Children's Healthcare Of Atlanta Research Articles

Association of Age with Acuity and Severity of Illness at Initial Presentation in Children, Adolescents, and Young Adults with Leukemia

Background: Adolescent and Young Adult (AYA) patients, historically defined by the National Cancer Institute as patients between 15-39 years, diagnosed with cancer are at risk for increased mortality compared with younger patients. Age 10 has been an alternate threshold for higher risk of treatment failure in leukemia. Acuity of illness - as defined by intensive care unit (ICU) level resource use at admission - in children with acute myeloid leukemia (AML) was shown in prior data from the Pediatric Health Information System to mediate excess mortality in Black patients (Winestone et al, Am J Hematol., 2017). This study aims to evaluate the acuity and severity of illness in patients presenting with new diagnoses of leukemia and evaluate whether AYAs have a higher acuity and severity of illness compared to younger patients. Methods: We performed a retrospective analysis of 771 patients aged 1-21 years with initial presentation of newly diagnosed leukemia at Children's Healthcare of Atlanta (CHOA) between 2010-2018. Exclusion criteria included those who received any treatment at another facility or those for whom information about the first 72 hours following presentation to the hospital was not available. High acuity of illness was defined (yes/no) as any ICU admission or ICU-level resource use - regardless of physical location in the hospital - by organ system (cardiovascular, respiratory, renal, hematologic, and neurologic) on manual chart review. High severity of illness was defined as initial white blood cell (WBC) count ≥ 50,000 cells/microliter or central nervous system (CNS) disease. Both acuity and severity were assessed during the first 72 hours following the point of diagnosis. Multivariable logistic regression was used to examine the association of age at diagnosis with acuity or severity of illness, controlling for sex, race/ethnicity, insurance status, and type of leukemia. Results: Among 536 patients with complete data to date, the median age was 6 years (interquartile range 3-12 years, 65.5% aged 1-9 years, 34.5% aged 10-21 years), 53.4% were male, 45.7% were Non-Hispanic White, and 65.7% had private insurance. Diagnoses included B-cell acute lymphoblastic leukemia (ALL) (70.5%), AML (16.4%), T-cell ALL (9.3%), and other leukemias (3.7%) (juvenile myelomonocytic, chronic myeloid, and Burkitt's). High acuity of illness was seen in 24.2% of patients and was more likely in older patients (10-21 years, 33.5% versus 1-9 years, 19.4%, p <0.01). Figure 1 illustrates involved organ systems by age group. High severity of illness was seen in 31.3% of cases, and although not significant, was also more likely in older patients (10-21 years, 32.5% versus 1-9 years, 27.9%, p=0.06). In multivariable analysis, compared to patients aged 1-9 years, patients aged 10-21 years had an increased odds of high acuity of illness [odds ratio (OR) 1.69, 95% confidence interval (CI 1.09-2.63)]. Patients with T-ALL also had an increased odds of high acuity of illness [OR 3.40 (CI 1.76-6.54); referent B-ALL]. The odds of high severity of illness did not differ by age [10-21 years OR 1.23 (CI 0.81-1.86)], but was significantly higher by Non-Hispanic Black race [OR 1.83 (CI 1.14-2.94); referent Non-Hispanic White] and type of leukemia [T-ALL OR 3.90 (CI 2.03-7.48); other leukemia OR 3.26 (CI 1.28-8.32); referent B-ALL] (Table 1). Discussion and Conclusions: AYA patients had an increased likelihood of high acuity of illness requiring ICU-level care at presentation, indicating a risk for disparities of outcomes from the point of initial presentation. Future directions will examine the association of acuity of illness with mortality in this cohort, as well as the role of race/ethnicity on severity of illness and mortality. This research will ultimately help inform strategies toward narrowing age disparities in outcomes of AYA leukemia.

Non-Invasive Screening for Sickle Cell Disease Using Only a Smartphone Application

Introduction Pediatric mortality due to Sickle Cell Disease (SCD) remains high and often occurs prior to diagnosis in most black and brown low- and middle-income countries (LMICs), due to the high prevalence of SCD and limited resources in these countries.To that end, we developed a low-cost non-invasive screening tool that can be widely available at the point of need. Our screening tool utilizes the power of both image analysis and predictive algorithms as a novel approach to SCD screening and diagnostics. In this new app, we have combined the quantitative analysis of fingernail pallor, an adaptation of our previously described smartphone app for estimating blood hemoglobin levels using smartphone images of the nail beds, with qualitative symptom and medical history data to compute an SCD Likelihood Score (SLS) that provides an individual's pre-diagnostic test potential for disease, a key metric in the disease screening and diagnostic process. Methods 417 subjects ages 0.5-20 years old were enrolled from Children's Healthcare of Atlanta (CHOA) hematology/oncology clinic; of these patients, 219 patients had SCD, and 198 did not. Pediatric patients were screened because they best simulated the target populations aimed for the app's use. Our analysis was restricted to the presence of specific pain, past stroke history and/or symptoms, splenic crises, neurodevelopment disorder, family history of SCD, due to their specificity to SCD and their ability to be answered without the need for resource-intensive diagnostics. The SLS was calculated by determining the presence of these medical history items and the patient's anemia status informed by the patient's most recent CBC. For each symptom a weighted score was generated by finding the difference between the percentage of subjects with SCD and those without SCD, who experienced the symptom. The presence of symptoms was binarized with 1 indicating the presence of a symptom and 0 indicating the absence of a symptom. The binary value of the symptom was then multiplied by the symptom's respective weighted value. These values were totaled and normalized to give the overall clinical profile-weighted score with the same weight as the anemia status. A receiver operating characteristic analysis was conducted to select a threshold for the SLS that provided the most accurate diagnosis in our training dataset. We incorporated this SLS model into our lab's previously described smartphone app predictive algorithm . Results A clinical assessment of the screening app was conducted at CHOA and enrolled 68 participants from both in-patient and out-patient settings. Patient hemoglobin measurements were obtained using our previously described hemoglobin screening app and patients (and their parents for those needing their assistance) were asked to provide responses to the symptom survey. Of the five medical history items queried, family history and a history of splenic crisis when present was found to contribute to 30% and 41%, respectively, of the overall clinical profile score. Whereas the presence of a history of pain, stroke history, and learning difficulties made up smaller portions of this portion of the SLS. The app performed with a sensitivity of 100% and a specificity of 75% in screening for SCD in patients <5 years of age, the target demographic for early screening. Conclusion This SCD screening application has the potential to expand and improve SCD outcomes in LMICs where access to the gold standard or any accurate diagnostic tools is limited. The app's non-invasive and easy-to-use nature and the ubiquity of smartphones globally make this screening tool a practical and feasible mode of screening. Early disease screening is paramount to survival for children with SCD and if this tool proves to be accurate in these settings, can be lifesaving for the hundreds of thousands of infants with SCD born in LMICs per year. Non-invasive modalities such as our SCD screening application can enable resource allocation and cost reduction by identifying those patients who do and do not have a high likelihood of disease, thus reserving tests for those who are high risk.

