CN114842976A - Premature brain injury prediction marker, prediction model and system - Google Patents

Abstract

The invention discloses a premature brain injury prediction marker, a prediction model and a system. After brain injury, due to the destruction of the blood brain barrier and cytoskeleton, various biological markers can be released into cerebrospinal fluid and reach peripheral blood through the blood brain barrier. In neonatal brain injury, biological markers can be detected from peripheral blood shortly after the injury occurs and predict the extent and prognosis of the brain injury. Invasive mechanical ventilation time, acidosis, frequent apnea, thrombocyte volume > 0.27 are postnatal independent risk factors for BIPI in preterm low birth weight infants. Invasive mechanical ventilation time (day), acidosis, frequent apnea, thrombocyte packing > 0.27, postnatal day 1 serum S100 β protein level (ng/ml), postnatal day 1, 7 serum B-FABP level (ng/ml), lvd (mm), ihd (mm) combined with constructing a regression model can better predict the occurrence of BIPI in premature infants.

Description

Premature brain injury prediction marker, prediction model and system

Technical Field

The invention belongs to a disease prediction model technology, and particularly relates to a premature brain injury prediction marker, a prediction model and a prediction system.

Background

Premature infants and infants with ultra-low birth weight still have high fatality rate and disability risk, and the incidence rate of diseases related to premature infants, such as premature infant brain injury (BIPI), bronchopulmonary dysplasia, neonatal necrotizing enterocolitis, and neonatal retinopathy, is gradually increased. BIPI is one of the main causes of death of children in the neonatal period, is closely related to diseases such as visual disturbance, cognitive behavior disturbance, and long-term cerebral palsy, brings heavy burden to families, society and health systems, seriously influences future life and learning of premature infants, diagnoses BIPI in the early period, and is particularly important for intervention and treatment as soon as possible.

BIPI is a cerebral hemodynamic disorder in premature infants caused by various infections, asphyxia, hypoxia, ischemia, etc., and has structural or functional damage to the brain to various degrees. Brain damage in premature infants can be classified according to neuropathology as: (ii) non-hemorrhagic brain damage: including periventricular leukomalacia (PVL), periventricular hemorrhagic infarction (PVHI), and late ventricular dilatation (ventriculomegaglay); ② hemorrhagic brain damage: including periventricular-intraventricular hemorrhage (PVH-IVH), parenchymal hemorrhage, subarachnoid hemorrhage (SAH), etc.; ③ other parts are damaged: such as cerebellar and brainstem injuries. PVL is further classified into cystic PVL, non-cystic PVL, and diffuse white matter lesions (WMD). PVH-IVH and PVL are common types of brain injury in premature infants, and the incidence of PVH-IVH has declined over the past few decades. Studies in the early 80 s of the 20 th century showed that very low birth weight infants weighing less than 1500g had a PVH-IVH incidence of 40% to 50%, whereas this data fell to 20% in the early 21 st century.

The occurrence of BIPI is currently believed to be related to a number of factors. Foreign research reports that the levels of Tumor necrosis factor alpha (TNF-alpha) and Interleukin 6 (IL-6) in cord blood of BIPI children are obviously higher than those of normal newborns, and the infection and inflammatory reaction are related to the generation of BIPI. It has been found that early onset sepsis is a common risk factor for hemorrhagic and non-hemorrhagic brain damage. Kidokoro et al found that gestational age is a risk factor for the development of BIPI and is negatively correlated with the development of BIPI. Since BIPI often lacks typical clinical manifestations and signs, diagnosis mainly depends on imaging examination such as craniocerebral ultrasound, Magnetic Resonance Imaging (MRI), and the like. Bedside craniocerebral ultrasound has the advantages of simple and convenient operation, no wound, low cost and the like, but has the limitations of unclear exploration, poor diagnosis specificity and the like. MRI has the advantages of non-invasiveness, no radiation, strong diagnostic sensitivity, strong specificity and the like, but has the disadvantages of complex operation, late examination time and difficult coordination of newborn infants, thereby limiting the clinical application of the infants. Therefore, a single examination method has certain inevitable disadvantages, and the joint diagnosis of the BIPI by multiple technologies is crucial.

Related researches show that the level of serum S100 beta protein is obviously increased after myocardial injury, which indicates that the specificity of the serum S100 beta protein as a brain injury marker is limited. In addition, due to the short duration in serum, the inconvenience of collecting cord blood specimen in NICU, etc., it is still necessary to utilize the characteristics of the serum more reasonably and to search for other serum markers with more specificity and longer duration. At present, no research report about the predictive significance of B-FABP in BIPI of premature infants exists. Magnetic Resonance Imaging (MRI) is a non-ionizing imaging technique with high soft tissue resolution that allows detailed assessment of the brain structure of premature infants.The white matter of the brain, the contrast of the grey matter, the myelination process, the development of ischemic pathologies originating from arterial abnormalities, vascular pathologies of the brain willis's rings and their branches, hemorrhagic pathologies, can be assessed by means of MRI techniques,>1mm focal lesion, diffuse WMD, etc. Therefore, the diagnosis value of MRI is the highest for WMD patients such as PVL and PVHI. Currently, the MRI sequences routinely used by the neonatal clinical evaluation brain are mainly T ₁ WI、T ₂ WI, Diffusion Weighted Imaging (DWI) and inversion recovery sequences (e.g., FLAIR sequences) diagnose brain damage via MRI signal abnormalities, but ignore subtle changes in brain structure and fail to identify brain damage early before signal changes. The application of the skull MRI quantitative detection technology in predicting and diagnosing the BIPI occurrence and recent nerve fate of the premature infant is not clearly determined.

Disclosure of Invention

The invention discloses the prediction effect of the serum marker combined quantitative MRI technology in the BIPI generation of premature low birth weight infants for the first time, and provides a theoretical basis for preventing and treating the BIPI in early stage and improving prognosis. The invention discloses the correlation between early change of peripheral blood B-FABP level and the generation of BIPI of premature infants for the first time.

Use of a predictor for the prediction of brain damage in a preterm infant, said predictor being one or more of invasive mechanical ventilation time (day), acidosis, frequent apneas, thrombocytic > 0.27, postnatal day 1 serum S100 β protein level (ng/ml), postnatal day 1, 7 day serum B-FABP level (ng/ml), lvd (mm), ihd (mm).

Use of a predictor for establishing a model for predicting brain injury in a preterm infant, said predictor being one or more of invasive mechanical ventilation time (day), acidosis, frequent apneas, thrombocytic > 0.27, postnatal day 1 serum S100 β protein level (ng/ml), postnatal day 1, 7 serum B-FABP level (ng/ml), lvd (mm), ihd (mm).

A predictive model of brain injury in premature infants, the regression equation being:P=1/[1+e-(-28.179)+0.458X ₁ +1.028X ₂ +1.932X ₃ +1.401X ₄ +1.489X ₅ +0.595X ₆ +0.410X ₇ +0.429X ₈ +1.740X ₉ ]wherein X is ₁ 、X ₂ 、X ₃ 、X ₄ 、X ₅ 、X ₆ 、X ₇ 、X ₈ 、X ₉ Invasive mechanical ventilation time (day), acidosis, frequent apneas, thrombocyte pressure > 0.27, postnatal day 1 serum S100 β protein level (ng/ml), postnatal day 1, day 7 serum B-FABP levels (ng/ml), lvd (mm), ihd (mm), respectively,Pis the Logistic model probability.

The application of the prediction model of the brain injury of the premature infant in predicting the BPD of the premature infant.

The method for establishing the model for predicting the brain injury of the premature infant comprises the following steps of collecting clinical relevant indexes of the premature infant; the correlation index is then substituted into the regression equation: p =1/[1+ e- (-35.032 +0.431X1+0.245X2+8.363X3-0.321X4+0.707X5+1.236X 6) ], yielding a predictive model for BPD in preterm infants; clinically relevant indicators of the premature infant include invasive mechanical ventilation time (day), acidosis, frequent apneas, thrombocyte packing > 0.27, post-natal day 1 serum S100 β protein level (ng/ml), post-natal day 1, 7 day serum B-FABP levels (ng/ml), lvd (mm), ihd (mm).

