Cross-sectional study covering the period from January 1, 2022 to May 31, 2022. Patients with an established diagnosis of sickle cell anemia, received in the emergency department, aged 0 to 21 years.

Our study was carried out at Albert Royer Children’s Hospital in the emergency department and in the emergency sector of the outpatient unit for sickle-cell adolescents and children at the same hospital. Albert Royer Children’s Hospital is a pioneering site in the care of children with sickle cell disease in Senegal, with an active file of about 4,000 patients.

Definitions and thresholds values; Infection was considered when:

Data sources were patient’s physical medical records and data collected were recorded on an excel file.

Data were analyzed using Excel and SPSS 28.0. A statistically significant association was defined by a p. value < 0.05.

Of these patients, 96.3% (315 patients) had follow-up for sickle cell disease, while the remaining 12 (3.7%) had no regular follow-up, or failed to keep follow-up appointments. Among patients admitted for infection, 98 out of 107 (91.5%) had regular follow-up.

All type of sickle cell disease (SCD) were represented, with homozygote predominating at 94.5% (309 patients), followed by SC sickle children, 14 patients (4.28%) and 4 Sβ thalassemia (1.22%). For infected patients, 99% were SS (106 patients) and only one was SC.

Median baseline hemoglobin was 8 g/dl, with extremes of 6 and 12 g/dl. 6 of the 327 patients (1.84%) were on hydroxyurea and 2 on transfusion therapy (0.6%).

Comorbidities found in these patients was dominated by asthma 13 patients (3.97%), then heart disease, 2 patients (0.62%) and other such as hypothyroidism, lupus and nephrotic syndrome. Asthma was the only comorbidity found in the proportion of infected sickle cell patients (5 patients or 4.6%).

Coverage of mandatory vaccines (from 0 to 15 months) was 93.88% (307 patients) among sickle-cell patients admitted to the emergency department. This coverage was higher for infected patients: 97% (104 patients).

The following graphs ( Figure 1 ) illustrate coverage of specific vaccines to sickle cell disease follow-up in particular pneumococcal polysaccharide vaccine, anti-typhoid vaccine and meningococcal vaccine.

43 patients (13.15%) were on antibiotic prophylaxis. In the 0 to 5 age group, 42 patients (60% of cases) were under antibiotic prophylaxis. In the infected sickle cell group, 15 patients (14%) were under antibiotic prophylaxis. This antibiotic prophylaxis was based on penicillin V for all patients. No patient was on malaria prophylaxis during the study period.

The number of patients who were both correctly vaccinated and on antibiotic prophylaxis was 39 (11.93%).

The main reason for consultation in the emergency department was pain 251 patients (76%), followed by fever 107 patients (32.7%) and asthenia 25 patients (7.6%).

For patients admitted at emergency department, vaso occlusive crisis (VOC) was the main diagnostic different diagnoses of patients admitted to the emergency department during the study period.

The most common clinical signs in patients admitted to the emergency department were fever above 38.5˚C, clinical anemia (pallor), auscultatory abnormality and respiratory distress. Hypoxia and bone swelling were present in less than 5% of cases. Four patients were malnourished.

Several types of infection were identified. The main site of infection was pulmonary (46 patients or 42%), comprising 42 cases of ACS, 3 cases of purulent pleurisy and one case of pulmonary tuberculosis. Pulmonary infections were followed by osteoarticular infections, with 5 cases of septic arthritis and 6 of osteomyelitis. There were as many digestive infections (gastroenteritis) as ENT infections (7 cases or 6.4%). ENT infections were all tonsillitis. 2 cases of transient erythroblastopenia were recorded based on severe anemia with low reticulocytes level. 2 patients presented with sepsis, one of which was complicated by septic shock leading to death. Malaria was found in 4 patients, with no signs of severity according to the WHO.

The median white blood cell count was 12,700/mm³ in patients admitted to the emergency department, and the median CRP level was 45.54 mg/l in the proportion of infected patients, with a maximum of 289.2 mg/l.

A total of 137 samples were taken from these patients. Bacteriological samples accounted for 54.7% (75 samples), including 46 blood cultures, 11 osteoarticular punctures, 3 pleural punctures, 1 lumbar puncture, 3 ECBU and 11 sputum PCRs for Mycobacterium tuberculosis. No viral samples were taken. Thick-film tests for plasmodium were carried out in 62 patients. The positivity rate for these samples was 4.3%. This corresponded to 5.6% of infected patients with an isolated germ.

6 cases of microbiologically documented infections. These were one case of Pneumococcal sepsis isolated by blood culture, one case of pulmonary tuberculosis isolated by PCR on sputum and 4 cases of malaria isolated by thick blood smear. No biological fluid puncture was used to isolate a germ.

Imaging made a considerable contribution to the diagnosis of infections, especially osteoarticular, pulmonary and soft-tissue infections. A total of 99 imaging examinations were carried out, including 59 chest X-rays with parenchymal infiltrates in 43 cases and 3 cases of pleural involvement. Ultrasound scans of the limbs were carried out in 8 patients, 5 of whom showed images of subperiosteal abscesses compatible with the diagnosis of osteomyelitis. 12 joint ultrasound scans were performed, 5 of which revealed a collection compatible with septic arthritis. All soft-tissue ultrasound scans revealed abscesses.

