Introduction

Sickle cell disease (SCD) has been acknowledged by the World Health Organization as a global public health problem, with estimates indicating that 300.000 to 500.000 children are born each year with severe hemoglobinopathy.¹

The term SCD refers to an heterogeneous group of inherited hemoglobinopathies in which the affected individuals have inherited an abnormal gene in the hemoglobin (the sickle hemoglobin [HbS]), either in homozygosis (HbSS or sickle cell anemia [SCA]) or in compound heterozygosis with another beta globin allele mutation, such as hemoglobin SC disease, sickle cell-beta thalassemia (which is divided into sickle cell-beta⁰ thalassemia and sickle cell-beta⁺ thalassemia), and others less frequent (sickle-alpha thalassemia, sickle-hereditary persistence of fetal hemoglobin, sickle-delta beta (0) thalassemia, sickle-Hb Lepore disease, sickle-HbD disease, sickle HbO Arab disease, and sickle-HbE disease).^2,3

HbS results from a mutation in the normal beta globin gene that leads to an abnormal hemoglobin, which under adverse circumstances, such as hypoxia, polymerizes into intracellular fibers, becoming less soluble and causing sickle cell deformity. The pathological polymerization of deoxygenated HbS is a key feature in the pathophysiology of the disease and explains its main clinical manifestations, such as vaso-occlusive phenomena and hemolytic anemia.^2-4

All forms of SCD are inherited in an autosomal recessive fashion, although there is considerable variability in severity according to the genotype. As such, individuals with HbSS and compound heterozygous for Sβ⁰ thalassemia have the most severe clinical phenotype, and those with SC and Sβ⁺ thalassemia have a milder clinical phenotype and increased life expectancy.^2,5

The sickle cell trait is a benign condition that confers a relative protection against Plasmodium falciparum malaria infection, and therefore carriers are less likely to get this disease. Although the sickle hemoglobin mutation is currently globally widespread due to migrations, this survival advantage has contributed to a high frequency of the condition in areas with a history of malaria, as sub-Saharan Africa.^2,3

About 75% of the global burden of SCD occurs in sub-Saharan Africa, where the resources to diagnose and manage this disorder are scarce. Therefore, many children in this area do not reach their fifth birthday and die undiagnosed due to disease complications.⁶ In well-resourced countries, a holistic approach with family education and well-structured comprehensive care programs has been instituted, which has enabled an increase in the average life expectancy of sickle cell children, with almost all surviving into adulthood. Therefore, as these patients age, other comorbidities manifest, such as pulmonary hypertension, diastolic heart dysfunction, and end-stage renal disease.^3,7

SCD testing is currently a standard component of newborn screening in some hospitals. Newborns with abnormal screening results should be referred to a pediatric hematologist, and a confirmatory testing should be performed.¹

SCD has a variable presentation in pediatric patients. The first symptoms usually manifest around six months of life, when fetal hemoglobin dissipates, and include pallor, anemia, vaso-occlusive episodes, acute chest syndrome, and susceptibility to bacterial infections, amongst others. Infection is the most common cause of mortality, particularly in very young children, with several preventive strategies developed throughout the years, such as antibiotic prophylaxis with penicillin and vaccination.¹

Antibiotic prophylaxis with penicillin or amoxicillin (used in Portugal, since the oral penicillin formula is not available) was shown to reduce the incidence of pneumococcal infections in children younger than five years and should be initiated at two months of age.¹

According to the Portuguese National Vaccination (PNV) Program, all children with SCD should receive routine childhood immunization. In addition, they should also receive the influenza vaccine annually since the age of six months, the 23-valent pneumococcal vaccine from the age of two years, and the meningococcal B vaccine and meningococcal ACWY conjugate vaccine since the age of two months.^1,4 Since October 2020, the meningococcal B vaccine has been included in the PNV Program for all children in the first year of life (not included at the time of this study). Daily folic acid supplementation is recommended to prevent bone marrow aplasia from folate deficiency.⁴ Hydroxycarbamide is used to increase fetal hemoglobin levels, thereby increasing oxygen-carrying capacity, decreasing sickle hemoglobin levels, and reducing several morbidities related to chronic hemolysis.¹

Stroke is a major known complication of SCD, primarily occurring in childhood. Routine screening with transcranial doppler ultrasonography (TCD) can identify vessel abnormalities by measuring the mean velocity artery flow that identifies children at high risk of stroke, and should be performed annually since the age of two years.¹

