical practices between regions. In clinical practice, predictive models necessitate both high performance and clinical interpretability [ 21 ]. In this study, our SHAP analysis quantified seven key predictors of PICC-CRBSI in preterm infants, ranked by descending importance : CRP, WBC, respiratory rate, blood oxygen saturation, number of punctures, catheterization age ≤ 7 days, and birth weight. Notably, CRP and WBC emerged as dominant factors, consistently occupying the highest positions in both feature importance rankings and SHAP summary plots(Figs. 2 and 3 ). CRP demonstrated exceptional discriminatory power, consistent with its established role as a rapid-response biomarker for neonatal sepsis[ 22 ]. Mechanistically, CRP amplifies host defenses by inducing cytokine cascades, activating complement pathways, and modulating phagocytic clearance—processes critical during bloodstream infections[ 23 ]. Although CRP lacks specificity for CRBSI alone, its elevated levels correlate strongly with culture-positive sepsis (versus culture-negative cases) [ 24 ], supporting its utility as a screening triage tool. In our cohort, abnormal CRP levels (> 10 mg/L) were observed in 79% of CRBSI infants versus 12% of non-CRBSI ( P < 0.001), reinforcing its diagnostic relevance. The SHAP value for CRP (+ 0.146) further confirmed its linear, dose-dependent association with infection risk ( P < 0.01). WBC abnormalities (leukocytosis or leukopenia) served as complementary predictors, reflecting dynamic immune dysregulation during sepsis [ 25 ]. The synergy between CRP and WBC enhanced early CRBSI detection (AUC: 0.91 vs. 0.83 for CRP alone), aligning with evidence that multimarker strategies improve diagnostic accuracy in neonatal infections [ 26 , 27 ]. These findings advocate for immediate clinical implementation of CRP-WBC monitoring protocols in neonatal intensive care units. High CRP levels was associated with an elevated risk of PICC-CRBSI who require escalated surveillance (including repeat cultures and consideration of empiric therapy). Particularly for high-risk subpopulations—notably those with catheterization within the first week of life or concurrent respiratory distress—the model supports twice per week CRP-WBC monitoring to enable early intervention. This strategy integrates contemporary sepsis detection principles with preterm-specific risk mitigation, offering two key advantages: (1) earlier identification of true CRBSI cases through biomarker-guided vigilance, and (2) decreased antibiotic overuse in stable infants by avoiding treatment based solely on suspicion. Our data suggest this dual benefit reflects the precision achievable through dynamic CRP-WBC monitoring. Following comprehensive feature selection and data preprocessing, our comparative analysis of four machine learning models revealed that the Random Forest (RF) algorithm achieved optimal performance in predicting PICC-CRBSI risk among preterm infants, demonstrating exceptional discriminative capacity (accuracy: 0.966, specificity: 0.976, AUC: 0.984). This superiority aligns with established literature documenting RF's ability to manage high-dimensional clinical data through ensemble-based decision tree aggregation, effectively balancing model complexity with generalization capacity [ 28 , 29 ]. More importantly, while conventional machine learning approaches often function as "black boxes," we addressed this translational limitation through systematic SHAP analysis. This advanced interpretability framework not only quantified the relative contribution of each risk factor but also revealed clinically meaningful nonlinear relationships. The incidence of PICC-related CRBSI directly reflects departmental infection prevention standards, with our results revealing decreased respiratory rate and blood oxygen saturation as key risk predictors. Bloodstream infections trigger systemic inflammation wherein mediators breach the underdeveloped blood-brain barrier of preterm infants, potentially disrupting brainstem respiratory centers and manifesting as apnea or desaturation[ 30 ]. Consequently, continuous monitoring of these vital signs is imperative for early infection detection and intervention. Furthermore, catheterization ≤ 7 days postnatally significantly increased risk (57% of CRBSI cases vs. catheterization > 7 days [43%]), attributable to concurrent deficits in innate and adaptive immunity alongside immature skin barrier function during this period [ 31 – 33 ]. To mitigate this risk, stricter catheterization protocols—real-time biomarker monitoring, aseptic technique reinforcement, and simulation-based staff training—must be implemented to reduce multiple punctures. Finally, birth weight remains a critical non-modifiable determinant; its inverse relationship with CRBSI risk (9.0% increase per 100g weight decrement [ 34 , 35 ]). A dedicated PICC management team should be set up for these infants. Improving the management of vascular access can help reduce catheter-related bloodstream infections. Healthcare teams require early identification of high-risk preterm infants with PICC to initiate preemptive interventions against CRBSI. We addressed this need by developing four machine learning models using routine clinical parameters, among which the RF algorithm demonstrated superior predictive performance (AUC: 0.984) for early risk stratification. Nevertheless, this study has methodological limitations. First, the model was validated only internally; external verification across multicenter cohorts with varying demographics and clinical practices is essential to confirm generalizability. Second, real-world implementation requires integrating this RF model into electronic health records (EHRs) to automate data extraction and generate real-time alerts for high-risk infants—a critical step toward enabling closed-loop, precision management of PICC-CRBSI that could significantly reduce infection rates. In summary, our study established a machine learning-based risk prediction tool for PICC-CRBSI in preterm infants, demonstrating the RF algorithm's superior predictive capability. The model identifies seven critical predictors stratified as inflammatory biomarkers, physiological indicators, demographic characteristics, and clinical procedural factors. The validated algorithm enables early identification of high-risk infants, facilitating targeted interventions. Prioritizing enhanced monitoring, proactive catheter care protocols, and preemptive antimicrobial stewardship for these patients may significantly reduce bloodstream infections burden while improving PICC safety outcomes. Declarations Funding Basic Research Project of Changzhi Municipal Bureau of Science and Technology Project No.:JC202439 Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Author Contribution YQG: Methodology, writing-original draft.YYD: Data clean and analysis，Model validation.WXS: Thesis guidanceLHW: Project administration.L.W: Supervision. Data availability statement The data used to support the findings of this study are not publicly available due to [reason], but can be obtained from the corresponding author upon reasonable request. Ethics statement This study was reviewed and approved by the Medical Ethics Committee of Changzhi Maternal and Child Health Hospital (Ethics Approval No: CZSFYLL2024.026) and complies with the ethical principles of the Declaration of Helsinki. As a prospective study, Written informed consent was obtained from all legal guardians prior to the participants' inclusion in the study,all personal information was anonymized using coded identifiers to protect participant privacy. References Teibel H, Hood K, Manasco K, Bhatia J. Antibiotic Administration Prior to Central Venous Catheter Removal in Neonates. J Pharm Pract. 2021;34:894–900. Ponticelli E, Clari M, Volpes M, et al. Complete Blood Count Collected Via Venipuncture Versus Peripherally Inserted Central Catheter in Hematological Patients: A Comparison of 2 Methods. Cancer Nurs. 2022;45:E36–42. van Tonder DJ, Keough N, van Niekerk ML, van Schoor AN. The position of the common facial vein in neonates: An alternate route for central venous catheter placement. Clin Anat. 2021;34:644–50. Brown R, Burke D. The hidden cost of catheter related blood stream infections in patients on parenteral nutrition. Clin Nutr ESPEN. 2020;36:146–9. Ren XL, Li HL, Liu J, et al. Ultrasound to Localize the Peripherally Inserted Central Catheter Tip Position in Newborn Infants. Am J Perinatol. 2021;38:122–5. Klinger G, Bromiker R, Zaslavsky-Paltiel I et al. Late-Onset Sepsis in Very Low Birth Weight Infants. Pediatrics 2023; 152. Bierlaire S, Danhaive O, Carkeek K, Piersigilli F. How to minimize central line-associated bloodstream infections in a neonatal intensive care unit: a quality improvement intervention based on a retrospective analysis and the adoption of an evidence-based bundle. Eur J Pediatr. 2021;180:449–60. Zhou X, Chen L, Liu H-X. Applications of Machine Learning Models to Predict and Prevent Obesity: A Mini-Review. Front Nutr 2022; 9. Triantafyllidis A, Polychronidou E, Alexiadis A, et al. Computerized decision support and machine learning applications for the prevention and treatment of childhood obesity: A systematic review of the literature. Artif Intell Med. 2020;104:101844. Fernandes M, Mendes R, Vieira SM, et al. Predicting Intensive Care Unit admission among patients presenting to the emergency department using machine learning and natural language processing. PLoS ONE. 2020;15:e0229331. Li XH, Yang XL, Dong BB, Liu Q. Predicting 28-day all-cause mortality in patients admitted to intensive care units with pre-existing chronic heart failure using the stress hyperglycemia ratio: a machine learning-driven retrospective cohort analysis. Cardiovasc Diabetol. 