A system for predicting brain injury of a premature infant comprises a prediction module, a data input module and a prediction output module, wherein the prediction module comprises a prediction model of the brain injury of the premature infant, the data input module is used for inputting relevant index data into the prediction model of the brain injury of the premature infant, and the prediction output module outputs the prediction probability of the brain injury of the premature infant according to the prediction model of the brain injury of the premature infant.

A computer loaded with the system for predicting brain injury of premature infant.

The area under the ROC curve of the analysis and risk assessment model of the risk factors related to the brain injury of the premature infant given by the prior art is 0.76, and the result of goodness of fit test of Hosmer. (χ 2 =4.016, P =0.675>0.05), the model is considered to have better risk factor distinguishing capability and fitting effect, the AUC of the prediction model for the generation of the BPD of the premature infant, which is constructed by combining 9 indexes and Logistic regression, is 0.998, the sensitivity is 97.6%, the specificity is 98.6%, and the prediction model is obviously higher than that of the existing prediction model.

The invention enters the NICU of Suzhou university subsidiary child hospital and the birth gestational age within 24 hours after the birth from the 07 th day 01 in 2018 to the 12 th day 31 in 2020<37 weeks, birth weight<2500g of premature infants with the hospitalization time of more than or equal to 14 days are taken as research objects, the serum S100 beta protein and B-FABP levels of the first, second and third days after the premature infants are dynamically detected by analyzing the clinical data of the premature infants and pregnant women, the head MRI examination is carried out at the corrected gestational age of 34-38 weeks, the brain structure is quantitatively measured, prospective clinical research is carried out, the prediction effect of the serum marker combined quantitative MRI technology in the BIPI generation of the premature low birth weight infants is explored, and the theoretical basis is provided for early BIPI prevention and treatment and prognosis improvement. The invention discloses the correlation between early change of peripheral blood B-FABP level and the generation of BIPI of premature infants for the first time. Data and methods. Taking invasive mechanical ventilation time (day), acidosis, frequent apnea, thrombocyte pressure > 0.27, postnatal day 1 serum S100 beta protein level (ng/ml), postnatal day 1, 7 serum B-FABP level (ng/ml), LVD (mm), IHD (mm) as independent variables, and combining Logistic regression to construct a BIPI prediction model, wherein the regression equation is as follows:P=1/[1+e-(-28.179)+0.458X ₁ +1.028X ₂ +1.932X ₃ +1.401X ₄ +1.489X ₅ +0.595X ₆ +0.410X ₇ +0.429X ₈ +1.740X ₉ ]wherein X is ₁ 、X ₂ 、X ₃ 、X ₄ 、X ₅ 、X ₆ 、X ₇ 、X ₈ 、X ₉ Invasive mechanical ventilation time (day), acidosis, frequent apneas, thrombocyte pressure > 0.27, postnatal day 1 serum S100 β protein level (ng/ml), postnatal day 1, day 7 serum B-FABP levels (ng/ml), lvd (mm), ihd (mm), respectively,Pis the Logistic model probability. The Logistic regression prediction model has a chi-square value of 2.689,Pand if the index is less than 0.001, the index has explanation capability on the prediction of the BIPI. Calculating the content of the content in the content file by using the classification interactive table, df =8,Pand the goodness of fit of the prediction model is good when the prediction model is more than 0.05. Taking the prediction variable as a check variable,the results show that the 9 indexes combined with the Logistic regression model show that the area under the ROC curve is 0.998, which is larger than that of each index independently predicted.

Drawings

FIG. 1 shows a method for measuring each index in magnetic resonance.

FIG. 2 is a ROC curve for predicting BIPI generation by serum S100 beta protein and B-FABP.

FIG. 3 is a ROC curve for prediction of BIPI for cranial MRI brain structure measurement indicators.

FIG. 4 is a ROC curve for the Logistic regression model to predict the occurrence of BIPI.

Detailed Description

(1) Invasive mechanical ventilation time, acidosis, frequent apnea, thrombocyte volume > 0.27 are postnatal independent risk factors for BIPI in preterm low birth weight infants.

(2) Serum S100 beta protein level at postnatal day 1 of BIPI premature infants and serum B-FABP levels at postnatal days 1 and 7 are obviously increased compared with non-BIPI premature infants, and the levels are positively correlated with the severity of BIPI; the risk of developing BIPI in preterm infants increases when the serum S100 β protein level at day 1 after the onset is greater than 6.325ng/ml, and the serum B-FABP level at days 1 and 7 is greater than 12.14ng/ml, 12.32ng/ml, respectively.

(3) B-FABP is a serum biomarker predictive of BIPI in preterm infants with a longer peak duration than S100 β protein.

(4) The LVD, IHD of BIPI premature cranial MRI are significantly larger than non-BIPI premature infants, and the risk of BIPI development is predicted to increase when measured at values greater than 13.75mm, 1.65mm, respectively.

(5) Invasive mechanical ventilation time (day), acidosis, frequent apnea, thrombocyte pressure > 0.27, postnatal day 1 serum S100 beta protein level (ng/ml), postnatal day 1, 7 serum B-FABP level (ng/ml), LVD (mm), IHD (mm) combined to construct a regression model can better predict the occurrence of BIPI in premature infants.

1. Inclusion criteria

The study subjects are premature infants who live in the subsidiary child hospital NICU of Suzhou university within 24 hours after the birth of the infant, the birth gestational age of the infant is less than 37 weeks, the birth weight of the infant is less than 2500g, and the hospitalization time of the infant is more than or equal to 14 days from the 07 th day 01 to the 2020 th day 31. The study was approved by the Suzhou university subsidiary Children hospital NICU ethical Committee (ethical No.: 2021CS 022).

Exclusion criteria:

(1) the hospitalization age is > 24 hours, and the hospitalization time is <14 days;

(2) those with neurological abnormalities or severe complications;

(3) genetic metabolic diseases or chromosomal abnormalities;

(4) bilirubin encephalopathy, intrauterine TORCH infection;

(5) presence of surgical illness or hospitalization with surgical operators;

(6) death within hours after hospitalization and active rescue treatment;

(7) imperfect clinical data.

2. Diagnostic criteria and grouping

2.1 diagnostic criteria for brain injury in premature infants

According to the expert consensus on brain injury diagnosis and prevention and treatment of premature infants made by the Association of Chinese physicians in 2012 ^[39] And expert group opinions on the diagnosis, prevention and comprehensive management of brain white matter damage of premature infants (expert group opinion on the comprehensive management of brain white matter damage of premature infants) made by the expert group of neurological forum of newborn in 2015 ^[40] And combining the 20 behavioral neural examination methods (NBNA) of the newborn in China ^[41] And (6) carrying out diagnosis. Premature infants need to meet both: imaging examination (skull ultrasound and/or MRI) abnormalities; abnormal NBNA scoring; including or not including: high risk factors causing brain damage; fourthly, clinical manifestations are shown; thus, the diagnosis of BIPI can be made.

