Acute Bilirubin Encephalopathy: A Propos of 151 Cases Collected during a Multicentric Study in Senegal ()
1. Introduction
Jaundice is a yellow discoloration of the mucous membranes and integuments due to hyperbilirubinemia and is the most common symptom in neonatology [1]. In the newborn, when this jaundice is predominantly unconjugated bilirubin, it exposes the child to a formidable complication responsible for a high mortality rate and disabling neurosensory sequelae: bilirubin encephalopathy. This occurs when unconjugated bilirubin not bound to albumin, which is fat-soluble, crosses the blood-brain barrier and attaches itself to the basal ganglia. The clinical manifestations combine a frank, non-cholestatic cutaneous-mucosal jaundice with non-specific neurological signs (disturbed consciousness, lethargy, inability to suck, abnormal movements, muscle tone disturbance, acute plaintive cry, apnea) [2]. Prevention of this complication is based on early detection and appropriate management of rhesus incompatibility in maternity wards by administering anti-D serum. At birth, the management of jaundice by blue light phototherapy is the main therapeutic intervention, with new intensive phototherapy techniques that have significantly reduced the risk of progression to kernicterus. Exchange blood transfusion remains a current therapeutic intervention in severe forms of free bilirubin jaundice [1].
Epidemiologically, the frequency of this neonatal complication has decreased significantly in developed countries in recent decades, thanks to advances in early detection and management, although a small upward trend has been noted recently [3] [4] [5] [6].
In contrast to our low-income countries, global estimates show that the prevalence of acute bilirubin encephalopathy is much higher, as in sub-Saharan Africa (SSA) with higher morbidity and mortality [7] [8] [9]. This much higher incidence can be explained by several factors: poor prevention in maternity wards (anti-D serum), lack of effective strategies for early detection, low awareness of jaundice risks among health care providers and communities, insufficient availability of phototherapy devices in neonatal units and maternity wards...
In Senegal, Faye reported 30 cases at the Centre Hospitalier National d’Enfants Albert Royer, corresponding to 1.14% of hospitalizations in neonatology [10]. However, in recent years, we have observed a resurgence of cases in neonatal units. This is why we undertook to carry out this study, the general objective of which was to determine the frequency of occurrence of this complication and to describe the diagnostic and prognostic aspects in Senegal.
2. Methodology
This was a multicentric study opened to all facilities with a neonatology unit at the national level. The cases were collected in 32 neonatal units of various levels, spread over the 14 regions of Senegal (Table 1). This was a retrospective (January to December 2002) and prospective (January to August 2021), descriptive and analytical, cross-sectional data collection. All newborns hospitalised for acute bilirubin encephalopathy were included in this study. The diagnostic criteria were the presence of jaundice associated with neurological signs (tonus disorders, abnormal movements, convulsions), with no other obvious etiology found (meningitis, asphyxia, inborn errors of metabolism, etc.). Newborns whose records were incomplete were not included. A data collection form was drawn up for this purpose. It was shared with all neonatal units. The parameters studied were: epidemiological data, maternal data, perinatal data, clinical data, biological data, therapeutic data and evolutionary data (see data collection form in the annex). The data were entered and analyzed using SPSS version 23 software. The significance threshold was retained for a p-value < 0.05 (Chi-square test).
3. Results
Epidemiologically: we collected 151 newborns, of which 14 records were not usable. In the end, the analysis concerned 137 patients. The average age at admission was 6 days (1 and 15 days). The majority of newborns (78%) were admitted during the first week of life. There was a male predominance (sex ratio 2.5). The majority of newborns came from the Dakar region (51%).
Regarding maternal data and background: the average age of the mothers was 26 years (13 years and 44 years) and the majority of the mothers were pauciparous (42%).
A family history of neonatal jaundice was found in 59% of newborns.
Regarding perinatal data: seventy-nine patients (57.6%) had a vaginal delivery, 22% were born prematurely and 32% had a low birth weight.
Regarding diagnostic aspects: half of the newborns were from home (50%), 87% were exclusively breastfed and the mean age of the patients at the time of admission was 6 days. Forty patients (33%) had early onset jaundice (Figure 1). The mean time to consultation was 3 days (1 and 12 days). Neurological signs were dominated by: decreased primitive reflexes (74%), abnormal movements (59%) and high-pitched cry (50%) (Table 2). Dehydration (30%) and anaemia (26%) were the most frequent associated signs.
Biologically, the mean bilirubin level was 416 umol/l and the majority (64%) had a bilirubin level above 350 umol/l (Figure 2). The Coombs test was performed in only 8% of the patients and the IAT in 4%. G6PD activity testing was not performed in any of our patients.
