Abstract
Background
The National Heart, Lung, and Blood Institute Consensus Expert Panel recommends emergency department evaluation for all patients with sickle cell disease who develop a fever. However, it is unclear what recommended components are present in institutional care plans for their patients with sickle cell disease and what variation exists in the evaluation of patients who present with sickle cell disease and fever across institutions. There are few studies evaluating this practice variation and little evidence concerning the appropriate work-up of these children. Our objective was to describe areas where significant practice variation exists in the care of children with sickle cell disease presenting with fever to established sickle cell centers.
Methods
We undertook a retrospective cross-sectional study of the actual care received at three comprehensive sickle cell centers at pediatric hospitals to understand the diagnostic evaluation, treatment and disposition of children ages 3 months to 21 years presenting to the emergency department with sickle cell disease and fever. Chart reviews were performed on all visits of children presenting with a diagnosis of sickle cell disease to the emergency department of one of the three participating sites between January 1, 2008 and December 31, 2012. All charts were reviewed, and any chart with a documented fever ≥ 38.5°C, either at home or in the emergency department, was included for analysis. Data abstraction included laboratory and radiographic evaluation as well as antibiotic use and disposition. All children pretreated with an antibiotic within the past 24 hours were excluded from the analysis as this could alter the diagnostic evaluation and disposition of these patients. Descriptive statistics were used to determine the percent of children who received a chest radiograph, blood culture, complete blood count, urinalysis, electrolytes, treatment with an antibiotic, and disposition of hospital admission. Due to the large sample sizes, relatively small differences in proportions were determined to be statistically significant; however, differences of less than 10 percent were not considered to necessarily be indicative of clinically meaningful differences in evaluation or treatment, and therefore differences smaller than 10% were reported as similar.
Results
Analysis to date reveals complete evaluation of 1251 visits [673 at the Children's Hospital of Wisconsin (CHW), 368 at Children's National Medical Center (CNMC), and 210 at Children's Hospital of Philadelphia (CHOP)]. The median age of the children at these visits was 3.4 years (interquartile range of 1.4 - 7.7). Analysis of diagnostic testing revealed approximately 98 percent of patients received a complete blood count and a blood culture, with no difference between sites. Ninety-three percent of patients were treated with an antibiotic, which also showed no meaningful difference across sites. Analysis of disposition revealed significant differences between sites, with 49%, 47%, and 100% of patients admitted to the inpatient units at CHW, CNMC and CHOP, respectively. Likewise, significant differences were seen in obtaining chest radiographs: 81%, 92%, and 29% at CHW, CNMC and CHOP, respectively. The percent of patients who received a urinalysis ranged from a high of 39% at CNMC to a low of 18% at CHOP. Electrolytes were obtained from 3%, 48% and 1% of patients at CHW, CNMC and CHOP, respectively.
Conclusion
Consistent with NHLBI guidelines, essentially all children with sickle cell disease presenting to the emergency department with fever receive a complete blood count, blood culture and antibiotics. These equal proportions suggest similar treatment guidelines across sites. There is significant variation between sites in the proportion of children who receive a chest x-ray, urinalysis, electrolytes and perhaps most importantly, admission to the hospital. These examples of practice variation may represent potential areas for quality improvement efforts to better define best care practices for children with sickle cell disease presenting to the emergency department for fever.
Disclosures:
No relevant conflicts of interest to declare.
HemoCue versus Complete Blood Count for Hemoglobin Measurement in Adults with Vaso-occlusive Crisis due to Sickle Cell Disease