Impact of Survivorship Care Plans (SCP) on Mortality in a Contemporary Cohort of Pediatric Patients Diagnosed with a Blood Cancer

Background: The 5-year overall survival for pediatric blood cancer (i.e., leukemia, lymphoma) is greater than 85%. However, blood cancers and their treatment predispose survivors to the risk of late effects and premature mortality. This study examines the association between the receipt of a personalized survivorship care plan (SCP) and mortality among survivors of childhood blood cancer. Methods: This retrospective cohort included survivors who were treated for a newly diagnosed blood cancer at Children's Healthcare of Atlanta (CHOA) between 2002-2016 and in remission for ≥2 years post-treatment (i.e., eligible to be seen in Aflac Cancer Survivor Clinic). Descriptive analyses were conducted to examine SCP receipt and mortality through 12/31/2020 based on linkage to the National Death Index. The cumulative incidence of death was estimated from the time of eligibility for survivor clinic until the date of death, 12/31/2020, or five years (y) post-eligibility date, whichever occurred earlier. A multivariable Cox proportional-hazard model was used, with SCP receipt as a time-dependent variable, to estimate the association between SCP receipt and 5-year mortality, controlling for age at diagnosis, sex, race/ethnicity, and cancer therapy exposure. Results: Among the 1,239 eligible survivors identified, 982 (79.3%) received a SCP at a median time of 0.50 y (IQR: 0.21, 0.84 y) from eligibility for survivor clinic. Recipients of a SCP (vs. non-recipients) were more likely to be female (44.8% vs. 38.1%; p=0.055), non-Hispanic White (56.6% vs. 47.5%; p=0.006), and younger at the initial cancer diagnosis (mean ± SD age: 8.4 ± 5.4 vs. 11.0 ± 5.8 y; p<0.001). Receiving a SCP differed by cancer type: 82.3% of patients diagnosed with acute lymphoid leukemia (ALL), 86.3% of acute myeloid leukemia (AML) patients, and 80.6% of other leukemia patients received a SCP vs. 68.6% of Hodgkin lymphoma patients and 76.1% of non-Hodgkin lymphoma patients (p<0.001). Therapy exposure differed among SCP recipients, with 75.6% receiving chemotherapy alone, 12.9% chemotherapy and radiation therapy, and 11.5% stem cell transplantation (SCT) The unadjusted 5-year survival was not significantly different between survivors who received a SCP and those who did not (98.8% [95% CI: 98.1, 99.6] vs. 96.4% [93.9%, 98.9%], p=0.065). In multivariable analysis, receipt of a SCP was marginally associated with a lower risk of death (adjusted hazard ratio [aHR]: 0.37 [95% CI: 0.14, 1.01], p=0.053; Table 1). Additionally, receipt of SCT (aHR: 9.16 [95% CI: 3.34, 25.1], p<0.001) was associated with a higher risk of death. Conclusions: Early evaluation of our cohort eligible for a survivorship clinic suggests an association between the receipt of a SCP and a higher adjusted 5-year survival in pediatric blood cancer survivors. Receiving an SCT remains an independent risk factor for mortality. Longitudinal follow-up of this cohort allows for the assessment of mortality trajectory and pattern. Capture of healthcare utilization will inform the impact of survivorship care and further define lifetime risk after contemporary treatment.

Causes of Death in Patients with Sickle Cell Disease from a Large Population-Based Cohort

Background Mortality rates and causes of death in children with sickle cell disease (SCD) have changed significantly over the past several decades. With ongoing improvements in standards of care, modern mortality rates and causes of death must be updated. Children's Healthcare of Atlanta (CHOA) houses one of the largest pediatric SCD programs in the country. This analysis reviews causes of death in children with SCD who were treated at CHOA between June 2010-June 2022. Methods We reviewed the CHOA's Sickle Cell Clinical Database (SCCD) for deceased patients. Demographics, SCD genotype, date and age at death, healthcare utilization, hydroxyurea (HU) use, chronic transfusion therapy (CTT), and bone marrow transplant (BMT) were obtained from the SCCD. Cause of death and history of significant comorbidities were abstracted from the medical record, death certificates and/or autopsy reports. Results A total of 4022 unique patients with SCD were seen at CHOA from June 2010- June 2022. Fifty-seven patients died during this period. All but 1 patient had been seen in the CHOA Sickle Cell Outpatient Clinic at least once. That patient received SCD care at a different institution, died from complications of a non-SCD related disease following transfer to CHOA, and was excluded from analysis. SCD-related causes accounted for 22 (39%) deaths. Acute SCD-related causes included bone marrow fat emboli, acute chest syndrome or pneumonia, cerebral hemorrhage in a person with history of stroke and moyamoya, and pulmonary hypertensive crisis. Of the infectious causes included Streptococcus pneumoniae, gram negative bacteria, unspecified streptococcal meningitis (by verbal report), and due to presumed septic shock complicated by severe anemia. Viral infections included respiratory syncytial virus, COVID-19 with severe anemia, and Parvovirus B-19 complicated by bone marrow fat emboli. Death due to complications of treatment included ceftriaxone-induced hemolysis and severe intrathoracic hemorrhage after apheresis catheter removal for treatment of ACS. BMT-associated complications included adenovirus infection after transplant with a matched unrelated donor and Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis after haploidentical transplant (see Figure 1). Non-SCD comorbidities or accidents accounted for 15 (27%) deaths. These included heart disease, advanced cancer or complications of cancer (including E. coli sepsis in a child with relapsed neuroblastoma), accidents, one due to motor vehicle accident and one drowning. Gastrointestinal causes included bleeding from undiagnosed inflammatory bowel disease and liver failure leading to coagulopathy and intracranial hemorrhage. There was a case of severe anemia related to systemic lupus erythematosus. The causes of death were unknown in the remaining 19 (34%) of individuals. Six (11%) were under 18 years of age. Five died at home or out of state and had no autopsy information, one had autopsy declined by the coroner. Of those who were 18 years of age or older, 7 (53%) were lost to follow-up, 3 (23%) had transitioned to adult care, and 3 (23%) were still being treated at CHOA. There were more deaths in the 0-2 year and 15-17 year groups, and the most among adults (n = 19, 34%). Death from congenital heart disease was more frequent in children 0-5 years old. Pneumococcal sepsis were in children 5-7 years old, after stopping penicillin and before the second Pneumovax vaccine. Other bacterial infections were evenly distributed by age. Non-infectious SCD-related deaths were more likely to occur in patients 12 years and older (n = 6, 86%) (Figure 2). Conclusions Continuing from a previous study by Lai et al. (2020) which reported the 10-year mortality rate at CHOA, this study described the causes of death of SCD patients during a 12-year period. Of direct SCD-related causes, bone marrow fat embolism was the most common and was also present in a patient who also had Parvovirus-B19 infection. Severe anemia was present in several patients at the time of death, associated with ceftriaxone-induced hemolysis, presumed septic shock, COVID-19, systemic lupus erythematosus, as well as G or intracranial hemorrhage. Patients beginning transition to adult care are at risk of becoming lost to follow-up and may experience difficulties transitioning to adult care. Future analyses will include review of regional hospital where available, and CDC's National Death Index. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Domain Adaptation Using Convolutional Autoencoder and Gradient Boosting for Adverse Events Prediction in the Intensive Care Unit.

More than 5 million patients have admitted annually to intensive care units (ICUs) in the United States. The leading causes of mortality are cardiovascular failures, multi-organ failures, and sepsis. Data-driven techniques have been used in the analysis of patient data to predict adverse events, such as ICU mortality and ICU readmission. These models often make use of temporal or static features from a single ICU database to make predictions on subsequent adverse events. To explore the potential of domain adaptation, we propose a method of data analysis using gradient boosting and convolutional autoencoder (CAE) to predict significant adverse events in the ICU, such as ICU mortality and ICU readmission. We demonstrate our results from a retrospective data analysis using patient records from a publicly available database called Multi-parameter Intelligent Monitoring in Intensive Care-II (MIMIC-II) and a local database from Children's Healthcare of Atlanta (CHOA). We demonstrate that after adopting novel data imputation on patient ICU data, gradient boosting is effective in both the mortality prediction task and the ICU readmission prediction task. In addition, we use gradient boosting to identify top-ranking temporal and non-temporal features in both prediction tasks. We discuss the relationship between these features and the specific prediction task. Lastly, we indicate that CAE might not be effective in feature extraction on one dataset, but domain adaptation with CAE feature extraction across two datasets shows promising results.

