A retrospective descriptive study was conducted from January 2018 to July 2020 (36 months) in the pediatric wards of Niamey National Hospital and Amirou Boubacar Diallo National Hospital, two referral hospitals in Niamey.

All sickle-cell children (SS or SC electrophoretic profile) aged between one (1) to 15 years admitted to the corresponding departments for suspicion of bacterial infection were included. All usable records containing at least one bacterial culture (urine, stool or blood) were studied. Children’s socio-demographic characteristics, clinical signs and culture results were the studied variables.

Urine was collected per micturition on the first micturition in a sterile jar in the morning. Stools were also collected aseptically in a sterile jar fitted with a sampling spatula. Blood cultures were taken in the event of a febrile peak (≥38.5˚C) or hypothermia (≤36.5˚C) in culture media. All samples were taken in hospital, and then transported to the laboratory in no more than one hour for processing. Lumbar puncture was performed if patients had neurological signs. Bacteriological analysis was carried out using the appropriate culture media, using BactAlert (Reference 4700003 BTA3D60 BioMérieux).

A data extraction sheet was used to collect information from patients’ hospitalization records, and from the biology laboratory’s registers. Data were entered and analyzed using Epi-Info7 version 7.2.1 software. Results were expressed as numbers and percentages for children’s variables and bacteriological data.

The study was approved by the Faculty of Health Sciences of Abdou Moumouni University of Niamey, and the management of both hospitals. Anonymity and medical confidentiality were respected.

Over 36 months, 350 children with sickle cell disease were admitted for suspected bacterial infection. Table 1 shows Children characteristics. The mean age was 10.9 months [6 months-14 years]. The sex ratio was 1.2. The SS electrophoretic profile was the most common (93.4%). Classical Expanded Program of Immunization (EPI) vaccination status (diphtheria, tetanus, pertussis, poliomyelitis, yellow fever, measles and pneumococcus) was up to date in 66% of patients. None had received vaccines outside the EPI. Fever was the most frequent reason for consultation (55.1%), followed by pallor (52.1%). C-reactive protein (more than 6 mg.L⁻¹ was positive in 84.4% of cases, and hyperleukocytosis, defined as a white blood cell count above 20.000 cells/µL, was found in 96.4%. The diagnosis of infection was confirmed in 62 patients (17.7%). The most frequent diagnoses (Table 2) were bacterial gastroenteritis (24.2%), urinary tract infection (19.4%), pneumonia (19.4%) and acute osteomyelitis (9.7%).

Bacteriological results are shown in Table 3. Blood cultures were taken from 46 children, with 28.3% of positivity. The main germs isolated were Salmonella typhi (13.0%) and Escherichia coli (8.7%). Urine cytobacteriological examination (UCBE) was carried out in 99 patients, with 12.1% of positive results. Klebsiella spp (7.1%) and Escherichia coli (5.0%) were found. Fifty-one (51) coprocultures were requested, with 29.4% positive results. Salmonella typhi (23.5%) and Escherichia coli (5.9%) were isolated. Cytobacteriological examination of cerebrospinal fluid revealed one (1) case of Streptococcus pneumoniae. Ceftriaxone combined with gentamicin was used in 96.3% in the treatment. Progression was favorable in the majority of patients (99.4%). One (1) death resulting from complications of anemia was observed.

The predominance of children under five years of age in this series has been reported by Diakité et al. [4] in Mali and Latoundji et al. [5] in Benin. Generally, the high susceptibility of sickle-cell patients to infection is well known, and the risk of infection is greatest in younger children, particularly infants [1] [2] . In addition to the physiological immaturity of the immune system in this age group, functional asplenia associated with abnormalities in immunoglobulins, leukocyte function and cell-mediated immunity could further weaken the means of infectious control in these children [6] [7] .

Knowledge of the local bacterial ecology is essential for effective antibiotic prophylaxis and vaccination prevention programs. In this study, Salmonella typhi and Escherichia coli dominated the bacteriological profile. In the series by Douamba et al. in Burkina Faso, Streptococcus pneumoniae (35.5%) and Salmonella spp (33.3%) were more frequently reported [8] . The predominance of Streptococcus pneumoniae found by these authors can be explained by the fact that brocho-pneumonia being the most frequent diagnosis. Indeed, pneumonia in children with sickle cell disease is essentially due to Streptococcus pneumoniae [7] [9] [10] [11] [12] . The relatively high EPI vaccination coverage in our context, considers this bacterium, and the systematic antibiotic prophylaxis with oral penicillin in all sickle-cell children, could explain the low proportion of pneumococcal infections observed. Generally, invasive pneumococcal infections have been a major cause of morbidity and mortality in sickle cell patients, especially in precarious living conditions. These infections are often brutal and severe, making curative treatments often ineffective [13] . While preventive antibiotic therapy with penicillotherapy has proved effective, vaccination remains the principal means of control [14] . Salmonella typhi, found in coprocultures in this study, is more likely to be implicated in osteoarticular infections [15] . According to the authors, this could be explained by their high susceptibility to osteoarticular infections. What’s more, these infections are endemic in the context of poor living conditions [3] . However, effective prevention through vaccination has been available for many years. This should be offered systematically to all children with a tare, generally from the age of two for the polysaccharide vaccine, and six months for the conjugate vaccine [16] . Other studies have also reported the predominance of Escherichia coli and Klebsiella spp in urinary tract infections. The perineum is highly colonized by enterobacteria of digestive origin, in particular Escherichia coli. In addition, this bacterium possesses specific uropathogenicity factors, thus favoring these infections [17] [18] .