All patients (107 patients) admitted for infection received probabilistic antibiotic therapy. 3rd generation cephalosporins were the molecules most frequently used 105 patients (98.1%) in infected patients. Followed by erythromycin, 42 patients (12.84%); aminoglycosides, 19 patients (5.7%) then ciprofloxacin, 8 patients (2.45%). Amoxicillin was used in 5 patients (1.53%) and vancomycin in one (0.31%). One patient received anti-tuberculosis treatment based on Rifampicin, Isoniazid, Ethambutol and Pyrazinamide.

The 4 patients with malaria infection and a positive thick smear without signs of severity according to WHO, received artesunate for the management of their malaria attacks.

Clinical improvement in 99% of cases, with no more fever and normalization of CRP. 1 case of death was recorded in the infected group: a 3-year-old female infant. 7 patients were taking hydroxyurea after hospitalization in the emergency department, including 2 in the infected group.

Quality of follow-up and infection

There was a statistically significant association (p < 0.005) between the quality of follow-up and the occurrence of infection ( Table 1 ). Infected patients had less regular follow-up of their disease than the non-infected group (Chi-square test).

*Khi-deux Pearson’s test = 0.003.

Preventive measures

There was a statistically significant association between sickle-cell-specific supplementary vaccines and the occurrence of infection. This association was not found for penicillin V antibiotic prophylaxis. Table 2 shows these different associations.

The main types of infection found were respiratory infections followed by osteoarticular, then digestive infections. This was superposable with the results of Thiam et al., in 2018 in Senegal. Because of their pathophysiology, we considered ACS to be infections. Moreover, the difficulty of identifying a germ in both pulmonary infections and ACS makes the distinction uncertain [18] . There was a significant proportion of highly suspected infections with fever but without bacteriological evidence (20.3%). This last category remains an indication for the initiation of broad-spectrum probabilistic antibiotic therapy recommended on the basis of several descriptive studies and expert opinion [12] [19] [20] . In our study, all these patients received antibiotic therapy, with an improvement in their biological inflammatory syndrome under antibiotic treatment.

Our study identified only one germ out of 46 blood cultures taken. There are few studies on the characterization of germs responsible for infections in sickle-cell patients in Africa [1] . The study by Bello et al. in Nigeria identified 39% Salmonella typhi infections, followed by Streptococcus pneumoniae at 14.7%, Salmonella para typhi and Staphylococcus aureus at 5.8%. In this study Gram negative bacteria remain higher than infection by Streptococcus Pneumoniae, this contrast with the most common pathogens found in higher income countries as the study by Yee ME et al. was Streptococcus pneumoniae 16%, followed by Staphylococcus and Streptococcus (viridans group and pyogens), Salmonella species at 6.4% [9] [17] . Several studies have shown a low positivity threshold for blood cultures, such as the study by B. S. Shihabuddin and C.A. Scarf, where out of 307 blood cultures, only 1 came back truly positive and 6 were contaminations [20] . A limitation of our study is the lack of information on antibiotic use prior to admission to the emergency department. In fact, the use of antibiotics could explain the low positivity of bacteriological samples. During the period of the study, viral tests have not been performed, which limits our understanding of the role of these viral infections in this study.

Due to the low positivity of blood cultures, empirical antibiotic therapy was initiated with 3rd generation cephalosporins in 98.1% of cases. This practice is in line with international recommendations, considering the functional asplenia and the high risk of sepsis in this condition [12] .

Penicillin prophylaxis was effective in under-5 years in 60% of cases. We found no association between this antibiotic prophylaxis and the absence of infection. Several studies have demonstrated the benefits of oral antibiotic prophylaxis with penicillin [4] [21] . However, this practice is not uniform in sub-Saharan Africa, with the study by Coria AL. et al. showing that 29% of practitioners did not prescribe antibiotic prophylaxis for their sickle cell patients [12] . In Studies conducted in Ghana, Tanzania and Nigeria; prescription of penicillin prophylaxis was effective in 7%, 10% and 17,2% respectively for sickle cell children [10] [22] - [24] . This was due to the irregular availability of the syrup form of penicillin V in Senegal, but also to the cost, as this treatment is not reimbursed. In addition, the infrequent isolation of Streptococcus pneumonia from febrile SCA children in sub-Saharan Africa could underlie this practice [6] [9] [25] . And finally, studies show a low adherence to penicillin V prophylaxis [26] [27] .

Malaria prophylaxis was not carried out in any of the patients included in this period of the study. This prophylaxis was systematic in the study by Coria et al. In Senegal, anti-malarial prophylaxis based on monthly doses of Sulfadoxin Pyrimethamin is recommended from June to October, corresponding to the peak malaria epidemic. However, the occurence of 4 malaria infections could suggest adopting year-round prophylaxis, as in other African countries [28] .

Immunization coverage was low (below 30%) for all specific sickle cell vaccines, such as meningococcal, anti-typhoid and pneumococcal. 20 years earlier Diagne I et al. show similar coverage for pneumococcal immunization and we note a slight improvement in anti-typhoid immunization 22% VS 8,4% in 2000 for similar population [29] . Better results were found in Brazil, notably for Pneumococcal vaccine PPSV23: 61% for the first dose [30] However, all these studies highlight under-vaccination rate in this selected high-risk group. This, could be linked to the cost of these vaccines, which are not reimbursed, as well as to their irregular availability. Subvention for sickle cell children by government should be offered by government.