Original Article | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sudan J Paed. 2020; 20(2): 99-110 SUDANESE JOURNAL OF PAEDIATRICS 2020; Vol 20, Issue No. 2 ORIGINAL ARTICLE Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/YemenHaifa Ali Bin Dahman (1)(1) Associate Professor of Pediatrics, Pediatric Department, Hadhramout University, College of Medicine, Hadhramout Governorate, Yemen Correspondence to: Haifa Ali Bin Dahman Associate Professor of Pediatrics, Pediatric Department, Hadhramout University, College of Medicine, Hadhramout Governorate, Yemen. Email: Bin_Dahman.H [at] hotmail.com Received: 07 December 2019 | Accepted: 25 May 2020 How to cite this article: Bin Dahman HA. Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast, Yemen. Sudan J Paediatr. 2020;20(2):99–110. https://doi.org/10.24911/SJP.106-1575722503 © 2020 SUDANESE JOURNAL OF PAEDIATRICS ABSTRACTPreterm birth (PTB) is a major determinant of neonatal mortality, morbidity and childhood disability. It has long-term adverse consequences for health. The causes of preterm delivery (PTD) are multifactorial. This study was conducted to determine the most common risk factors for PTB at Mukalla Maternity and Childhood (MCH) Hospital, Yemen. A retrospective case-control study was conducted. It involved the records of 100 women with live PTD as cases and 400 women with live term delivery as controls at Mukalla MCH Hospital in 2018. All the data were analysed using the Statistical Package for the Social Sciences (version 24). The statistical relationship between independent variables and PTB was studied by the Chi-square test in bivariate analysis and logistic regression in multivariate analysis. The strength of association was confirmed by using odds ratios (ORs) with a 95% confidence interval (CI). Risk factors with statistically significant association with premature birth were a family history of PTD (adjusted OR [AOR] 2.353; CI 1.3-4.258, p-value 0.005), pre-eclampsia (AOR 4.120; CI 1.818-9.340, p-value 0.001), parity (AOR 2.139; CI 1.249-3.662, p-value 0.006), premature rupture of membranes (AOR 4.161; CI 2.323-7.456, p-value 0.000) and abnormal amniotic fluid volume (AOR 4.534; CI 1.364-15.071, p-value 0.014). An early recognition of preterm risk factors will help medical staff and healthcare workers to identify women with a high-risk pregnancy. KEYWORDSPremature; Preterm; Premature rupture of membranes; Hypertension; Risk factor; Mukalla; Yemen. INTRODUCTIONThe World Health Organisation (WHO) defined the premature birth or preterm birth (PTB) as birth occurring after 20 weeks and before 37 weeks of gestation [1]. It can be further subdivided into 5% extremely preterm (<28 weeks), 10% very preterm (28-<32 weeks) and 85% moderate preterm (32-<37 completed weeks of gestation) [2–5]. PTB is a major public health issue and the most important clinical problem in obstetrics and neonatal medicine [2]. It is considered as the leading cause of high perinatal and postnatal morbidity and mortality in the developed and developing countries [6–10]. Each year, 15 million premature babies are born in the world. Preterm complications are responsible for death in approximately 1 million preterm babies yearly [3]. Prematurity is the second cause of infant deaths after congenital abnormalities [11]. PTB rates varied widely between countries [12] with higher incidences in developing countries than in developed countries [8,12]. At a national level, the estimated PTB rate ranged from about 5%-18%; the highest rates of PTB were in Southeastern Asia and Sub-Saharan Africa (13.5% and 12.3% of all live births, respectively) [12]. An increment of 21% in PTB rates was reported in the United States in comparison to the 1990s [13]. Cerebral palsy, sensory deficit, learning disabilities and respiratory illnesses occur more frequently amongst children born with prematurity [2,4,14]. The morbidity associated with PTB results in enormous physical, psychological and economic costs in later life [15]. Moreover, PTB carries both a financial and social burden on their families. This is related to the expenses from providing intensive care for extremely preterm newborns [13,16,17], prolonged hospital stay and sudden loss of preterm babies [2]. PTB aetiology is incompletely understood. It is thought to involve various obstetric, genetic, environmental and demographic factors that may act independently or interact to cause PTB [8]. The risk of PTB is increased when a combination of two or more factors is present [18]. Most of the risk factors for PTB are modifiable [19]. More than 75% of PTB deaths can be prevented without intensive care [2]. Over the past decade, seven countries have halved their deaths due to PTB by ensuring that the frontline workers are skilled in the care of