Health Disparity in Hematopoietic Cell Transplantation for Sickle Cell Disease: Analyzing the Association of Insurance and Socioeconomic Status Among Children Undergoing Hematopoietic Cell Transplantation

OBJECTIVE: The impact of racial and socioeconomic status (SES) on access to hematopoietic cell transplantation (HCT) are well known. Even when controlled for SES, racial minorities have lower access to HCT than the racial majority groups. Since the majority of patients with sickle cell disease (SCD) in the US are of African descent, HCT for SCD provides a unique opportunity to study the impact of SES on access to HCT in a relatively uniform racial population. The objectives of this study were to explore whether insurance status and SES are associated with the HCT for SCD.METHODS: The clinical database from the SCD program at Children's Healthcare of Atlanta (CHOA), which contains over 3,100 patients, was used to identify SCD patients who have undergone HCT between January 1, 2012 and December 31, 2016 at CHOA. We then identified non-transplanted SCD patients matched for age at transplant, gender, and SCD genotype. Three matched controls were randomly selected for each case except for one older patient who only had two controls. Non-Georgia residents during the year of transplant were excluded; one sickle cell patient who had BMT due to acute lymphoblastic leukemia was also excluded. Among both the transplanted patients and their controls, insurance status was assessed as well as socioeconomic status was estimated based on block-level census data, specifically household income, home value, percent single parent households, percent unemployment, and education level. Conditional logistic regression analysis using backwards elimation was performed to determine the odds of receiving HCT.RESULTS: Of 34 patients with SCD who underwent HCT at CHOA during this time period, 27 (66.7% females and 33.3% males) were Georgia residents. Twenty-five of the 27 patients received a matched-sibling donor transplant and two had matched-unrelated donor transplant. Among the HCT patients, the median age at transplant was 8 ± 4.7 years (range 1-20 yr). Of the 27 HCT patients, 59.3% had Medicaid and 40.7% had Private insurance at the time of transplant. Among the controls, 81.3% were covered by Medicaid insurance, 17.5% with Private insurance, and 1.3% self-pay. Using a conditional logistic regression model, we found insurance status to be a statistically significant factor in predicting whether a patient undergoes HCT. The model showed that those with Private Insurance are 2.96 (95% Confidence Interval: 1.02, 8.54) times more likely to undergo HCT compared to patients with Medicaid and self-pay. No other socioeconomic status variables were associated with receiving HCT.CONCLUSION: The data suggest that there may be health disparities in the application of HCT for SCD and that SCD patients who have private insurance are more likely to undergo HCT than those with Medicaid insurance. Important limitations of this study include absence of information about the availability of matched-sibling donors among the non-transplanted patients. A more detailed analysis of the impact of disease severity on the application of HCT for SCD is warranted. Similarly, these data provide the rationale for a larger study of national trends on the impact of the type of insurance and SES on access to HCT for SCD as well as factors that may be contributing to this health disparity. DisclosuresNo relevant conflicts of interest to declare.

Increased Proportion of Sickle Cell Disease Patients Develop Cardiovascular Disease Risk Factors Following Successful Hematopoietic Cell Transplantation

Background: Hematopoietic cell transplantation (HCT) is currently the only proven curative treatment for sickle cell disease (SCD), with >90% event-free survival using a matched related donor (MRD). Long-term follow up in patients transplanted for malignant disease demonstrates a 4-fold increase in risk of developing cardiovascular disease (CVD). Risk factors for CVD include obesity, hypertension, diabetes, and dyslipidemia, which are reported in a higher incidence in both survivors of HCT and childhood malignancy. While patients with SCD typically have lower weight and blood pressure (BP) for age, cure of SCD may lead to decreased metabolic demand as well as increased weight and BP, thus uniquely predisposing post-transplant SCD subjects to CVD risk. Further, because of premature mortality due to SCD, true risk of CVD in these patients may be unknown.Objective: To investigate the development of CVD risk factors in SCD subjects who underwent HCT with myeloablative conditioning.Methods: We performed a retrospective data collection on SCD subjects who underwent allogeneic HCT at Children's Healthcare of Atlanta (CHOA) and survived to ³1 year post-HCT. Data on baseline patient and transplant characteristics as well as CVD risk factors (body mass index [BMI], BP, and cholesterol) were collected. BMI was categorized per CDC definition as underweight (<5th%), healthy weight (5th% to <85th%), overweight (85th% to <95th%), or obese (≥95th%). BP was categorized as >50th% and >90th% for patients with SCD, as per Table 1 in Pegelow et al. Pre- and post-HCT values were compared using Wilcoxon signed-rank tests or McNemar's tests, with p-values <0.05 considered statistically significant.Results: Sixty-seven SCD subjects underwent allogeneic HSCT from December 1993 to March 2016 at CHOA, 63 of whom survived to ≥1 year post-transplant. Fourteen patients did not have pre- or post-HCT BMI data available, thus were excluded from the analyses. Majority had clinically severe SCD (77.8%) and received bone marrow (93.3%) from a matched sibling donor (91.1%). Conditioning was myeloablative in all, majority using a combination of busulfan, cyclophosphamide, and anti-thymocyte globulin ± fludarabine. Patients underwent HCT at a median age of 8.8 years (range 1.8-20.3), at which time 40 (88.9%) were healthy weight, while 3 of each (6.7%) were underweight, overweight, and obese (Table 1). At last follow up (median age 10.3 years; range, 3.8-21.8), 27 patients (60.0%) were healthy weight, while 6 (13.3%) were underweight, 5 (11.1%) overweight, and 7 (15.6%) obese. When pre- and post-transplant BMI categories were compared, 66.7% of patients had no change, whereas 17.8% increased and 15.6% decreased. Of the 8 patients whose BMI category increased, 6.7% became overweight and 11.1% obese. Of the 7 patients who BMI category decreased, 2 obese patients (28.6%) became overweight, 2 overweight patients (28.6%) became obese, and 3 normal weight patients (42.9%) became underweight. BMI significantly increased from a median of 16.2 kg/m2 pre- to 19.0 kg/m2 post-HCT (p<0.001). There was no increase in the proportion of patients exceeding either the 50th or 90th percentile for systolic BP post-HCT (Figure 1). The proportion with diastolic BP exceeding 50th or 90th percentile for age for SCD was increased post-HCT compared to baseline. Five patients were diagnosed with hypertension ≥6 months post-transplant, 4 of whom were receiving treatment for chronic graft-versus-host disease (GVHD; steroids, cyclosporine [CSA], or steroids plus CSA) and 1 of whom was overweight. Routine cholesterol testing was rare, consistent with a pediatric cohort, with only 2 patients having hypercholesterolemia, both of whom were receiving treatment for chronic GVHD. Two patients had mildly elevated hemoglobin A1c levels post-HCT and none were diagnosed with diabetes. Ten patients developed chronic GVHD (22.0%), 8 of whom remained in a stable BMI category, with 1 of each either increasing or decreasing.Discussion: Following successful HCT, patients with SCD developed relative diastolic hypertension and increased BMI, both of which are known risk factors for CVD. Patients who developed chronic GVHD were more likely to be diagnosed with hypertension and/or hypercholesterolemia, although BMI category was not impacted. These data provide the rationale for longer term follow up for CVD risk factors and CVD in patients with SCD following HCT. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Patterns of Pediatric Facial Fractures.

