Epidemio-Clinical Characteristics of Perinatal Anoxia and Immediate Outcome of Patients at Hospital Teaching Gabriel Touré of Bamako

Introduction: Neonatal asphyxia (NA) is one of the most likely causes of neuro-developmental abnormalities in children. In Mali it is responsible for half of the early deaths and the third of neonatal mortality. Updated data would help understand and improve intervention strategies to reduce mortality. Objective: It is the study of epidemiological and clinical characteristics, the immediate outcome and the factors associated with newborn (NB) mortality with NA. Material and Methods: This was a prospective cross-sectional study from June 27 th to September 3 rd 2016 about the NBs admitted for NA in the Hospital Teaching Gabriel Touré of Bamako. The clinical and biological data including the prognosis were collected from the


Introduction
Neonatal asphyxia is defined by the World Health Organization (WHO) as the failure to establish or initiate normal breathing at birth [1].
It is characterized by metabolic acidosis with an arterial pH of less than 7 and a basal deficit greater than or equal to 12 mmol/L. Its occurrence is related to a severe alteration of uteroplacental gas exchange leading to fetal hypoxia and immediate respiratory acidosis and metabolic acidosis, reflecting an alteration of cellular metabolism. It combines clinical and biological criteria such as cardiorespiratory and neurological depression (APGAR score < 7 in the fifth minute life) and evidence of acute hypoxia responsible for severe acidemia (arterial blood pH < 7 and/or baseline deficiency ≥ 12 mmol/l) [2].
WHO estimates that four million children born with NA other the one hundred and thirty million births annually. Among them one million die, and nearly the same number survive with severe neurological sequelae [1]. NA is the third leading cause of neonatal mortality worldwide (23%), after low birth weight (30%) and neonatal infections (25%) [1]. In the United States of America, the incidence is 6‰ live births [3], while in Africa, its overall rate would be 42‰ [4].
In Cameroon, the Demographic and Health Survey (DHS) in 2011 reported an incidence of 31‰ births [4]. In Burkina Faso, the intra-hospital incidence of perinatal anoxia (PA) was 19.8% [5]. In Mali, PA accounted for 42% of early neonatal mortality and 30% of the overall mortality [6].
The NA is one of the most likely causes of neuro-developmental abnormalities in children [7]. In France in 2009, Meau-Petit et al. pointed out that 40% of motor disabilities of cerebral origin (MDCO) were attributable to cerebral anoxia [8]. In Cameroon, Mbonda et al. in 2010 observed that 43.68% of MDCOs were due to the NA; and these MDCOs were associated in 41.5% with epilepsy [9].
If the part of neonatal anoxia in neonatal mortality is known in Mali [6], however, its epidemio-clinical characteristics had never been studied before.
This work, which is unique in this domain in Mali, will permit to understand perinatal anoxia in Mali and to improve its management.
The objective of this work is to study the epidemiological and clinical characteristics of PA, as well as the immediate outcome of newborns included.

Material and Methods
This was a cross-sectional prospective study, which was conducted from June It had included all neonates from 0 to 72 hours of life who have been hospitalized for NA for at least 37 weeks.
The diagnosis of neonatal asphyxia was essentially based on clinical criteria: neurological abnormalities at birth, an APGAR score < 7 at 5 minutes of life and a neonatal resuscitation at birth.
The severity of asphyxia was measured according to the classification of SARNAT (SARNAT I, II, III) [10]: -Grade I or minor encephalopathy with resolved moderate hypotonia and hyperexcitability in less than 48 hours; -Grade II or moderate encephalopathy leading to convulsions, impairment of consciousness, tone and abnormal movements as well; -Grade III or severe encephalopathy corresponding to a deep coma with loss of brainstem reflexes. Neonatal infection was defined from the following arguments: -Anamnestic: maternal fever per partum, stained amniotic fluid, prolonged rupture of membranes for more than 12 hours, chorioamnionitis; -Clinics: refusal of suckling, digestive disorders, thermal instability, apnea-bradycardia-desaturations syndrome, respiratory distress, tachycardia; -Biological: Increase of C-Reactive Protein (CRP) above 20 mg/L, signs of infection on Complete blood count (CBC); to bacteriological by the positivity of the blood culture. Respiratory distress was diagnosed: in the presence of signs of respiratory control, apnea, tachypnea with respiratory rate above 60 cycles per minute, bradypnea under 30 cycles per minute.
The parameters studied were: -The mother's data: age, parity, gravidity, level of schooling, occupation, obstetric events prior and post partum. -Newborn data: the term, the chronological age, the sex, the course of delivery, APGAR, the resuscitation, clinical and biological data, transfer. The data were collected from the pregnancy monitoring records, the liaison sheets in the case of transfer, from the parents/caregivers and the medical register.
This study was conducted in accordance with the ethical principles of Health Act of Mali concerning research. Permission to conduct the study was granted by the local research committee of our institution. Parents' informed consent was obtained before the inclusion of newborns.
The collected data was recorded and analyzed with the software Epi info version 3.5.1. The statistical tests of Chi squared student were used to compare the qualitative results, the significance level was less than 0.05.

