Obesity in Pediatric Sickle Cell Disease: A New Phenomenon

SIGNIFICANCE Sickle Cell Disease (SCD) is the most common genetic disorder in the world and in the United States (U.S.), with about 1 in 13 African Americans born with sickle cell trait and approximately 100,000 individuals currently living with SCD. In this disease, irregular shaped red blood cells (RBC) termed sickle RBC, get trapped in the microcirculation thereby causing ischemic organ damage throughout the body. Due to the chronic hemolysis (breaking of RBCs), patients with SCD have traditionally had a high basal metabolic rate and increased energy consumption (measured by resting energy expenditure) resulting from rapid production of new RBCs. Therefore, patients with SCD tended to be underweight and did not experience weight-related comorbidities such as hypertension, dyslipidemia and hyperglycemia. In recent years, there has been significant improvement in the medical management for patients with SCD, which has resulted in improved morbidity and mortality and increased life span. With these advancements in medical management, recent research has begun to describe increased rates of overweight and obesity in both pediatric and adult patients with SCD. Our research team's pilot data in Mississippi shows that 24% of our pediatric patients with SCD are overweight or obese, defined as body mass indexes (BMI) ≥ 85th percentile. Overweight and obesity is a new medical phenomena in SCD.

The two leading causes of death in individuals with SCD are cardiovascular disease and stroke. In 1973, Lemuel Whitley Diggs estimated a median survival of 14.3 years for individuals with SCD, with 20 percent of deaths occurring in the first 2 years of life. However, as a result of improved medical management, patients with SCD are living longer. In 1994, life expectancy had increased to 42-48 years for the most severe SCD genotype HbSS, and 60-68 years for the less severe SCD genotype HbSC. Since patients with SCD are no living longer and overweight and obesity is a new phenomenon in SCD there is very little knowledge regarding its clinical impact on precipitant risk factors (i.e., hypertension, hyperlipidemia, and hyperglycemia) for cardiovascular disease, stroke, and overall morbidity.

The two published studies to date have been in adults with SCD and document that overweight and obese weight status is associated with metabolic syndrome, hypertension and dyslipidemia resulting in a higher risk for cardiovascular disease and stroke. Type 2 Diabetes (T2D) has historically not been clinically documented in patients with SCD. Yet, recent studies have shown that the prevalence of T2D in adult patients with SCD now approaches that of their healthy African American counterparts. The same study showed that the risk of stroke and overall morbidity for patients with both SCD and T2D was significantly higher compared to patients with just one disease. It is critically important to assess and characterize these potential increased risk factors for cardiovascular disease and stroke in pediatric patients with SCD.

Accurate Measurement of Outcomes in SCD In order to accurately assess weight-related risk factors for cardiovascular disease and stroke (i.e., hypertension, hyperlipidemia, and hyperglycemia) in pediatric patients with SCD, research is needed that appropriately adjusts for disease-related blood processes. Specifically, individuals with SCD have lower systemic blood pressures compared to individuals without SCD. In adults with SCD, clinical problems are observed with blood pressures at or greater than the 90th percentile. When utilizing the 90th percentile, rates of hypertension are similar in adults with SCD (40%) as their healthy counterparts. It is thus recommended that blood pressure values be assessed relative to the lower values expected for patients with SCD, and that a lower threshold of ≥ 90th percentile be used to define 'relative hypertension' in individuals with SCD. Unfortunately, there are no current published age, sex and height-matched norms for blood pressure in youth with SCD. To robustly characterize the association between weight status and elevated blood pressure, the current study will utilize published age, sex and height blood pressure norms for healthy youth without SCD but define relative hypertension (primary outcome) ≥ 90th percentile.

Overweight/obesity is associated with impaired glucose intolerance and is a strong predictor of development of T2DM, which increases the risk for cardiovascular disease and stroke. Specifically, early development of T2DM leads to significant microvascular and macrovascular complications in early adulthood. The greatest risks for impaired glucose tolerance in children is weight status (overweight and obesity) and family medical history of overweight and obesity. Screening for hyperglycemia and insulin resistance in high-risk youth is recommended, as children are often asymptomatic. Expert committees and the American Diabetes Association have endorsed the use of fasting plasma glucose or HbA1c for screening those who are overweight or obese (BMI > 85th percentile), staring at age 10 years of age. However, the use of HbA1c for glucose tolerance testing in SCD is not reliable due to the shortened life span of RBCs in SCD. Therefore, fasting glucose levels and serum fructosamine are recommended for patients with SCD. Fructosamine is a glycated protein and predicts glucose levels over a period of approximately 3-4 weeks and will be used as the primary measure of hyperglycemia (secondary outcome) in the current study.