The aim of this study was to assess patterns of maxillofacial trauma in the pediatric population in Atlanta. This information is important to help guide management and allocate resources for treatment of maxillofacial injuries at Children's Healthcare of Atlanta (CHOA). This study was a retrospective chart review of children who presented from 2006 to 2015. Inclusion criteria were: (1) age 18 years old or younger, (2) presentation to emergency department, (3) diagnosis of maxillofacial fractures, and (4) evaluation by Oral and Maxillofacial Surgery, Otolaryngology, or Plastic Surgery services. Medical records were reviewed to record demographic, mechanism of injury, fracture location, and yearly incidence of injury. Descriptive statistics were computed to summarize findings and overall trends. During the study period, 39,833 patients were identified. Of them, 1995 met the inclusion criteria. The majority were male (n = 1359, 68%) with an average age of 9.4 years old (range of 1 month to 18 years old). Mechanisms of injury were motor vehicle collisions (MVC) (n = 597, 29.9%), fall (n = 565, 28.3%), sports injury (n = 317, 15.9%), pedestrian struck (n = 215, 10.8%), assault/abuse (n = 204, 10.2%), other (n = 81, 4.1%), or gunshot wound (n = 16, 0.8%). Fracture sites were mandible (n = 519, 26%), complex (n = 479, 24%), nasal (n = 419, 21%), dentoalveolar (n = 279, 14%), orbital (n = 259, 13%), and maxilla (n = 40, 2%). Males had a higher incidence of assault than females (n = 185, 91% of assaults). The incidence of maxillofacial trauma increased with age with a peak incidence in 13 to 16-year-olds (n = 566, 28.3%). During the years examined, there was an upward trend in MVCs as the etiology with a peak incidence of facial fractures due to MVCs occurring in 2015. All other mechanisms remained constant during this time period. There was an increase in pediatric facial fractures secondary to motor vehicle collisions from 2007 to 2015 despite improvements in regulations, traffic safety, and technology.

10-Year Mortality in Patients with Sickle Cell Disease from a Large Population-Based Cohort

Background Mortality rates and causes of death in children with sickle cell disease (SCD) have changed significantly over the past several decades. With ongoing improvements in standards of care, modern mortality estimates must be updated. Children's Healthcare of Atlanta (CHOA) houses one of the largest pediatric SCD programs in the country. This analysis reviews mortality in children with SCD who were treated at CHOA between June 2010-June 2020. Methods We reviewed the CHOA's Sickle Cell Clinical Database (SCCD) for deceased patients. Demographics, SCD genotype, date and age at death, healthcare utilization, hydroxyurea (HU) use, chronic transfusion therapy (CTT), and bone marrow transplant (BMT) were obtained from the SCCD. Cause of death and history of significant comorbidities were abstracted from the medical record. Mortality rates were calculated using person-time for all CHOA SCD patients and population-based mortality was obtained from CDC WONDER Online Database. Results A total of 3,698 patients with SCD were seen at CHOA from June 2010-June 2020 and accounted for 19,998 person-years. Of the 46 patients who died during that time, all but 1 patient had been seen in the CHOA Sickle Cell Outpatient Clinic at least once. That patient received SCD care at a different institution, died from complications of a non-SCD related disease following transfer to CHOA, and was excluded from analysis. Of the 45 remaining patients (178 person-years), the majority were sickle cell anemia genotypes (n=37, 82%; Hb SS or Hb S β0 thalassemia), followed by Hb SC (n=5, 11.1%) and Hb S β+ thalassemia (n=3, 6.7%); 53% (n=24) were female. The average age at death was 12.8 years (1.0-22.8 years). Twenty-one (46.7%) patients had ever been treated with HU and 11 (24.4%) were currently on HU at the time of last contact (either death or last CHOA encounter). Eleven (24.4%) patients had ever been treated with CTT, and 3 (8.9%) were being treated with CTT at time of last contact. In comparison, during the same time period an average of 58% and 12% of the overall CHOA SCD population were being treated with HU and CTT, respectively. Forty-one patients had at least 1 SCD-related encounter at CHOA within 12 months prior to death. For the 4 patients who had not been seen at CHOA within 12 months, the average time since last contact was 1.72 years (1.03 - 2.7 years). Eighty percent (n=8) of the 10 patients who died at age &amp;gt;19 had either recently transitioned to adult care or had documentation of a transition plan. During this 10-year period, the crude death rate was 2.3 per 1,000 person-years. The majority of deaths (62%, n=28) were attributable SCD-related causes and corresponded to a cause-specific mortality rate of 1.40 per 1,000 person-years. The remaining 38% (n=17) of deaths were caused by complex non-SCD comorbidities or accidental trauma and corresponded to a non-SCD related mortality rate of 0.85 per 1,000 person-years. In comparison, the mortality rate for African American persons age &amp;lt;22 in Georgia from 2009-2018 was 0.9 per 1,000 person-years. Of the 28 patients who died due to an SCD-related cause: 12 (27%) experienced an acute illness related to SCD, 4 (9%) succumbed to acute bacterial infections, 3 (7%) died from complications of SCD treatment (1 procedure-related, 2 drug-induced hemolytic anemia), and 1 (2%) died from complications of BMT. Nine (20%) had either recently transitioned to adult care or were lost to follow-up and had no significant non-SCD comorbidities, therefore cause of death was unknown. Of the 17 non-SCD related deaths, 16 (36%) were due to complex non-SCD comorbidities (including cancers, autoimmune disorders, and congenital heart disease) and 1 (2%) from an accidental trauma. Conclusions To our knowledge, this is the largest and most recent study of mortality in patients with SCD from a single institution. Overall, the demographics of deceased patients were similar to those of the CHOA SCD population as a whole. This cohort indicates that trends in mortality of children with SCD have largely shifted away from bacterial infections to other complications of SCD, subsequent treatments, or comorbidities. Our data confirm that patients nearing transition to adult care are at high risk of mortality. This study is limited to deaths that are known to the CHOA system, so future analyses will expand this cohort to include data from death certificates and CDC's National Death Index. Disclosures Lane: FORMA Therapeutics: Membership on an entity's Board of Directors or advisory committees; Global Blood Therapeutics: Membership on an entity's Board of Directors or advisory committees.