Results
Out of 324 hospitalized newborns hospitalized during the study period, 76 were Open Journal of Pediatrics for a PA, or 23.45%. There were 50 male newborns (66%) and 26 females (34%), a sex ratio of 1.94. The majority, 68 newborns were admitted in less than 24 hours of life (89.5%), the average weight was 2876 ± 689 g with extremes (1600 and 4000 g), the average height was 50 ± 2.9 cm with extremes (43 and 58 cm), the cranial perimeter was 34.11 ± 1.8 cm with extremes (29 and 38 cm). The majority of newborns were referred (78.5%), those of CHU Gabriel Toure maternity unit accounted for (18.9%) and those who come by themselves were (2.6%).
Cranial ultrasonography performed in 24 neonates (31.6%) was normal in most cases 79.2%, however an increase in the index of resistance of the cerebral artery was observed in 4 cases.
The average hospital stay was 5.6 days with extremes ranging from (1 and 16 days). The outcome at discharge was favorable in 51 patients (67.1%). We have recorded 25 deaths (32.9%).
The evolution according to the stage of encephalopathy and the score relationship of Sarnat-Death is recorded in (Figure 2).

Discussion
Perinatal asphyxia is a relatively frequent situation, which is serious because of the cerebral pain it causes. It is one of the most common causes of neuro-developmental abnormalities in children [11] and a leading cause of death [12].
The PA may result in multi-visceral failure or anoxic-ischemic encephalopathy (AIE) [13]. and the most represented group was between 2500 and 4000 g [8]. In our series the male predominance is substantial with 66% (sex ratio: 1.9). This result is similar in all the studies we have gone through [6] [8]. According to Badawi [21], male sex increases the risk of perinatal asphyxia by 50%. Johnston reports that sex hormones, including estrogen, would be protective against anoxic-ischemic lésions [22].
The young age of mothers, age group or fertility rate was highest and primiparity was significantly associated with the PA. Because of their inexperience, they would tend to be less able to monitor their pregnancy [19]. In our series, the mean age of mothers was 24.17 ± 5.5 years with extremes (16 and 44 years), near half of our mothers 48.6% (n = 37) were primiparous. A maternal age under 25 was also found by Aslam et al. in Pakistan in 2012 [23]. Rehana et al. in India, on the other hand, found that Asphyxia increased with maternal age > 35 years [24].
The WHO recommends a minimum of 4 ANC for the proper monitoring of pregnancies. The 4 repositioned ANC must be well planned over time with specific objectives. They enable not only the prevention, detection and early treatment of complications, but also the preparation of childbirth and the promotion of health [25]. Additional consultation may be required in case of complications or specific need for control.
In our study, mothers performed more than 3 ANC in 45. The low frequentation of health facilities during pregnancy in our study can be explained by the illiteracy of mothers (54%), the difficult socio-economic conditions and the absence of a generalized health insurance system.
Capacity building for obstetricians and midwives to improve the quality of pregnancy monitoring and delivery management are key levers for preventing PAs. In our study more than half of newborns 76% (n = 58) were born vaginally, among which 9% by instrumental low way (forceps 3%, suction cup 6%). Caesarean birth accounted for 24% (n = 18). This rate of caesarean section is comparable to the rate obtained by a Congolese study (25.9%) [16], but higher than that found by Daniel, K.K. et al., Cameroon (18.6%) [8]. The vaginal delivery was not a risk factor in the study in Benin, unlike cesarean delivery [19].
Postnatal neurological assessment is fundamental because the importance and duration of neurological signs are the best long-term prognostic criteria. In our study, clinical symptomatology was dominated by neurological signs (94.7%), followed by respiratory and hemodynamic signs with 81.6% and 52.6%, respectively. Neurological signs frequently found were disruption of archaic reflexes (86.8%), hypotonia (57.9%) and convulsion (13.2%). In the series of Ouedrago S.O. et al. the main neurological manifestations pointed out were archaic reflex anomalies (92.9%), hypotonia (88.2%) and convulsions (35.3%) [6]. This observation is noticed by many other African authors [16].
