Maternal obesity and intrapartum obstetric complications among pregnant women: Retrospective cohort analysis from medical birth registry in Northern Tanzania

Summary Background In the last decade, Tanzania has observed a dramatic increase in overweight and obesity among women of childbearing age, a demographic shift that has been associated with intrapartum obstetric complications in high‐income countries. Similar increases in maternal morbidity including postpartum haemorrhage, hypertensive disorders of pregnancy, and rates of caesarean delivery have not yet documented in Tanzania. This analysis describes intrapartum obstetric complications associated with maternal obesity among pregnant women delivering at teaching hospital in Northern Tanzania. Methods A retrospective cohort analysis was conducted using the hospital's antenatal care (ANC) and birth registries from 2000 to 2015. The World Health Organization (WHO) body mass index (BMI) categories were applied to classify BMI status of pregnant women within 16 weeks of gestational age at their first ANC visit. Relative risk (RR) of obstetric complications with corresponding 95% confidence intervals (CIs) were estimated using multivariable log‐binomial regression, adjusting for clustering effect for the correlation between multiple deliveries of the same woman. Results Among 11 873 women who delivered babies in the hospital during the study period, 3139 (26.5%) fit the definition of overweight and 1464 (12.3%) women with obesity. Compared with women with normal weight, women with obesity were at over 2.6 times at risks of experiencing pre‐eclampsia/eclampsia (RR: 2.66; 95% CI, 2.08‐3.40), pregnancy‐induced hypertension (RR 2.13; 95% CI, 1.26‐3.62), and postpartum haemorrhage (RR 1.22; 95% CI, 1.00‐1.49). Additionally, women with obesity had also higher risk of either elective (RR 2.40; 95% CI, 1.88‐3.06) or emergency (RR: 1.53; 95% CI, 1.34‐1.75) caesarean delivery. Conclusion Maternal obesity is an emerging health problem in Tanzania. This study clearly demonstrates an association between increased risk of intrapartum complications and obesity. A review of guidelines around ANC screening and intrapartum care practices considering BMI, as well as appropriate messages for women with obesity, should be considered to improve maternal and newborn outcomes.


Summary
Background: In the last decade, Tanzania has observed a dramatic increase in overweight and obesity among women of childbearing age, a demographic shift that has been associated with intrapartum obstetric complications in high-income countries.
Similar increases in maternal morbidity including postpartum haemorrhage, hypertensive disorders of pregnancy, and rates of caesarean delivery have not yet documented in Tanzania. This analysis describes intrapartum obstetric complications associated with maternal obesity among pregnant women delivering at teaching hospital in Northern Tanzania.

Methods:
A retrospective cohort analysis was conducted using the hospital's antenatal care (ANC) and birth registries from 2000 to 2015. The World Health Organization (WHO) body mass index (BMI) categories were applied to classify BMI status of pregnant women within 16 weeks of gestational age at their first ANC visit. Relative risk (RR) of obstetric complications with corresponding 95% confidence intervals (CIs) were estimated using multivariable log-binomial regression, adjusting for clustering effect for the correlation between multiple deliveries of the same woman.

Conclusion:
Maternal obesity is an emerging health problem in Tanzania. This study clearly demonstrates an association between increased risk of intrapartum complications and obesity. A review of guidelines around ANC screening and intrapartum care practices considering BMI, as well as appropriate messages for women with obesity, should be considered to improve maternal and newborn outcomes.

