In the whole pediatric ward of CHU-P, 24,597 children were admitted during the study period, including 6238 newborns. Among the latter, 6224 had a medical record. Among those 6224 medical records, 20 did not get information on exclusion modalities. In total, the study focused on 6204 newborns who were live babies admitted to the unit and with usable medical records. Newborns thus accounted for 25.2 % of admissions in the ward.

The mean age for mothers was 26.3 ± 9.1 years. Mothers aged 20 to 35 years were the predominant age group (81.1%). Among the mothers, 3641 were housewives (58.7%). Among all the newborns, 3530 were boys (56.9%) and 2674 were girls (43.1%) i.e. a sex ratio of 1.32. Mean age for newborns on admission was 6.01 ± 5.39 days. The newborns admitted were aged 24 hours and younger in 73.0% of cases. Four thousand three hundred and four (4304) newborns were born in CHU-P maternity i.e. a frequency of 69.4 % and 4238 (68.3%) had a presumed date of childbirth between 37 and 41 weeks of amenorrhea. Mean birth weight of newborns was 2608.85 ± 727.67 g. Table 1 shows the distribution of newborns admitted from 2010 to 2016 to the CHU-P neonatal care unit according to sociodemographic, obstetrical and fetal characteristics.

We note an overall trend of increase in admissions over the years until 2013; it was followed by a decline. Figure 1 illustrates the chronological trends in admissions and deaths during the study period.

As regards diseases and syndromes concerning all the newborns, 3357 had a clinically obvious neonatal-infection (54.1%), 2158 were premature (34.8%), and 1863 had perinatal asphyxia (30.0%). Table 2 shows the distribution of newborns admitted from 2010 to 2016 to the CHU-P neonatal care unit according to diseases and syndromes.

Among the 6204 newborns admitted to the CHU-P neonatal care unit from 2010 to 2016, 1045 died i.e. gross mortality of 16.8%. Over the years, the following death rates were recorded: 18.4%, 21.2%, 21.3%, 13.6%, 13.9%, 16.8% and 15.1% respectively in 2010, 2011, 2012, 2013, 2014, 2015 and 2016. Table 3 indicates the trends in admissions, deaths and death rates in the CHU-P neonatal care unit from 2010 to 2016.

Deaths occurred in 62.1% of cases in the first 24 hours of life. Figure 2 shows the distribution of newborns deaths in the unit according to their age.

As presented in the Table 4, the factors associated with neonatal death, were the mothers’ age (<20 years), the mothers’ occupation (housewife, /craftwoman), the newborn sex (male), the newborns’ age (>1 day), the gestational age (<37 weeks), and the birth weight (<2500 g).

*Reference terms.

Figure 2. Distribution of newborns deaths in the CHU-P neonatal care unit from 2010 to 2016 according to their age (N = 6204).

During the study period, intra-hospital-based neonatal morbidity accounted for 25.2% of general morbidity in the pediatric ward. This rate indicates the importance of neonatal care unit in that second referral ward. It points out the level of demands of neonatal care by the beneficiary population. This rate is higher than the one found out by Noudamadjo et al. in a similar study carried out from 2002 to 2009 in the same ward which was estimated at 17.8% [4]. Therefore, there is an upward trend in neonatal care needs expressed by the population.

