This study was initiated by the sickle cell disease management section of the Hematology-Oncology unit of the pediatric department of the Centre Hospitalier Universitaire Sylvanus Olympio (CHU SO) in Lomé.

The pediatric department of the CHU Sylvanus Olympio has a capacity of 170 beds and cribs spread over ten hospitalization pavilions, including two continuing care pavilions. It is the national reference service for child healthcare in Togo. In the hierarchy of health structures in the country, the pediatric department is ranked as a mid-level. Each year, approximately nine thousand children from age 0 to 18 years seek medical care in the pediatric department, and a third of those patients are hospitalized. The hemato-oncology unit has an oncology subunit and a hematology subunit that house the sickle cell management section.

Over the period of 13 years (2006-2018), 1062 files of patients with sickle cell disease were reviewed. Among those, 1055 files met the criteria for the study, and therefore they were retained for the study. In 7 files the medical records files were incomplete and rejected. Our study population represented 99.3% of registered patients.

The majority of patients were aged between 5 and 24 years. Those under 5 years of age represented 9.3% of the total number of patients (Figure 1). There were 566 (53.6%) male patients and 489 (46.4%) female patients with a sex ratio of 1.16. The sickle cell patients followed in our unit came from all 6 regions of Togo, Lomé—Commune (54.5%), Maritime (43.4%), Plateaux (1.5%), Kara (0.3%), Savanes (0.2%) and Central (0.1%).

The age at the diagnosis was specified in 813 patients (77.1%). The average age was 5.4 ± 4.6 years. Only 21.3% were discovered before the age of 1 year. The circumstances of discovery were reported in 946 patients. Osteo-articular VOC was the first circumstance of discovery (28.9%), followed by decompensated anemia (19%) and hand-foot syndrome (15.6%). Only 5.4% of patients were discovered systematically. In 4% of cases, the discovery was made during a family investigation.

The mean age of the patients who were systematically screened was 14.1 ± 7.9 years. Of the 1055 sickle cell patients who followed up, SS homozygotes were in the majority (64.5%) followed by SC and Sβthalassemia (0.4%). For 8.2% of the patients, it was an SF profile. One patient had an SD profile.

Osteoarticular VOC was the primary finding in SS and SC forms of sickle cell disease (Table 1).

The initial assessment (Table 2) revealed that 26% of sickle cell patients with G6PD deficiency and 18% with cardiomegaly.

The distribution of patients according to the number of follow-up visits during the last 24 months revealed 190 patients who had attended at least one

*Vaso-occlusive crisis.

appointment, 67 patients who had attended two appointments, and 118 patients who had attended three or more appointments. In the last 24 months, 95 patients had one emergency visit, 33 patients had two visits and 31 patients had three or more visits. Two hundred and thirty-two (232) hospitalizations were counted in the last 24 months for 140 patients, i.e. 1 hospitalization per patient per year.

The majority of the sickle cell patients followed (70.3%) had never been transfused. Of those who were transfused, 24% were transfused less than twice. Among the complications found in sickle cell patients during the whole follow-up period, VOC represented 68%, infectious complications 35.7%, thrombotic complications 6.9%, chronic complications 9.3% and other complications 4.5% (Table 3). Anemic complications were not documented.

Twenty-five (25) sickle cell patients died, representing a mortality rate of 2.4%. The causes of death were anemia (n = 7), sepsis (n = 5), acute chest syndrome/respiratory distress (n = 2), heart failure (n = 1), renal failure (n = 1), stroke (n = 1), suicide (n = 1), HIV/AIDS infection (n = 1) and severe VOC of

undetermined etiology (n = 6). The mean age of the patients at death was 12.28 ± 6.66 years. The average number of deaths per year was 2.27 ± 0.90. Thirteen (13) male and 12 female sickle cell patients died. Of the 25 patients who died, 24 were SS homozygotes and one had a SF hemoglobin profile.

The reorganization of the unit with the increase in the number of consultation posts and the various IEC activities carried out within the care section has made it possible to show the importance of the patients already being followed up who, by word of mouth, spread information on the quality of sickle cell disease care in the hospital. On the other hand, there has been renewed interest in sickle cell disease in the media since 2005, when the WHO devoted an agenda item to sickle cell disease and recommended several measures at its 59^th annual congress [1].

Our study population not only was mainly pediatric-aged children but also some adolescents and young adults as in most pediatric chronic disease management units [8]. This can be explained by the difficulty that the patients have in leaving their pediatrician who seems to understand their problem better and who has been helping them to cope with their suffering for many years. Furthermore, in our hospital, there is no sickle cell unit for adults.

Adequate management of sickle cell disease must be early, therefore early screening must be encouraged especially in countries where there is no neonatal screening [1].

The male predominance found in our series has no particular explanation, especially since sickle cell disease is an autosomal disease.