| INTRODUCTION
The global prevalence of obesity has increased considerably over the past two decades, and currently, about two billion people are either overweight or obese. 1 Obesity is a rapidly emerging public health problem for women, with around 17.9% of pregnant women in first trimester presently classed with obesity. 2 The current increase of prevalence among women with obesity in Africa is steeper than European and Asian countries. 3 This has a significant public health concern considering the region is also experiencing high prevalence of undernutrition among women of reproductive age. 4 The new analysis based on data from Demographic and Health Surveys (DHS) from 27 countries in sub-Saharan Africa (SSA) show that 20% of women with overweight or obesity. 5 While obesity used to be more prevalent in high-income countries, since the 2000s, an increase in obesity has been documented in low-and middle-income countries (LMICs), particularly in SSA, where the pooled prevalence of overweight/obesity among general population from 32 SSA countries estimated to be 15.9%. 6 Tanzania has observed a dramatic increase in trend of women of childbearing age (WCBA) (15-49 years) who are overweight or obese.
The prevalence of overweight or obesity among WCBA in Tanzania increased from 11% in 1992 to 28% in 2016, with higher urban obesity prevalence than rural settings. 7 Moreover, a household survey conducted in Dar es Salaam revealed an overall prevalence of obesity of 19.2% with women having significantly higher levels of obesity (24.7%) compared with men (9%). 8 Tanzania is among the countries with the highest maternal mortality ratio (MMR) in SSA. Maternal deaths in Tanzania account for a ratio of 556 per 100 000 women in 2015 to 2016 and represent 18% of all deaths of WCBA. 7 The main direct causes of maternal death are haemorrhages, infections, unsafe abortions, hypertensive disorders, and obstructed labours. 9 Previous studies have reported on the association between higher body mass index (BMI) and increased risk of adverse maternal outcomes and obstetric intrapartum complications. 3,10 These complications include hypertensive disorders in pregnancy such as eclampsia/preeclampsia, 11,12 gestational diabetes mellitus (GDM), [11][12][13][14][15][16] thromboembolic disorders, 17 postpartum haemorrhage (PPH), 19,20 and maternal deaths 21,22 as well as services indicative of complications include induction of labour 18 and caesarean delivery. 19,20 In addition, the evidence from 14 countries in SSA shows that obesity in pregnancy is associated with increased risks of adverse labour, child, and maternal outcomes. 2,10 The negative effects of obesity on intrapartum obstetric complications have not yet investigated thoroughly in the LMIC setting.
In Tanzania, for example, no studies have assessed the contribution of overweight/obesity status of the mother on obstetric intrapartum complications. Furthermore, Tanzania has adopted focused antenatal care (ANC) guidelines from World Health Organization (WHO), although there is no portion of ANC service devoted for optimal maternity care (clinical guidelines) specific for screening and care for women with obesity. To address this gap, this study describes intrapartum complications associated with early pregnancy maternal obesity among women delivered at a referral hospital in northern Tanzania. Findings could be useful to policy makers, healthcare providers, and program planners who seek to improve obstetric outcomes in Tanzania.

| Study design, study area, and study population
The study was a cohort analysis using retrospective record review of women who delivered at Kilimanjaro Christian Medical Centre The study population included women who delivered singleton babies at KCMC who had also BMI (both body weight and height) measured at their initial ANC visit. Only women who attended their first ANC visit at most 16 weeks of pregnancy gestational age of were included in the analysis, to exclude women who's BMI might be affected by a later stage of pregnancy. Women with twin pregnancy/multiple gestations (3057) and those referred to the facility for obstetric complications (2319) were excluded from this analysis to minimize overrepresentation of women at high risk of intrapartum complications. Women who delivered more than once at KCMC Hospital during the 15 years period were included in the analysis, and demographic characteristics of these women were described by their first birth. Our final analysed sample was 11 873 of women whom were recorded to have one or more births during the analysis period ( Figure 1). These women had 12 759 additional deliveries in the follow-up period; of them, 11 074 women delivered once, 712 twice, and 87 thrice in KCMC hospital during the analysis period.

| Data sources and data set
We extracted data from KCMC birth registry. The birth registry, which has been operational since 2000, is a computerized database, which records information on all deliveries that take place at KCMC hospital.
Data from ANC cards and exit interviews are entered into the database within 24 hours of delivery. Information used in this analysis (date of first ANC visit, body weight, and height and date of last menstrual period) are drawn from ANC cards, and information on labour and delivery (mother's weight at admission, date and time of delivery, gestational age, plurality, and mode of delivery) comes from medical registry database. The details of data collection methods and tools have been described elsewhere. [23][24][25][26] In summary, the data entered in the registry-included parents' socio-demographic characteristics, maternal health before and during the current pregnancy, complications during labour and delivery, and regarding the mother's previous pregnancies.

| Study variables and variable definitions
The variable on early pregnancy maternal BMI was derived from measured height and weight recorded at the first ANC visit. WHO categories were used to classify BMI: underweight (<18.5), normal weight (18.5-24.9), overweight (25-29.9), and obesity (≥30). Levels of education were classified as no education, primary school, secondary school (both ordinary and advanced level), and postsecondary education.

| Intrapartum complications and mode of delivery
The mode of delivery was classified as vaginal delivery without complications, caesarean section (elective or emergency), or an assisted delivery (by vacuum or forceps). PPH was defined as blood loss greater than or equal to 500 ml within 24 hours, as visually estimated by the midwife after a vaginal delivery. New onset hypertension of systolic blood pressure (SBP) greater than or equal to 140 mmHg or diastolic blood pressure (DBP) greater than or equal to 90 mm Hg after 20 weeks gestation on two separate occasions was classified as pregnancy-induced hypertension (PIH). Eclampsia and pre-eclampsia were analysed in one category of severe hypertensive disorders in pregnancy. Pre-eclampsia was defined as new onset hypertension accompanied by greater than or equal to 300 mg of proteinuria in urine over 24 hours, measured on two occasions, at least 6 hours apart, after 20 weeks of gestation age in a previously normotensive woman.

