Abstract: Background：With an increasing number of very preterm infants (VPI, <32 weeks' gestation) treated in China, central vascular catheterization has become a common technique in Chinese neonatal intensive care units (NICUs). Inappropriate use of central vascular catheters(CVCs) may occur in the treatment, but there is still lack of relative data. Objective：To retrospectively analyze the firstyear data from the standardized database of the Chinese Neonatal Network (CHNN) and preform hospitallevel questionnaires, to reveal the current problem of central vascular catheterization among VPIs in Chinese NICUs and to provide baseline data for future quality improvement program. Design：Crosssectional study. Methods：The study described the current situation of central catheterization across gestational age (GA) weeks, including rate, duration and site variation, using the firstyear data of CHNN (from Jan. 1st 2019 to Dec. 31st 2019) from the participating NICUs. Questionnaires were collected on hospital level in terms of the regulation and management of CVC wards, indications of insertion and removal, and related complications of central catheterization. Infants with GA of 24+0~31+6 weeks, admitted into CHNN database from Jan. 1st 2019 to Dec. 31st 2019 within 24 hours after birth were included. Those infants with major congenital anomalies, transferred to other hospitals or discharge against medical advice were excluded. Incomplete and substandard data were also excluded. Subgroup analysis were done for infants with GA at 2428 weeks and 2931 weeks. Research sites were classified into children's specialized hospitals, maternal and children's healthcare centers and general hospitals. Questionnaires were designed to investigate the regulation, management, indications, maintenance and related complications of central catheterization, which were filled by department directors or senior neonatologists authorized by the director. Main outcome measures：The type, rate and duration of central catheterization. Results：A total of 6,532 VPIs from 57 CHNN participating sites were included in the analysis. A total of 69.9% (4,563/6,532) cases received central catheterization. Overall, 38.8% (2,532/6,532), 5.6% (368/6,532), 59.6% (3,895/6,532) and 0.8% (55/6,532)infants received umbilical venous catheters(UVC), umbilical artery catheters(UAC), peripherally inserted central catheters(PICC) and surgical central venous catheters(SCVC). Infants with central catheterization had smaller GA and lower birthweight, and were more likely to be smallforGA, multiple birth, outborn and with 5min Apgar score less than 7, compared to noncatheterized infants (P <0.01). The proportion of pregnancy hypertension, antenatal hormone use, and cesarean section was also higher in the mothers of catheterized infants. The mortality rate did not differ between catheterized and noncatheterized infants but catheterized infants had higher rate of each morbidity and longer hospitalization (P <0.01). Rate of any type of central catheterization generally decreased along with the increasing GA, and the rates were 87.8% and 63.1% for 2428 GA weeks and 2931 GA weeks respectively. Rates of UAC and UVC also decreased along with the increasing GA. Rate of PICC was the highest at 2528 GA weeks, and it was still as high as 52.8% for infants at 2931 GA weeks. The rate of using two or more catheters(UVC/PICC/SCVC) was 29% at 2528 GA weeks, which decreased to 21.8% at 2931 GA weeks. The median duration of UAC, UVC and PICC were 6.0 (IQR: 4.08.0) days, 7.0 (IQR: 4.09.0) days and 22 (IQR: 15.031.0)days, respectively. There were significant variations among NICUs with different hospital types on the catheterization rates and the duration. Response rate of the questionnaire was 91.2% (52/57). Only 50% (26/52) hospitals have hospital guidelines for central vascular catheterization. There were also significant variations on the indications of insertion and removal, and the longest duration as well. A total of 62% (32/52) hospitals had central linerelated leakage and 23% (12/52) had thrombosis within one year. Conclusion：Central vascular catheterization has been commonly used in Chinese NICUs. However, overuse and insufficiency are both existing. Other major problems are the exceeding length of PICC and significant site variation. Guidelines and regulations are on demand. National quality improvement efforts are needed to promote the rationale and standardized use of central catheters for VPIs in NICUs.

Key words: Very preterm infants, Umbilical artery catheter, Umbilical venous catheter, PICC, Central venous cannulation, Duration of catheterization