2025;24:10. Rahmani K, Garikipati A, Barnes G, et al. Early prediction of central line associated bloodstream infection using machine learning. Am J Infect Control. 2022;50:440–5. Sakai H, Iwata M, Terasawa T. External validation of the Michigan PICC catheter-associated bloodstream infections score (MPC score) for predicting the risk of peripherally inserted central catheter-associated bloodstream infections: A single-center study in Japan. Infect Control Hosp Epidemiol. 2023;44:480–3. The Lancet Respiratory M. Opening the black box of machine learning. Lancet Respir Med. 2018;6:801. Pourhoseingholi MA, Vahedi M, Rahimzadeh M. Sample size calculation in medical studies. Gastroenterol Hepatol Bed Bench. 2013;6:14–7. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1–45. Gordon A, Greenhalgh M, McGuire W. Early planned removal versus expectant management of peripherally inserted central catheters to prevent infection in newborn infants. Cochrane Database of Systematic Reviews. 2018; 2018. Rosenthal VD, Bat-Erdene I, Gupta D, et al. International Nosocomial Infection Control Consortium (INICC) report, data summary of 45 countries for 2012–2017: Device-associated module. Am J Infect Control. 2020;48:423–32. Lake JG, Weiner LM, Milstone AM, et al. Pathogen Distribution and Antimicrobial Resistance Among Pediatric Healthcare-Associated Infections Reported to the National Healthcare Safety Network, 2011–2014. Infect Control Hosp Epidemiol. 2017;39:1–11. Jansen SJ, Broer SDL, Hemels MAC, et al. Central-line-associated bloodstream infection burden among Dutch neonatal intensive care units. J Hosp Infect. 2024;144:20–7. Wang J, Chen H, Wang H et al. A Risk Prediction Model for Physical Restraints Among Older Chinese Adults in Long-term Care Facilities: Machine Learning Study. J Med Internet Res 2023; 25. Sproston NR, Ashworth JJ. Role of C-Reactive Protein at Sites of Inflammation and Infection. Front Immunol 2018; 9. Poggi C, Lucenteforte E, Petri D, et al. Presepsin for the Diagnosis of Neonatal Early-Onset Sepsis: A Systematic Review and Meta-analysis. JAMA Pediatr. 2022;176:750–8. Povoa P, Garvik OS, Vinholt PJ, et al. C-reactive protein and albumin kinetics after antibiotic therapy in community-acquired bloodstream infection. Int J Infect Dis. 2020;95:50–8. Gilfillan M, Bhandari V. Biomarkers for the diagnosis of neonatal sepsis and necrotizing enterocolitis: Clinical practice guidelines. Early Hum Dev. 2017;105:25–33. Maestraggi Q, Lebas B, Clere-Jehl R et al. Skeletal Muscle and Lymphocyte Mitochondrial Dysfunctions in Septic Shock Trigger ICU-Acquired Weakness and Sepsis-Induced Immunoparalysis. Biomed Res Int. 2017; 2017: 7897325. Li Q, Gong X. Clinical significance of the detection of procalcitonin and C-reactive protein in the intensive care unit. Exp Ther Med. 2018;15:4265–70. Cruz AF, Herrmann J, Ramcharran H, et al. Sustained vs. Intratidal Recruitment in the Injured Lung During Airway Pressure Release Ventilation: A Computational Modeling Perspective. Mil Med. 2023;188:141–8. Jin Y, Lan A, Dai Y, et al. Development and testing of a random forest-based machine learning model for predicting events among breast cancer patients with a poor response to neoadjuvant chemotherapy. Eur J Med Res. 2023;28:394. Al-Matary A, Abozaid S, Al Suliman M et al. Correlation between Bronchopulmonary Dysplasia and Cerebral Palsy in Children: A Comprehensive Analysis Using the National Inpatient Sample Dataset. Child (Basel) 2024; 11. Bliss JM, Wynn JL. Editorial: The Neonatal Immune System: A Unique Host-Microbial Interface. Front Pediatr. 2017;5:274. Slater A, Straney L, Alexander J, et al. The Effect of Imputation of PaO2/FIO2 From SpO2/FIO2 on the Performance of the Pediatric Index of Mortality 3. Pediatr Crit Care Med. 2020;21:520–5. Emeriaud G, Lopez-Fernandez YM, Iyer NP, et al. Executive Summary of the Second International Guidelines for the Diagnosis and Management of Pediatric Acute Respiratory Distress Syndrome (PALICC-2). Pediatr Crit Care Med. 2023;24:143–68. Papoff P, Caresta E, Luciani S, et al. The starting rate for high-flow nasal cannula oxygen therapy in infants with bronchiolitis: Is clinical judgment enough? Pediatr Pulmonol. 2021;56:2611–20. Yildizdas D, Yontem A, Iplik G, et al. Predicting nasal high-flow therapy failure by pediatric respiratory rate-oxygenation index and pediatric respiratory rate-oxygenation index variation in children. Eur J Pediatr. 2021;180:1099–106. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 24 Dec, 2025 Editor invited by journal 02 Dec, 2025 Editor assigned by journal 28 Nov, 2025 Submission checks completed at journal 28 Nov, 2025 First submitted to journal 20 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8162982","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":565106027,"identity":"7e951d66-24f8-4bf6-85a5-f22d2f7364a4","order_by":0,"name":"Yongqin Guo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIie3RsQrCMBCA4SuBm6JdI0p9hUihz9OgdFLQrUOHQKUOIq76Ft3UUQqZ4u5Y6QvopuBgd8XUzSHfnJ/kLgCW9YfQTYvq8Uyo26+OZRgn5qTNVOQ7UnkdGQ15qZU58WDMu44kPpfjoHOZkwYPA8392R5FDiqKhURwF8vwe0KyabXVPbFzUnUW+x4wfcoNtxQ5p4jikGJ0FhqBs4kpCTmjSESuaDAVGWmSjHi3ldXjaxpAs6Re8mCj6yVvcMhCrahxlv46LcprXH8lI8fbPU48d7H6nryhvx23LMuyPnoBdqdLgJ6PuxEAAAAASUVORK5CYII=","orcid":"","institution":"Changzhi Maternal and Child Health Hospital","correspondingAuthor":true,"prefix":"","firstName":"Yongqin","middleName":"","lastName":"Guo","suffix":""},{"id":565106028,"identity":"4961541d-7d97-4cf9-8e7b-43eb0393f919","order_by":1,"name":"Yingying Dou","email":"","orcid":"","institution":"Changzhi Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yingying","middleName":"","lastName":"Dou","suffix":""},{"id":565106029,"identity":"2c4fd146-f468-449c-a700-a84904decc0b","order_by":2,"name":"Wenxia Song","email":"","orcid":"","institution":"Changzhi Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wenxia","middleName":"","lastName":"Song","suffix":""},{"id":565106030,"identity":"d9173c37-5673-474e-a344-b68797ed199e","order_by":3,"name":"Lihong Wang","email":"","orcid":"","institution":"Changzhi Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Lihong","middleName":"","lastName":"Wang","suffix":""},{"id":565106031,"identity":"c23ac728-48be-4c2e-82fb-7cf6e7c294fe","order_by":4,"name":"Li Wang","email":"","orcid":"","institution":"Changzhi Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-11-20 10:14:41","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8162982/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8162982/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99219733,"identity":"c58dad52-c4e0-480e-9af9-ef9a6c7d9880","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":241284,"visible":true,"origin":"","legend":"","description":"","filename":"Figure.docx","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/b80c4d547c0bd7db50e63121.docx"},{"id":99219744,"identity":"af6cacd5-82a4-4156-8210-b51587f0a61c","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96155,"visible":true,"origin":"","legend":"","description":"","filename":"DevelopmentandValidationofaMachineLearningBasedRiskPredictionModelforPICCRelatedBloodstreamInfectionsinPrematureInfantsUsingSHAPInterpretability.docx","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/c8b1f3e12010ffb26f5b1fff.docx"},{"id":99219737,"identity":"db47949d-690a-48b5-a758-0d321a237ba7","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28156,"visible":true,"origin":"","legend":"","description":"","filename":"Table.docx","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/203e6bac4f61d03b3e1fd2e7.docx"},{"id":99219736,"identity":"3f5fd035-9763-4aaa-9d51-1a57d6854863","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"json","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6590,"visible":true,"origin":"","legend":"","description":"","filename":"4d49d1b9734c4a48a2d3bf6563adef46.json","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/3b761c254c63a0730dd4383f.json"},{"id":99219739,"identity":"77d850ff-dbe0-47ce-bc92-ff19228d8ca5","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":478003,"visible":true,"origin":"","legend":"","description":"","filename":"Ethicalreview.