The most common diseases were respectively neonatal infections (54.1%), prematurity and low birth weight (34.8%) and perinatal asphyxia (30%). Neonatal infections were the most common disease, with 54.1% of admissions. This frequency of neonatal infections may be due either to the poor follow-up of pregnancy and lack of asepsis during childbirth or to manipulations during neonatal sepsis. For instance, in addition to classical maternal and fetal infections, the traditional practices which consist in having the umbilical cord fall down before the 72nd hour of life, would predispose those newborns to get infections which start from the umbilical cord. These findings are identical to the ones reported by Noudamadjo et al. between 2002 and 2009 in a study conducted in the same unit which showed that 57.8% of the hospitalizations were decided for neonatal infections [4]. The predominance of neonatal infections among newborn diseases has been reported by several African authors with rates of 62% according to Kedy Koum et al. and 63.7% according to Thomson et al. [5] [6]. Other authors reported lower rates. Those authors were Nagalo et al. who found 23.5% and Akaffou et al. 30.3% [7] [8]. Prematurity and low birth weights were the second cause of hospitalization. This part of prematurity in the structure of diseases raises many problems such as content and quality of care inputs ranging from health staff to consumables trough equipment. That equipment is particularly insufficient and even non-existent in our context, thus predisposing the extremely premature newborns to a very high risk for death. This result is similar to the one of Kedy Koum et al. in Douala (28.9%), the one reported by Azoumath et al. in Togo (22.2 %), and also similar to the one reported by Bobossi in the Central African Republic (37.1%) [5] [9] [10]. Noudamadjo et al. (27.57%) found a rate close to the one of this study [4]. In contrast, Nagalo et al. in Burkina Faso found a lower rate (17.8%) [7]. Perinatal asphyxia was also a common cause of admission to the unit. It was the third cause of hospitalization. For instance, 30% of admissions were decided for perinatal asphyxia. This high rate of perinatal asphyxia raises the problem of the quality of perinatal medicine in Benin. In fact, the prevention of asphyxia depends on the good follow-up of pregnancy, particularly in the last quarter, a systematic and proper actual use of the partograph in the management of labor and on good neonatal resuscitation according to the concept of Help Babies Breathe (HBB) if necessary. Nagalo et al. reported a rate lower than the one of this study (20.9%) [7], as well as Thomson et al. in South Sudan (4.7%) [6]. As regards Kedy Koum et al., they reported a rate close to the one of this study (27.4%) [5], whereas Noudamadjo et al. reported a higher ratio (42%) [4].

During the study period, the gross rate of neonatal mortality in the CHU-P was 16.8%. This high rate may be due to the method of recruitment of newborns. This rate is lower than the one found by Noudamadjo et al. (21.1 %) [4]. Kouéta et al. had found a neonatal mortality rate close to the one of this study (15.3%) [11]. Nagalo et al. in Burkina Faso (13.1%) as well as Noria et al. in Algeria (5.3%) reported rates lower than the one of this study [7] [12]. In the West African sub-region, in Nigeria for example, Omoigberale et al. reported in 2010 a neonatal mortality of 20.3%; Mabiala-Babela et al. reported in 2009 in Congo neonatal mortality of 36.8% [13] [14]. The death rate has gradually declined from 21.3% to 13.6% during the seven years of the study. The two extremes were observed in 2012 with the highest death rate (21.3%) and in 2013 with the lowest rate (13.6%). That gradual decline of the death rate during the study may be due to the sustained and continued efforts of the medical staff to improve care provision. Actually, hygiene was improved within the unit since 2009 through the systematization of the use of disposable diapers for any newborn who was therein admitted. In addition, the training of inmate trainees and maternities’ staffs in emergency obstetrical and neonatal care (EmONC) contributed to reducing that neonatal mortality rate. Within the CHU-P neonatal care unit, 77.9% of neonatal deaths were recorded during the first week of life. Deaths in the first 24 hours of life accounted for 62.1% of cases. This result is higher than the one reported by Noudamadjo et al. who had found 75% of neonatal mortality occurred in the first week of life and 48.3% in the first 24 hours of life [4]. Kouéta et al. in Ouagadougou in 2007, Noria et al. in Algeria in 2015 and Sylla et al. in Mali in 2009 reported neonatal mortalities in the first week of life respectively estimated at 97.3%, 96.85% and 91.10% [11] [12] [15]. These results give rise to many reflections. At first sight, one would think that newborns were admitted in a very precarious health condition; this raises the issue of management of pregnant women, labor, and newborns within the community before admission; each of those three components should be improved to reduce neonatal mortality. Then, despite the precarious health condition of newborns on their admission, if the installations, the medical and technical equipment, the consumables, the drugs, and staff professional skills were at a high level, a certain number of dead newborns could have recovered. This raises once again the issue of financing of neonatal care units in Sub-Saharan Africa in general and especially in Benin.

This study shows that the following factors were associated with death: the age of the mother (less than 20 years), the mother occupation (poorly paid), the male gender of the newborn, the newborns’ age less than a day, the place of birth outside the reference hospital (out born), low birth weight and prematurity. These factors related to death are those described in the literature [4] [7] [9] [13] [15]. It is by acting on these factors that we can reduce the still high death rate in our unit.