| Statistical analysis
All analyses were performed using STATA 15 (College Station, Texas, USA). One-way ANOVA for continuous variables and Pearson's chi-square test for categorical variables were used to examine socio-F I G U R E 1 Schematic presentation of the number of deliveries, inclusions, and exclusions during the study period. Data from the Kilimanjaro Christian Medical Centre (KCMC) birth registry T A B L E 1 Demographic and socio-economic characteristics of study sample among women delivered at KCMC Hospital with BMI recorded at ANC before 16 weeks gestational age (N = 11 873) therefore, we did not seek patient consent. To protect women's privacy, confidentiality, and anonymity, we de-identified client identification biodata during analysis.    (Table 3).

| DISCUSSION
The an overestimation of prevalence. The prevalence of obesity observed in our study was also lower compared with the studies in Europe, 19,29 Asia, 30 and Australia. 31 This may concur with the reported upsurge of maternal obesity prevalence in SSA. 2 Differences in prevalence between high-income countries and low-income countries is more likely to be attributed by population health differences and awareness of risk factors of obesity. Highincome countries have facilities and programs to address obesity in both pregnant and nonpregnant women, and some are incorporated into ANC services. 32 The differences might also be attributed the timing of attendance at ANC during the first trimester of pregnancy. Because the current analysis only included women who were less than or equal to 16 weeks of gestational age at first ANC visit, only a quarter (24%; 12 759/53 662) of the total deliveries that occurred at KCMC in the study period were eligible for inclusion. To rule out the differences in our group compared with the whole, we conducted subgroup analysis among the women included and excluded from the study and found no difference in terms of maternal age, education level, parity, area of residence, and marital status.
In our study, women with obesity were at increased risk of eclampsia/preeclampsia. This is similar to previous studies, which F I G U R E 2 Trends in pregnant women delivered at Kilimanjaro Christian Medical Centre (KCMC) by body mass index (BMI) status. Data from the KCMC birth registry reported an association between pre-pregnancy obesity and preeclampsia, reportedly due to high fat composition among women with obesity. 2,18,19,29,31,[33][34][35] These findings underscore the importance of frequent monitoring blood pressure during prenatal/ANC since the first sign of preeclampsia is commonly a rise in blood pressure, calling for strengthening routine blood pressure checks in ANC services keeping BMI categories in mind.
This study revealed that maternal overweight and obesity is a potential screening risk factor for PPH during delivery. This is in agreement with the previous studies, which elicited PPH as a pregnancy complication associated with maternal obesity. 36,37 Whereas a cohort study conducted in Denmark found that PPH was unaffected by BMI, 40 studies from the Netherlands, Brazil, and Oman revealed that women with obesity were more likely to have PPH as compared with women with normal weight. 30,38,39 In this study, we found that women with obesity had increased risk of caesarean deliveries compared with normal-weight women.
Similarly, studies elsewhere have shown that overweight or women with obesity had an elevated risk of both elective and emergency caesarean deliveries. 10,18,19,40,41 This could be explained by comorbid conditions such as diabetes mellitus, high blood pressure, preeclampsia, foetal presentation, and macrosomia of the baby. The higher frequency of emergency caesarean deliveries is a major concern due to complications associated with recurrence of caesarean deliveries in next pregnancies. Although this study conducted at zonal referral hospital, which would be expected to have more complicated cases, which need emergence caesarean deliveries, but during analysis, we tried to minimize selection bias by excluding all referrals. Therefore, guidelines around intrapartum care in Tanzania might be more appropriate to today's obstetric population if they were tailored to address obesity in pregnancy and particularly caesarean deliveries among women with obesity.
Although we did not have data available to identify the primary cause of PPH, pre-eclampsia, and indication of CS, the study findings suggest obesity as one risk factor for these conditions. PPH and preeclampsia are among the major causes of maternal death in majority of developing countries including Tanzania. 6  presented here could be useful for policy makers and practitioners seeking to improve the quality of ANC and intrapartum care in Tanzania or similar LMIC settings that are experiencing increase in prevalence of women with overweight or obesity.

| CONCLUSION
Maternal obesity is an emerging health problem in Tanzania with increased risks of adverse pregnancy outcomes. This has implications around awareness of maternal obesity by both women with pregnancy intentions or who are pregnant, as well as healthcare providers.
ANC services should utilize client information on BMI as part of management of the pregnant women. Additionally, policy makers may consider revisiting or revising ANC and intrapartum care guidelines to reflect additional risk among women with obesity.

ACKNOWLEDGEMENTS
We would like to thank Birth Medical Birth Registry staffs at obstetrics and gynaecology department and Norwegian government who supported the establishment of KCMC medical birth registry.
Our deepest gratitude goes to all of the women delivering at KCMC during the study period for providing their information. We also acknowledge and thank Marya Plotkin for her review of the manuscript.