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/5793c17ccfe3bdebc0135a44.pdf"},{"id":99219741,"identity":"ca126a89-09df-4acb-96c1-8ff8a0fd8827","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117344,"visible":true,"origin":"","legend":"","description":"","filename":"4d49d1b9734c4a48a2d3bf6563adef461enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/c64fbcab48ae118f2a5eb483.xml"},{"id":99219727,"identity":"9fb63c6b-a782-43f2-8e74-e0a9eb3756f5","added_by":"auto","created_at":"2025-12-30 09:31:08","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31598,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/7873b5848b40f1f1fb3daffd.png"},{"id":99219734,"identity":"282bc9a5-5d04-44aa-8bae-3db590a7f40d","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16900,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/42c59d2faa4a9d09c64a1e79.png"},{"id":99219740,"identity":"e3e79cec-1af3-4259-b246-d728d6a32dd0","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22078,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/e0d73de7f4dd2da4fda741b4.png"},{"id":99219728,"identity":"4ab481b4-0a05-4721-be20-aa62d3956f36","added_by":"auto","created_at":"2025-12-30 09:31:08","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114151,"visible":true,"origin":"","legend":"","description":"","filename":"4d49d1b9734c4a48a2d3bf6563adef461structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/3e13560dc42ad4b2b99f52af.xml"},{"id":99219742,"identity":"e4284bc1-c2da-4e77-b815-05e20110b972","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127571,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/ef220611ea029200e1ed2454.html"},{"id":99318293,"identity":"e61d7be9-f492-4cd2-8c48-5195124d5172","added_by":"auto","created_at":"2025-12-31 16:32:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":119161,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve of internal validation\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/41e841a382afccb399e3d452.png"},{"id":99219735,"identity":"f5ea1de3-efec-4197-869b-93383f8da15d","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":48841,"visible":true,"origin":"","legend":"\u003cp\u003eVariable importance ranking plot of the RF model\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/23986cd7b6fcc63f1a882d88.png"},{"id":99219738,"identity":"a8765bc4-3c47-48bf-98bb-5cf4adaa528e","added_by":"auto","created_at":"2025-12-30 09:31:09","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":60194,"visible":true,"origin":"","legend":"\u003cp\u003eSHAP summary plot of variables in the RF model\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/b779433f0affe3776faf4f9f.png"},{"id":99323723,"identity":"4aefbaa0-748a-4e1e-9f2e-345af7d77721","added_by":"auto","created_at":"2025-12-31 16:46:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1166743,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8162982/v1/c6446cd9-dfc3-40ed-885d-cd41387126ba.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and Validation of a Machine Learning-Based Risk Prediction Model for PICC-Related Bloodstream Infections in Premature Infants Using SHAP Interpretability","fulltext":[{"header":"Introduction","content":"\u003cp\u003eA peripherally inserted central catheter (PICC) is a technique that involves puncturing peripheral superficial veins with a catheter whose tip is placed in the superior or inferior vena cava. PICC is widely used in newborns, especially preterm infants. It avoids the pain of repeated punctures, protects peripheral veins, and serves as a vital route for parenteral nutrition and drug delivery [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Because PICC is invasive, it can lead to catheter-related bloodstream infections (CRBSI) in preterm infants, who have underdeveloped immune systems. CRBSI raises the risk of death, extends hospital stays, and increases medical expenses. It also worsens both short- and long-term outcomes [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. The incidence of PICC-CRBSI in preterm infants ranges from 7.25% to 13.78% [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Bloodstream infections in preterm infants often present atypically and are hard to distinguish from other neonatal diseases; misdiagnosis rates can reach 20\u0026ndash;30% [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Therefore, early identification of high-risk infants for CRBSI is thus crucial in managing PICC in preterm infants.\u003c/p\u003e \u003cp\u003eCurrently, blood culture is the gold standard for the diagnosis of bloodstream infection, but it is slow and has limited sensitivity. Limited blood volume in preterm infants further lowers positive detection rates, complicating early diagnosis [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Thus, building an objective, high-performance risk prediction model for PICC-CRBSI in preterm infants is clinically important. Machine learning has been increasingly used in medicine [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], with advantages in disease prediction [\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Most research on machine learning to predict bloodstream infection risk has focused on adults [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], with few studies involving neonates. In addition, machine learning models can be challenging to interpret, making it hard to understand how predictions are made [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTo address this, we aim to build a risk prediction model for PICC-CRBSI in preterm infants using machine learning and interpret the optimal model with shapley additive explanations (SHAP) to reveal key risk factors and enhance clinical interpretability and practical value.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy population\u003c/h2\u003e \u003cp\u003eThe current study prospectively included preterm infants who received PICCs in the Neonatal Intensive Care Unit of a tertiary hospital in China between June 2023 and December 2024.Inclusion criteria: (1) Gestational age\u0026thinsp;\u0026lt;\u0026thinsp;37 weeks, birth weight\u0026thinsp;\u0026lt;\u0026thinsp;2000g; (2) Admitted to the hospital within 12 hours after birth; (3) Underwent PICC placement in this department; (4) No bloodstream infection before catheter placement; (5) Family members had been fully informed and signed the informed consent form. Exclusion criteria: (1) Those who had the PICC removed within 48 hours after insertion; (2) Those with incomplete clinical data or medical records. The study protocol was approved by the Institutional Review Board at the Medical Ethics Committee of Changzhi Maternal and Child Health Hospital (Committee Name:Ethics Committee of Changzhi Maternal and Child Health Hospital.Ethics Number: CZSFYLL2024.026) and complies with the ethical principles of the Declaration of Helsinki..We thoroughly explained the study's purpose, procedures, potential benefits and risks, as well as the right to withdraw at any time to all parents or guardians of eligible children. After ensuring their full understanding, we obtained written informed consent from each guardian.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSample size calculation\u003c/h3\u003e\n\u003cp\u003eThe sample size was calculated using the formula n = [Z\u003csub\u003eα\u003c/sub\u003e\u0026sup2; (1-P)\u0026times;P]/δ\u0026sup2;, with an allowable error (δ) of 0.05 and Z\u003csub\u003eα\u003c/sub\u003e of 1.96 [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. With a reported PICC-CRBSI incidence rate of 13.78% in preterm infants and a 20% allowance for data loss, the calculation gave a required sample of 220. To strengthen the results, 490 preterm infants were included and then randomly split into a training set (n\u0026thinsp;=\u0026thinsp;343) and a validation set (n\u0026thinsp;=\u0026thinsp;147) in a 7:3 ratio.\u003c/p\u003e\n\u003ch3\u003eAssessment of variables\u003c/h3\u003e\n\u003cp\u003eAfter conducting a literature review and discussion, the candidate variables were identified as follows. The first category, general information, included gestational age, birth weight, gender, 5-minute Apgar score, mode of delivery, parity, neonatal respiratory distress syndrome, and the need for mechanical ventilation. The second category, catheter-related factors, included age at the time of catheter insertion, insertion site, number of punctures, need for repositioning, duration of catheter placement, frequency of catheter maintenance, episodes of occlusion, use of a connector wrap, and any displacement. The third category, vital signs, consisted of axillary temperature, heart rate, breathing rate, mean blood pressure, and blood oxygen level. The fourth category, laboratory tests, included red blood cell count (RBC), white blood cell count (WBC), platelet count (PLT), neutrophil percentage (NEUT), lymphocyte percentage (LYMPH), and C-reactive protein (CRP). CRBSI was diagnosed based on clinical practice guidelines of the infectious diseases society of America [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Laboratory data collected within 72 hours prior blood culture collection was defined as Specimen 1, while data collected on the day of blood culture collection was defined as Specimen 2. Comparing laboratory indicators collected prior to infection with those obtained at the onset of infection enables quantification of host immune dynamics and facilitates the identification of specific biomarkers. All Vital sign monitoring data are the average date collected within 24 hours on the day when the blood culture specimens were obtained.\u003c/p\u003e\n\u003ch3\u003eData collection methods\u003c/h3\u003e\n\u003cp\u003eAfter systematic training, two researchers independently used standardized data collection forms to review and collect data. Disease-related and laboratory information were extracted from the electronic hospital medical records, while catheter-related factors and vital signs came from nursing electronic documentation. The two researchers checked and verified data together. Variables with over 20% missing values were excluded from analysis. The team discussed and resolved any disagreements during data collection or entry.\u003c/p\u003e\n\u003ch3\u003eLaboratory Indicator Monitoring\u003c/h3\u003e\n\u003cp\u003eBlood Culture Specimen monitoring: Two blood samples are required, one from a venous catheter and the other obtained via peripheral venipuncture. The volume of each blood draw should be at least 2 milliliters. Blood cultures are performed using a bilateral method, with HP/PYP blood culture bottles, and the blood specimens are incubated. Bacterial species identification is conducted using an automated microbial analysis system manufactured by bioM\u0026eacute;rieux (France). NICU physicians and hospital infection control department doctors collaborate to exclude contaminated specimens and determine whether the pathogenic microorganisms in the blood specimen are the cause of CRBSI.