During the study period, 258 patients were seen in the department. The records of 38 children were unusable, so 220 children with renal disease were analysed. RF was present at admission in 60% of children (n/N = 132/220). At consultation, prevalence was 59.1% (n/N = 78/132) for AKI and 40.1% (n/N = 54/132) for CKD with RF (Figure 1).

For children with AKI, the mean age was 6.53 ± 4.77 years with a sex ratio of 1.8. The mean serum creatinine level was 27.22 mg/L (240.9 μmol/L) and BUN at 0.85 g/L. Socio-demographic, clinical and biological characteristics of the patients are shown in Table 1. Aetiologies were dominated by pre-renal AKI complicating nephrotic syndrome in 37.2% (n/N = 29/78), dehydration from stomach flu in 17.6% (n/N = 9/51) and sepsis in 7.8% (n/N = 4/51). Intrinsic AKI was represented by post-infectious acute glomerulonephritis (PIAGN) 16.7% (n = 13/78) including group A streptococcus and plasmodium falciparum. Acute pyelonephritis (APN) accounted for 77.7% (n/N = 7/9) of acute tubulointerstitial

nephritis (ATN). The topographical classification of AKI aetiologies is shown in Figure 2. Indications for dialysis were present in 18% (n/N = 14/78) dominated by hyperkaliemia (62%), anuria (32%) and acute pulmonary oedema (21%). Half of the children with AKI had dialysis access. Among them, 71.4% were on haemodialysis (HD) and 28.6% on peritoneal dialysis (PD). The unmet need for dialysis was 50% (n/N = 7/14). The overall mortality of AKI was 14.1% (n/N = 11/78) (Table 1). AKI mortality was 78.5% (n/N = 11/14) in case of dialysis indications and 85.7% (n/N = 6/7) in case of no access to dialysis treatment.

In CKD with kidney failure, the mean age of children was 8.68 ± 4.74 years with a sex ratio of 2.6. The mean serum creatinine level was 36.56 mg/L (323.5 μmol/L) and the BUN level was 0.99 g/L. Socio-demographic, clinical and biological characteristics of the patients are shown in Table 1. Half of the children were classified as having CKD stage II (early stage). The other half were classified as CKD

stage III to V of which 48.1% (n/N = 13/27) were from the outset admitted to dialysis (Table 2). The aetiologies were dominated by primary focal segmental glomerulosclerosis (FSGS) at 53.7% (n/N = 29/54), congenital renal hypoplasia at 22.2% (n/N = 12/54) and reflux nephropathy at 29.4% (n/N = 5/17) (Figure 3). Late-onset access to dialysis treatment was 54% (n/N = 7/13), of which 71.4% were on haemodialysis and 28.6% on continuous ambulatory peritoneal dialysis (CAPD). The need for dialysis was unmet in 46% (n/N = 6/13). CKD global mortality was 29.6% (n/N = 8/27) (Table 1). In case of no access to dialysis treatment, mortality was 67% (n/N = 4/6).

CKD: chronic kidney disease; RF: renal failure.

Prevalence

GFR was impaired in 60% of the children in our study. AKI was present in slightly more than half of the cases. The prevalence of AKI is variously assessed in the literature. Indeed, most data come from intensive care units, post-surgical intensive care units, neonatology units or nephrology units. The type of children recruitment there is different from one unit to another [13]. In that respect, the prevalence of AKI in an unselected population is particularly high, around 40%, and requires renal replacement therapy (RRT) in just under 20% of cases [12]. In a meta-analysis, the prevalence of AKI in children in sub-Saharan Africa was estimated at 33.7% [14]. These results from literature inspire the need to carry out multicentre studies of an adequate level of evidence in the paediatric population not only at the community level, but also in reference services such as paediatric nephrology units in sub-Saharan Africa, while standardising research protocols. It will be necessary to adapt the definition of operational variables to our context, including the RIFLE criteria for AKI, the method of measuring serum creatinine level, the formulae for calculating GFR and their interpretation according to KDIGO [14] [15].

Age and sex

The mean age of children in our series was 6.5 ± 4.77 years. This result is similar to Bunchmann’s study in Alabama, which found a mean age of 6.16 years [16]. It is therefore desirable to develop preventive and screening strategies in our context, to avoid certain forms of curable acute kidney disease that can potentially affect the child’s vital prognosis or worsen the functional prognosis of the kidney with the risk of chronicity. The male predominance is respected in this study as found in most of the literature [17].