Healthcare Utilization Among Children and Adolescents with Sickle Cell Disease during the COVID-19 Pandemic

BackgroundEarly experience with the COVID-19 pandemic showed disproportionately high morbidity and mortality among individuals with certain chronic medical conditions. Individuals with sickle cell disease (SCD) are at high risk for pulmonary and other complications including acute chest syndrome (ACS) and have high rates of hospitalization from other viral respiratory infections, raising concern that COVID-19 would be associated with higher morbidity, mortality and health care utilization among those with SCD. Public health interventions such as social distancing, avoidance of large group activities, and widespread use of masks have been shown to reduce the transmission of COVID-19 in the general population but have been inconsistently implemented. In Georgia, COVID-19 restrictions, including school closures, were implemented in mid-March, and on-site school instruction was replaced by virtual instruction for the remainder of the school year. At our institution, most routine, non-urgent outpatient clinic visits were cancelled or postponed from mid-March through May in order to minimize COVID-19 exposure risk. Efforts to initiate the use of telemedicine as an alternative to in-person office visits were rapidly instituted. We hypothesized that adherence to public health restrictions, especially sheltering in place, would be high among patients and families with SCD, and sought to measure the impact of COVID pandemic on healthcare utilization in children and adolescents with SCD in the Atlanta area.MethodsThe SCD Program at Children's Healthcare of Atlanta (CHOA) provides comprehensive outpatient, emergency and inpatient services at 3 locations in metropolitan Atlanta. CHOA's Sickle Cell Clinical Database (SCCD) contains prospectively collected demographic, diagnostic, treatment and other clinical information on all patients with SCD beginning in 2010, including all outpatient clinic, emergency department (ED) and inpatient hospital utilization. To assess the impact of COVID-19 on healthcare utilization, we tracked clinic, ED and inpatient utilization for the 4-month period (March through June) 2020 compared with the same 4-month period in 2018 and 2019.ResultsThe figure shows utilization patterns for each four-month period from 2018-2020. As expected, face to face outpatient clinic visits fell dramatically from February to April 2020 (-25% in March, -64% in April) and then returned to pre-COVID levels by June. The addition of telemedicine visits raised total outpatient visits in June 2020 to above pre-COVID levels. Total utilization during the 4-month period in 2020 were compared to the mean for the same periods in 2018 and 2019. Face to face clinic visits decreased from 2971.5 to 2023 (-32%), ED visits from 1,217 to 687 (-44%), and total inpatient admissions from 699 to 410 (-41%). Admissions with a primary discharge diagnosis of pain decreased from a mean of 407 in 2018-2019 to 173 (-57%), fever/infection from 67.5 to 40 (-41%), and ACS from 101 to 75 (-26%). Patients with chronic pain and/or history of high utilization (>5 admissions in a given year) showed decreases in utilization similar to all other patients.SummaryThese data describe the significant changes in utilization among pediatric patients with SCD during the COVID-19 pandemic. Face to face outpatient clinic visits decreased during March and April but returned to pre-COVID levels in June. Unexpectedly, ED and inpatient hospital utilization for acute illness decreased dramatically through April and remained low through June. In March there was a significant decrease in the clinic setting due to a large number of cancelled or rescheduled outpatient visits, despite many being rescheduled as telemedicine visits. However, the largest unexpected decrease was seen in emergency department visits and hospitalizations for acute events, specifically fever and pain events. It is also important to note the decreased utilization of patients with chronic pain who are typically high utilizers. During clinic encounters, families mentioned that less stress from school, reduced respiratory infections, and better medication adherence with parents at home, were possible contributors to reduced sickle cell symptoms while sheltering in place. These observations will guide the development of a patient survey with the goal of obtaining qualitative data to explain the reasons for decreased utilization during the pandemic.FigureDisclosuresLane:FORMA Therapeutics: Membership on an entity's Board of Directors or advisory committees; Global Blood Therapeutics: Membership on an entity's Board of Directors or advisory committees.

Obtaining a Von Willebrand Evaluation at Time of Acute Heavy Menstrual Bleeding Presentation Leads to Overestimation of Von Willebrand Levels

Background: Acute heavy menstrual bleeding (HMB) is common for adolescent females, with about a quarter of menstruating females seeking care for HMB over a 3-year time period (O'Brien et al, Blood 2018). Inherited bleeding disorders are common in this adolescent population, identified in 24.6% referred for hematologic evaluation (Zia et al, Blood 2016). The timing and contents of the hemostatic workup for acute HMB in adolescents is extrapolated from adults, although the causes of acute HMB varies significantly between adult women and adolescents. A consensus statement by the American College of Obstetrics and Gynecology recommends obtaining a variety of hemostatic tests including CBC, von Willebrand studies, factor VIII, prothrombin time, partial thromboplastin time, fibrinogen, and thyroid stimulating hormone at the time of presentation (Committee Opinion 557, ACOG 2011). Factor VIII and Von Willebrand studies are known to be increased in the setting of physiologic stress and supplemental estrogen use, questioning their diagnostic accuracy in the setting of acute bleeding. Repeat testing is often required for diagnosis of von Willebrand disease A von Willebrand factor antigen (VWF:Ag) or von Willebrand factor ristocetin cofactor level over 100IU/dL has been shown to have a negative predictive value (NPV) of 95%.(Doshi et al, ASH 2018). Methods: As part of a quality improvement initiative to improve the evaluation and management of adolescents with HMB at Children's Healthcare of Atlanta (CHOA), we instituted an acute HMB protocol for emergency department (ED) and inpatient use. This protocol was implemented at all CHOA emergency departments in metropolitan Atlanta. Subjects were included if they presented with acute HMB as determined by an adapted Philip Menorrhagia Screening Tool. Subjects with a previously diagnosed bleeding disorder, ITP, active rheumatologic disease, cancer, or anticoagulant use were excluded. Descriptive statistics were used to summarize demographics and clinical characteristics. Patients with a positive Philip screen underwent a uniform bleeding inventory and a standardized set of laboratory tests based on the adult consensus statement. Inpatient and outpatient treatments were standardized by hemoglobin level and symptomology. Follow up with hematology and gynecology was encouraged for all. Data was extracted using various heavy menstrual bleeding ICD-10 codes from January 1, 2017 to December 31, 2018. Individuals with von Willebrand studies at baseline and follow up were identified. T-tests and Wilcoxon rank sum tests were utilized to compare VWF:Ag, VWF:RCo and Factor VIII as baseline and follow up. Results: Over a 2-year period, 232 adolescent girls were seen in CHOA EDs for acute HMB with 88 (37.9%) requiring admission and 6 (2.6%) requiring intensive care. The population was primarily African American (63%) with a median age at presentation of 14.8 years (IQR 13.1-16.7). The majority of adolescents had the core hemostatic labs drawn (55.6%) as described per protocol. Thirty-six individuals had baseline and follow up VWD studies. Those with repeat VWD studies were younger (median 13.2 years vs 15.0 years), more commonly white (44.4% vs 21.2%), were more likely to have been admitted (83.3% vs 29.6%) and more likely to have had a hematology follow up appointment (63.4% vs 7.8%). Mean and median VWF:Ag, VWF:RCo and Factor VIII were significantly higher at presentation with HMB than at follow up. Of those with a baseline VWF:Ag and/or VWF:RCo &amp;gt;100, there was a 96.4% NPV for the diagnosis of VWD. For individuals whose initial VWF:Ag and VCWF:RCo were both &amp;gt;100, there was 100% NPV. Conclusions: Among the adolescents cared for at our institution with acute HMB who had confirmatory VWD testing, initial VWF:Ag and VWF:RCo &amp;gt;100 ruled out VWD based on repeat testing. However, poor adherence with hematology or gynecology follow-up may give false reassurance against a diagnosis of VWD. Further improvements of our quality improvement initiative will include a limited hemostatic workup at presentation with a focus on improved adherence to follow up and subsequent hemostatic evaluation. Disclosures White: National Hemophilia Foundation: Other: Shire Clinical Fellowship Program. Sidonio:Kedrion: Research Funding; Novo Nordisk: Membership on an entity's Board of Directors or advisory committees; Takeda-Shire: Membership on an entity's Board of Directors or advisory committees, Research Funding; Bioverativ: Membership on an entity's Board of Directors or advisory committees, Research Funding; Genetech: Membership on an entity's Board of Directors or advisory committees, Research Funding; Octapharma: Membership on an entity's Board of Directors or advisory committees, Research Funding; Grifols: Membership on an entity's Board of Directors or advisory committees, Research Funding; Biomarin: Membership on an entity's Board of Directors or advisory committees; Uniqure: Membership on an entity's Board of Directors or advisory committees.

