Birth asphyxia in Northern Uganda: Associated factors, intrapartum practices and the predictive value of lactate

Abstract

Introduction: Birth asphyxia is responsible for approximately 24% of all neonatal deaths globally, with an even higher share in Uganda. Birth asphyxia is associated with immediate neurological and long-term neurodevelopmental complications among newborns who survive which affects their quality of life and places a big burden on families and society. There is limited literature about birth asphyxia and intrapartum care interventions in Uganda. This study highlights the prevalence and associated factors of birth asphyxia, intrapartum care practices and the predictive value of umbilical artery lactate for adverse outcomes. Methods: Quantitative and qualitative approaches were used. For sub-study I, a cross-sectional descriptive study design was used to determine the prevalence and factors associated with birth asphyxia in 2930 women. To determine the predictive value of umbilical artery lactate for adverse newborn outcomes (sub-study IV) a diagnostic accuracy study design was done among 2655 newborns. For sub-study II, a sequential explanatory mixed methods study design was used to describe the foetal heart rate monitoring practices of health care workers. A qualitative descriptive study design was used to explore health care workers’ experiences of managing foetal distress and birth asphyxia (Sub-study III). Women presenting in labour at Gulu Regional Referral Hospital and St. Mary’s Hospital Lacor in Gulu district were screened for inclusion in sub-study I and IV. The principal investigator and research assistants attended births to determine the Apgar score at 1 and 5 minutes of birth, and also obtained a blood sample for umbilical artery lactate measurement. The newborns were observed for one hour to record the outcomes. Foetal monitoring practices by health care workers were studied using reviews of 251 maternity records, 42 observations of monitoring practices for women in labour, 11 interviews and 2 focus group discussions with health care workers. To document experiences in managing foetal distress and birth asphyxia, interviews were conducted with 16 health care workers at maternity units in different health facilities in Gulu District. The prevalence of birth asphyxia was determined as a frequency of newborns who had an Apgar score <7 at five minutes out of the total number of live births. Factors associated with birth asphyxia were determined using logistic regression analysis. Receiver operator curves (ROC) were used to determine the diagnostic accuracy of umbilical artery lactate. Content analysis was used to summarise qualitative data for both sub-study II and III. Results and findings Sub-study I: A total of 2,930 mother-newborn pairs were included, and the prevalence of birth asphyxia was 154 [5.3% (95% confidence interval: 4.5- 6.1)]. Factors associated with birth asphyxia were; maternal age ≤19 years [adjusted odds ratio (aOR) 1.92 (1.27-2.91)], syphilis infection [aOR 2.45(1.08-5.57)], and a high white blood cell count [aOR 2.26 (1.26-4.06)], while employment [aOR 0.43 (0.22-0.83)] was protective. Additionally, referral [aOR1.75 (1.10-2.79)], induction/augmentation of labour [aOR 2.70 (1.62-4.50)], prolonged labour [aOR 1.88 (1.25-2.83)], obstructed labour [aOR 3.40 (1.70-6.83)], malpresentation/ malposition [aOR 3.00 (1.44-6.27)] and assisted vaginal delivery [aOR 5.50 (2.30-13.30)] were associated with birth asphyxia. Male newborns [aOR 1.92 (1.28-2.88)] and those with a low birth weight [aOR 2.20 (1.07-4.50)], were also more likely to develop birth asphyxia. Sub-study IV: The median umbilical artery lactate levels were 3.5 mmol/l (Inter-Quartile Range (IQR) 2.7-4.7). Newborns with high lactate levels were more likely to have adverse outcomes. The area under the curve was consistently above 80% for all the outcomes assessed (Apgar score<7 at 5 minutes, resuscitation with bag and mask and admission to neonatal care unit). The umbilical artery lactate cut-off value of 5.5 mmol/l was associated with the best area under the curve values of between 76% to 83% for the four adverse outcomes. Sub-study II: According to record reviews and observations, almost all health care workers examined the foetal heart rate at admission. The number examined dropped during monitoring in the first and second stages of labour. Whereas the records indicated that 76.8% of women were monitored, the observations showed only 68.0%. The records showed that foetal heart was monitored every 30minutes (IQR 30-120) but an interval of 139 minutes (IQR 87-662) was observed. Barriers to foetal heart rate monitoring were both institutional and health worker related. Health care workers acknowledged that their practice was different from the recommended guidelines. Sub-study III: Health care workers knowledge of managing foetal distress and birth asphyxia was variable, with a few misconceptions such as wrong cut-offs of FHR in the definition of foetal distress and the perceived need to use drugs during newborn resuscitation. Health care workers find themselves in a difficult situation with inadequate resources, and internal pressures that affect their ability to offer good quality care. The lack of resources included; oxygen for use during neonatal resuscitation at Health centre III, a special care area for newborns with birth asphyxia at Health centre III and IV and blood for transfusion during caesarean section at Health centre IV and the hospital. Challenges with referral included inadequate resources at each level of care and problems with transportation. Health care workers also felt under pressure to perform beyond their abilities due to high expectations from women and their surrounding community. The findings also show that women in labour were blamed and abused by relatives when they gave birth to newborns with asphyxia. Conclusions: One in 20 newborns had birth asphyxia. Factors associated with birth asphyxia included: maternal socio-demographic characteristics such as young maternal age, intrapartum complications and newborn characteristics such as low birth weight and male sex. We recommend that women at risk of birth asphyxia be given priority for close monitoring and emergency obstetric care. Umbilical artery lactate was found to be a good predictor of adverse newborn outcomes and should be considered as a complementary test during diagnosis of birth asphyxia at a cut off of 5.5mmol/l. Foetal heart rate monitoring by the health care workers was found to be sub-optimal with discrepancies between what was being done and what was actually documented. The Ministry of Health should endeavour to provide the required monitoring devices, practice guidelines and regular training.