Specific diagnostic criteria are as follows:

(1) high risk factors causing brain damage: hypoxia ischemia, hemodynamic disorders, infection, blood coagulation dysfunction, history of abnormal labor, pregnancy complications or complications;

(2) the clinical manifestations are as follows: atypical, may be accompanied by central apnea, bradycardia, hypotension, hypertension, lethargy, dysphoria, convulsion, dystonia, abnormal primary reflex, etc., and may also have no clinical symptoms;

(3) ultrasonic manifestation of the skull: including ependymal hemorrhage, SAH, subdural hemorrhage, IVH, PVH and infarction, PVL etc., wherein IVH is fractionated by Papile method under ultrasound ^[42] The method comprises four stages: stage I: bleeding from the layer of the hair growing matrix under the unilateral or bilateral ventricular canals; and II, stage: the ventricular subintimal hemorrhage breaks into the ventricles of the brain, namely, the intracerebroventricular hemorrhage without the enlargement of the ventricles of the brain; grade III: intracerebral hemorrhage with ventricular dilatation; IV stage: IVH is associated with periventricular hemorrhagic infarction. The PVL is divided into four stages under ultrasound according to a de Vries method ^[42] : stage I: bilaterally symmetrical hyperechoic masses around the ventricles of the brain last for more than or equal to 7 days, and no cystic cavity appears behind the hyperechoic masses; and II, stage: local hyperechoic globes around the ventricles on both sides, forming a small sac cavity around the ventricles at least 2 weeks after the postnatal period; grade III: extensive hyperechoic around ventricles on both sides, which changes into extensive capsule cavities around ventricles after several weeks (the earliest 2 weeks), and the capsule cavities can be fused into tablets; IV stage: bilateral periventricular widespread hyperechoic and white matter involvement, with a shift to periventricular and subcortical superficial cortical diffuse cystic changes after several weeks (first 2 weeks).

(4) MRI examination: according to 2012' expert consensus on brain injury diagnosis and prevention and treatment of premature infant ^[39] The MRI abnormal expression of BIPI children patients comprises: early stage of white matter non-hemorrhagic damage T ₁ WI appears as a high signal in the white matter region, T ₂ WI is low or equal signal; late stage T ₁ Disappearance of WI signals or low signals or reduction of white matter volume, T ₂ WI is a high signal, and the morphology of the ventricles is altered in severe cases. Early hemorrhagic injury at T ₁ WI appears as a high signal, T ₂ WI is low signal and later at T ₁ WI、T ₂ WI are both high signals.

(5) Neonate 20 items behavioral neuro testing method (NBNA score) ^[41] : the division into 5 parts, namely 6 items of behavior ability, 4 items of passive muscle tension and active muscle tension, 3 items of original reflex and 3 items of general reaction. Each score has 3 points, i.e., 0 (failed to elicit and significant abnormal), 1 (minor abnormal) and 2 (completely normal), with a total of 40 points,<35 are classified as abnormal. Details are shown in the attached table1。

2.2 BIPI reference scale

Mild brain damage ^[43,44] : first, only one sulcus or lobe of the brain is involved in bleeding; (ii) one or more intracerebroventricular hemorrhage without ventricular dilatation; and thirdly, the white matter damage focus in MRI is less than or equal to 2mm and the range is less than or equal to 3 parts.

Severe brain injury ^[43,44] : (ii) bleeding involves two or more brain lobes; ② intracerebral hemorrhage with ventricular dilatation; ③ white matter damage focus in MRI>2mm and/or range>3, part (b); in severe cases extensive white matter is affected subcortically.

2.3 diagnostic criteria for primary disease and common complications in premature infants

(1) Neonatal Respiratory Distress Syndrome (NRDS) ^[45] : newborn babies appear in 12h after birth and manifest by progressive aggravated dyspnea with groan, shortness of breath, irregular respiration and inspiratory trihydroxy, and X-rays can show that the transparency of two lungs is reduced, the bronchia is inflated, and severe babies show white lungs.

(2) Necrotizing enterocolitis Neonatorum (NEC) ^[46] : clinically, abdominal distension, hematochezia and vomiting are mainly shown, severe cases are accompanied by shock and multi-organ functional failure, the abdominal X-ray plain film is characterized by partial intestinal cyst-like pneumatosis, and the pathology is characterized by necrosis of the far end of ileum and the near end of colon.

(3) Premature bronchopulmonary dysplasia (BPD) ^[47] : premature or very low birth weight infants are still receiving mechanical ventilation, CPAP or FiO 28 days after birth and at 36 weeks corrected gestational age ₂ >BPD was diagnosed at 30%. If FiO ₂ And (3) the oxygen saturation is less than or equal to 30 percent, and the oxygen saturation is 90-96 percent, the patient is subjected to an oxygen withdrawal test, and the BPD is diagnosed by the oxygen withdrawal loser.

(4) Respiratory failure (respiratory failure) ^[48] : the arterial blood-gas analysis shows the oxygen partial Pressure (PO) by taking the respiratory frequency as the expressions of early acceleration, late deceleration, labored respiration, nasal wing fanning, trisection and the like ₂ ) Less than or equal to 60mmHg with or without partial pressure of carbon dioxide (PCO) ₂ ) > 50mmHg, is a common respiratory systemSerious complications.

(5) Apnea of premature infant ^[49] : the breath stopping time is more than or equal to 20 seconds, or the breath stopping time<20 seconds with heart rate slowing<When the blood is taken 100 times/min, skin bluish purple, muscle tension reduction, reaction hypofunction and oxygen saturation reduction can occur. Frequent apnea refers to repeated apnea more than or equal to 3 times/hour.

(6) Retinopathy of prematurity (ROP) ^[50] : stage 5 by severity: stage I: a boundary line appears between the temporal periphery and the avascular zone of the fundus retina; and stage II: the division line of the fundus is raised, and ridge-like changes occur; stage III: retinal vasodilatation proliferation with fibrous tissue proliferation; and IV, period: a tractional retinal detachment from the peripheral portion to the posterior pole; stage V: the retina was completely peeled off.

(7) Neonatal sepsis (neonatal septicema) ^[51,52] : mainly takes the symptoms of fever or non-rising body temperature, poor reaction, no eating, jaundice, hepatosplenomegaly, shock and the like, and simultaneously satisfies any 1 item in the following conditions: the nonspecific examination of blood is more than or equal to 2 positive items: increased or decreased white blood cell count, a number of rod-shaped nucleus cells/neutrophils of 0.16 or more, and a platelet count<Increased 100X 109/L, C reaction protein and increased procalcitonin; ② the cerebrospinal fluid is checked to be the pyogenic meningitis change; and detecting pathogenic bacteria DNA in blood. Wherein Early-onset septicemia (EOS) is mainly vertically transmitted by parent pathogenic bacteria when the onset time is less than or equal to 3 days old; late-onset septicemia (LOS) is usually caused by patients with onset time > 3 days, and is usually nosocomial infection or community-acquired infection.

(8) Neonatal pulmonary hemorrhage (pulmony hemorrhage of the newborn, PHN) ^[53] : it means that massive hemorrhage of lung, at least 2 lung lobes are involved, and clinically the infant can suck hemorrhagic liquid in nasal cavity or trachea, and can eliminate hemorrhage or intragastric hemorrhage caused by operation injury.

(9) Hyperbilirubinemia of newborn ^[54] : hyperbilirubinemia is diagnosed after birth of a newborn when the serum bilirubin level exceeds 95% of the neonatal hour bilirubin profile.

(10) Parenteral Nutrition Associated Cholestasis (PNAC) ^[55] : the compound is a common and serious complication in the process of parenteral nutrition of premature infants, the clinical continuous parenteral nutrition is over 14 days, the patients are accompanied by jaundice, the color of stool is lightened, the serum direct bilirubin is increased (more than 34 mu mol/L), and the cholestasis caused by other reasons is eliminated, so the diagnosis is considered to be PNAC.

(11) Extrauterine growth retardation (EUGR) ^[56] : the premature infant does not reach the expected intrauterine growth after birth, and the growth and development metering index is below the tenth percentile level of the infant of the same age at the time of discharge or 36 weeks of corrected gestational age according to a 2013-Fenton curve.

(12) Ventilator Associated Pneumonia (VAP) ^[57] : pulmonary infection of a trachea cannula patient child after receiving mechanical ventilation for 48 hours or after withdrawing the trachea, accompanied by respiratory condition change, pulmonary rale and new radiological abnormality.