The etiology was found in only 72 cases (52%). Neonatal infections (19%) and rhesus (16.7%) and ABO (8.7%) incompatibilities were the main causes found (Table 3).
Table 1. Characteristics of the different structures.
Figure 1. Repair of patients by age of onset of jaundice.
Table 2. Distribution of cases by neurological signs.
Figure 2. Distribution of cases by total bilirubin level.
Table 3. Distribution of cases according to the cause found.
Table 4. Correlation between deaths and signs.
Therapeutic and evolutionary aspects: 45 patients (33%) had received intensive phototherapy, and only 2% had received exchange blood transfusion. Fourteen neonates (10%) did not receive phototherapy because it was unavailable. The case fatality rate was 48.9%. Deaths were significantly correlated with prematurity (p = 0.029). The occurrence of opistotonic hypertonia (p = 0.050) and respiratory pauses (p = 0.037) were significantly found to be poor prognostic signs (Table 4).
4. Discussion
We recorded 151 cases of acute bilirubin encephalopathy. In comparison, the reported prevalence in Western countries are much lower [11] [12] [13]. From 2011 to 2012, the National Reference Centre for Perinatal Haemobiology had identified only 5 cases of kernicterus in Ile de France [12]. In the United States From 1992 to 2004 over 13 years, a Pilot Registry of nuclear jaundice was set up and recorded 125 cases of nuclear jaundice [3]. In Canada, from 1999 to 2000, Suad F. recorded 12 cases of kernicterus [14]. However, global estimates show that the prevalence of acute bilirubin encephalopathy is much higher in low-income countries, particularly in Sub-Saharan Africa (SSA), with a higher morbidity and mortality rate [7] [15]. In Nigeria, one study found a 15.3% incidence of bilirubin encephalopathy in 159 newborns treated for hyperbilirubinemia [9]. This high incidence of kernicterus in our study may have several explanations: poor prevention of rhesus alloimmunisation in maternity wards, lack of effective strategies for early detection of kernicterus in maternity wards, coupled with a policy of early discharge of newborns from maternity wards, poor awareness of the risks of kernicterus among health care providers, insufficient availability of phototherapy devices in neonatal units and maternity wards (Table 1).
There was a male predominance (sex ratio of 2.5). This result is in line with that of several authors [14] [16] [17]. There may be a genetic factor explaining this tendency to male predominance [18].
The majority (70%) of cases were full term newborns and 22% of cases were preterm newborns. But this reflects the fact that detection was not systematically performed in preterm infants. Preterm and low birth weight newborns have an increased risk of developing bilirubin encephalopathy compared to normal birth weight newborns. Indeed, hyperbilirubinemia is more frequent and its evolution more prolonged in the premature newborn compared to the full-term newborn, due to the immaturity of the purification systems, in particular at the hepatic level, and its toxicity for the central nervous system is more marked because the blood-brain barrier is itself more permeable due to the immaturity of the cell junctions. Cases of kernicterus have been reported for low bilirubin levels in premature newborns [19] [20].
The mean age at admission was 6 days of life. This mean age is generally reported as [3] [9]. Jaundice was early in onset in 40% of the newborns and half of the newborns (50%) had come from home. This is generally reported in the literature [3] [14] [17]. This can be explained by a lack of screening and follow-up of neonatal jaundice in neonatology units for multiple reasons: home births, short length of stay in maternity wards, delay between discharge from maternity wards and the first post-natal consultation which often exceeds the first week of life, delay in identification of jaundice in black skin by parents, socio-cultural beliefs, non-availability of neonatology units in some areas. Thus, in our series, the average delay of consultation was 3 days. This figure is comparable to that found by most authors [16] [19].
Neurological signs were dominated by: decreased archaic reflexes (74%), abnormal movements (59%), high-pitched cry (50%). The same signs were described by Faye P. M. in different proportions [10].
The mean bilirubin level was 416 umol/l. Several authors have looked at the indirect bilirubin concentration at which treatment should be initiated [20] - [25]. However, there is not necessarily a correlation between bilirubin levels and the occurrence of kernicterus, especially in preterm infants.