Obstructive sleep apnea syndrome (OSAS) is a common pediatric disorder with an estimated prevalence of 1−5%.⁸ It causes episodic upper airway collapse, which disrupts ventilation and therefore interferes with sleep quality. Hypoxia is one of the main triggers for vaso-occlusion, and desaturation is common in both OSAS and SCD.⁸ According to several studies, the prevalence rates of OSAS are higher in children with SCD, reaching 41% when considering a cut point of obstructive apnea hypopnea index (AHI) superior to one.⁸ Considering that SCD is a complex disease with many comorbidities and that OSAS is a treatable condition that may be associated with multiple adverse health outcomes, it is of utmost importance to implement measures for screening and managing this disease, particularly in such a vulnerable population as that of children under the age of five years.

The purpose of this study was to assess the sociodemographic and clinical characteristics, as well as management and follow-up of all children up to the age of five years with a diagnosis of SCD in the Portuguese hospital Professor Doutor Fernando Fonseca. This is a level II hospital in the region of Amadora and western Sintra covering a total population of 568,069 people, of whom 90,761 are children under 15 years old.⁹ The latest data point to a foreign resident population of 8.65% in Sintra and 10% in Amadora, of whom 55.6% and 62% are African residents, respectively.^10,11 This demographic feature makes this region particularly interesting for the study of SCD.

Methods

A descriptive retrospective observational study was conducted between January 2010 and December 2019, with data collected from computerized inpatient records.

The following variables were assessed: sociodemographic and clinical parameters, comorbidities (including concomitant diseases and follow-up at other medical appointments), and disease management. Sociodemographic parameters included age, gender, date of birth, origin, family’s birthplace, and surveillance during pregnancy. Pregnancies was considered well monitored if having at least six medical appointments, blood tests with serology in the three trimesters, and at least three ultrasounds performed. Clinical parameters retrieved included neonatal screening (which, since 2013, has become a standard approach in newborns of mothers with sickle cell or ancestry from countries with high risk for SCD), age at diagnosis, type of sickle cell disease, and steady-state hemoglobin. The hemoglobin analysis methods consisted of capillary electrophoresis in cases up to six months of age and high-performance liquid chromatography (HPLC) thereafter.

Management and surveillance of the disease included transcranial doppler ultrasonography screening, sleep evaluation (only from the age of two years), use of prophylactic medication, and compliance with the PNV Program and other recommended vaccines. For vaccines not included in the PNV Program, the following eligibility criteria were considered: age over two years for the Pneumococcal Polysaccharide vaccine (PPSV23); age over six months for the Influenza vaccine; age over two months for the Meningococcal B and ACWY conjugate vaccines (free for children with SCD since 2016); and age over 12 months for the hepatitis A vaccine.

OSA severity classification was established through the obstructive apnea hypopnea index (AHI): mild if AHI 1-5/h, moderate if AHI 5-10/h, and severe if AHI>10/h.

As this study aimed to describe sociodemographic and clinical features, as well as management and comorbidities of children up to five years of age with a diagnosis of SCD and at least two years of hematologic follow-up, children who were lost to follow-up at the study hospital for any reason were excluded from the analysis.

Statistical analysis was performed in Excel Microsoft 365, 2020 Microsoft Corporation©.

Ethics permission for the study was obtained from the Ethics Committee of the hospital.

Results

A total of 86 patients aged until five years were included in the study, 46 (53.5%) of whom were female. Most were of Portuguese (85%) followed by Angolan (8%) nationality. Regarding parents’ origin, the most frequent was Angola, corresponding to 32.6% of cases, followed by Guinea-Bissau (16%) and São Tomé e Principe (11.6%; Table 1) Sixty-three (73.3%) pregnancies were monitored in Portugal.

The predominant type of SCD in the study cohort was HbSS (SCA), present in 80 (93%) cases, followed by HbSC (n=3; 3.5%). Only one case of HbSβ⁺ thalassemia (1.2%), HbSβ⁰ thalassemia (1.2%), and alpha thalassemia association (1.2%) each were identified. In one child, the simultaneous presence of glucose-6-phosphate dehydrogenase deficiency was identified.