(13) Anemia of newborn ^[58] : the hemoglobin (Hb) concentration is less than 145g/L within 7 days after birth, wherein 120 g-145 g of the hemoglobin is mild anemia, 90 g-119 g of the hemoglobin is moderate anemia, and less than 90g of the hemoglobin is severe anemia.

(14) Hypoglycemia (hypoglycemia) ^[59,60] : the whole blood sugar of the newborn can be diagnosed when the whole blood sugar is less than or equal to 2.2mmol/L no matter the gestational age or the day age of the newborn. Among them, hypoglycemia can be classified by severity as: mild (> 1.7-2.2 mmol/L), moderate (> 1.0-1.7 mmol/L) and severe (. ltoreq.1.0 mmol/L).

(15) Hyperglycemia (hyperglycemia) ^[60] : at present, the whole blood glucose of more than 7.0mmol/L is taken as a diagnostic standard in China.

(15) Neonatal Emergency physiology score peri-life supplement II (score for neural access physiology, neural extension, version II, SNAPPE-II) ^[61] : SNAP-II based scoring criteria including mean blood pressure, minimum body temperature, PO, as established by Richardson 2001, were used ₂ /FiO ₂ Minimum pH, per hourUrine per hour, twitch, birth weight, 5min Apgar score, and 9 variables for small for gestational age. The scoring range is as follows: the score is 0-9 mild, 10-19 moderate, and more than 19 severe, the score is higher, and the disease condition is more critical. See attached table 2 for details.

2.4 antimicrobial drug Classification

The national ministry of health 2015 issued the guidelines for clinical application of antibacterial drugs ^[62] The antibacterial drugs are classified into three classes, i.e., non-limited class, limited class and special use class according to their efficacy, safety and indication. Meanwhile, referring to the antibacterial medicine clinical application hierarchical management catalogue in Jiangsu province and the relevant regulations of the antibacterial medicine clinical application hierarchical management catalogue in subsidiary child hospitals of Suzhou university, the antibacterial medicine hierarchical standard is as follows:

the non-limiting antibacterial drug is safe, effective and relatively low-cost antibacterial drug proved by long-term clinical application. Such as penicillin, amoxicillin, ampicillin, and the like.

The limited antibacterial drug is an antibacterial drug which is proved to be safe and effective after long-term application, but has limitations in safety, curative effect, influence on bacterial drug resistance, drug price and the like, and needs to be limited according to the condition of a patient, population, indication, prescription amount of the drug and the like. Such as cefmetazole, moxalactam, amoxicillin/sulbactam, cefoperazone/sulbactam, azithromycin injection and the like.

The special antibacterial drugs are new antibacterial drugs on the market, and are expensive, and the clinical data of the curative effect or the safety of the special antibacterial drugs are less or not better than those of the existing drugs. Such as vancomycin, meropenem, imipenem, linezolid and the like.

2.5 study groups

According to the BIPI diagnostic standard, the premature infant which develops the BIPI is used as a BIPI group, and in the premature infant which does not generate the BIPI, the premature infant of which the general data is matched with the BIPI group is used as a non-BIPI group; according to the result of the imaging examination ^[43,44] The BIPI components are divided into mild and severe BIPI subgroups; according to the MRI abnormal condition of the skull ^[39] The BIPI fraction was divided into an MRI abnormal group and an MRI normal group.

3. Research method

3.1 Observation studies were performed on preterm infants meeting inclusion criteria.

(1) Clinical data were collected for comparison between preterm and pregnant women: general condition of two groups of premature infants: gender, gestational age of birth, birth weight, head circumference, mode of production, 1 minute Apgar score, 5 minutes Apgar score, whether it is a double or multiple fetus, whether it is a tube baby, whether it is a small for gestational age baby; ② general condition of two groups of preterm mothers: maternal age, fetal frequency, parity of labor, amniotic fluid pollution, placental prematurity, placenta hominis, premature rupture of fetal membrane for more than or equal to 18h, cholestasis, gestational diabetes, gestational hypothyroidism, gestational anemia, severe preeclampsia, whether glucocorticoid is used before delivery, whether magnesium sulfate is used before delivery, and infection before delivery; ③ two groups of premature infants have primary basic diseases and complications: NRDS, NEC, BPD, PHN, respiratory failure, frequent apneas, ROP, EUGR, EOS, LOS, VAP, hyperbilirubinemia, infectious pneumonia, septic shock, central nervous system infection, hypoglycemia and its severity, duration of hypoglycemia, hyperglycemia, PNAC, anemia, left-right bypass Patent Ductus Arteriosus (PDA), PDA ≧ 1.5mm, Atrial Septal Defect (ASD), pulmonary hypertension, acidosis, SNAPPE-II score; four groups of treatment measures for premature infants during hospitalization: the number of hospitalization days, whether invasive mechanical ventilation is used or not, invasive mechanical ventilation time, oxygen therapy time, high-frequency ventilation, normal-frequency ventilation, oxygen absorption concentration more than or equal to 30%, before machine withdrawal, caffeine use or not, machine withdrawal failure and re-uploading probability, PICC (peripherally inserted central catheter) tube placement rate and duration, PICC complications (complicated blood-borne infection, tube removal, tube blockage, breakage or tube placement failure and the like), umbilical vein tube placement rate, venous nutrition time more than or equal to 30 days, blood transfusion times (more than or equal to 3 times), venous fluid infusion glucose speed (the 1 st, 3 th and 7 th days), antibiotic use condition, pig lung surfactant use condition and mouse nerve growth factor use condition; laboratory examination of two groups of premature infants when admitted to hospital: white blood cell count (× 10) ⁹ ,/L), platelet count (× 10) ⁹ /L), width (%) of platelet distribution, thrombocyte volume (%), PO ₂ （mmHg）、PCO ₂ (mmHg), hematocrit (%), serum sodium (mmol/L), serum potassium (mmol/L), serum calcium (mmol/L), lactic acid (mmol/L), albumin (g/L), prealbumin (mg/L); sixthly, nutrition supply conditions of two groups of premature infants during hospitalization: oral liquid intake, intravenous liquid intake, total liquid intake, intestinal caloric intake, parenteral caloric intake, total enteral nutrition time, and the like on postnatal days 1, 3, 7, 14, 21.

(2) And (3) performing peripheral venipuncture after admission of the two groups of premature infants, respectively reserving 0.5-1 ml of peripheral venous blood on the 1 st, 7 th and 14 th days after admission, centrifuging at the room temperature of 20 ℃ at the rotating speed of 3000r/min for 5 minutes, taking the upper layer serum, placing the upper layer serum into a dry EP (ethylene propylene) tube, and placing the sample into a refrigerator at the temperature of-80 ℃ within 1 hour for storage to be subjected to unified detection. This assay uses a double antibody one-step sandwich Enzyme-linked immunosorbent assay (ELISA) to measure serum S100. beta. protein, B-FABP levels, and compares the differences between the two groups.

(3) Completing skull B-ultrasonic examination of all premature infants on postnatal days 1, 3 and 7, and rechecking every week after abnormality exists, or rechecking every 2 weeks until discharge; NBNA scoring after 40 weeks corrected gestational age; and perfecting the MRI examination of the skull in 34-38 weeks of corrected gestational age: measurement: lateral Ventricular Diameter (LVD), double parietal width (BPW), Interhemispheric distance (IHD), Third ventricular diameter (V) ^3rd ) The diameter of The Cerebellum (TCD), the Thickness of the cortex (Cortical Thickness), the Thickness of the corpus callosum (Thickness of the corpius callosum). And comparing the difference of MRI brain structure measurement results between the BIPI group and the non-BIPI group and between the BIPI group and the three groups, namely the MRI normal group, the MRI abnormal group and the non-BIPI group of premature infants.