We found only 52% of the causes. These etiologies were dominated by neonatal infections (19%) followed by fetomaternal incompatibilities in the rhesus system (16.7%) and ABO system (8.7%). These same causes were reported by Faye P.M. in different proportions: neonatal infections 33.3%, ABO incompatibilities 33.3% rhesus 20%. [10]. Similar proportions have been reported by several authors in Africa [7] [16] [19]. Haemolysis is the main cause of severe jaundice, whether it is linked to fetomaternal incompatibility or maternal-fetal infection or to a corpuscular anomaly [6] [26] [27]. G6PD deficiency is described as a frequent cause of severe jaundice in developed countries and in the Maghreb [3] [5] [14] [28]. In our patients, few investigations were performed in this sense but there would certainly be cases of G6PD deficiency among the 65 newborns (48%) whose etiologies were not explored. This is certainly due to the insufficient means of investigation in our hospitals, but also to the low socio-economic level.
Concerning the therapeutic and evolutionary aspects: only 33% of patients had benefited from intensive phototherapy against 58% under conventional phototherapy and only 2% from exchange transfusion. In the FAYE study, exchange transfusion was performed in 7 newborns (23.3%) [10]. The low use of intensive phototherapy in our study can be explained by the unavailability of phototherapy equipment in some neonatal units, especially in the periphery, as indicated in Table 1. Exchange transfusion is an intervention that requires a certain technicality, training, blood and pharmaceutical products, and consumables, which are often not available in some centers.
The case fatality rate was 48.9%. This figure is comparable to the 44.8% found by Merine S. [29] and 36.5% by Ogunlesi [30]. In contrast, Diala found 22% deaths in his study [9]. Deaths were significantly correlated with prematurity (p = 0.029).
Preterm newborns are considered to be at increased risk of kernicterus compared to term newborns [31] [32]. Ogunlesi previously reported prematurity, low birth weight and severe anaemia as important risk factors for mortality in patients with bilirubin encephalopathy [29]. This high mortality could be explained by the precariousness of the clinical condition of the premature baby and above all by the lack of equipment necessary for their overall management.
Our study has certain limitations. Our study was hospital-based, which did not allow us to deduce a general prevalence of this complication in Senegal. Nevertheless, we consider that this incidence of cases is excessive in view of the seriousness of kernicterus, if we consider that there is probably an underestimation, as many cases go unnoticed, especially when the newborn is preterm. As a retrospective study, we have not been able to assess the long-term evolution of bilirubin encephalopathy
5. Conclusion
Bilirubin encephalopathy remains a major public health concern, due to the high mortality and numerous neurological sequelae. Infections and fetomaternal incompatibility are the most frequent causes in our context. Progress in terms of early detection and adapted management must be obtained urgently on a national scale.
ANNEX
FILE NUMBER DATE......./......./.......
Survey Sheet
Characteristics of the Structure
• Name: ...............
• Type: ................
• Number of births per year: ...............
• Neonatology capacity: ................
• Phototherapy device:
o Conventional:
Number: ................
Functional number: .................
o Intensive equipment (Tunnel, Reflector) Number: .....................
Functional number: ..................
• Number of paediatricians: ...................
o Number of nurses: ...................
Epidemiological data
• Full name: ..................
• Sex: M F
• Age: ...............
• Telephone: ...............
• Address: ............
Maternal data
• Mother’s age ...............;
• Management ...............;
• Parity ...............;
• History of neonatal jaundice. ...............;
Perinatal data
• Exact wording: ....................
• Prematurity ; Full term ; Post term
• Lower abdomen ; Caesarean section
• Cry at birth: yes no
• Birth weight...............gram
• Breastfeeding: Exclusive breastfeeding ; Exclusive formula ; Mixed
Clinical data
• Mode of admission: Transfer Home
• Data on jaundice
o Age of onset...................days
o Consultation period...................days
• Neurological signs
· Refusal to suckle
· High-pitched cry
· Head rejection
· Upward gaze paralysis
· Breathing pauses
· Apnea
· Convulsions
· Lethargy
· Consciousness disorders
· AR drop ;
· Abnormal movements: Reptation ; Rolling ; Pedalling
· Hypertonia in opistotonos ;
· Hypotonia ;
• Other associated signs:
· Pallor
· Fever
Paraclinical
· Dehydration
· Undernutrition
• Maximum bilirubin...................umol/L
• G6PD assay
• Coombs test
• RAI
Probable or proven cause of jaundice Hemolytic causes
• ABO alloimmunisation
• Rhesus alloimmunisation
• Alloimmunisation in subgroups
• G6PD deficiency
• Pyruvate kinase
• Infectious jaundice
Gluco-conjugation defect
• Mother’s milk jaundice
• Undernutrition
• Prematurity
• Crigler and Najjar
Cause not found
Treatment methods
• Phototherapy
Type of device: Intensive ; Conventional
Duration: ..........................
• Exchange transfusion
• Transfusion
Evolution
Length of stay: ............
➢ Deaths
➢ Survival
Transfer