Endocrine and metabolic complications in children and adolescents with Sickle Cell Disease: an Italian cohort study

Sickle cell disease (SCD) is an inherited disease due to a single-point mutation on the β-globin subunit of hemoglobin (Hb) determining polymerization of the mutant HbS and resulting in sickling of erythrocytes. It is characterized by a high clinical variability because of inflammation, hemolysis, and micro-vascular obstruction leading to unpredictable acute complications and chronic organ damage [1, 2]. The HbS mutation can be inherited in homozygosis (HbSS) or in heterozygosis with other β-globin qualitative (HbSC) or quantitative (HbSβ0 and HbSβ+) defects. Subjects affected by HbSS and HbSβ0 have the most severe phenotype while the other forms have milder clinical manifestations [3].

In high-income countries the great and continuous rise of the SCD survival rate demonstrated in the last decades was mainly due to newborn screening programs [4‐6], advances in the supportive care, and a better use of disease modifying agents such as Hydrxyurea (HU) [7]. However, the reduction of mortality has led to an increase of long-term complications, including also metabolic and endocrine ones.

Specifically, poor growth and delay of pubertal development are the most frequent disorders observed in children and adolescent with SCD. Children with SCD have lower height, weight, and body mass index (BMI) than healthy controls [8]. However, published data on endocrine and metabolic disorders during childhood and adolescence, such as gonadal insufficiency, thyroid dysfunction, and bone and glycemic metabolism, are really few [9‐11]. The pathophysiology of these complications is not yet fully understood. Endocrine disorders appear to be related to vaso-occlusive and ischemic events, rather than iron overload resulting from frequent transfusion [9]. According to the literature, the prevalence of endocrine and metabolic disorders in children with SCD varies in different populations depending on the literacy rate, socioeconomic status, and access to appropriate treatment [7, 9, 10, 12]. In children with HbSS, the treatment with HU has been demonstrated to allow growth rates similar to patients with HbSβ⁺ or healthy controls [13].

Aims of the present study were to define the growth pattern, endocrine complications, and metabolic alterations in children and adolescents with SCD and to evaluate the role of therapeutic regimens in improving anthropometric, endocrine, and metabolic parameters.

This study reports data from 52 children and adolescents (29 males, 55.7%) with SCD having mean age of 11.1 ± 4.6 years. Thirty-eight subjects (73%) presented HbSS genotype while others were HbSC. Among our population, 50% of patients was pubertal; 9 out of 23 female had menarche (average age of menarche 12.8 years). Analyzing patients according to the genotype (HbSS and HbSC), 42% of subjects with HbSS and 70% of subjects with HbSC were pubertal.

HbSS showed lower level of Hb and higher level of HbS %, WBC, PTL, LDH, and bilirubin than HbSC ones (Table 1).

Survival rate among children with SCD has increased especially in the recent decades, due to an earlier diagnosis and a better quality of care. Consequently, the incidence of long-term complications, such as metabolic and endocrine disorders, is increasing in SCD population and it has become a main concern to treat them properly, improving their prognosis and their quality of life. In this study we demonstrated a high prevalence (92%) of endocrine complications and metabolic alterations in the pediatric SCD population, mainly represented by vitamin D insufficiency or deficiency, insulin resistance, and to a lesser extent GHD, subclinical hypothyroidism, and hypogonadism. In literature, it is really difficult to understand the cumulative incidence of these disorders in subjects with SCD because of published studies evaluated mainly one single endocrine alteration. Specifically, growth impairment and delayed puberty are the most frequent disorders observed among SCD pediatric patients [10, 11].

Özen et al. reported that 50% of the examined population show endocrine disorders mainly represented, as in our study, by insufficiency/deficiency of vitamin D and to a lesser degree of osteopenia, hypoplasia/testicular atrophy, hypogonadism, hypothyroidism, and insulin resistance [22].

In our study the prevalence of endocrine complications was even higher that those reported by Özen et al. [22], but it is important to consider that the majority of our subjects were immigrants, coming mainly from Africa (96%) with socio-economic conditions that may influence the anthropometric, endocrine and metabolic parameters.