Similar to lower systemic blood pressure, individuals with SCD have historically been shown to have low cholesterol levels due to high levels of hemolysis and consumption of the lipids for new membrane synthesis. The estimated overall prevalence of dyslipidemia in all children in the US is 20.3% and increases to 42.9% in youth with obesity. With current treatments decreasing hemolysis and overall turnover of RBCs it is unknown if the medical community will start to see an increase in hyperlipidemia (secondary outcome) in pediatric patients with SCD.

Key Risk Factors for Overweight/Obesity Though reasons for the development of overweight/obesity in the general pediatric population are multifaceted, energy imbalance (e.g., consuming more calories than exerted through physical activity) has the most research support and identified areas for intervention. Children spend about 15 minutes a day engaged in moderate-to-vigorous physical activity, far less than national recommendations of 60 minutes a day. Children also spend over 85% of their time in sedentary activities, which is associated with increased weight status. Additionally, very few children eat a diet consistent with national guidelines and instead consume large portions of high-calorie, nutrient-poor foods and sugar-sweetened beverages, all of which are associated with weight gain and obesity. These problems are especially severe in Mississippi, which has the highest rates of physical inactivity and obesity in the U.S. In addition, African American youth demonstrate the highest rates of sedentary activity. Finally, family history of obesity, hypertension, hyperlipidemia, T2D, and maternal history of diabetes during pregnancy are significantly associated with increased risk for being children being overweight/obese.

To begin addressing these gaps in knowledge, the investigators will evaluate blood pressures, hyperlipidemia, and hyperglycemia in pediatric patients with SCD who are overweight/obese compared to pediatric patients with SCD who are underweight/normal weight. We will further evaluate diet, exercise, and family medical history as key risk factors for overweight/obesity, hypertension, hyperlipidemia, and hyperglycemia in this population. If an increased risk for hypertension, hyperlipidemia and/or hyperglycemia is found related to overweight/obesity, it will be important for future prospective studies to examine the clinical impact on traditional complications of SCD, including cardiovascular disease, stroke, vaso-occlusive crises, acute chest syndrome, and neuropathies. This knowledge will be critically important as hypertension, hyperlipidemia and hyperglycemia are all risk factors that can be modified with pharmacologic intervention and/or behavioral lifestyle changes such as dietary counselling and exercise. This new knowledge can further lead to updated screening recommendations, as there are currently no standardized guidelines for screening for hypertension, hyperlipidemia and hyperglycemia in pediatric patients with SCD.

Planning Phase: The first phase of the study will include 2 months for Research Nurse training, obtaining study supplies, and finalizing IRB approval (which will be submitted within two weeks of grant submission). Our team anticipates starting participant enrollment in October 2020 and plan to complete enrollment by August 2021. Study recruitment will occur for 8 months of the 1 year study period. The final two months of the project will be dedicated to data analysis, manuscript preparation and additional grant applications.

Study Participants (N=200): The UMMC Pediatric Sickle Cell Clinic is the only specialty pediatric SCD clinic in Mississippi. It is one of the largest pediatric SCD programs in the nations, serving approximately 800 patients annually throughout the state of Mississippi. Pediatric SCD clinic is held twice weekly on the UMMC main campus in Jackson, MS. There is also a multidisciplinary clinic twice monthly and satellite clinics in Hattiesburg, Meridian and Gulfport, MS. Pediatric patients and a primary caregiver/parent (parent) (>21 years old) will be recruited from the UMMC main campus. Approximately 300 pediatric patients with SCD ages 10-19 years were seen in clinic in 2019. Dr. McNaull and Dr. Karlson have an approximate 80% recruitment rate to single-visit research studies in this population. Thus, the investigators do not anticipate any difficulty with recruitment for the current cross-sectional study.

Our team will recruit 60 youth who are underweight/normal weight and 60 youth who are overweight/obese, along with a parent for each youth.

Biological Variable: The following biological variables will be extracted from the participants electronic health record (EHR): age, sex, SCD genotype, and pubertal status according to Tanner Stage criteria. Tanner stage will be determined by the regular medical provider at the clinic visit and extracted from the EHR. SCD occurs equally in males and females. Thus, males and females will be recruited in equal numbers.