\u003c/p\u003e \u003cp\u003eComplete blood count (CBC) samples are collected simultaneously with blood cultures. About 2 milliliters of venous blood is collected in EDTA-K2 anticoagulant tubes, which prevent clotting, for testing. CBCs are analyzed using an automated hematology machine that measures blood components through both impedance (electrical resistance) and laser scattering methods. The CRP test was performed using the nephelometric method with the automatic specific protein analyzer BN-11.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eData Preprocessing\u003c/h2\u003e \u003cp\u003eFirst, records with over 20% missing data are removed. If less than 20% of values are missing, gaps are filled using the mean for the same gestational age. The type of each variable is checked: gender and twin status are set as categories, while vital signs and lab results are treated as continuous variables. To prevent differences in scale, all numerical features are normalized to between 0 and 1. This keeps features on the same scale and stops any single feature from dominating the model. After cleaning, 490 cases remained for analysis.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStatistical Methods\u003c/h3\u003e\n\u003cp\u003eNormally distributed quantitative data are shown as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation and compared between groups with the independent samples t-test. Non-normally distributed quantitative data are shown as median (P25, P75) and compared using the rank sum test. Qualitative data are summarized as frequency and percentage, and compared using the chi-square test or Fisher\u0026rsquo;s exact test. Analyses were done in SPSS 24.0. Results were considered statistically significant at P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\n\u003ch3\u003eMachine learning algorithms\u003c/h3\u003e\n\u003cp\u003eAll samples were randomly shuffled using R 4.2.2 software and then allocated to the training set and validation set in a 7:3 ratio. Three feature selection methods, namely LASSO regression, Boruta algorithm, and recursive feature elimination (RFE), were employed. Four models were developed: logistic regression (LR), decision tree (DT), random forest (RF), and light gradient boosting machine (LightGBM). Each model was trained and tested with ten-fold cross-validation. Hyperparameters were adjusted using grid search. Finally, the performance of the four machine learning models was evaluated and compared using AUC, precision, recall, specificity, accuracy, and F1 score (F1-Score).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eA total of 490 preterm infants meeting the study criteria were enrolled. Of these, 68 developed CRBSI (incidence rate 13.9%). A total of 76 pathogenic bacteria were isolated from 68 positive blood cultures, with Gram-negative bacteria being the majority (56 strains, 73.68%). Klebsiella pneumoniae was found most often (31 strains, 40.79%), followed by Escherichia coli (18 strains, 23.68%), Pseudomonas aeruginosa (4 strains, 5.26%), and three other Gram-negative bacteria (3.95%). The 20 detected Gram-positive bacteria (26.32%) included Staphylococcus epidermidis (10 strains, 13.16%), Staphylococcus aureus (6 strains, 7.89%), Group B Streptococcus (2 strains, 2.63%), and Enterococcus spp. (2 strains, 2.63%).\u003c/p\u003e \u003cp\u003eIn the data of 490 children, univariate analysis showed that gestational age, weight, 5-minute Apgar score, mechanical ventilation, catheterization age, puncture times, catheter dwell time, catheter maintenance frequency, axillary temperature, respiratory rate, blood oxygen saturation, RBC, WBC, NEUT, and CRP had statistically significant differences (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eGeneral information of premature infants\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTotal (n\u0026thinsp;=\u0026thinsp;490)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNon-CRBSI (n\u0026thinsp;=\u0026thinsp;422)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCRBSI (n\u0026thinsp;=\u0026thinsp;68)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eStatistic\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eFetal age,[n (%),(W)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9.012\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e34\u0026thinsp;\u0026le;\u0026thinsp;Fetal age\u0026lt;37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e93 (18.98)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e85 (20.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8 (11.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e28\u0026thinsp;\u0026le;\u0026thinsp;Fetal age\u0026lt;34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e328 (66.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e285 (67.54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e43 (63.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026lt;28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e69 (14.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52 (12.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e17 (25.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eBirth weight,[n(%),(g)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e18.306\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e1500\u0026thinsp;\u0026le;\u0026thinsp;Birth weight\u0026thinsp;\u0026lt;\u0026thinsp;2000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99 (20.20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e92 (21.80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 (10.29)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e1000\u0026thinsp;\u0026le;\u0026thinsp;Birth weight\u0026thinsp;\u0026lt;\u0026thinsp;1500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e314 (64.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e275 (65.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e39 (57.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;1000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77 (15.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e55 (13.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22 (32.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eGender, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.919\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eGirl\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e237 (48.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e205 (48.58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e32 (47.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eBoy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e253 (51.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e217 (51.42)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36 (52.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e5-minute Apgar score,[n(%),score]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026gt;7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e433 (88.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e379 (89.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e54 (79.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57 (11.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43 (10.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14 (20.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMode of delivery,n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.356\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNatural childbirth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e229 (46.73)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e200 (47.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e29 (42.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eCesarean section\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e261 (53.27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e222 (52.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e39 (57.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNumber of fetuses,n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.415\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.664\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eSingle fetus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e396 (80.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e344 (81.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e52 (76.47)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eTwins\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94 (19.18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e78 (18.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e16 (23.53)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNRDS,n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.392\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.732\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e185 (37.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e163 (38.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22 (32.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e305 (62.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e259 (61.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e46 (67.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMechanical ventilation,n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9.023\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e405 (82.