探讨药物治疗早产儿动脉导管未闭（PDA）的时机。

回顾性分析2016年1月至2018年12月四川锦欣妇女儿童医院新生儿科收治的符合入选标准的早产儿（胎龄<32周）151例。根据布洛芬的使用情况分成3组，其中症状时治疗组43例、症状前治疗组32例、对照组76例。根据是否诊断为有血流动力学意义的PDA（hsPDA），分为hsPDA组和非hsPDA组，症状时治疗组属于hsPDA组，症状前治疗组及对照组属于非hsPDA组。出生第1天检测血常规、生化、血气分析等。出生满3 d筛查超声心动图，记录当日脉压差及尿量。两治疗组，口服布洛芬10 mg/kg，24 h及48 h后各予5 mg/kg，治疗72 h后复查超声心动图。对照组，未予布洛芬治疗，出生后7 d复查超声心动图。复查PDA患儿，出生后30 d再次复查超声心动图。采用χ²检验比较症状时治疗组、症状前治疗组、对照组3组间性别、孕母使用过激素的比例、胎膜早破>18 h比例、剖宫产比例、小于胎龄儿比例、动脉导管7 d关闭率、动脉导管30 d关闭率、脓毒症发生率及呼吸窘迫综合征、脑室内出血、支气管肺发育不良、坏死性小肠结肠炎发生率等的差异；采用单因素方差分析比较胎龄、出生体质量、尿量、脉压差、血pH值的差异；采用Kruskal-Wallis H法比较5 min Apgar评分、正压通气时间、吸氧时间等的差异。

性别、胎龄、孕母使用过激素、胎膜早破、剖宫产、出生体质量、小于胎龄儿，5 min Apgar评分、血pH值、尿量、呼吸窘迫综合征发生率，动脉导管生后30 d关闭率、正压通气时间、吸氧时间，发生脑室内出血、支气管肺发育不良、坏死性小肠结肠炎情况，在症状时治疗组、症状前治疗组、对照组3组之间比较，差异均无统计学意义（P>0.05）。症状前治疗组、症状时治疗组和对照组3组间脉压差分别为（22.13±13.83）mmHg（1 mmHg=0.133 kPa）、（24.24±9.72）mmHg、（16.22±8.81）mmHg，症状前治疗组高于对照组，差异无统计学意义（t=0.732、P=0.639），症状时治疗组高于症状前治疗组和对照组，差异均具有统计学意义（t=3.25、4.710，P=0.002、<0.001）。3组脓毒症发生率分别为9.38%（3/32）、27.91%（12/43）和10.53%（8/76），症状前治疗组低于对照组，差异无统计学意义（χ²=0.033，P=0.856）；症状时治疗组高于症状前治疗组和对照组，差异均具有统计学意义（χ²=5.933、4.230，P=0.015、0.040）。动脉导管生后7 d关闭率，症状时治疗组和症状前治疗组均高于对照组［65.63%（21/32）vs 60.47%（26/43）vs 32.89%（25/76）］，差异均具有统计学意义（χ²=8.524、9.866，P=0.004、0.002）。动脉导管生后30 d关闭率组间差异均无统计学意义（P>0.05）。

口服布洛芬可以促进动脉导管早期闭合，但对远期闭合率并无明显优势。预防性布洛芬干预不能缩短PDA早产儿的正压通气、吸氧时间及降低脑室内出血、支气管肺发育不良、坏死性小肠结肠炎等并发症的发生率。而对hsPDA，布洛芬干预能否减少呼吸支持依赖及并发症的发生仍有待于进一步的研究。

To explore the timing of drug therapy for patent ductus arteriosus (PDA) in preterm infants.