premature babies and improving supplies of life-saving commodities and equipment [2]. Despite the advances in prenatal medicine and preterm preventive measures, other countries showed an increment in PTB incidence with related morbidity and mortality [20]. The objective of this study was to determine the most common risk factors contributing to preterm deliveries at Mukalla Maternity and Childhood (MCH) Hospital, Yemen. MATERIALS AND METHODSThe study was designed as a retrospective case-control (1:4) study at Mukalla MCH Hospital, Hadhramout Coast, Yemen, during January-May 2018. The records of 591 live-born deliveries in the Gynaecology and Obstetrics Ward, Mukalla MCH Hospital, were divided into two groups. One of these was a case group (mothers with preterm delivery [PTD] and their infants) consisting of 100 preterm deliveries by vaginal delivery or caesarean section (CS), out of a total of 107 preterm deliveries during the study period. Here, seven medical records were excluded due to incomplete data. Preterm labour was approved by their medical care records, ultrasound reports and notes on gestational age assessment of the newborn. The last menstrual period (LMP) was not eligible as this information was not documented in most of the medical records. The other was a control group (mothers with full-term delivery and their infants) consisting of 400 full-term deliveries (gestational age between 37 and 42 weeks) by vaginal delivery or CS, out of a total of 464 full-term deliveries, selected through a simple random sampling during the study period. Here, 20 medical records were excluded due to incomplete data (Figure 1). Data collectionThe study variables collected from the medical reports included background information of the mother (maternal age, education, current occupation, parity and gravidity), current pregnancy details (interval between the birth of the newborn baby and the previous delivery, total number of antenatal care (ANC) visits during the current pregnancy, presence of maternal medical illnesses [hypertension (HTN), diabetes, anaemia], history of vaginal infection, urinary tract infection (UTI), preeclampsia, premature rupture of membrane (PROM), abnormal amniotic fluid volume (polyhydramnios and oligohydramnios), antepartum haemorrhage (placenta previa and abruptio placenta) and mode of delivery). It also included past obstetric history (history of PTD and abortion), family history of PTD and newborn information (foetal position, multiple pregnancy, weight and sex). Figure 1. The study design including the records of 591 live-born deliveries. FTNB, full-term newborn baby; PTNB, preterm newborn baby. Definitions
Statistical analysisThe data were entered and analysed using the Statistical Package for the Social Sciences (version 24). We derived descriptive statistics (frequency and percentage) to determine the distribution of demographic information including maternal age, education and employment amongst case and control groups. The continuous data were compared between the two groups using Mann-Whitney U-test. The statistical relationship between the risk factors and PTB was assessed by applying the Chi-square (χ2) test (bivariate analysis), taking a level of significance of p < 0.05. The strength of association was assessed by calculating the odds ratio (OR) with a 95% confidence interval (CI). The variable was considered to have a statistically significant association with PTB if the χ2 p-value was <0.05, and the CI for the OR did not include the value 1.0. In case of inconsistency between χ2 and the OR, the result of the former was taken. The Fisher’s exact test was considered in the case of independent variables with a small size sample (number <5 in any cell). All variables displaying a significant relationship with PTD (p < 0.05) were entered into binary logistic regression (multivariate analysis). The results were reported as adjusted OR (AOR) and 95% CI along with the p-values. The independent variables with a small size sample were not enrolled in the model. RESULTSIn this study, maternal age for cases ranged between 16 and 45 years with a median (SD) of 25 (6.378) years, whereas, in the control group, the maternal age ranged between 14 and 45 years with a median (SD) of 26 (5.798) years. No statistically significant difference between the median age of the case and control groups (Z: −0.640, p-value: 0.522). Educated mothers form 77.4% of the study sample, 95.8% were housewives and only 4.2% were working mothers. In the case group (mothers with PTD and their infants, Tables 1-3), PTD was common in the maternal age group of 14-25 years (51%) followed by >25-35 years (39%) and >35-45 years (10%). Illiterate mothers constituted 27%, 48% had primary education, 21% had secondary education and 4% had