3.2 skull B-ultrasound examination

A bedside craniocerebral ultrasound examination of premature infants was performed by a sonographer using a color Doppler ultrasound diagnostic apparatus model TOSHIBA TA700, Japan, with a probe frequency of 3 MHZ. When the infant is asleep or in a quiet state, the patient takes the supine right position, and the examiner performs continuous ultrasonic scanning on the brain of the infant to observe the abnormal echo condition of the brain structure, blood flow and brain tissue. After the examination is finished, the ultrasonic brain image is judged by a sonographer.

3.3 skull MRI examination

The infant patient is sedated by 10mg/kg intravenous bolus injection of phenobarbital sodium 15-30 minutes before the examination. After the infant is asleep, an imaging physician performs a cranial MRI examination of the infant using an GE3.0T superconducting magnetic resonance apparatus, selects orthogonal cross loops, and performs a conventional MRI, DWI scan. Axial direction T covering the entire head ₁ The WI scan parameters are: TE31ms, TR3316ms, FOV200mm, matrix 512 × 512, slice thickness 5.0 mm; axial direction T covering the entire head ₂ WI scan parameters were TE103ms, TR6443ms, FOV200mm, matrix 512 × 512, slice thickness 5.0mm, gap 5.0 mm; the DWI scan parameters were TE78ms, TR3954ms, FOV200mm, matrix 512 × 512, slice thickness 5.0mm, gap 5.0 mm. Measuring method ^[37,63] : measuring left/right LVD (L-LVD/R-LVD) on the lateral ventricle horizontal level of the coronal section of the brain, and adding the left/right LVD and the L-LVD; measuring BPW and IHD on the coronal section by taking the connecting line of the cochlea on the two sides and the basal osteotomy as a horizontal line; ③ V ^3rd Is at an axial position T ₁ On the WI sequence, the sequence is obtained by front and back combined level measurement; measuring the maximum TCD on the coronal section of the lateral ventricle; measuring the distance between the inner surface and the outer surface of the cortex (the inner surface of the cortex is a white matter boundary, and the outer surface is a boundary between the grey matter and the cerebrospinal fluid) to obtain the thickness of the cortex; sixthly, measuring the thickness of the corpus callosum through three positions of the knee, the middle and the pressing part of the midsagittal section of the cerebrum, and calculating the average value; see figure 1 for details. After the examination is finished, the radiologist reads and diagnoses the infant under the condition of unknown infant.

3.4 NBNA score

After 40 weeks of preterm infant corrected gestational age, 20 NBNA scores were performed by the health care physician. The examination is carried out in the middle of two times of nursing of the infant patient, the environment is quiet, the light is semi-dark, the room temperature is set between 22 ℃ and 27 ℃, and the examination is completed within 10 minutes. After the examination is finished, the children patients are scored by the doctors of the rehabilitation department.

3.5 specimen detection

(1) The specimen to be detected is taken out of a refrigerator at-80 ℃, thawed at room temperature for 20 minutes, and simultaneously 20 Xwashing buffer and distilled water are mixed according to the weight ratio of 1: and (5) diluting by 20.

(2) And arranging standard substance holes and sample holes on the lath, wherein 50 mu L of standard substances with different concentrations are added into the standard substance holes, 50 mu L of samples to be detected are added into the sample holes, and blank holes are not added.

(3) In each well except the blank well, 100. mu.L of detection antibody labeled with horseradish peroxidase was added, the reaction well was sealed with a sealing plate, and incubated in a 37 ℃ incubator for 60 minutes.

(4) Discarding the liquid, patting dry on absorbent paper, filling 350 μ L of washing solution into each hole, 1min later, throwing off the washing solution and patting dry, and washing the plate for 5 times in this way.

(5) 50. mu.L of substrate A, B was added to each well and incubated for 15 minutes at 37 ℃ in the absence of light.

(6) Stop solution (50. mu.L) was added to each well, and the OD of each well was measured at a wavelength of 450nm using a microplate reader within 15 minutes.

(7) And (3) taking the OD value of the measured standard substance as an abscissa and the concentration value of the standard substance as an ordinate, drawing a standard curve by using ELISA Calc software, obtaining a linear regression equation, substituting the OD value of the sample into the equation, and calculating the concentration (ng/ml) of the sample.

4. Main experimental reagent and equipment

5. Statistical method

Statistical analysis of all data using SPSS 25.0 statistical processing software, with the data being measured in terms of mean + -SD with normal distribution (

+/-s), wherein independent sample t test is adopted for comparison between two groups, Leven method is adopted for comparison between three groups for testing whether the homogeneity of the variance is met, variance analysis is adopted for the satisfied persons, and Krusal-Wallis H test is adopted for the unsatisfied persons; is not in compliance withMedian (25 th, 75 th percentile) of normal distribution [ M (P) ₂₅ ，P ₇₅ )]As shown, the Mann Whitney U test is adopted for comparison between two groups, and the Krusal-Wallis H test is adopted for comparison between three groups; the counting data is expressed in percentage (%), and the two groups are compared by adopting chi-square test or continuity correction chi-square test (n is more than or equal to 40, T is more than or equal to 1) _min 5) analysis; performing Logistic regression analysis on factors with statistical difference in single factor analysis to obtain independent risk factors of BIPI generation,P<0.05 the difference was considered statistically significant; drawing the prediction efficiency of the evaluation index of the ROC curve graph, and calculating the area under the curve; john's index = sensitivity + specificity-1, using the maximum john's index to calculate the cut-off value; the continuity variable adopts pearson correlation analysis, the classification variable adopts Spearman correlation analysis to obtain a correlation coefficient r,P< 0.05A correlation between the two variables was considered.

The result is

1. Clinical basic information and data

1.1 general comparison of preterm infants in two groups

During the study, 238 premature infants with gestational age of less than 37 weeks were observed, 9 cases of death or automatic discharge within hours of active rescue in admission were excluded, 7 cases of genetic abnormality were present, 2 cases of chromosomal abnormality were present, 8 cases of patients who were transferred to surgery during hospitalization, 1 case of congenital small intestine blockade, 1 case of esophageal blockade, wherein 43 cases of BIPI were observed, and 77 premature infants whose general data matched with the BIPI group among 167 premature infants without BIPI occurrence were regarded as a non-BIPI group, 120 cases in total in two groups, 54 cases in men, 66 cases in women, and average gestational age (32.03 ± 2.18) weeks and average birth weight (1870.3 ± 548.83) g. 32 mild brain injuries and 11 severe brain injuries in the BIPI group; 14 cases of MRI abnormal group and 29 cases of MRI normal group. No significant difference is observed between BIPI group and non-BIPI group in terms of gender, gestational age, birth weight, head circumference, production method, whether it is a test tube infant, whether it is multiple or double, whether it is small for gestational age, and whether the score of 1min and 5min Apgar is less than or equal to 7 (a)P>0.05) as detailed in table 1.

Note: a is chi-square value obtained by chi-square test ² The value b is obtained by t testtThe value is obtained.

1.2 comparison of high-risk factors of two groups of premature infants

1.2.1 general comparison of pregnancy between two groups of preterm infants

Two groups of premature infant mothers are more than 35 years old or less than 20 years old, the gestation period, the delivery time, amniotic fluid pollution, the placental premature peeling, the placenta hominis is placed in front, the rupture of the fetal membrane is more than or equal to 18h, cholestasis, gestation diabetes, hypothyroidism in gestation period, anemia in gestation period, severe preeclampsia, whether glucocorticoid is used before delivery, whether magnesium sulfate is used before delivery and the infection condition before delivery are compared without obvious difference (the method has no obvious difference in the conditions that (the method is used for treating the diseases of pregnancy and the like)P>0.05) as detailed in table 2.

Note: a is chi's obtained by chi-square test ² The value c is chi-square obtained by continuous correction chi-square test ² Value, -is not estimated.