Hemoglobin Levels Vary in Infants in Relation in Occurrence of Influenza Virus Infections

Iron deficiency anemia (IDA) screening in children, and its treatment is standard in general pediatrics. IDA has been linked to developmental concerns. However, iron supplementation recently has been shown to have risks in resource limited environments as it seems to be associated with increased risks of infections. The role of iron in bacterial survival is well established. However more difficult to explain are reports linking iron supplementation to higher rates of respiratory illness in children. Recent studies have identified hepcidin as an important mediator between inflammation from common infections, and iron repletion status (Prentice et al 2019). In the current study we hypothesized that overall hemoglobin levels in infants may fluctuate depending on birth cohorts, as 1) exposure to respiratory illnesses peaks in winter months, and 2) children aged 6-12 months have the highest incidence of respiratory illness of all age groups (up to 6-8 upper respiratory infections per year). Institutional review board permission (Children Health Care of Atlanta), was obtained and the billing registry, laboratory and patient records from an multiethinic urban clinic (n= 7,616) were reviewed. Influenzas' infection was defined as a patient presenting with an acute illness and a positive rapid influenza test. Hemoglobin was obtained at 9 to 12 months routine visit by capillary draw and point of care measure (HemoCue hb). A total of 1236 patient files were reviewed from 2007 until 2018. Each year was examined separately and then overlapped. Influenza rates were lowest in the summer, rose starting September, and peaked during the winter months. In the first year of life, hemoglobin levels peaked in March, with lowest levels observed in July and December. Infants with a diagnosis of influenza had lower hemoglobin levels and more pronounced fluctuations. At age 7-9 months, medial hemoglobin was 12.1 (95% confidence interval [CI] ±0.026) g/dl and for those infected with influenza 9.8% 95% CI ±0.138). These observations have implications regarding screening, lower limits of normal may vary during the year. supplement. Other viruses also occur during the winter months and could play a role. These results require further validations. Figure Disclosures No relevant conflicts of interest to declare.

Adverse Events during Induction Therapy in Pediatric Acute Lymphoblastic Lymphoma (ALL)

Introduction Advances in pediatric Acute Lymphoblastic Leukemia (ALL) have led to substantial improvement in remission rates. However, there is a lack of data characterizing morbidity during induction. During this period, patients are stratified into "standard risk," receiving a 3 drug induction regimen of steroids, vincristine, and asparaginase, or "high risk" receiving anthracycline as a fourth chemotherapeutic agent. An understanding of adverse events (AEs) may facilitate supportive care and predict and guide hospitalization needs during this phase of treatment. There is potential benefit in comparing AE rates between the two treatment groups. This study aimed to identify rates of clinically significant AEs, hospital readmissions, and intensive care unit (ICU) admissions in pediatric patients with ALL during induction. Methods This retrospective cohort included patients ages 1 through 21 who initiated induction therapy for ALL between January 1, 2013 and May 13, 2018 at Children's Healthcare of Atlanta (CHOA). Algorithms to identify and grade AEs according to Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 definitions were determined a priori. Using these algorithms, manual chart abstraction to identify AEs occurring from diagnosis to day 34 after the start of induction was performed for 19 AEs: pancreatitis, hepatotoxicity, hypoxia, adult respiratory distress syndrome (ARDS), hyperglycemia, thromboembolic event, stroke, infection, sepsis, fever, hypertension, hypotension, seizure, ileus, anaphylaxis, hyponatremia, and constipation. Chart abstraction additionally identified ALL risk classification, admissions to the intensive care unit (ICU), and readmissions. Demographic data were extracted in an automated fashion from the electronic health record. Rates of each AE, readmissions, and ICU admissions were calculated for the induction period. The highest grade of each AE was captured for these analyses and clinically relevant grade 2-5 AEs are reported. Development of each AE, readmission, and ICU admission were stratified by therapy risk group and analyzed by chi square, Fisher's exact, or Kruskal Wallis tests as appropriate. Results AE grading and abstraction are complete and reported on 286 of 512 eligible patients; median age at diagnosis was 5.7 years; 45.1% were female, 47.6% were non-Hispanic white, 20.5% were black, and 23.6% were Hispanic. Among those discharged before the end of induction (283/286), 36.0% were readmitted. Additionally, 12.9% of all patients were admitted to the ICU at some point during induction. Overall, 72.4% of patients had at least one AE &gt; grade 3 during induction. The most common clinically relevant AEs were: ALT elevation (36.3%); hypertension (33.9%); and infection (17.8%) (Table 1). Four of the 286 patients died during induction; 3 deaths were attributed to respiratory failure, and 1 to sepsis. Overall, 36% of patients required readmission with no difference in readmission rates by treatment group. However, patients with high risk treatment were significantly more likely to have an ICU course (18.0% vs 8.2%, p=0.01). A larger proportion of high risk patients experienced sepsis (6.3% vs 3.8%, p=0.05) or hypotension requiring medical intervention (6.6% vs 3.5%, p=0.05). Conclusions This study demonstrates that AEs ≥ grade 3 are common in induction and that more than a third of patients require readmission during induction after initial discharge. While capture of grade 3-5 AEs indicate severe events, several grade 2 AEs are clinically relevant to morbidity and patient management, such as thromboembolism and fever. These results highlight the AEs driving morbidity during induction therapy, and can be used to guide patients regarding expected AEs. Work is ongoing to analyze the full cohort. Further analyses will compare rates of the highest grade AEs, length of hospitalization, readmission, and ICU admission by chemotherapy regimen. Disclosures No relevant conflicts of interest to declare.

Vaccination Adherence in Pediatric Patients Post-Hematopoietic Stem Cell Transplant