Etiological aspects

Pre-renal and renal causes were the main aetiologies of AKI in our study. This result was found in Africa and Senegal literature [15] [17]. We noted that most causes of RF such as post-infectious acute glomerulonephritis, haemolytic uraemic syndrome (HUS) and stomach flu, are accessible to prevention [18]. To this end, we need to develop simple and low-cost prevention strategies in terms of public health.

Unmet need for dialysis and mortality

AKI mortality in our study was 14.1%. It was 34% in the results of Olowu et al in sub-Saharan Africa [15]. Despite its reversible nature, AKI contributes to approximately 1.7 million deaths each year worldwide. Thus, the International Society of Nephrology has made the initiative “0by25” (zero preventable deaths from acute kidney injury by 2025), a human right and particularly a child’s right [19]. The Save Young Live (SYL) programme has been launched to save children’s lives by providing access to renal replacement therapy through training in peritoneal dialysis acutely and dissemination of the technique [19]. This programme can be implemented in targeted and trained health facilities, but especially in paediatric nephrology units where the need for dialysis is certainly increased due to the numerous referrals. In our situation, the unmet need for dialysis was 50%. In case of need for dialysis, AKI mortality was 73% in Olowu’s study and 78.5% in ours. When dialysis is available, most receive haemodialysis (HD). This can be explained by the absence of peritoneal dialysis in many centres [15]. It is, therefore, necessary to implement this SYL programme in Senegal, especially since consumables are free for children and families. This will require a good organisation of the activity and a multidisciplinary involvement, notably paediatrician, nephropaediatrician, paediatric surgeon, urologist, resuscitator, psychologist, paramedics, dietician...).

Prevalence

Epidemiological data on CKD in children are limited [20] [21] [22]. Some data have been reported from Nigeria and South Africa [22] [23]. The prevalence of CKD with RF was 24.5% in our study. A better understanding of the prevalence of CKD in the paediatric population would allow the development of prevention and screening policies.

Age and sex

In our study, the mean age of children was 8.68 ± 4.74 years. The mean age during CKD was 10.6 ± 3.46 years in Senegal in 2014 [24]. After the infantile period, children more frequently develop acquired kidney disease or express a late-diagnosed congenital and/or hereditary kidney disease phenotype. It is therefore desirable to extend prevention and screening programmes to older children to avoid some forms of chronic kidney disease [24] [25].

The sex ratio in our work was 2.6. In general, there is a male predominance in childhood CKD (sex ratio = 1.5 to 3 depending on the study) reflecting the preponderance of boys among children with CAKUT (Congenital abnormalities of the kidney and urinary tract) [26]. The male predominance is respected in this study as found in most of the literature worldwide regardless of the type of kidney disease [17].

Etiological aspects

Paediatric CKD aetiologies are different from those of adults. Congenital malformations and hereditary nephropathies were the leading causes in the United States in 2012. FSGS was the main renal lesion found in chronic glomerulopathy. The same aetiologies were found in Europe [26]. Congenital abnormalities of the kidney and urinary tract (CAKUT) account for 50% - 60%. The other major causes are glomerular kidney disease (5% - 15%) and hereditary kidney disease (10% - 20%) [26]. In Turkey, the causes were represented by malformative uropathies, renal hypoplasia and hereditary nephropathies [27]. Infectious causes have been reported in some low-income countries [28]. In our study, the main aetiologies were primary FSGS, renal hypoplasia and uropathy.

Unmet need for dialysis and mortality

The incidence of paediatric stage 5 CKD (ESRD) requiring dialysis or renal transplantation varies widely between countries but is around 5 to 10 new cases per million children and year [26] [29]. Lately, dialysis access was 54%, with 71.4% on haemodialysis and 28.6% on continuous ambulatory peritoneal dialysis (CAPD). The unmet need for dialysis was 46%. The overall mortality of CKD was 29.6% in the hospital. The one-year survival rate in chronic haemodialysis was estimated to be low in Senegal (34.3%) [24], due to lack of adapted equipment, low availability and high cost of erythropoiesis-stimulating agents. Also, the absence of nutritional and psycho-affective support could largely explain this low survival in pediatric chronic dialysis in Dakar [24] [30].

The monocentric characteristic of the study is a limitation of the epidemiological study. The data obtained reflect hospital prevalence. Diuresis has not been taken into account as criteria for defining renal failure.