1.2.2 comparison of Primary basal diseases and complications in two groups of premature infants

Compared with the non-BIPI group, the BIPI group has obvious higher scores in NRDS, BPD, PHN, respiratory failure, frequent apnea, EUGR, EOS, VAP, septic shock, central nervous system infection, hypoglycemia, severe hypoglycemia, hypoglycemia duration, anemia, acidosis incidence and SNAPPE-II than the non-BIPI group, and the difference has statistical significance ((P< 0.05); while the two groups have no significant difference in NEC, ROP, LOS, mild hypoglycemia, moderate hypoglycemia, hyperglycemia, hyperbilirubinemia, infectious pneumonia, PNAC, left-to-right shunting type PDA, PDA ≥ 1.5mm, ASD, and pulmonary hypertension (b)P>0.05) as detailed in table 3.

TABLE 3 comparison of the onset of the primary disease and complications in two groups of premature infants

Note: a is chi's obtained by chi-square test ² The value c is chi-square obtained by continuous correction chi-square test ² The value d is obtained by Mann-Whitney U testZValue, -is not estimated.

1.2.3 comparison of treatment status of two groups of premature infants during hospitalization

The number of hospitalization days, invasive mechanical ventilation time, oxygen therapy time, high-frequency ventilation, oxygen uptake concentration of more than or equal to 30 percent of the BIPI group, caffeine before withdrawal, venous nutrition time of more than or equal to 30 days, sugar rate of 1 day, blood transfusion of more than or equal to 3 times, pig lung surfactant and special antibacterial drugs are obviously more than those of the non-BIPI group, and the difference has statistical significance (the difference is that: (the total number of the medicines is more than that of the non-BIPI group)P< 0.05); the two groups have no significant difference in constant-frequency ventilation, failure in withdrawal and recovery, PICC (peripherally inserted central catheter), PICC time, PICC complications, umbilical vein catheterization, 3 rd day and 7 th day sugar rate, use of non-limiting antibacterial drugs and use of limiting antibacterial drugs (the dosage of the drug is not limited by the dosage of the drug (the dosage is not limited by the dosage of the drug)P>0.05) as detailed in table 4.

Note: a is chi's obtained by chi-square test ² The value b is obtained by t testtThe value c is chi-square obtained by continuous correction chi-square test ² The value d is obtained by Mann-Whitney U testZThe value is obtained.

1.2.4 comparison of laboratory test results for admission to premature infants in both groups

In the laboratory examination results on the day of admission, two groups had leukocytes > 20X 10 ⁹ the/L ratio and its median level, platelets > 300X 10 ⁹ the/L ratio and its median level, the distribution width of platelets > 16.5% and its median level, the thrombocyte pressure > 0.27 and its median level, the hematocrit, the lactate, the albumin, the PO ₂ ≤60mmHg、PCO ₂ The difference in ≦ 24mmHg is statistically significant (P< 0.05); two groups showed < 4X 10 white blood cells ⁹ ratio/L platelet < 100X 10 ⁹ Wide distribution of platelets and/or the ratio ofThe ratio of degree less than 12%, median level thereof, thrombocyte volume less than 0.1, serum sodium, serum potassium, serum calcium, prealbumin, PCO ₂ No significant difference > 50mmHg (II) ((III))P>0.05) as detailed in table 5.

Note: a is chi's obtained by chi-square test ² The value b is obtained by t testtThe value d is obtained by Mann-Whitney U testZValue, -is not estimated.

1.3 two groups premature infants nutritional status comparison

1.3.1 enteral Nutrition comparison of two groups of premature infants

The time for starting enteral feeding and the time for complete enteral feeding of the BIPI group are both significantly later than those of the non-BIPI group, and the differences have statistical significance (P< 0.05), see table 6 for details.

Note: d is the result of Mann-Whitney U testZThe value is obtained.

1.3.2 comparison of fluid intake in two groups of preterm infants

The oral liquid amount and the total liquid amount of the BIPI group at 3 days after the birth are obviously less than those of the non-BIPI group, the intravenous liquid intake amount at 7 and 14 days is obviously higher than that of the non-BIPI group, and the difference has statistical significance (P< 0.05); the venous fluid amount, the oral fluid amount and the total fluid amount on the 3 rd day, the 7 th day and the 14 th day of the two groups have no significant difference (P>0.05) as detailed in table 7.

Note: b is obtained by t testtThe value d is obtained by Mann-Whitney U testZThe value is obtained.

1.3.3 caloric intake comparison of two groups of preterm infants

BIPI group intestinal caloric and total caloric of 3 days after birthLess than that of the non-BIPI group, and the intestinal exocentric caloric value at postnatal days 7 and 14 is obviously higher than that of the non-BIPI group, and the difference has statistical significance (P< 0.05); there was no significant difference between the intestinal caloric index at postnatal day 3, intestinal caloric index at postnatal days 7 and 14 and total caloric index (P> 0.05). See table 8 for details.

Note: b is obtained by t testtThe value d is obtained by Mann-Whitney U testZThe value is obtained.

1.4 Logistic multifactorial regression analysis of postnatal BIPI in two groups of preterm infants

Taking whether the BIPI occurs as a dependent variable, taking the factor with statistical significance of difference after single factor comparison in the clinical data as an independent variable, and carrying out Logistic multi-factor regression analysis, wherein the result is as follows: invasive mechanical ventilation time (OR =2.135, 95% CI: 1.253-3.639), acidosis (OR =9.943, 95% CI: 1.782-55.481), frequent apneas (OR =7.950, 95% CI: 1.697-37.241), platelet aggregation > 0.27 (OR =28.207, 95% CI: 2.286-296.228) are independent risk factors for BIPI after preterm infants, as detailed in Table 9.

Comparison of S100 beta protein and B-FABP in two groups of preterm infants

2.1 comparison of serum S100 beta protein in two groups of premature infants

Comparing results among groups: the serum S100 beta protein level at the 1 st day after the premature infant in the BIPI group is obviously higher than that in the non-BIPI group, and the difference has statistical significance (P< 0.05); while there was no significant difference between the two groups at postnatal days 7 and 14 for serum S100 β protein level ((P> 0.05); see table 10 for details.

Comparing results in the group: no significant difference in serum S100 beta protein levels at days 1, 7 and 14 in the non-BIPI group (II)P> 0.05). The serum S100 beta protein level at the 1 st day after the BIPI group is obviously higher than that at the 7 th and 14 th days,the difference is statistically significant (P< 0.05), no significant difference in serum S100 beta protein levels at days 7 and 14: (P> 0.05); see table 10 for details.

Note: *: compared with the comparison of the day 14 after the birth,Pless than 0.05; b is obtained by t testtThe value, m, is obtained by analysis of varianceFThe value is obtained.

2.2 prediction of BIPI Generation by serum S100 beta protein

A ROC curve was drawn using postnatal day 1 serum S100 β protein (ng/ml) levels as variables and non-BIPI as references to evaluate the predicted potency of serum S100 β protein levels for BIPI generation, with an AUC of 0.931 (95% CI: 0.888-0.973,P< 0.05), with a cut-off value of 6.325ng/ml, sensitivity of 95% and specificity of 81%, see FIG. 2 for details.

2.3 comparison of serum B-FABP levels in two groups of preterm infants

Comparing results among groups: the serum B-FABP level of the premature infants of the BIPI group at 1 day and 7 days after birth is obviously higher than that of the non-BIPI group, and the difference has statistical significance (P< 0.05); and the serum B-FABP level at the 14 th postnatal day has no significant difference in comparison between the two groups (P> 0.05); see table 11 for details.

Comparing results in the group: no significant difference in serum B-FABP levels at postnatal days 1, 7 and 14 in non-BIPI group preterm infants: (P> 0.05). The serum B-FABP level of the premature infants in BIPI group at 1 day and 7 days is obviously higher than that of the premature infants at 14 days, and the difference has statistical significance (P< 0.05); the serum B-FABP level on the 1 st day and the 7 th day is not obviously different (P is more than 0.05); see table 11 for details.