finished college. Only 3% were formally employed. The regular ANC (≥4 times) was recorded in 72% of cases, 38% were primigravida and 62% had a previous pregnancy. Premature delivery was common in mothers with gravida of ≤4 times (79%) and an interpregnancy spacing of ≥2 years (50.8%). Previous premature delivery, previous abortion and recurrent abortion (two or more) were noticed in 10%, 11% and 9% of cases, respectively. A positive family history of premature delivery was detected in 28% of cases. Medical illnesses were recorded in 51% of cases. Chronic HTN was noticed in 18% of cases, and 26% of mothers in the premature group had preeclampsia during this pregnancy. Anaemia and diabetes mellitus were noticed in 32% and 3% of mothers, respectively. UTI (59%) was more common than vaginal infection (45%). PROM was present in 34% of cases, and 24% had an abnormal amniotic fluid volume (7% with polyhydramnios and 17% with oligohydramnios). Placental abnormalities (4%) were also detected with placenta previa in 3% of cases. Vaginal delivery (75%) was more common than CS (25%), and only 11.3% had a history of previous CS. A single foetus was the outcome in 93% of premature deliveries, 6% with twin delivery and 1% had triple babies. The male:female ratio of the premature newborns was 1:1.17 with 94% presenting with cephalic presentation. In the control group (mothers with full-term delivery and their infants, Tables 1-3), full-term delivery was common in the maternal age group of 14-25 years (49%, n = 196), followed by >25-35 years (43%, n = 172) and >35-45 years (8%, n = 32). Illiterate mothers constituted 21.5% (n = 86), 49% (n = 196) had primary education, 23% (n = 92) had secondary education and 6.5% (n = 26) finished college. Only 4.5% (n = 18) were formally employed. A regular ANC (≥4 times) was recorded in 76.8% (n = 307) of mothers, 22.8% (n = 91) were primigravida and 77.3% (n = 309) had a previous pregnancy. A full-term delivery was common in mothers with a gravida of ≤4 times (78%, n = 312) and an interpregnancy spacing of <2 years (50.2%, n = 155). Mothers in the full-term control group had a history of previous premature delivery (3.3%, n = 13), previous abortion (6.5 %, n = 26) and recurrent abortion (two or more) (5.8%, n = 23). A positive family history of premature delivery was detected in 14.3% (n = 57) of mothers. Table 1. Comparison between cases and controls by selected demographic and medical variables.
HTN, hypertension; OD, odds ratio; PTB, preterm birth; χ2, Chi-square. aVariables with statistical significance in relation to PTB. bVariables with cell count less than 5. Medical illnesses were recorded in 36.8% (n = 147) of cases in the control group. Chronic HTN was recorded in 3.8% (n = 15) of mothers, and 5.3% (n = 21) of mothers in the full-term control group had pre-eclampsia during this pregnancy. Anaemia and diabetes mellitus were recorded in 30.5% (n = 122) and 1% (n = 4) of mothers, respectively. Table 2. Comparison between cases and controls by selected maternal past obstetric history.
CS, caesarean section; OD, odds ratio; χ2, Chi-square; yrs, years. aVariables with statistical significance in relation to PTB. bVariables with cell count less than 5. cMissing cases were not included in the total sample count. UTI (51.5%, n = 206) was common than vaginal infection (31.8%, n = 127). PROM was present in 12.8% (n = 51) of mothers, and 9.5% (n = 38) had an abnormal amniotic fluid volume (polyhydramnios in 1.8% and oligohydramnios in 7.8%). Placental abnormalities (2.8%, n = 11) were also detected with placenta previa in five mothers (1.3%). Vaginal delivery (78%, n = 312) was more common than CS (22%, n = 88), and only 9.7% (n = 36) had a history of previous CS. The single foetus was the outcome in 98.3% (n = 393) of full-term deliveries, and 1.8% (n = 7) of mothers had twin delivery. The male:female ratio of the full-term newborns was 1:1, and 94.5% (n = 378) presented with cephalic presentation. Statistical significanceIn bivariate analysis (Tables 1-3), the risk factors for premature delivery considered statistically significant were primigravida mothers (p = 0.002, OR 2), family history of premature delivery (p = 0.001, OR 2.34), previous premature delivery (the Fisher’s exact test 0.008, OR 3.31), presence of medical illness (p = 0.009, OR 1.79), chronic HTN (p = 0.000, OR 5.63), pre-eclampsia (p = 0.000, OR 6.34), vaginal infection (p = 0.013, OR 1.76), PROM (p = 0.000, OR 3.53), abnormal amniotic fluid volume (p = 0.000, OR 3) [polyhydramnios (the Fisher’s exact test 0.011, OR 4.23) and oligohydramnios (p = 0.005, OR 2.44)], and multiple pregnancy (more than one foetus) (the Fisher’s exact test 0.011, OR 0.237). Table 3. Comparison between cases and controls by selected maternal recent pregnancy.
