Introduction
The WHO defines preterm birth (PTB) as births before 37 completed weeks of pregnancy.1 It has been estimated that PTB occurs in 10.6% of worldwide deliveries.2 A recent prospective cohort study showed that PTB incidence in China stood at 5.2% from 2017 to 2018.3 PTB represents the leading cause of perinatal morbidity and mortality.2 4 Furthermore, PTB has been linked to long-term health conditions, such as poorer neurodevelopmental outcomes, higher rates of hospital admissions, and behavioural, social-emotional and learning difficulties in childhood, consuming prodigious health resources and placing a heavy burden on society.5–7
Risk factors associated with PTB include multiple medical, genetic, environmental and socioeconomic factors.4 However, certain results from epidemiological studies have been inconsistent. PTB can be divided into spontaneous preterm birth (SPTB) and iatrogenic preterm birth (IPTB). SPTB refers to the spontaneous initiation of regular uterine contractions during labour and cervical changes, independent of premature rupture of membranes. IPTB, including labour induction and caesarean delivery without spontaneous labour, mainly results from maternal or fetal complications. The proportion of IPTB varies between regions and countries, ranging from 11.8% to 31.7%.8 Existing studies have shown that IPTB is associated with poorer neonatal outcomes than SPTB.9–14 In addition, differences between aetiologies for PTB have been investigated.15 16 Black race, smoking, nulliparity, body mass index (BMI) and prior caesarean birth were significant risk factors for SPTB but had a small or insignificant effect on IPTB. For IPTB, multiple clinical conditions (eg, pre-eclampsia, chronic hypertension) were the primary influencing factors, while other risk factors had much smaller or inconsistent contributions. The above evidence has identified that IPTB and SPTB might represent a different aetiological or causal pathway resulting in preterm delivery. Thus, the failure to focus on the distinction between PTB subtypes may explain the inconsistencies in previous epidemiological studies.
Accurate phenotypic classification is essential to epidemiological study and developing effective methods for preventing PTB. So far, a comprehensive comparison of aetiological heterogeneity between SPTB and IPTB is still lacking. Therefore, the study aims to identify the difference between SPTB and IPTB in terms of the aspects of associated risk factors and pregnancy outcomes. It seeks to investigate the varying degrees of association between multiple risk factors and SPTB versus IPTB, the differences in risk factor profiles for each PTB subtype, and the variations in the severity of adverse pregnancy outcomes between the two forms of PTB.
Materials and methods
Study design and population
A case-control design was used for this study. Patients were recruited at the International Peace Maternity & Child Health Hospital of Shanghai (IPMCH), China, a tertiary obstetrics and gynaecology hospital, which was conducted in October 2020. Inclusion criteria were as follows: patients were singleton natural pregnancies delivery from August 2018 through October 2020, the case group delivered before 37 weeks, and controls were randomly selected at IPMCH by matching the time of first inspection with the cases, with a ratio of 2 controls per case, for deliveries between 370/7 and 410/7 weeks. The enrolled subjects were from Shanghai Province and shared the same ethnic background. Patients who met one of the following criteria were excluded: (1) cases with multiple pregnancies; (2) in vitro fertilisation; (3) stillbirths; and (4) missing information. Finally, a total of 1676 eligible samples were recruited (1189 controls with 487 preterm deliveries). In the case of preterm deliveries, 343 were SPTB and 144 were IPTB. All included samples were neither smokers nor drinkers.
Definition and data collection
The primary outcome was the type of preterm birth: spontaneous preterm birth (SPTB, due to preterm labour with regular uterine contractions and cervical changes, with or without preterm membrane rupture) or indicated preterm birth (IPTB, due to any maternal or fetal medical complication necessitating early delivery). Gestational age was established based on first-trimester ultrasound estimations for the majority of patients and a small minority of patients by last menstrual bleeding prior to pregnancy. The characteristics that were chosen for investigating the relationship between two subtypes of PTB included non-medical and medical factors that were available in the sample: maternal age, education level, fetal sex, BMI at the beginning of pregnancy, previous obstetric history (gravida, para, miscarriage), scarred uterus, myoma of uterus, hypertension, pre-eclampsia, gestational diabetes mellitus (GDM), anaemia in pregnancy, intrahepatic cholestasis of pregnancy (ICP), reproductive tract infections, fetal growth restriction (FGR), fetal distress, oligohydramnios, placenta abnormality(placenta previa, placenta implantation, velamentous placenta and battledore placenta unified), placental abruption, antenatal haemorrhage. Secondary outcomes included newborn birth weight, fetal macrosomia, 1-minute Apgar score, placenta weight and delivery mode.
Statistical analysis
Data were analysed using SPSS V.17 software. Categorical data were compared using either the χ2 test or Fisher’s exact test, and continuous data were analysed with Student’s t-tests. Two multivariable logistic regression models were generated, which included all the variates of non-medical and medical factors mentioned above to calculate ORs of each subtype of PTB for each risk factor. To check for multicollinearity among the factors included, two linear regression models were constructed, and the variance inflation factors (VIFs) for all factors were calculated. The results showed that all VIF values were less than 2, indicating that there is no multicollinearity issue among the included factors. P value <0.05 was considered statistically significant.