Background: Over 13,000 hematopoietic stem cell transplants (HSCT) are performed each year to treat malignant and non-malignant hematologic diseases in the United States. Infections are among the leading causes of mortality after HSCT and vaccination is one of the most important preventive strategies available for infectious diseases. The Infectious Disease Society of America (IDSA) has guidelines regarding re-vaccination, starting as early as 6 months post-HSCT based on the type of HSCT and clinical condition of the patient. There are however limited available data on adherence with these guidelines. As such the immune dysregulation and pathobiology working group report of the NIH HSCT late-effects initiative identified the study of the adherence with consensus guidelines for post-HSCT as a key research recommendation.Objective: To determine the adherence for all recommended vaccines based on the 2013 IDSA guidelines up to 2 years post-HSCT in a retrospective study at a large pediatric HSCT programMethods: A retrospective chart review using electronic medical records (EMR) of all patients who have undergone HSCT or have received post-HSCT care at Children's Healthcare of Atlanta (CHOA) from January 2010 to December 2016. We collected demographic information (age, gender, race, and socioeconomic status) as well as the date of HSCT, date of last clinic visit, and date of vaccine administration. We reviewed vaccinations and dates administered using the Georgia Registry of Immunization Transactions and Services (GRITS), and determined if these were given per the IDSA guidelines. Vaccines were administered according to standardized protocol and recommendation communicated by HSCT physician to the referring oncologist and primary care physicians except for pneumococcal vaccines and meningococcal vaccines in patients with sickle cell disease (SCD) which are administered in the HSCT clinic during follow up in the multidisciplinary “ Ex-Sickle” late-effects clinic. We compared vaccination rates in SCD post-HSCT to vaccination rates for other indications. We used univariate logistic regression to identify risk factors for low compliance. Our final model included age group, gender, and health insurance coverage and clinic type. Values of p <0.05 were considered statistically significant. The analyses were carried out using R version 3.4.2.Results: Of 237 patients surviving at least two years post-HSCT, greater than 94% of patients did not receive one or more vaccines in accordance with the 2013 IDSA guidelines. Table 1 provides the detailed disease specific breakup of vaccine adherence. Given the small sample sizes for vaccine adherence, the remainder of our results focus on whether patients received any of doses of the recommended vaccinations, regardless of time since HSCT. SCD patients had significantly higher odds of receiving pneumococcal polyscaaharide vaccine (PPV23) (OR: 0.12; 95% CI: 0.05, 0.29) and meningococcal vaccine (MCV )(OR: 0.40; 95% CI: 0.16, 0.96) compared to patients with Leukemia/Lymphoma (LL), higher odds of receiving Hep B (OR: 0.40; 95% CI: 0.17, 0.94 ), Pneumococcal conjugate vaccines (PCV 13) (OR: 0.17; 95% CI: 0.05, 0.45), PPV 23 (OR: 0.10; 95% CI: 0.04, 0.26), MCV (OR: 0.11; 95% CI: 0.04, 0.28), and HPV (OR: 0.14; 95% CI: 0.03, 0.53) vaccines relative to patients with Solid tumor/Brain tumor (ST/BT), and significantly higher odds of receiving PCV 13 (OR: 0.25; 95% CI: 0.07, 0.84) and PPV 23 (OR: 0.10; 95% CI: 0.02, 0.34) relative to patients with other hematologic diseases (Hem), and SCD patients had significantly higher odds of receiving PPV 23 (OR: 0.06; 95% CI: 0.01, 0.24) and MCV (OR: 0.08 ; 95% CI: 0.02, 0.29) vaccines relative to the patients with Immune Deficiency (ID).Discussion: These data have uncovered a major potential gap in service delivery in a large pediatric transplant program. SCD patients had higher rates of adherence with vaccination post-HSCT suggests the benefit of follow up in a multidisciplinary late-effects clinic. Further studies are needed to identify the factors impacting the vaccine adherence in order to increase the compliance to improve outcomes among HSCT patients. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Trends in Hospital Utilization for Acute Illness in a Large Population-Based Cohort of Children and Adolescents with Sickle Cell Disease (SCD): 2010–2017

Disease severity and healthcare utilization varies widely among persons with sickle cell disease (SCD). Hydroxyurea (HU) has been demonstrated to reduce rates of pain and acute chest syndrome, the leading causes of inpatient utilization in patients with Hb SS and S β°-thalassemia in clinical trials. We recently reported that HU was clinically effective in reducing rates of pain and acute chest syndrome in patients who initiate treatment. Use of hydroxyurea in SS/S β°-thalassemia has increased markedly since 2010. Thus we sought to detemine trends of hospital utilization for acute illness during an 8 yr in which HU utilization increased markedly (2010-2017). Data from years 2010-201 were obtained from the SCD database and patient records at Children's Healthcare of Atlanta (CHOA). Utilization data were restricted to acute care admissions. Admissions for elective procedures, non-SCD related discharge diagnoses, rare SCD genotypes, and patients who had undergone bone marrow transplant were excluded. Patients were compared based on number of hospitalizations, age, sex, SCD genotype (SS/S β°-thalassemia vs Hb SC/S β+thalassemia), and discharge diagnosis. A total of 3,116 patients had at least one encounter between 2010 and 2017; 2,947 patients met inclusion criteria. From 2010-2017 the total number of active patients per year increased from 1,546 patients to 1,789 patients (+16%), while the total number acute care admissions increased from 1,295 admissions to 1,609 admissions (+24%). There were no significant differences in the proportion of patients with genotypes SS/S β° thalassemia genotypes (67.0% vs 63.9%, p=0.06). Overall patients with SS/S β° thalassemia had higher admission rates compared to SC/S β+ thalassemia patients (0.94 vs 0.57 admissions per patient per year). During the study period. overall admission rates in SCD (acute illness hospitalizations/patient/yr) increased from a low of 0.74 in 2011 to a high of 0.90 in 2017. The proportion of admissions attributed to SS/S β°-thalassemia patients decreased (79.2% in 2010 vs 72.3% in 2017, p<0.0001). However, admission rate in SC/S β+ thalassemia increased (0.53 to 0.69 admissions per patient per year). Overall, over 60% of patients were not admitted in any given year, and the proportion of patients with 0-1 admissions in a given year remained unchanged. However, the proportion of super high hospital utilizers (SHHU), patients with 8 or more admissions in a given year, increased by 185%. In 2001 this group made up 0.6% of all patients and accounted for 7.3% of admissions; in 2017 SHHU had increased 1.8% of patients and 24.3% of admissions. There was no difference in genotype or sex between SHHU and non-SHHU patients. SHHU were older (>90% of patients over age 8 years), and had greater percentages of admissions for pain and acute chest syndrome then non-SHHU. In conclusion, during a period in which HU utilization in SS/S β°-thalassemia increased significantly, hospital utilization for acute illness in SS/S β° thalassemia decreased as expected. However, during the same period there was an unexpected increase in overall hospital utilization for acute illness in SCD. This increase in hospital utilization was the result of 1) a marked increase in SHHU and 2) an overall increased utilization in SC/S β+ thalassemia. DisclosuresDampier:Pfizer: Research Funding.

Hydroxyurea effectiveness in children and adolescents with sickle cell anemia: A large retrospective, population-based cohort.

The clinical efficacy of hydroxyurea in patients with sickle cell anemia (SCA) has been well established. However, data about its clinical effectiveness in practice is limited. We evaluated the clinical effectiveness of hydroxyurea in a large pediatric population using a retrospective cohort, pre-post treatment study design to control for disease severity selection bias. The cohort included children with SCA (SS, Sβ0 thalassemia) who received care at Children's Healthcare of Atlanta (CHOA) and who initiated hydroxyurea in 2009-2011. Children on chronic transfusions, or children with inadequate follow up data and/or children who had taken hydroxyurea in the 3 years prior were excluded. For each patient healthcare utilization, laboratory values, and clinical outcomes for the 2-year period prior to hydroxyurea initiation were compared to those 2 years after initiation. Of 211 children with SCA who initiated hydroxyurea in 2009-2011, 134 met eligibility criteria. After initiation of hydroxyurea, rates of hospitalizations, pain encounters, and emergency department visits were reduced by 47% (<0.0001), 36% (P = 0.0001) and 43% (P < 0.0001), respectively. Average hemoglobin levels increased by 0.7 g/dl (P < 0.0001). Hydroxyurea effectiveness was similar across gender, insurance types and age, although there was a slightly greater reduction in hospitalizations in younger children. Am. J. Hematol. 92:77-81, 2017. © 2016 Wiley Periodicals, Inc.