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e358 (84.83)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e47 (69.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85 (17.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64 (15.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e21 (30.88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eAge in days at catheter placement,[n(%),(d)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8.615\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e198 (40.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e159 (37.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e39 (57.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026gt;7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e292 (59.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e263 (62.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e29 (42.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eCatheter insertion site, n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.081\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.037\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eHead\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76 (15.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68 (16.11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8 (11.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eLower limbs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e133 (27.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e107 (25.36)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e26 (38.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eupper limbs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e281 (57.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e247 (58.53)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e34 (50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNumber of punctures,[n(%),(time)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e28.618\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e374 (76.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e340 (80.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e34 (50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026gt;2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e116 (23.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e82 (19.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e34 (50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eReadjust the catheter, n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.428\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.627\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e256 (52.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e224 (53.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e32 (47.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e234 (47.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e198 (46.92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36 (52.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eThe indwelling time of PICC,[n(%),(d)]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e4.502\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e235 (47.96)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e211 (50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e24 (35.29)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026gt;14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e255 (52.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e211 (50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e44 (64.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eFrequency of dressing change, n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.371\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003etimes per seven days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e428 (87.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e375 (88.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e53 (77.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026lt;times per seven days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e62 (12.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e47 (11.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e15 (22.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eCatheter blockage, n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.617\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e444 (90.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e384 (91.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e60 (88.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46 (9.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (9.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8 (11.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eCatheter connector wrapping, n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.291\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.114\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57 (11.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46 (10.90)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11 (16.18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e433 (88.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e376 (89.10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e57 (83.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eCatheter displacement, n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e332 (67.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e287 (68.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e45 (66.18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e158 (32.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e135 (31.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e23 (33.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eAxillary Temperature, [\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),℃]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36.8 (36.7, 36.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36.8 (36.7, 36.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36.8 (36.7, 37.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e11768.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eHeart Rate,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),bpm]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e146 (142, 148)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e146 (142, 148)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e145 (139.5, 148)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.176\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e15806\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eRespiratory Rate ,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),bpm]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44 (42, 44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44 (42, 44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e42 (40, 44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e17000.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMean Blood Pressure,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),mmHg]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e39 (36, 41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39 (36, 41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e38 (36, 40.25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e16440.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eBlood oxygen saturation,\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e93 (92, 94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e93 (92, 94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e92 (91, 94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e17941\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eRBC,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),10\u003csup\u003e12\u003c/sup\u003e/L]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.2 (4.1, 4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.2 (4.1, 4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.3 (4.1, 4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.084\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e12485\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.3 (4.18, 4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.3 (4.2, 4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.25 (4.1, 4.36)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e17744\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eWBC,[n(%),10\u003csup\u003e9\u003c/sup\u003e/L]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.074*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (5.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21 (4.98)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5(7.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u0026thinsp;~\u0026thinsp;25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e459 (93.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e399 (94.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e61(89.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (0.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (0.47)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (2.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e68 (13.88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29 (6.87)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e39 (57.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u0026thinsp;~\u0026thinsp;25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e401 (81.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e384 (91.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e17 (25.