ANC, antenatal care; CS, caesarean section; OD, odds ratio; PROM, premature rupture of membranes; UTI, urinary tract infection. aVariables with statistical significance in relation to PTB. bVariables with cell count less than 5. Other independent variables with no statistically significant association as a risk factor of premature delivery include maternal age, educational level of the mother, occupation, ANC ≤4 times, gravida more than four, interpregnancy spacing <2 years, history of previous abortion, recurrent abortion, anaemia, diabetes mellitus and UTI, presence of placental abnormalities, CS as a mode of the recent or previous delivery, foetal sex and foetal malposition. Table 4 shows the results of logistic regression (multivariate analysis). A family history of PTD (AOR 2.353; CI 1.3-4.258, p-value 0.005), pre-eclampsia (AOR 4.120; CI 1.918-9.340, p-value 0.001), parity (AOR 2.139; CI 1.249-3.662, p-value 0.006), PROM (AOR 4.161; CI 2.323-7.456, p-value 0.000) and abnormal amniotic fluid volume (AOR 4.534; CI 1.364-15.071, p-value 0.014) were all significantly associated with PTD. DISCUSSIONThe data included in the present study revealed that a family history of PTD, preeclampsia, parity, PROM and abnormal amniotic fluid volume were statistically significant risk factors for PTD. To the best of author’s knowledge, this is the first study conducted in Yemen studying a large number of independent variables as risk factors concerning PTB. The risk factors of PTB are multifactorial. Some of which are preventable and curable, whereas, in most of the cases, the exact causes of preterm labour are unknown [18]. Genetic susceptibility may play a key role in the aetiology of PTB [22]. The epidemiological studies demonstrate that mothers, sisters and daughters share the risk for PTB [22]. A family history of PTB is an independent risk factor for PTB in subsequent generations [23,24]. It is not only limited to the mother herself but also can include her other relatives (siblings and sisters) [19,23,24]. In this study, a family history of PTD was considered as a statistically significant risk factor for PTD (OR 2.353, CI 1.3-4.258, p = 0.005). A similar finding was reported by Boivin et al. [25] and Sherf et al. [23]. Table 4. Logistic regression of risk factors affecting PTD.
df, degrees of freedom; Confidence interval; OD, odds ratio. aStatistically significant. In this study, medical illness and HTN (essential and pregnancy-induced) showed an association with PTB in bivariate analysis. Preeclampsia-eclampsia was a significant risk factor for PTB in bivariate analysis (OR 6.34, p = 0.000) and multivariate analysis (AOR 4.120, CI 1.818-9.340, p = 0.001). According to the study of Goldenberg et al. [8], maternal complications related to HTN might be associated with an increased risk of PTD. Bakhteyar et al. [26] and Davies et al. [27] reported a more than four-fold increase in the risk of PTB in mothers with pre-eclampsia. The other studies support the association between pre-eclampsia and PTB [28–32]. A parity as a significant risk factor of PTD was shown in this study (AOR 2.139, CI 1.249-3.662, p = 0.006). This finding was supported by the other studies that reported an association between primiparity and PTB [33–35]. Ananth et al. [36] considered primiparity as an intermediate risk factor for spontaneous and medically indicated PTB. Mayo et al. [37] reported an increased risk of spontaneous PTB in nulliparous teenagers, especially those 16 years or younger. In comparison between nulliparous and parous women, Koullali et al. [38] found that high parity, as well as nulliparity, is considered as a risk factor for spontaneous PTB, whereas Waldenström et al. [39] and Chan and Lao [40] suggested that advanced maternal age influences the risk of PTB regardless of parity. PROMs proceed about 20%-30% of PTBs [5,8]. Mothers with PROM had 5.11 times higher risk for PTB (CI 2.69-9.69, p ≤ 0.001) [26]. This study found that PROM is a statistically significant risk factor for PTD (AOR 4.161, CI 2.323-7.456, p = 0.000). The same finding was supported by the other studies [32,41]. The reverse was shown by Carter et al. [31], who reported no significant association between PROM and PTB (AOR 1, CI 0.91-1.11). In this study, abnormality in amniotic fluid volume was reported in mothers with PTD (24%) and showed a statistically significant association as a risk factor for PTB in both bivariate analysis (p = 0.000) and multivariate analysis (AOR 4.534, CI 1.364-15.071, p = 0.014). Studying the effect of oligohydramnios and polyhydramnios separately on PTB revealed a statistical significance in bivariate analysis ([oligohydramnios (p = 0.005) and polyhydramnios (Fisher’s exact test 0.011)] but not in multivariate analysis. This might be related to the small sample size in this study. Goldenberg et al. [8] reported an association between extremes in the volume of amniotic fluid (oligohydramnios or