Regarding diagnosis, 44.2% (n=38) of patients were diagnosed through neonatal screening (hemoglobin electrophoresis at birth), 34.9% (n=30) in the context of hospitalization, and 17.4% (n=15) in the context of a medical appointment with laboratory confirmation. The diagnostic setting could not be ascertained in three cases that had been diagnosed in Angola (n=2) and Cape Verde (n=1). Considering only children born after 2013 (n=38), 68.4% (n=26) had been diagnosed through neonatal screening, while only 25% (n=12) of children born before 2013 (n=48) had been diagnosed through this procedure at the study hospital.

The median age at diagnosis was five months (interquartile range [IQR] 1-12 months). Analyzing only children born before 2013, when neonatal screening was not implemented at the study hospital, the median age at diagnosis was eight months (IQR 3-17 months), and considering only children born after 2013, the median age at diagnosis was one month (IQR 0-6 months).

Hemoglobin levels were assessed by considering 1 g/dL intervals, with 41.9% of children presenting a baseline hemoglobin level between 7-8 g/dL (Table 2).

All patients were regularly followed in Hematology outpatient consultation. In addition, 74% were also followed in Cardiology, 30% in Otorhinolaryngology, and 28% in Sleep Disorders consultation (Table 3).

Regarding vaccination status, the vast majority of children (96.5%) were compliant with the PNV Program. Concerning vaccines not included in the Program but recommended, 85% had received the Pneumococcal polysaccharide vaccine (PPSV23), 87% the Influenza vaccine, and 80.6% the Meningococcal B vaccine. The lowest immunization rate was observed in children eligible for Meningococcal ACWY conjugate vaccine (39%). Additionally, 52.9% of the study population had also received the Hepatitis A vaccine. Patients were only assessed for vaccines they were eligible to, which is why denominators are smaller for some immunizations compared to others. The complete immunization adherence rate, including recommended vaccines (all the previously mentioned, excluding Hepatitis A vaccine) was 49% (Table 4).

All patients were taking acid folic supplementation (n=86) and the majority was also receiving daily amoxicillin (n=84; 98%) as prophylaxis. One child was taking clarithromycin instead of amoxicillin due to allergic reaction to the latter. Ten children (12.6%) underwent treatment with an angiotensin-converting enzyme inhibitor, nine due to dilated cardiomyopathy (one of whom also had hypertension) and one due to persistent microalbuminuria. Six patients (6.9%) were under hydroxycarbamide therapy. All these patients had at least seven hospitalizations (average 9, minimum 7−maximum 17 hospitalizations) prior to treatment start, with at least two being due to vaso-occlusive crisis. One patient was hospitalized for acute chest syndrome. After therapy introduction, only one patient was hospitalized (two times, one for vaso-occlusive crisis). All patients started therapy after the year 2016 (the youngest was three years and seven months old).

Fifty-eight children (68.6%) performed transcranial doppler and one showed alterations, specifically middle cerebral artery stenosis.

Sixty-two children (72.1%) performed at least one echocardiogram. Left ventricle dilatation was the most frequent pathological finding, observed in 13 children (Table 5).

Twenty-one children (24.4%) aged between 2.3−4.9 years underwent overnight polysomnography, 20 (95.2%) of whom had OSAS. Most children (n=12; 57.1%) had mild disease, six children (28.6%) had moderate disease, two children (9.5%) had severe disease, and only one child (4.8%) had a normal AHI (Table 6). Seven children underwent tonsillectomy, two in the context of severe OSA.

On average, each child experienced four hospitalizations during the five years. Sixteen children had no hospitalizations. The main acute complication requiring hospitalization during the study period was vaso-occlusive crisis (38.3%), followed by fever of unknow origin (23.4%), upper respiratory infection (17.8%), bacterial pneumonia and splenic sequestration (10.7% each), and worsening anemia (9.5%).

Discussion

According to the literature, the most prevalent form of SCD is by far HbSS (SCA), followed by HbSC. This was confirmed in the present study, with the vast majority of cases (93%) presenting homozygosity for the S allele. The second most common form of SCD identified in the study was HbSC.

In agreement with previous reports from several other studies, a high prevalence of SCD was identified in families with origins in sub-Saharan Africa: 74.5%.^1,5,6,7

Proper management of SCA begins with a correct diagnosis early in life, ideally during the newborn period. A significant proportion of cases in this study (44.2%) was diagnosed through neonatal screening, a number that is even higher (68.4%) if considering only the period after selective screening started (2013). Neonatal diagnosis allows early initiation of preventive interventions, including prophylactic antibiotics and vaccination, which help prevent complications, namely overwhelming sepsis, and thus the number of hospitalizations. In addition, it also allows prompt education of families in the recognition of early symptoms, enabling appropriate approaches and interventions.