Methods: Recruitment: All study procedures will be completed at youth's regularly scheduled hematology appointment. A SCD research nurse coordinator will call potentially eligible families (based on age, SCD genotype, and clinic visit history) one to three days prior to their clinic visit as youth need to come fasting for study-related lab draws at their appointment. If families are interested in the study, then the procedures will be explained over the phone. The research nurse coordinator will be present in clinic the day of the youth's appointment to explain procedures in detail, enroll youth participants and a parent, and obtain written informed consent, permission, and assent. Procedures: After providing consent, the parent will complete two electronic questionnaires in REDCap on iPads (5-15 minutes to complete). One questionnaire will be the Family Nutrition and Physical Activity (FNPA) screening tool to assess youth's nutrition and exercise. The second questionnaire will be a family history evaluation of the most common family risk factors for overweight/obesity and its co-morbidities. Blood pressure measurement will be taken three times in concordance with the 2004 New Task Force Report on High Blood Pressure in Children and Adolescents. Youth participants will then complete study blood collection during regularly scheduled clinical venipuncture blood draws (15 minutes).

Assessments: Body mass index z-score (BMI z-score) will be calculated at the study visit using height (cm) and weight (kg). Trained nurses will take height to the nearest 0.1cm using a standard Stadiometer and weight to the nearest 0.1kg on a calibrated electronic scale while the children are wearing light clothing and no shoes. BMI will then be calculated by dividing the weight in kilograms by the height in square meters (kg/m2) and then be converted to BMI z-score based on age and gender norms. The BMI percentile will be charted using current Pediatric Center for Disease Control guidelines, with underweight defined as <5th percentile, normal weight 5-85th percentile, overweight defined at BMI >85th percentile and obese defined as >95th percentile.

Blood Pressure Measurements: Systolic and Diastolic blood pressures will be collected according to recommendations of the 2004 New Task Force Report on High Blood Pressure in Children and Adolescents using the proper cuff size with routinely calibrated automated blood pressure monitors (SphygmoCor version 8.0, AtCor Medical). Youth will be seated comfortably at rest for 5 minutes with feet flat on the floor. Measurements will be made on the non-dominant arm supported at heart level and will be taken 3 times, one minute apart. The first measure will be used for adaptation purposes and the mean of the second and third measures will be recorded and used for analyses The median and 90th percentile blood pressure in over 1500 children with SCD has been previously published although it did not take into consideration height or include data for each year of age individually. Pre-hypertension, termed relative hypertension in SCD, is defined as blood pressure greater than the 90th percentile but less than the 95th percentile. This new clinical recommendation of using the 90th percentile blood pressure to define hypertension in pediatric SCD will be used in the current study.

Blood pressures for all participants will be classified using the National High Blood pressure Education Program (NHBPEP) Working Group on Children and Adolescent. For our study, (1) Normotensive will be assigned to patients with systolic blood pressure <90th percentile and (2) hypertensive will be reported as any patient with blood pressure >90th percentile for age, sex and height.

Laboratory Blood Draws: Peripheral blood (10 mL) will be collected at scheduled blood draws in green and red top heparinized tubes. Blood will be sent to the laboratory at UMMC for analysis and send out lab analysis to Mayo Clinic Laboratories. The following laboratory analyses will be obtained: CBC, Reticulocyte count, lipid panel, fasting glucose and fructosamine. CBC, reticulocyte count and glucose are laboratory tests that are taken at regular scheduled appointments as a part of maintenance SCD care. Lipid panel and fructosamine are additional tests.

Hyperlipidemia: Fasting lipid panel which includes total cholesterol, LDL, HDL and Triglycerides will be obtain from blood samples from each participant. Each participant will be classified into 1 of 3 risk groups for total cholesterol and LDL lipid panel. Risk groups are defined as follows: (1) Acceptable level <170mg/dL and<110mg/dL for LDL (2) borderline level 170-199mg/dL and 110-129 mg/dL and (3) High level >200mg/dL and >130mg/dL for total cholesterol and LDL respectively. HDL and triglyceride levels will be stratified by percentile based on age and sex. All reference levels are in concordance with lipid screening guidelines from the AAP which were adapted from the Lipid Research Clinic Pediatric Prevalence Study.

Hyperglycemia: Both fasting glucose and serum fructosamine will be drawn to assess for hyperglycemia. Participants will be categorized by fasting glucose: (1) Normal glucose 80-100mg/dL (2) Elevated glucose 101-125mg/dL (3) Hyperglycemia >126mg/dL. Impaired fasting glucose included any subject with glucose >100mg/dL. Serum fructosamine levels will be categorized into (1) Normal levels 203-263umol/L and (2) Elevated >263 umol/L.