Note: *: compared with the comparison of the day 14 after the birth,Pis less than 0.05. b is obtained by t testtThe value, m, is obtained by analysis of varianceFThe value is obtained.

2.4 prediction of BIPI Generation by serum B-FABP

Respectively taking the serum B-FABP level at the postnatal day 1 and the postnatal day 7 as variables, drawing ROC curves by taking a non-BIPI reference object, evaluating the prediction efficiency of the serum B-FABP level at the day 1 and the day 7 on the BIPI generation, wherein the AUC is 0.876 (95% CI: 0.808-0.945 respectively,P＜0.05）、0.902（95%CI：0.845～0.909，P< 0.05), the cut-off values were 12.14ng/ml and 12.32ng/ml, respectively, the sensitivities were 79% and 79%, and the specificities were 84% and 90%, as shown in FIG. 2.

3. Comparison of MRI brain Structure measurements of two groups of premature infants

In the study, the BIPI group premature infant MRI abnormal group accounts for 14 cases, accounting for 32.6% (14/43), which are cystic PVL 2 cases, diffuse WMD 3 cases, PVL + PVH 1 cases, PVL + ventricular dilatation 2 cases, PVHI 1 cases, severe HIE 3 cases, IVH + ventricular dilatation + hydrocephalus 1 cases, and PVH-IVH + SAH 1 cases; BIPI group preterm MRI normal group accounted for 67.4% in 29 cases (29/43).

3.1 comparison of MRI brain Structure measurements of different groups of premature infants

LVD, IHD and V of brain of premature infant in BIPI group are obtained by brain structure measurement of MRI image ^3rd Is significantly greater than that of the non-BIPI group, the callus thickness is significantly less than that of the non-BIPI group, and the difference has statistical significance: (P< 0.05); there was no significant difference in BPW, TCD, cortical thickness between the BIPI and non-BIPI groups of preterm infants: (P>0.05) as detailed in table 12.

Note: b is obtained by t testtThe value d is obtained by Mann-Whitney U testZThe value is obtained.

The BIPI group was further divided into BIPI group MRI abnormal group of premature infants and BIPI group MRI normal group of premature infants, and comparison between the three groups with non-BIPI group gave: BIPI group LVD, IHD, V of premature infant MRI abnormal group and normal group ^3rd Significantly greater than the non-BIPI group, and the callus thickness significantly less than the non-BIPI group: (P< 0.05), whereas between the abnormal and normal MRI groups LVD, IHD, V ^3rd The callus thickness has no significant difference (P> 0.05); the MRI abnormal group and the normal group and the non-BIPI group in the BIPI group have no significant difference in comparison of BPW, TCD and cortical thickness ()P>0.05) as detailed in table 13.

Note: #: compared with the non-BIPI group,Pless than 0.05; e is the result of Kruskal-Wallis H testHThe value, m, is obtained by analysis of varianceFThe value is obtained.

3.2 prediction of BIPI occurrence by magnetic resonance brain Structure measurements

Drawing an ROC curve by taking the LVD as a variable and taking a non-BIPI as a reference object, and evaluating the prediction efficiency of the LVD on the BIPI occurrence, wherein the result is as follows: AUC is 0.625 (95% CI: 0.519 to 0.731,P< 0.05), cutoff value of 13.75mm, sensitivity of 60.5% and specificity of 66.2%, see figure 3 for details.

Drawing an ROC curve by taking IHD as a variable and non-BIPI as a reference object, and evaluating the prediction efficiency of the IHD on the BIPI generation, wherein the result is as follows: AUC is 0.677 (95% CI: 0.571 to 0.782,P< 0.05), cutoff value of 1.65mm, sensitivity of 65.9%, specificity of 67.6%, see figure 3 for details.

③ with V ^3rd For variables, ROC curves were plotted against non-BIPI as reference, and V was evaluated ^3rd The predicted potency for BIPI development, the results are: AUC is 0.642 (95% CI: 0.536-0.749,P< 0.05), cutoff value of 2.35mm, sensitivity of 32.6% and specificity of 91.9%, see figure 3 for details.

And fourthly, drawing a ROC curve by taking the thickness of the callus as a variable and taking non-BIPI as a reference, evaluating the prediction efficiency of the thickness of the callus on the occurrence of BIPI, and obtaining a result: AUC is 0.606 (95% CI: 0.498-0.715,P< 0.05), cutoff value of 2.31mm, sensitivity of 73.0% and specificity of 48.8%, see figure 3 for details.

4. Index combined Logistic regression construction BIPI prediction model

The levels of S100 beta protein (n) in serum at day 1 after the onset of invasive mechanical ventilation (day), acidosis, frequent apnea, thrombocytic pressure > 0.27g/ml), serum B-FABP level (ng/ml) at postnatal day 1 and 7 days, LVD (mm), IHD (mm) are independent variables, a BIPI prediction model is constructed by combining Logistic regression, and the regression equation is as follows:P=1/[1+e-(-28.179)+0.458X ₁ +1.028X ₂ +1.932X ₃ +1.401X ₄ +1.489X ₅ +0.595X ₆ +0.410X ₇ +0.429X ₈ +1.740X ₉ ]wherein X is ₁ 、X ₂ 、X ₃ 、X ₄ 、X ₅ 、X ₆ 、X ₇ 、X ₈ 、X ₉ Invasive mechanical ventilation time (day), acidosis, frequent apneas, thrombocyte pressure > 0.27, postnatal day 1 serum S100 β protein level (ng/ml), postnatal day 1, day 7 serum B-FABP levels (ng/ml), lvd (mm), ihd (mm), respectively,Pis the Logistic model probability. The Logistic regression prediction model has a chi-square value of 2.689,Pand if the index is less than 0.001, the index has explanation capability on the prediction of the BIPI. Calculating the content of the content in the content file by using the classification interactive table, df =8,Pand the goodness of fit of the prediction model is good when the prediction model is more than 0.05. And (3) by taking a prediction variable as a test variable, the result shows that the area under the ROC curve is 0.998 by the 9 indexes combined with the Logistic regression model, and the area is larger than that of each index for independent prediction. See table 14, table 15, fig. 4 for details.

Neurodevelopmental disorder caused by BIPI is still a main problem left by survived premature babies, the generation factors of BIPI are not uniformly determined at home and abroad at present, and the study at abroad considers that the gestational age and the birth weight are independent risk factors of BIPI generation and have negative correlation with the morbidity. Other risk factors include male, twin or multiple pregnancy, hypoglycemia, chorioamnionitis, low Apgar score, etc. It is currently believed that the development of BIPI is associated with postnatal hypoxia, ischemia, infection, malnutrition, carbohydrate metabolism and other factors in premature infants.

Frequent apnea and acidosis are independent risk factors for the occurrence of the BIPI, and the reasons are considered as follows: frequent apnea causes collapse of end-expiratory alveoli, reduced gas exchange, and long-term apnea with PO ₂ Decrease, PCO ₂ Elevation, reflexively, leads to bradycardia and hypoxemia, and hypoxia can cause abnormal discharge in cerebral cortex, inducing or aggravating brain injury. Simultaneous hypoxia presents PCO ₂ The increase of the pH value and the decrease of the pH value cause the accumulation of acidic substances due to energy metabolism disorder, and metabolic acidosis can increase the permeability of capillary vessels and the fragility of blood vessels, further influence the autonomous regulation function of cerebral blood flow, and cause the disturbance of cerebral hemodynamics, thereby causing brain injury.

The concept that the prenatal use of magnesium sulfate may reduce the risk of developing cerebral palsy was first proposed by Nelson et al in the 90 s of the 20 th century ^， Subsequently, Doyle LW et al applied magnesium sulfate to pregnant women at risk of preterm birth to evaluate its neuroprotective effect, and the results showed that preterm infants who used magnesium sulfate prenatally had a reduced risk of cerebral palsy compared to the control group of approximately 1/3. However, the mechanism of the neuroprotective effect of magnesium sulfate is not clear, and the reason may be that magnesium sulfate can block a glutamate receptor, and the blood vessel expansion effect of magnesium sulfate can increase uterine blood flow perfusion, so that the incidence rate of hypoxia ischemic brain injury in the perinatal period is reduced, but the application of magnesium sulfate in the two groups of premature infants before delivery has no obvious difference.