It has been demonstrated that children with SCD had a poorer growth compared to matched healthy subjects [23]. Near two thirds of SCD patients experience a decline in one or more growth parameters (height, weight, and BMI) and the incidence of growth retardation (defined by the presence of one or more of anthropometric parameters below the 5th percentile) could reach the 38% during the follow-up [24]. In our population, the prevalence of growth alterations was about 3.8% when height was considered <-2DS ​​and 9.6% when BMI-SDS was considered <-2SD. The discrepancy between our results and previous published data [24] could be explained by differences in the study design (longitudinal vs. transversal study).

The underline mechanism on growth delay in SCD is very complex and probably influenced by many variables, such as hematologic and cardiovascular status, socio-economic factors, endocrine function, metabolic function and nutritional status [25].

It has been shown that the mean height SDS of children with SCD is comparable to those of children with constitutional growth delay but it is higher than those of children with GHD [26, 27]. In agreement with published data, our study demonstrated that growth was more affected in subjects with HbSS genotype than in subjects with HbSC genotype.

According to the therapeutic regimen, significant differences were found with respect to BMI-SDS and sitting height/height ratio. Although not expected, patients treated with HU for more than one year had lower BMI-SDS and sitting height/height ratio. The reason of these findings is likely related to the more severe phenotype of patients treated with HU for more than a year. In fact, in our opinion, the disease severity could influence these anthropometric parameters. However, it would be important to continue these evaluations in order to assess whether treated patients may have an improvement in growth parameters over time by a reduction in clinical severity. In fact, the prospective use of HU can both improve clinical outcome of the disease and also positively influence growth and development, reducing the risk of iron overload due to the chronic transfusion regimen.

Our data showed significant correlations between clinical parameters of disease severity and anthropometric parameters: children with better control of the disease (expressed as lower number of hospital admissions in 2016, lower number of days of hospitalization in 2016 and lower ACS) had higher values of height-SDS adjusted for TH. Subjects with HbSS genotype showed a negative correlation between the number of ACS and the values of height-SDS adjusted for TH. A good clinical control of the disease seems not only to affect the survival but also to reduce the long-term comorbidity.

In our SCD population, vitamin D insufficiency was demonstrated in 63.5% while 21.2% had a deficient level. In a study conducted by Buison et al., 65% of children with SCD had levels of vitamin D lower than those of healthy children [28]. Jackson et al [29] reported that 96% of SCD patients had vitamin D level between 10 and 20 ng/ml. Severe vitamin D deficiency (< 10 ng/ml) was found in 64% of subjects and it was demonstrated to be associated with age and reduction in lung function but not with pain and/or ACS episodes. In a Spanish study on vitamin D status in 78 children with SCD, Garrido et al. [30] report that near to 80% and 56.4% had vitamin D level < 30 ng/ml and < 20 ng/ml, respectively.

The vitamin D metabolism is complex because of the involvement of different organ including skin, intestines, liver, kidney, and parathyroid [9]. Patients with SCD have some peculiar characteristics that can lead to the development of vitamin D deficiency such as decreased appetite or reduction of nutrients absorption due to the intestinal mucosa damage. Continuous red blood cells production to compensate anemia characterize SCD and causes an increase of basal metabolic rate with higher nutritional demands [8, 31]. Moreover, in SCD patients with renal impairment conversion of vitamin D to its active form can be reduced. Finally, vitamin D binding protein levels can be low being SCD an inflammatory disease [28]. The importance of vitamin D assessment in patients with SCD is supported by the demonstration that vitamin D deficiency is more frequent among children with SCD than in controls [32].

The different prevalence of vitamin D deficiency/insufficiency demonstrated between African Americans and Caucasians populations can be explained by the decreased synthesis of vitamin D in the skin [33] and differences in dietary habits [34]. A better absorption of dietary calcium and lower levels of vitamin D binding protein have been demonstrated in African Americans subjects compared to Caucasians [35, 36]. This suggests that neither the optimal Vitamin D threshold for Caucasians nor levels suggested for healthy African Americans are applicable to patients with SCD [32]. It is therefore really important to identify the optimal level of vitamin D in children and adults with SCD, in particular in patients of African origin living in European Countries, as patients enrolled in our study.

In our population, vitamin D levels showed an inverse and statistically significant correlation with the number of admissions and hospitalizations in 2016 and the average number of admissions in the last 5 years, suggesting that this deficit could adversely affects the clinical severity of the disease.