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026gt;25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21 (4.29)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9 (2.13)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12 (17.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003ePLT,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),10\u003csup\u003e9\u003c/sup\u003e/L]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e289 (246, 316)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e289 (246, 316)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e284 (246.75, 316)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.881\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e14186\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e258 (214, 306)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e256 (214, 306)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e265 (216, 308)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.179\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e12894\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eNEUT0,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),%]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47.4 (44.82, 49.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e47.4 (44.97, 49.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e48.1 (40.5, 52.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.995\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e14355.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46.8 (41.3, 48.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46.5 (41.2, 48.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e47.6 (44.77, 50.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e10166.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eLYMPH0,[\u003cem\u003eM\u003c/em\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003e25\u003c/sub\u003e,\u003cem\u003eP\u003c/em\u003e\u003csub\u003e75\u003c/sub\u003e),%]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47.8 (43.82, 50.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48.2 (45.37, 50.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e46.4 (41.05, 50.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e16317\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47.9 (42.6, 50.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48.25 (43.6, 50.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e44.2 (40.05, 51.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.084\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e16223\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eCRP(\u0026gt;10mg/L),n(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.212\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.555\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e448 (91.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e389 (92.18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e59 (86.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e42 (8.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33 (7.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9 (13.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e151.553\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e384 (78.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e370 (87.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14 (20.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e106 (21.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52 (12.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e54 (79.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eNote: *Fisher's exact test was used.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThree methods\u0026mdash;Lasso regression, Boruta algorithm, and RFE\u0026mdash;were used to screen baseline variables. The intersection of their results identified seven predictive variables, including birth weight, age at catheterization, number of punctures, respiratory rate, oxygen saturation, WBC, and CRP.\u003c/p\u003e \u003cp\u003eFour machine learning algorithms, including LR, DT, RF, and LightGBM, were used to construct models and perform internal validation. The results are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. All four models were evaluated. The random forest (RF) model performed best, with an area under the ROC curve of 0.984 (0.965\u0026ndash;1.000), precision of 0.857, recall of 0.900, specificity of 0.976, accuracy of 0.966, and an F1 score of 0.878. The detailed information was shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance evaluation of four machine learning algorithms\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC(95%Cl)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eF1\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.902(0.802-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.552\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.800\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.884\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.653\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.843(0.738\u0026ndash;0.948)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.550\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.733\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.932\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.912\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.629\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.984(0.965-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.857\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.900\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.976\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.966\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.878\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLightGBM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.614(0.493\u0026ndash;0.734)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.231\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.300\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.843\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.769\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.261\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e illustrates the sequence of variable importance ranking in the RF model, from highest to lowest: CRP, WBC, respiratory rate, blood oxygen saturation, number of punctures, age at catheterization, and birth weight.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e illustrates the SHAP summary plot, which first displays the feature importance of each variable. In this plot, each point represents a sample. The horizontal axis shows the magnitude of each feature's influence on the prediction. The greater a point's distance from the central line (zero point), the more significantly that feature influences the model's prediction.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003ePICC catheters provide long-term intravenous access in preterm infants but are associated with substantial CRBSI risks due to immature immune systems. Globally, neonatal CRBSI rates range from 2% to 30%, with higher rates in developing countries[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. In our study, the PICC-CRBSI rate was 13.9% (68 out of 490), which is similar to the rate reported by Ren et al. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Notably, most infections in this study were caused by Gram-negative bacteria, primarily Klebsiella pneumoniae and Escherichia coli, whereas data from developed countries indicate that Gram-positive bacteria are the primary cause[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. These differences may reflect variations in infection control and clinical practices between regions.\u003c/p\u003e \u003cp\u003eIn clinical practice, predictive models necessitate both high performance and clinical interpretability [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. In this study, our SHAP analysis quantified seven key predictors of PICC-CRBSI in preterm infants, ranked by descending importance : CRP, WBC, respiratory rate, blood oxygen saturation, number of punctures, catheterization age\u0026thinsp;\u0026le;\u0026thinsp;7 days, and birth weight. Notably, CRP and WBC emerged as dominant factors, consistently occupying the highest positions in both feature importance rankings and SHAP summary plots(Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCRP demonstrated exceptional discriminatory power, consistent with its established role as a rapid-response biomarker for neonatal sepsis[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Mechanistically, CRP amplifies host defenses by inducing cytokine cascades, activating complement pathways, and modulating phagocytic clearance\u0026mdash;processes critical during bloodstream infections[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Although CRP lacks specificity for CRBSI alone, its elevated levels correlate strongly with culture-positive sepsis (versus culture-negative cases) [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], supporting its utility as a screening triage tool. In our cohort, abnormal CRP levels (\u0026gt;\u0026thinsp;10 mg/L) were observed in 79% of CRBSI infants versus 12% of non-CRBSI (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), reinforcing its diagnostic relevance. The SHAP value for CRP (+\u0026thinsp;0.146) further confirmed its linear, dose-dependent association with infection risk (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01).