A total of 151 premature infants (gestational age <32 weeks) who were admitted to the Neonatology Department of Sichuan Jinxin Women and Children's Hospital from January 2016 to December 2018 and met the inclusion criteria were retrospectively analyzed. According to the use of ibuprofen, the patients were divided into three groups: symptomatic treatment group (n=43), pre-symptomatic treatment group (n=32), and control group (n=76). According to whether hemodynamically significant PDA (hsPDA) was diagnosed or not, the patients were divided into either an hsPDA group or a non-hspda group. The symptomatic treatment group belonged to the hsPDA group, while the pre-symptomatic treatment group and control group belonged to the non-hspda group. Routine blood tests, blood biochemistry, and blood gas analysis were performed on the first day of birth. Echocardiography was performed 3 days after birth, and pulse pressure difference and urine output were recorded. The two treatment groups were initially given ibuprofen orally at 10 mg/kg, followed by 5 mg/kg at 24 h and 48 h, respectively. Echocardiography was reexamined 72 hours after treatment. The control group was not treated with ibuprofen and echocardiography was reviewed 7 days after birth. Echocardiography was reexamined 30 days after birth in patients with unclosed ductus arteriosus. The chi-square test was used to compare gender, proportion of pregnant mothers who used hormone, rate of premature rupture of membranes >18 h, cesarean section rate, proportion of neonates less than gestational age, closure rates of ductus arteriosus at 7 d and 30 d, and the incidences of sepsis, respiratory distress syndrome, intraventricular hemorrhage, bronchial pulmonary hypoplasia, and necrotizing enterocolitis between the symptomatic treatment group, pre-symptomatic treatment group, and control group. Univariate analysis was used to compare the differences in gestational age, birth weight, urine volume, pulse pressure, and blood pH. Kruskal-Wallis H method was used to compare the difference of 5 min Apgar score, positive pressure ventilation time, and oxygen absorption time.

Gender, gestational age, proportion of pregnant mothers who used hormone, rate of premature rupture of membranes >18 h, cesarean section rate, birth weight, proportion of neonates less than gestational age, 5 min Apgar score, blood pH, urine output, the incidence of respiratory distress syndrome, closure rate of ductus arteriosus at 30 d, positive pressure ventilation time, oxygen time, intraventricular hemorrhage, bronchial pulmonary hypoplasia, and necrotizing enterocolitis did not differ significantly among the three groups (P>0.05). The pulse pressure differences in the symptomatic treatment group, pre-symptomatic treatment group, and control group were (22.13±13.83) mmHg (1 mmH=0.133 kPa), (24.24-9.72) mmHg, and (16.22-8.81) mmHg, respectively; although there was no significant difference between the pre-symptomatic treatment group and control group (t=0.732, P=0.639), the pulse pressure difference in the symptomatic treatment group was significantly higher than those of the pre-symptomatic treatment group and control group (t=3.25 and 4.710, P=0.002 and <0.001, respectively). The incidence of sepsis in the symptomatic treatment group, pre-symptomatic treatment group, and control group was 9.38% (3/32), 27.91% (12/43), and 10.53% (8/76), respectively; although there was no significant difference between the pre-symptomatic treatment group and control group (χ²=0.033, P=0.856), the incidence of sepsis in the symptomatic treatment group was significantly higher than that of the pre-symptomatic treatment group and control group (χ²=5.933 and 4.230, P=0.015 and 0.040, respectively). The closure rate of ductus arteriosus at 7 d was signficantly higher in both the symptomatic treatment group and pre-symptomatic treatment group than in the control group [(65.63% (21/32) vs 60.47% (26/43) and 32.89% (25/76), χ²=8.524 and 9.866, P=0.004 and 0.002, respectively].

Oral ibuprofen can promote the early closure of ductus arteriosus, but has no significant advantage in terms of long-term closure rate. For premature infants with PDA, preventive ibuprofen intervention could not shorten positive pressure ventilation and oxygen absorption time or reduce the incidence of intraventricular hemorrhage, bronchial pulmonary dysplasia, and necrotizing enterocolitis. However, for premature infants with hsPDA, whether ibuprofen intervention can reduce respiratory support dependence and the occurrence of complications remains to be further studied.