It was hypothesized that inadequate levels of vitamin D could be linked to a condition of chronic inflammation, as well as low levels of HDL-C [37]. This finding is confirmed also in our population study. Dividing our patients into two groups according to HDL-C levels, we found that vitamin D values ​​were significantly lower in the group of patients with lower HDL-C values (< 40 mg/dl). In addition, HDL-C values ​​of our population showed a negative relationship (p < 0.05) to neutrophils percentage, LDH and ferritin values, particularly in subjects with HbSS genotype. Seixas et al. [37] found a negative association between LDH and HDL-C levels, showing how low HDL-C levels could be a prognostic marker of hemolysis and endothelial dysfunction in view of their anti-inflammatory, anti-oxidant, anti-aggregating, anti-coagulant and pro-fibrinolytic role. Patients with SCD and high HDL-C levels had fewer reticulocytes, WBC, monocytes, PTL, and erythroblasts and a lower concentration of HbS as well as a lower concentration of hemolytic markers. Our data confirm that HDL-C and vitamin D could play an important role in inflammatory condition such as SCD.

The 3.8% of our population showed GHD. An impairment of the GH-IGF1-IGFBP3 axis was demonstrated in SCD subjects [38, 39]. Children with SCD have significantly decreased IGF-1 concentrations compared to children with constitutional delay of growth. The poor synthesis of IGF-1 could depend on a primitive defect of the axis, but also from malnutrition and hypermetabolic status of these patients [40]. In some cases, however, there is a real GHD due to a pituitary vascular insult during vaso-occlusive crises [41, 42]. These patients could benefit from a human recombinant GH replacement therapy [43]. Our data showed that mean values ​​of both IGF-1 and IGFBP-3 were lower in subjects with HbSS genotype compared to subjects with HbSC genotype and that IGF1-levels had a positive correlation with Hb and Hb mean values ​​of 2016 and a negative correlation with average LDH and LDH mean values ​​of 2016. These data underline how the clinical severity of the disease, the number of vaso-occlusive crises and chronic hemolysis could adversely affect GH-IGF-1-IGFBP-3 axis in SCD patients.

In our population 11.5% of patients had pathological HOMA index. In literature there is evidence of insulin resistance among patients with SCD [10]. A multicenter study by Fung et al. [44] revealed that for every 10 years of transfusion therapy, subjects with SCD have a 2.5 times greater probability to develop diabetes (while patients with thalassemia have a double risk). However, there are cases of insulin resistance in patients with normal oral glucose tolerance test. In these patients BMI values were above 85° percentile and none of the patients with normal weight had an insulin resistance condition [22]. However, it is important to point out that our study included a pediatric population with a normal BMI and this could also explain the reduced rate of diabetes. There is currently no agreement on the causes of this complication and further investigations are needed.

We found a case of hypergonadotropic hypogonadism and one case of ovarian failure in HbSS genotype group. The etiology of hypogonadism in SCD is not fully understood yet: in some cases, primitive gonadal failure is related to structural anomalies, resulting from chronic tissue damage associated with chronic anemia condition and local vaso-occlusive crises [45]. According to this hypothesis, our study demonstrated a direct correlation between Hb levels and testosterone average values in males, demonstrating how clinical control can affect reproductive function.

This study has some important limitations. This is a single center study with a small sample size. Secondly there are some confounders factors (i.e. genotype, gender…). The main outcome of this study was to report clinical features of our patients, in a cross sectional way, to better understand the actual prevalence of both metabolic alterations and endocrine complications. Surely, a longitudinal study design of these parameters will provide us more information on the natural history of these complications in SCD. However, it must be considered that in our country SCD is a rare disease and, at the best of our knowledge, this is the first Italian study on these topics. In literature there are very little data available on these condition in pediatric patients with SCD mainly in European Countries. We think that it is important to evaluate these conditions in different environmental setting.

Subjects with SCD show a high prevalence of metabolic alterations and endocrine complications. However, our results suggest that through the achievement of a good clinical control the SCD patients can obtain a positive impact on growth, metabolic and endocrine function.

Consequently, it is crucial to perform periodic anthropometric and endocrine evaluations, especially during puberty, and to have a comprehensive approach to this disease in order to reduce its long-term complications.