Are we missing the mark? Fever, respiratory symptoms, chest radiographs, and acute chest syndrome in sickle cell disease

Are we missing the mark? Fever, respiratory symptoms, chest radiographs, and acute chest syndrome in sickle cell disease

Hydroxyurea Effectiveness in Children and Adolescents with Sickle Cell Anemia: A Large Retrospective, Population-Based Cohort Study

Background: The clinical efficacy of hydroxyurea (HU) in patients with sickle cell anemia (SCA) is well established. In clinical trials, HU has been demonstrated to alter the clinical course of SCA; HU decreases rates of pain, dactylitis, acute chest syndrome (ACS), transfusions, hospitalizations and improves hemoglobin levels. Other studies have shown that HU improves quality of life and decreases mortality. While evidence of efficacy has been well demonstrated, published data about clinical effectiveness is limited. Additionally, attempts to compare outcomes of patients taking HU to those not taking HU have been limited by the inherent selection bias for greater disease severity in patients on HU. We sought to examine the clinical effectiveness of HU in the large population of patients with sickle cell disease (SCD) at Children's Healthcare of Atlanta (CHOA). The program provides comprehensive care to >1,700 active patients; because they include about 95% of all children and adolescents with SCD in the greater Atlanta Metropolitan Area, the program provides a population-based sample. Currently 57% of the 922 children ≥ 1 yr of age with SCA (SS and Sβ0 thalassemia) who are not on chronic transfusions are receiving HU.Methods: Using a retrospective cohort, pre-post treatment study design to control for disease severity selection bias we evaluated the clinical effectiveness of HU in patients with SCA who received care at CHOA and who first initiated HU in 2009-2011. Children on chronic transfusions, or children with inadequate follow up data and/or who ever took HU in the 3 years prior were excluded. Clinical guidelines for dosing HU have been standardized locally and recommend initial dosage of 20mg/kg/day, followed by dose escalation every 2 months to maximum tolerated dose. For each patient, healthcare utilization, laboratory values and clinical outcomes for the 2-year period prior to HU initiation were compared to those for two years after initiation. Medians were compared using the Wilcoxon Signed Rank test and means using T test. Rate ratios were computed using unadjusted Poisson regression. Interactions testing whether the effect of HU varied by age, sex, or insurance status, were assessed using multivariable Poisson regression.Results: Of 211 children with SCA who initiated HU in 2009-2011, 134 met eligibility criteria. After initiation of HU, the rate of hospitalization was 0.53 the rate before HU, a 47% reduction (Table). The number of inpatient days, emergency room (ER) visits, pain encounters, episodes of acute chest syndrome (ACS), and transfusions were also significantly reduced (Table). The hemoglobin level (mean +/- SD) pre- and post-HU initiation was 7.3+/-2.7 g /dl and 8.1+/- 2.9 respectively; an increase of 0.8+/-1.1, p<0.0001. Similarly, pre-HU mean corpuscular volume increased from 82.9+/-7.5 fl pre-HU to 95.7+/-10.7 post-HU; an increase of 12.8+/-7.2, p <0.0001. There was a statistically significant interaction between the effect of HU and age of initiation of the drug on hospitalization rates; patients < 7 years of age had a greater reduction in hospitalizations than older children (58% vs 33%, p=0.03). The effect of HU on hospitalizations did not vary by sex or insurance type.Table 1Effect of HU initiation on clinical outcomes using Poisson RegressionRate Ratio (Confidence Limits)P ValueHospitalizations0.53 (0.43 - 0.66)<0.0001Inpatient Days0.50 (0.40 - 0.63)<0.0001ER visits0.57 (0.49 - 0.67)<0.0001Pain Encounters0.64 (0.51 - 0.81)0.0001ACS0.57 (0.39 - 0.83)0.0036Blood Exposure0.43 (0.29 - 0.64)<0.0001Conclusions: HU is clinically effective in children and adolescents with SCA. HU decreased hospitalizations, ER visits, pain encounters, ACS, use of transfusions and improved hemoglobin levels. HU effectiveness was similar across gender, insurance types and age, although there was a slightly greater reduction in hospitalizations in younger patients. These results are important as they parallel results obtained from HU efficacy studies, thus demonstrating that in 'real life' settings, even without the additional monitoring and adherence incentives of clinical trials, HU improves short-term outcomes in pediatric SCA. DisclosuresOff Label Use: Hydroxyurea use as disease modifying therapy in pediatric patients with sickle cell disease.

Improving inpatient cystic fibrosis pulmonary exacerbation care: two success stories

BackgroundPulmonary exacerbations (PEx) in cystic fibrosis (CF) are a frequent cause of hospitalisations and lead to long-term decline in pulmonary function. Successful CF inpatient care requires the coordination of multiple...

Population-Based Analysis Of High Hospital Utilization (HHU) Among Children and Adolescents With Sickle Cell Disease (SCD)

BackgroundSickle cell disease (SCD) is characterized by marked heterogeneity in clinical manifestations, severity and utilization of health care services. This heterogeneity is particularly evident with regards to utilization of inpatient services, which account for a large percentage of total health care costs and significant absence from school and work. Previous studies of utilization, based on analysis of administrative datasets, have been limited by inaccuracies in ICD-coding, inability to differentiate among SCD genotypes, and/or inability to capture individuals never hospitalized. Single center studies have been limited by potential disease severity referral bias, inability to capture service utilization at outside facilities, and relatively small numbers. ObjectivesWe sought to identify children and adolescents with SCD with high hospital utilization (HHU) and to determine the relationship of HHU with age, SCD genotype, and insurance coverage. We also sought to examine trends over a three year period (2010-2012) and explore the relationship of any trends in hospital utilization over time. MethodsThe clinical database of the SCD Program at Children's Healthcare of Atlanta (CHOA) was used to identify all patients with SCD, ages ≥ 5 years, who received comprehensive outpatient, acute care, and inpatient services at three CHOA facilities in metro Atlanta during 2010-2012. SCD genotype was confirmed for each patient by review of hematologic and clinical data, including results of diagnostic hemoglobin electrophoresis. All outpatient clinic, emergency department, and inpatient utilization was captured as well as insurance (payer) status at each encounter. HHU was defined as ≥3 admissions during a calendar year. Data were analyzed for significance by Chi-square. The extent to which the data were population-based was determined from the Georgia Hospital Association database with includes DRG-level data for all inpatient hospitalizations in Georgia. ResultsHospitalizations at CHOA accounted for 96.1% of all SCD admissions for children within the 28-county greater metro Atlanta area. During 2010-2012, 1348 individual children and adolescents ≥ 5yr of age with SCD received comprehensive care at CHOA (1076 in 2010, 1112 in 2011, 1159 in 2012): 928 (68.8%) had SS/S β°thal, 334 (24.8%) SC, 86 (6.4%) S β+thal. Of 1159 active patients in 2012, 441 (38.1%) were 5-9 yr of age, 419 (36.2%) 10-14 yr, 299 (25.8%) ≥ 15 yr; 474 (40.9%) had traditional medicaid (MC), 205 (17.7%) managed-care medicaid (MMC), 374 (32.3%) private payer (PP), and 105 (9.1%) multiple payers (MP). HHU (≥ 3 admissions/yr) decreased from 119 of 1076 (11.1%) in 2010 to 96 of 1112 (8.6%) in 2011 and 95 of 1159 (8.2%) in 2012 (p <.01). The Table shows percentage HHU for each year by age, genotype, and payer.TableAge (years)SCD GenotypePayerYear5-910-14≥15PSS/S βoSCS β+PMCMMCPPMPPTotal20108.19.616.9< .0113.35.84.6< .01154.47.619.8< .0111.120115.57.214.1< .0110.86.64.30.1211.85.85.611.30.018.620123.610.212< .018.68.15.80.6510.74.35.912.4< .018.2 ConclusionsFor children and adolescents with SCD, HHU decreased during the three year period. HHU was significantly associated with older age and with MC and MP insurance coverage during all three years. HHU was significantly associated with the SS/S β°thal genotype in 2010, but not in 2011-12. The extent to which the decline of HHU among patients with SS/S β°thal genotype was related to increasing use of hydroxyurea is being investigated. Disclosures:Off Label Use: Hydroxyurea is not FDA approved for use in children.