Numerous research experiments have shown that the role of infection in BIPI development is becoming more and more important. The study of Hassanein et al shows that IL-6 is more than or equal to 1.077x10 ^-7 Is an independent risk factor for PVL generation, and considers that the release of inflammatory factors is increased after infection to influence the formation of cerebral vessels, thereby causing the disorder of the hemodynamics. Researches at home and abroad discover that EOS can increase the risk of BIPI generation. The invention proposes that EOS and septic shock are related to the generation of BIPI, but the EOS and the white blood cell count is more than 20 multiplied by 10 ⁹ the/L is not an independent risk factor for the occurrence of BIPI.

The invention discloses the relationship between the serum S100 beta protein and the BIPI for the first time, and the result shows that the serum S100 beta protein level of the premature infants in the BIPI group is obviously higher than that of the premature infants in the non-BIPI group at the postnatal 1 day, the serum S100 beta protein levels of the two groups at the postnatal 7 day have no obvious difference, and meanwhile, the change of the S100 beta protein level is positively correlated with the severity of the BIPI. The S100 beta protein is mainly present in glial cells of the central and peripheral nervous systems, and under normal conditions, the S100 beta protein cannot penetrate the blood brain barrier and is released into the peripheral blood. When the premature infant has brain injury, glial cells are necrotic, the blood brain barrier is damaged, S100 beta protein penetrates the blood brain barrier and is released into peripheral blood, and is discharged through the kidney, and the half-life period is about 30-100 minutes. In the present study, the ROC curve shows that the area under the curve of serum S100 β protein predicting BIPI is 0.931, and when the protein level of S100 β is 6.325ng/ml, the sensitivity of predicting BIPI generation is 95.3%, and the specificity is 81.8%, which indicates that the prediction efficiency of serum S100 β protein on BIPI generation is good. The invention shows that the serum B-FABP level of the BIPI group premature infant gradually rises and reaches a peak value at the postnatal day 1, the postnatal day 7 still maintains a high level state, a descending trend appears later, the change of the serum B-FABP level is positively correlated with the severity of the BIPI, an ROC curve shows that the AUC of the serum B-FABP level prediction BIPI at the 1 st day and the 7 th day is respectively 0.876 and 0.902, the critical values are respectively 12.14ng/ml and 12.32ng/ml, the sensitivity is 79 percent and 79 percent, the specificity is 84 percent and 90 percent, the B-FABP is a serum biological marker which has longer peak duration than S100 beta protein, but the ROC prediction efficiency is not better than the S100 beta protein, and a large sample and multi-center research is needed to verify the value of the B-FABP in BIPI prediction.

MRI scan analysis of 97 premature infants with gestational age < 30 weeks by Kidokoro shows that IHD and LVD of PVL patients are obviously increased and TCD is reduced compared with a control group. It has been found that lateral ventricle enlargement and cerebellar volume reduction are associated with neurodevelopmental disorders, wherein cerebellar injury can cause cognitive and behavioral deficits in children. Cerebellar injury is related to reasons such as hypoxia and ischemia, supratentorial hemorrhage can cause blood decomposition products (heme and iron) to deposit in cerebrospinal fluid, iron deposition and catalysis of free radical attack caused by active oxygen can also cause cerebellar atrophy, and TCD is reduced. There was no significant difference in TCD between the two groups of preterm infants of the present invention.

The invention discloses a novel combined prediction model of BIPI generation of premature infants, and ROC curve analysis shows that the areas under curves for predicting BIPI generation are respectively 0.755, 0.611, 0.696, 0.657, 0.931, 0.876, 0.902, 0.646 and 0.677 by taking invasive mechanical ventilation time (day), acidosis, frequent apnea, thrombocyte deposition > 0.27, postnatal day 1 serum S100 beta protein level (ng/ml), postnatal day 1 serum B-FABP level (ng/ml), postnatal day 7 serum B-FABP Level (LVD) (mm) and IHD (mm) as independent variables; and further performing Logistic regression analysis by combining multiple indexes, wherein the result shows that the area under the ROC curve of the combined prediction model is 0.998, the sensitivity is 97.6%, the specificity is 98.6%, and more than 9 indexes are independently predicted. The model can better predict the occurrence of the BIPI, can help to identify and diagnose the BIPI at early stage in clinic, improve the prognosis of BIPI sick children and reduce the possibility of bad neurodevelopment fate brought by the BIPI.

In summary, the occurrence of BIPI is associated with a variety of factors. BIPI occurs as a result of infection, hypoxia, ischemia, malnutrition, abnormal carbohydrate metabolism, and changes in blood viscosity. The change of the levels of serum markers S100 beta protein and B-FABP is closely related to the generation of BIPI and positively correlated with the severity thereof, namely LVD, IHD and V of MRI brain ^3rd The callus thickness measurement can provide possibility for MRI prediction of BIPI from the quantitative angle, and the multi-index combined regression model can better predict the occurrence of BIPI.

Claims (10)

1. Use of a predictor for the prediction of brain injury in a preterm infant, wherein the predictor is one or more of invasive mechanical ventilation time, acidosis, frequent apneas, thrombocytic pressure > 0.27, postnatal day 1 serum S100 β protein level, postnatal day 1, day 7 serum B-FABP level, LVD, IHD.

2. Use of a predictor factor in the establishment of a model for predicting brain injury in a preterm infant, wherein the predictor factor is one or more of invasive mechanical ventilation time, acidosis, frequent apneas, thrombocytic pressure > 0.27, postnatal day 1 serum S100 β protein level, postnatal day 1, day 7 serum B-FABP level, LVD, IHD.

3. Use according to claim 1 or 2, characterized in that the unit of invasive mechanical ventilation time is a day.

4. A predictive model of brain injury in premature infants, characterized by the regression equation:P=1/[1+e-(-28.179)+0.458X ₁ +1.028X ₂ +1.932X ₃ +1.401X ₄ +1.489X ₅ +0.595X ₆ +0.410X ₇ +0.429X ₈ +1.740X ₉ ]wherein X is ₁ 、X ₂ 、X ₃ 、X ₄ 、X ₅ 、X ₆ 、X ₇ 、X ₈ 、X ₉ Respectively invasive mechanical ventilation time, acidosis, frequent apnea, thrombocyte pressure > 0.27, postnatal serum S100 beta protein level, postnatal serum B-FABP level, LVD, IHD at day 1 and 7,Pis the Logistic model probability.

5. Use of the predictive model of brain injury in preterm infants according to claim 4 for predicting BPD in preterm infants.

6. The method of establishing a predictive model of brain injury in a premature infant of claim 4, comprising the steps of collecting clinically relevant indicators of the premature infant; the correlation index is then substituted into the regression equation: p =1/[1+ e- (-35.032 +0.431X1+0.245X2+8.363X3-0.321X4+0.707X5+1.236X 6) ], yielding a predictive model for BPD in preterm infants; the clinically relevant indexes of the premature infant comprise invasive mechanical ventilation time, acidosis, frequent apnea, thrombocyte volume more than 0.27, postnatal day 1 serum S100 beta protein level, postnatal day 1 and 7 serum B-FABP level, LVD and IHD.

7. A system for predicting brain injury of a premature infant comprises a prediction module, a data input module and a prediction output module.

8. The system of claim 7, wherein the prediction module comprises a prediction model of the brain injury of the premature infant of claim 4.

9. A computer loaded with the system for predicting brain injury in a premature infant of claim 8.

10. The computer of claim 9, comprising a display, a keyboard, and a hard disk.

Images