Questionnaires: Parent participants will complete the below questionnaires. Food and Exercise Survey: Parents will complete the 20 Question Family Nutrition and Physical Activity (FNPA) screening tool to assess youth's nutrition and exercise. The FNPA instrument was developed at Iowa State University in collaboration with the Academy of Nutrition and Dietetics following a comprehensive evidence analysis examining the influence of physical activity, sport participation, television viewing and video games for their influence on child adiposity. Initial validation was performed in a sample of 854 children and their parents in Des Moines Iowa. Correlational analysis and logistic regression were used to examine associations among FNPA factors and the relationship between these factors and child weight status. Significant correlations were found between BMI and seven of the ten constructs (breakfast/family meals, modeling nutrition, high calorie beverage intake, television in the bedroom, parent physical activity, child physical activity and sleep schedule) as well as between BMI and the overall FNPA score.

Family History Survey: Parents will complete a short family history survey of the most common family risks for overweight and obesity which include: maternal history of BMI greater than the 85th percentile, maternal history of gestational diabetes, family history of overweight/obesity, family history of hypertension, cardiovascular disease, high cholesterol, and T2DM.

Medical Chart Review: In addition to the biological variables above, the following variables will be extracted from the participants EHR from the current clinic visit: demographics (race, insurance payer), history of acute chest, asthma, obstructive sleep apnea, avascular necrosis and stroke, medications, disease severity score calculated based on number of ED visits and hospitalizations during the past 3 years. BMI and clinical visit laboratory results (hemoglobin, reticulocyte count, random non-fasting glucose) over the last 3 years.

Potential Problems and alternative strategies: Recruitment and retention. Despite COVID-19, all pediatric SCD are operating in-person at full capacity. To maximize recruitment and retention, the study nurse coordinator will customize recruitment materials and strategies (e.g., study brochure, visual aids, and confirming Medicaid transportation to appointment). The investigators will use compensation and methodologies (research paired with medical appointments) effective for ≥80% recruitment in previous cross-sectional studies with this population [e.g. Physical Activity and Pain in Youth with Sickle Cell Disease. While the investigators do not anticipate problems with recruitment, the team will plan to add the Meridian and Hattiesburg satellite clinics if 30% below recruitment goals in Month 3. Age range. Another potential concern is the broad age range of 10-19 year olds, with differences in weight potentially being due to pubertal Tanner stage. To address these biological confounds, age will be stratified for in equal numbers during recruitment (10-14 years and 15-19 years) and pubertal Tanner stage will be controlled for in analyses. Finally, youth participants may refuse blood draw, weight/height measurement, or blood pressure measurement. However, these are standard medical procedures that this pediatric patient population is very familiar with by 10 years of age. Our clinic routinely uses ultrasound guided blood draws to reduce the number of needle sticks needed to obtain adequate blood sample. Additionally, every effort to use lidocaine numbing cream and distraction during blood draws will be made for youth who express anxiety regarding venipuncture blood draw.

Statistical Analyses Power and Sample Size: Statistical analyses will be conducted with the MCCTR Biostatistical Core (Seth Lirette, PhD). G*Power 3.0.10 chi-squared a-priori power analyses specified that a sample size of 100 youth (50 per weight-status group) will provide greater than 90% power for detecting differences in the prevalence rate of relative hypertension (primary outcome) in youth with SCD who are overweight/obese versus youth with SCD who are underweight/normal weight. An anticipated large effect size (.8) for power analyses is based on the prevalence rate of 4.5% for hypertension in African American youth 10-19 years who were obese compared to a prevalence rate of 40.2% for relative hypertension in African American young adults 18-20 years with SCD.

Attrition and Missing Data: The investigators anticipate a small portion (approximately 5%) of unusable laboratory data due to clinical difficulties with blood draw (e.g., nurse not able to draw enough blood). Based on the above power analyses, the total target sample of N=100 youth will provide a large enough sample size to account for this type of random missing laboratory data. Questionnaires will be checked by the study nurse coordinator for completeness prior to study completion. Random missing questionnaire data will be handled or imputed using author recommendations or mean imputation strategies.

Analytic plan Aim 1: Statistical analyses will be calculated in SPSS 26.0 or Stata 15.0. Prevalence rates of hypertension, hyperlipidemia and hyperglycemia will be calculated for the two study groups. A series of non-parametric exact Fisher tests will examine the prevalence rate of hypertension, hyperlipidemia and hyperglycemia in pediatric patients with SCD who are overweight or obese (i.e., have a BMI ≥ 85th percentile) compared to their underweight and normal weight (i.e., have a BMI 1st to 84th percentile) counterparts, controlling for Tanner pubertal stage.

Analytic plan Aim 2: A series of logistic regressions will examine the differential impact (odds ratio) of diet, exercise, and family history of overweight/obesity on the risk for weight status, hypertension, hyperlipidemia, and hyperglycemia in youth with SCD, controlling for Tanner pubertal stage.