\u003c/p\u003e \u003cp\u003eWBC abnormalities (leukocytosis or leukopenia) served as complementary predictors, reflecting dynamic immune dysregulation during sepsis [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. The synergy between CRP and WBC enhanced early CRBSI detection (AUC: 0.91 vs. 0.83 for CRP alone), aligning with evidence that multimarker strategies improve diagnostic accuracy in neonatal infections [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThese findings advocate for immediate clinical implementation of CRP-WBC monitoring protocols in neonatal intensive care units. High CRP levels was associated with an elevated risk of PICC-CRBSI who require escalated surveillance (including repeat cultures and consideration of empiric therapy). Particularly for high-risk subpopulations\u0026mdash;notably those with catheterization within the first week of life or concurrent respiratory distress\u0026mdash;the model supports twice per week CRP-WBC monitoring to enable early intervention. This strategy integrates contemporary sepsis detection principles with preterm-specific risk mitigation, offering two key advantages: (1) earlier identification of true CRBSI cases through biomarker-guided vigilance, and (2) decreased antibiotic overuse in stable infants by avoiding treatment based solely on suspicion. Our data suggest this dual benefit reflects the precision achievable through dynamic CRP-WBC monitoring.\u003c/p\u003e \u003cp\u003eFollowing comprehensive feature selection and data preprocessing, our comparative analysis of four machine learning models revealed that the Random Forest (RF) algorithm achieved optimal performance in predicting PICC-CRBSI risk among preterm infants, demonstrating exceptional discriminative capacity (accuracy: 0.966, specificity: 0.976, AUC: 0.984). This superiority aligns with established literature documenting RF's ability to manage high-dimensional clinical data through ensemble-based decision tree aggregation, effectively balancing model complexity with generalization capacity [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. More importantly, while conventional machine learning approaches often function as \"black boxes,\" we addressed this translational limitation through systematic SHAP analysis. This advanced interpretability framework not only quantified the relative contribution of each risk factor but also revealed clinically meaningful nonlinear relationships.\u003c/p\u003e \u003cp\u003eThe incidence of PICC-related CRBSI directly reflects departmental infection prevention standards, with our results revealing decreased respiratory rate and blood oxygen saturation as key risk predictors. Bloodstream infections trigger systemic inflammation wherein mediators breach the underdeveloped blood-brain barrier of preterm infants, potentially disrupting brainstem respiratory centers and manifesting as apnea or desaturation[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Consequently, continuous monitoring of these vital signs is imperative for early infection detection and intervention. Furthermore, catheterization\u0026thinsp;\u0026le;\u0026thinsp;7 days postnatally significantly increased risk (57% of CRBSI cases vs. catheterization\u0026thinsp;\u0026gt;\u0026thinsp;7 days [43%]), attributable to concurrent deficits in innate and adaptive immunity alongside immature skin barrier function during this period [\u003cspan additionalcitationids=\"CR32\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. To mitigate this risk, stricter catheterization protocols\u0026mdash;real-time biomarker monitoring, aseptic technique reinforcement, and simulation-based staff training\u0026mdash;must be implemented to reduce multiple punctures. Finally, birth weight remains a critical non-modifiable determinant; its inverse relationship with CRBSI risk (9.0% increase per 100g weight decrement [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]). A dedicated PICC management team should be set up for these infants. Improving the management of vascular access can help reduce catheter-related bloodstream infections.\u003c/p\u003e \u003cp\u003eHealthcare teams require early identification of high-risk preterm infants with PICC to initiate preemptive interventions against CRBSI. We addressed this need by developing four machine learning models using routine clinical parameters, among which the RF algorithm demonstrated superior predictive performance (AUC: 0.984) for early risk stratification. Nevertheless, this study has methodological limitations. First, the model was validated only internally; external verification across multicenter cohorts with varying demographics and clinical practices is essential to confirm generalizability. Second, real-world implementation requires integrating this RF model into electronic health records (EHRs) to automate data extraction and generate real-time alerts for high-risk infants\u0026mdash;a critical step toward enabling closed-loop, precision management of PICC-CRBSI that could significantly reduce infection rates.\u003c/p\u003e \u003cp\u003eIn summary, our study established a machine learning-based risk prediction tool for PICC-CRBSI in preterm infants, demonstrating the RF algorithm's superior predictive capability. The model identifies seven critical predictors stratified as inflammatory biomarkers, physiological indicators, demographic characteristics, and clinical procedural factors. The validated algorithm enables early identification of high-risk infants, facilitating targeted interventions. Prioritizing enhanced monitoring, proactive catheter care protocols, and preemptive antimicrobial stewardship for these patients may significantly reduce bloodstream infections burden while improving PICC safety outcomes.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eBasic Research Project of Changzhi Municipal Bureau of Science and Technology\u003c/p\u003e \u003cp\u003eProject No.:JC202439\u003c/p\u003e \u003cp\u003eConflict of interest\u003c/p\u003e \u003cp\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.\u003c/p\u003e \u003cp\u003ePublisher\u0026rsquo;s note\u003c/p\u003e \u003cp\u003e All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eYQG: Methodology, writing-original draft.YYD: Data clean and analysis，Model validation.WXS: Thesis guidanceLHW: Project administration.L.W: Supervision.\u003c/p\u003e\u003ch2\u003eData availability statement\u003c/h2\u003e \u003cp\u003eThe data used to support the findings of this study are not publicly available due to [reason], but can be obtained from the corresponding author upon reasonable request.\u003c/p\u003e \u003cp\u003eEthics statement\u003c/p\u003e \u003cp\u003e This study was reviewed and approved by the Medical Ethics Committee of Changzhi Maternal and Child Health Hospital (Ethics Approval No: CZSFYLL2024.026) and complies with the ethical principles of the Declaration of Helsinki. As a prospective study, Written informed consent was obtained from all legal guardians prior to the participants' inclusion in the study,all personal information was anonymized using coded identifiers to protect participant privacy.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTeibel H, Hood K, Manasco K, Bhatia J. Antibiotic Administration Prior to Central Venous Catheter Removal in Neonates. J Pharm Pract. 2021;34:894\u0026ndash;900.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePonticelli E, Clari M, Volpes M, et al. Complete Blood Count Collected Via Venipuncture Versus Peripherally Inserted Central Catheter in Hematological Patients: A Comparison of 2 Methods. Cancer Nurs. 2022;45:E36\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Tonder DJ, Keough N, van Niekerk ML, van Schoor AN. The position of the common facial vein in neonates: An alternate route for central venous catheter placement. Clin Anat. 2021;34:644\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown R, Burke D. The hidden cost of catheter related blood stream infections in patients on parenteral nutrition. Clin Nutr ESPEN. 2020;36:146\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRen XL, Li HL, Liu J, et al. Ultrasound to Localize the Peripherally Inserted Central Catheter Tip Position in Newborn Infants. Am J Perinatol. 2021;38:122\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKlinger G, Bromiker R, Zaslavsky-Paltiel I et al. Late-Onset Sepsis in Very Low Birth Weight Infants. Pediatrics 2023; 152.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBierlaire S, Danhaive O, Carkeek K, Piersigilli F. How to minimize central line-associated bloodstream infections in a neonatal intensive care unit: a quality improvement intervention based on a retrospective analysis and the adoption of an evidence-based bundle. Eur J Pediatr. 2021;180:449\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou X, Chen L, Liu H-X. Applications of Machine Learning Models to Predict and Prevent Obesity: A Mini-Review. Front Nutr 2022; 9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTriantafyllidis A, Polychronidou E, Alexiadis A, et al. Computerized decision support and machine learning applications for the prevention and treatment of childhood obesity: A systematic review of the literature. Artif Intell Med. 2020;104:101844.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFernandes M, Mendes R, Vieira SM, et al. Predicting Intensive Care Unit admission among patients presenting to the emergency department using machine learning and natural language processing. PLoS ONE. 2020;15:e0229331.