Neonatal resuscitation is an important intervention to improve the prognosis of children with NA. Depending on the time of prenatal asphyxia, per natal or post natal, the causal pathology, the time of the beginning of resuscitation, the technical platform, neonatal resuscitation may be effective or not to prevent complications in children. In our study resuscitation at birth involved 75.7% of newborns, the average duration was 16.6 minutes with extremes of (5 mn and 30 mn). In the series of Ouedrago S.O. et al., resuscitation concerned all newborns at birth with an average duration of ten minutes [6]. In Congo, Okoko A.R. et al.
found out an average duration of 4.3 minutes with extremes of (1 min and 26 mn), in 62.5% [16].
Perinatal asphyxia is responsible for poor adaptation to ectopic life with an Apgar score < 7 in the fifth minute, but this sign is not specific. This score makes it possible to assess the immediate adaptation of the newborn at birth [6] [26]. A child with an Apgar score ≥ 7 at 5 minutes of life got adapted well. The Apgar rating, a simple and available descriptive element, is the indispensable tool for diagnosing PA in Africa. In view of these advantages, the rating of the Apgar score should not be neglected by healthcare providers [6] [26]. In our study Apgar score was not made in nearly half of newborns. In 24% of cases it was less than 7 in the 5th minute. This result is lower than that of Danièle, K.K. in Cameroon 64% [6]. It suggests that 76% of our newborns with a PA had an Apgar score at birth that did not reflect central nervous system depression.
Most newborns with a score below 4 to 5 minutes developed encephalopathy, 14% for those with an Apgar score between 4 and 6. It should be noted that a low Apgar score of 5 minutes is correlated to the immediate future of the child (risk of encephalopathy) but this score has a rather low prognostic value for long-term outcome [12].
The neonatal encephalopathy testifies in case of asphyxia of the cerebral anoxia. The most widely used clinical and prognostic classification is Sarnat classified in three grades.  [28].
The management of the NA includes several aspects such as: medication management, a good hydro-electrolyte balance, the correction of metabolic disorders, respiratory assistance, parenteral nutrition and even therapeutic hypothermia may be necessary [29]. This management can only be done on the basis of significant clinical, biological and neuroradiological monitoring. In our context, the means of treatment and surveillance remain very limited hence the high mortality. Prevention therefore remains an essential element. It involves well-monitored ACs, adequate fetal monitoring, early management of the NA and the mastery of neonatal resuscitation techniques in the delivery room [8].
Beyond its high mortality, the NA is one of the most likely pathologies of neuro-developmental sequelae in children [11]. The spectrum of these developmental abnormalities is large and can be evaluated by several scores among them Denver score [30]. The Denver score allows health professionals to assess the development of children's psychomotor skills between birth and 6 years old.
This study was a prospective study about perinatal anoxia and included only newborns who met the clinical criteria of perinatal asphyxia that is the Apgar score. The non-availability of PH in the cord and basic defict in our structure has not permitted us to master these parameters in our patients. That could, in fact, make the number of diagnosed infra-clinical cases underestimate in our cohort. New studies, taking also into account, the biological criteria of anoxia, could fill up this gap of selection.

Conclusion
The PNA is a public health issue in Bamako because of its frequency and severity as shown by the high morbidity and mortality observed in the neonatology unit.
In a context of low resources, it is necessary to focus on prevention by monitoring pregnancies, delivery assisted by skillful personnel, mastery of neonatal resuscitation techniques in the delivery room. The evolution of PA is closely related to the severity of encephalopathy.