SCA requires treatment and follow-up by a multidisciplinary team. Adherence to the PNV Program in SCD patients is essential to protect them from vaccine preventable diseases. This study showed that most children (96%) are compliant with the vaccination program. The lowest immunization adherence rates occurred with Meningococcal ACWY conjugate vaccine − the newest vaccine, only available for free to SCD patients since 2016 − and with Hepatitis A vaccine. The complete immunization adherence rate, including all recommended vaccines (excluding the Hepatitis A vaccine), was 48.8%. This contrasts with a U.S. study about immunization adherence in children with SCD, where the adherence rate to the recommended immunization schedule was 6% and the lowest immunization rate was observed in children eligible for the meningococcal B vaccine (25%).¹² In the present study, compliance with Meningococcal B vaccine was 80.6%. This emphasizes the need to reinforce the importance of vaccination and educate families and caretakers. Quality improvement measures should focus on increasing immunization adherence.

Hydroxycarbamide is the only approved treatment for SCD and the recommendations for treatment initiation have changed over time. For many years, its use was reserved for children older than two years and had specific eligibility criteria that included three or more hospitalizations for vaso-occlusive crisis; two or more hospitalizations for acute thoracic syndrome; or any combination of at least three hospitalizations with these two diagnoses per year or one episode of severe acute thoracic syndrome, priapism, stroke, or any other severe complication. More recently, there was a paradigm shift, with recommendations to introduce this therapy earlier in life and several sources, namely the British Society for Hematology, advising treatment start around the ninth month of life, regardless of other factors.¹³ The present study was conducted over nine years, and hence most children were not under hydroxycarbamide therapy. Still, it should be noted that about 6.9% of children under five years of age were already under hydroxycarbamide therapy for having had at least three hospitalizations for vaso-occlusive crisis. Knowing that this treatment is, not only safe and effective, but also one of the only disease-modifying therapies available, it should be used early to prevent complications. Therefore, it can be anticipated that a change will be seen in upcoming years towards the introduction of hydroxycarbamide earlier in life. In this context, it would be interesting to conduct a study investigating if earlier treatment start influences hospitalizations and outcomes in children with SCD.

Cardiac complications are a common feature in SCD and an important cause of morbidity and mortality associated with the disease. A study of echocardiographic screening in children with ≥ 10 years of age and SCD documented 25% of left ventricle dilation.¹⁴ In the present study, 21% of patients had left ventricle dilatation, which represents a non-negligible proportion in children under five years old.

The prevalence of OSAS in children with SCD appears to be higher than in the general pediatric population.^8,15 In this study cohort of children younger than five years only, about a quarter underwent night polysomnography and OSAS was identified in the vast majority (95.2%, 38.1% with moderate to severe forms). This finding is in agreement with the literature, perhaps even more than the previous study that only included older children, between the ages of four and 18 years. ⁸ Since OSAS is a treatable condition and is associated with various adverse health outcomes (including minimum peripheral oxygen saturation significantly lower than non-SCD counterparts, behavioral problems, cognitive deficits, and cardiovascular changes, namely hypertension), this finding highlights the importance of early screening of the disease, since it has a direct impact on patients’ quality of life, both in child and adulthood.¹⁵

One of the main limitations of this study was its small sample size, which makes it difficult to extrapolate results to the general population, namely the correlation between the type of SCD and disease severity.

A prompt diagnosis is the first step towards improving the outcomes of individuals with SCD, and is key for parental education, namely regarding symptom recognition and seek for immediate medical care.

Authorship

Inês Filipa Mendes - Data curation; Formal analysis; Investigation; Methodology; Software; Validation; Writing - original draft; Writing - review & editing

Adriana Costa - Data curation; Formal analysis; Investigation; Methodology; Software; Writing - original draft

Joana Lage - Data curation; Formal analysis; Investigation; Methodology; Software; Writing - original draft

Bernardo Monteiro - Data curation; Formal analysis; Investigation; Methodology; Software; Writing - original draft

Teresa Ferreira - Conceptualization; Methodology; Validation; Writing - original draft; Writing - review & editing

Helena Cristina Loureiro - Conceptualization; Methodology; Validation; Writing - original draft; Writing - review & editing