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi XH, Yang XL, Dong BB, Liu Q. Predicting 28-day all-cause mortality in patients admitted to intensive care units with pre-existing chronic heart failure using the stress hyperglycemia ratio: a machine learning-driven retrospective cohort analysis. Cardiovasc Diabetol. 2025;24:10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRahmani K, Garikipati A, Barnes G, et al. Early prediction of central line associated bloodstream infection using machine learning. Am J Infect Control. 2022;50:440\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSakai H, Iwata M, Terasawa T. External validation of the Michigan PICC catheter-associated bloodstream infections score (MPC score) for predicting the risk of peripherally inserted central catheter-associated bloodstream infections: A single-center study in Japan. Infect Control Hosp Epidemiol. 2023;44:480\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Lancet Respiratory M. Opening the black box of machine learning. Lancet Respir Med. 2018;6:801.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePourhoseingholi MA, Vahedi M, Rahimzadeh M. Sample size calculation in medical studies. Gastroenterol Hepatol Bed Bench. 2013;6:14\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGordon A, Greenhalgh M, McGuire W. Early planned removal versus expectant management of peripherally inserted central catheters to prevent infection in newborn infants. Cochrane Database of Systematic Reviews. 2018; 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosenthal VD, Bat-Erdene I, Gupta D, et al. International Nosocomial Infection Control Consortium (INICC) report, data summary of 45 countries for 2012\u0026ndash;2017: Device-associated module. Am J Infect Control. 2020;48:423\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLake JG, Weiner LM, Milstone AM, et al. Pathogen Distribution and Antimicrobial Resistance Among Pediatric Healthcare-Associated Infections Reported to the National Healthcare Safety Network, 2011\u0026ndash;2014. Infect Control Hosp Epidemiol. 2017;39:1\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJansen SJ, Broer SDL, Hemels MAC, et al. Central-line-associated bloodstream infection burden among Dutch neonatal intensive care units. J Hosp Infect. 2024;144:20\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang J, Chen H, Wang H et al. A Risk Prediction Model for Physical Restraints Among Older Chinese Adults in Long-term Care Facilities: Machine Learning Study. J Med Internet Res 2023; 25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSproston NR, Ashworth JJ. Role of C-Reactive Protein at Sites of Inflammation and Infection. Front Immunol 2018; 9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoggi C, Lucenteforte E, Petri D, et al. Presepsin for the Diagnosis of Neonatal Early-Onset Sepsis: A Systematic Review and Meta-analysis. JAMA Pediatr. 2022;176:750\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePovoa P, Garvik OS, Vinholt PJ, et al. C-reactive protein and albumin kinetics after antibiotic therapy in community-acquired bloodstream infection. Int J Infect Dis. 2020;95:50\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGilfillan M, Bhandari V. Biomarkers for the diagnosis of neonatal sepsis and necrotizing enterocolitis: Clinical practice guidelines. Early Hum Dev. 2017;105:25\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaestraggi Q, Lebas B, Clere-Jehl R et al. Skeletal Muscle and Lymphocyte Mitochondrial Dysfunctions in Septic Shock Trigger ICU-Acquired Weakness and Sepsis-Induced Immunoparalysis. Biomed Res Int. 2017; 2017: 7897325.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Q, Gong X. Clinical significance of the detection of procalcitonin and C-reactive protein in the intensive care unit. Exp Ther Med. 2018;15:4265\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCruz AF, Herrmann J, Ramcharran H, et al. Sustained vs. Intratidal Recruitment in the Injured Lung During Airway Pressure Release Ventilation: A Computational Modeling Perspective. Mil Med. 2023;188:141\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJin Y, Lan A, Dai Y, et al. Development and testing of a random forest-based machine learning model for predicting events among breast cancer patients with a poor response to neoadjuvant chemotherapy. Eur J Med Res. 2023;28:394.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Matary A, Abozaid S, Al Suliman M et al. Correlation between Bronchopulmonary Dysplasia and Cerebral Palsy in Children: A Comprehensive Analysis Using the National Inpatient Sample Dataset. Child (Basel) 2024; 11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBliss JM, Wynn JL. Editorial: The Neonatal Immune System: A Unique Host-Microbial Interface. Front Pediatr. 2017;5:274.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlater A, Straney L, Alexander J, et al. The Effect of Imputation of PaO2/FIO2 From SpO2/FIO2 on the Performance of the Pediatric Index of Mortality 3. Pediatr Crit Care Med. 2020;21:520\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEmeriaud G, Lopez-Fernandez YM, Iyer NP, et al. Executive Summary of the Second International Guidelines for the Diagnosis and Management of Pediatric Acute Respiratory Distress Syndrome (PALICC-2). Pediatr Crit Care Med. 2023;24:143\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePapoff P, Caresta E, Luciani S, et al. The starting rate for high-flow nasal cannula oxygen therapy in infants with bronchiolitis: Is clinical judgment enough? Pediatr Pulmonol. 2021;56:2611\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYildizdas D, Yontem A, Iplik G, et al. Predicting nasal high-flow therapy failure by pediatric respiratory rate-oxygenation index and pediatric respiratory rate-oxygenation index variation in children. Eur J Pediatr. 2021;180:1099\u0026ndash;106.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-pediatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bped","sideBox":"Learn more about [BMC Pediatrics](http://bmcpediatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bped/default.aspx","title":"BMC Pediatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Machine learning, Preterm infants, PICC, Bloodstream infection, Prediction model","lastPublishedDoi":"10.21203/rs.3.rs-8162982/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8162982/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eTo develop a risk prediction model for peripherally inserted central catheter-related bloodstream infections (PICC-CRBSI) in preterm infants.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe conducted a prospective study of 490 preterm infants with PICC placement in a Chinese tertiary neonatal ICU (June 2023\u0026ndash;December 2024). The data were split into training and validation sets at a 7:3. CRBSI was the primary outcome. Feature selection was performed using LASSO regression, the Boruta algorithm, and recursive feature elimination (RFE), and prediction models were constructed using four machine learning algorithms, including logistic regression (LR), decision tree (DT), random forest (RF), and light gradient boosting machine (LightGBM). The optimal model was selected through performance comparison.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eCRBSI occurred in 68 cases (13.9%), predominantly caused by Gram-negative bacteria (e.g., Klebsiella pneumoniae, Escherichia coli). The random forest model performed best among the four machine learning models, with an area under the receiver operating characteristic curve (AUC) of 0.984, precision of 0.857, recall of 0.900, specificity of 0.976, accuracy of 0.966, and F1-score of 0.878. The SHAP summary plot identified the top 7 most important features in order: C-reactive protein (CRP), white blood cell count (WBC), respiratory rate, blood oxygen saturation, number of punctures, age at catheterization, and birth weight.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eOur study developed and interpreted a risk prediction model for PICC- CRBSI in preterm infants. It is expected to assist clinicians in timely risk stratification and targeted interventions, thereby reducing the incidence of CRBSI and improving the prognosis of preterm infants.\u003c/p\u003e\u003ch2\u003eClinical Trial Number:\u003c/h2\u003e \u003cp\u003eNot applicable.\u003c/p\u003e","manuscriptTitle":"Development and Validation of a Machine Learning-Based Risk Prediction Model for PICC-Related Bloodstream Infections in Premature Infants Using SHAP Interpretability","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-30 09:31:01","doi":"10.21203/rs.3.rs-8162982/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2025-12-24T08:42:36+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-02T08:04:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-28T15:34:40+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-28T15:34:04+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Pediatrics","date":"2025-11-20T09:35:08+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-pediatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bped","sideBox":"Learn more about [BMC Pediatrics](http://bmcpediatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bped/default.aspx","title":"BMC Pediatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6b41bcd2-bd66-4cd9-8f42-7ef4f40518ef","owner":[],"postedDate":"December 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-12-30T09:31:01+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-30 09:31:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8162982","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8162982","identity":"rs-8162982","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}