Homozygous SS sickle cell patients were predominant in our series. In contrast, a study in Burkina Faso, among a population of newborns found that the prevalence of Hb C was 16.7% as opposed to 7.1% Hb S [9]. This can be explained by the fact that hemoglobin C originates in the Volta basin (Burkina Faso, northern Togo, and northern Ghana) [10].

Sickle cell disease was discovered in 67.3% of our patients before the age of 5 years. Only 31.6% of those under 5 years of age were found before the age of 1 year and more than half (55.2%) were between 4 and 5 years of age. Ideally, newborn screening should be carried out to ensure adequate treatment. But failing that, early detection is necessary. In the 1960s, John Dacie described sickle cell disease as an essentially pediatric disease, as the majority of deaths related to sickle cell disease occurred between the ages of 1 and 3 [11]. Despite the progress made in the management of the disease, the risk of premature death still exists. An estimated 5% of deaths in children under 5 years of age are related to sickle cell disease in Africa [1]. This proportion rises to 9% in West Africa and is as high as 16% in some countries. In Jamaica, the majority of deaths from sickle cell disease occur between 6 and 12 months of age [1]. The WHO estimates that half of all sickle cell patients die before the age of 5 years. As the mean age of diagnosis in our series is 5.4 ± 4.6 years, it is highly likely that many patients die without being diagnosed with sickle cell disease.

The mean age of systematic screening is even older (14.1 ± 7.9 years) and systematic screening only concerns 5.4% of our study population.

In most countries where sickle cell disease is a major public health concern, routine screening is not commonly practiced and the diagnosis is usually made when the patient presents with a serious complication [1].

Evidence suggests that newborn screening for sickle cell disease, when linked to timely diagnostic testing, parental education, and comprehensive care, can significantly reduce morbidity and mortality in the first year of life and in infancy [12].

The absence of neonatal screening in our series is due to the lack of a national policy incorporating a neonatal screening and follow-up program for sickle cell disease. This explains the late detection and management of the disease in most of our patients.

Sickle cell disease is usually discovered in Africa during complications, and in our series, osteoarticular VOC were the first circumstances of discovery in both homozygous SS and SC types of sickle cell disease. The VOC were followed by anemia and hand-foot syndrome. These first 3 circumstances of discovery were significantly more related to homozygous sickle cell disease SS than to SC. For the other circumstances of discovery, including abdominal CVO, there was no difference between the 2 sickle cell genotypes.

Our results are contrary to those of Ayéroué et al. in Burkina Faso who compared SS sickle cell patients to SC sickle cell patients and found that VOC were equivalent in mean number per year and mean duration per year. Regarding anemia, they also found that it was more frequent in SS than in SC [9].

CVO was also the first type of complication found in our patients. Shapiro et al. had already reported the high frequency of CVO in sickle cell patients, estimated at 30% of pain days in a year [13]. Paul Telfer et al. found similar results [14].

Malaria ranked second in our series whereas Paul Telfer et al. ranked chest syndrome second [14]. This difference is related to the endemicity of malaria in our part of Africa and the vulnerability of sickle cell patients to this condition whereas sickle cell traits would be protected. In the African series, the first reason for the consultation has always been VOC, followed by infectious and anemic complications [7] [15].

Priapism was the third most common complication in our patients (5.8%). Since 1934, priapism has been recognized as a complication of sickle cell disease [16]. In our unit, Gbadoé et al. found a prevalence of 26.3% in 2001 [7]. Our prevalence is much lower than that of Gbadoé et al. whose figure is probably closer to reality as it was based on the interview of each patient. As priapism is a complication that often occurs at night and is managed outside the follow-up unit, episodes were not always noted in the medical record. Sickle cell disease is responsible for 65% of priapisms in children [17].

In our study, salmonella was the most common germ found in infections. This is a germ often found in sickle cell patients [18].

Determination of the baseline hemoglobin level is of vital importance in sickle cell disease. Its knowledge allows the practitioner to judge the evolution of the disease from the various blood count controls and, in case of anemia, to indicate or not a blood transfusion. Unfortunately, this level was only determined in 27 of our patients and was on average 7.67 ± 1.64g/dl. This level is similar to the 1999 study by Gbadoé et al. in SS homozygous patients (7.4 ± 1.4 g/dl) but is far from that of the SC type found by the same authors (10.7 ± 2.4 g/dl) [7]. The baseline hemoglobin level was reported to be higher in the order of 10 to 12 g/dl in the Senegal haplotype [19]. This indicates the severity of the signs in the benign haplotype. The low rate of hospitalization of our patients in the last 24 months, on average one hospitalization per patient per year, is a good reflection of the importance of follow-up.

The mean age of our patients at death was 12.3 ± 6.7 years. The lack of neonatal screening means that many children with sickle cell disease die without access to follow-up units. Half of all sickle cell patients die before the age of 5 [1] [20]. In our series, as in all the others, anemia and infection were the main causes of death.

We had registered death in children under 5 years of age, unlike Paul Telfer et al. in England who had no deaths in the same age group [14]. In their series, this was due to neonatal screening and early management, as this follow-up program was absent in our series.