polyhydramnios) with PTB and PROM. Several studies supported an increased risk of preterm labour with oligohydramnios [28,32,42]. Lo et al. [43] reported an ten-fold increase in the risk for early preterm labour in women with oligohydramnios. On the contrary, Ahumada-Barrios and Alvarado [30] reported no significant association between oligohydramnios and PTB. In a study carried by Many et al. [44], preterm labour was related to the underlying cause of polyhydramnios rather than the excess fluid amount. Mahapula et al. [45] reported an increased risk of PTD in women with polyhydramnios (AOR 8.3, CI 1.7-40.2, p-value 0.008). Strength and limitationsThe strength of this study relies on being conducted in Mukalla MHC Hospital which is the oldest hospital in the area and considered as a referral hospital for high-risk pregnancies and women with complications. Another supporting point is the lack of similar previous studies conducted in this area. Hence, the results extend knowledge on the common risk factors influencing PTD in Mukalla. Several limitations should be considered when the results of this study are interpreted. The first main limitation of the present study was the use of retrospective cross-sectional data to determine the influence of risk factors in PTD. Hence, further longitudinal analyses with larger samples will provide stronger evidence in this topic and allow a better generalisation of the results to other populations. Second, the rate of PTB was underestimated as the study was conducted in one hospital, and stillbirths were excluded from the study. Deliveries in private hospitals and home deliveries were not included. Third, the risk factors such as maternal malnutrition, presence of congenital infection, uterine anomalies, use of herbal therapy, history of abdominal trauma and medical illnesses as bronchial asthma, heart disease and renal disease were not assessed either due to the lack of information in the records or due to low prevalence amongst the studied population. Finally, gestational age assessment was based on data from medical records, ultrasound reports and notes on gestational age assessment of the newborn. This is because the date of the first day of the LMP was not included in most of the medical records. CONCLUSIONSIn this study, a family history of PTD, pre-eclampsia, parity, PROM and abnormal amniotic fluid volume are the risk factors for premature delivery. Recognising the most common risk factors for PTD will help to increase the awareness about high-risk pregnancy, improve the preventive measures of preterm risk factors and modify preterm care protocol in nurseries. CONFLICT OF INTERESTThe authors declare that there is no conflict of interest regarding the publication of this article. FUNDINGNone. ETHICAL APPROVALThe study was approved, and an ethical clearance was given by the Ethics Committee of Hadhramout University College of Medicine (HUCOM), Yemen. The study is a retrospective (record-based) case-control study. No names were used for data collection. Participants’ consent is not required according to the approved research guidelines of HUCOM, and confidentiality was ensured at all levels. REFERENCES
|
How to Cite this Article |
Pubmed Style Haifa Ali Bin Dahman. Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. Sudan J Paed. 2020; 20(2): 99-110. doi:10.24911/SJP.106-1575722503 Web Style Haifa Ali Bin Dahman. Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. http://www.sudanjp.com/?mno=77001 [Access: January 22, 2021]. doi:10.24911/SJP.106-1575722503 AMA (American Medical Association) Style Haifa Ali Bin Dahman. Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. Sudan J Paed. 2020; 20(2): 99-110. doi:10.24911/SJP.106-1575722503 Vancouver/ICMJE Style Haifa Ali Bin Dahman. Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. Sudan J Paed. (2020), [cited January 22, 2021]; 20(2): 99-110. doi:10.24911/SJP.106-1575722503 Harvard Style Haifa Ali Bin Dahman (2020) Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. Sudan J Paed, 20 (2), 99-110. doi:10.24911/SJP.106-1575722503 Turabian Style Haifa Ali Bin Dahman. 2020. Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. Sudanese Journal of Paediatrics, 20 (2), 99-110. doi:10.24911/SJP.106-1575722503 Chicago Style Haifa Ali Bin Dahman. "Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen." Sudanese Journal of Paediatrics 20 (2020), 99-110. doi:10.24911/SJP.106-1575722503 MLA (The Modern Language Association) Style Haifa Ali Bin Dahman. "Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen." Sudanese Journal of Paediatrics 20.2 (2020), 99-110. Print. doi:10.24911/SJP.106-1575722503 APA (American Psychological Association) Style Haifa Ali Bin Dahman (2020) Risk factors associated with preterm birth: a retrospective study in Mukalla Maternity and Childhood Hospital, Hadhramout Coast/Yemen. Sudanese Journal of Paediatrics, 20 (2), 99-110. doi:10.24911/SJP.106-1575722503 |