Emergency Department Management of Children with Sickle Cell Disease.

Abstract

Vaso-occlusive crises (VOC) are a common cause of emergency department (ED) visits for children with sickle cell disease (SCD). To better understand our patient population and compare with reports from other centers, we sought to describe the presentation, management, and disposition of children with VOC at our center's ED. We also aimed to identify predictors of hospital admission. We retrospectively reviewed hospital records of all patients with SCD, age 8–19 years, who presented to our urban pediatric ED in 2003 with a chief complaint of pain. We identified all subjects diagnosed with VOC and not another cause of pain. We obtained the following data for each: SCD genotype; duration of VOC and treatment prior to ED presentation; the nature of analgesia and use of intravenous fluids (IVF) in the ED; hemoglobin (Hgb) concentration; and disposition (admitted, discharged, discharged with subsequent ED visit for same crisis). Categorical variables were evaluated by the χ2 and Fisher exact tests and continuous variables by the t-test. Odds ratios (OR) and 95% confidence intervals (CI) were calculated where appropriate. In 2003, there were 320 ED visits for patients with SCD and pain. Among these, there were 279 diagnoses of VOC in 105 individual patients: 45 had one visit, 25 two visits, and 16 ≥5 visits. Mean number of visits per patient was 2.7; 23 (22%) patients accounted for 145 (55%) visits. Homozygous sickle cell anemia (Hgb SS) was present in 73/105 patients, accounting for 222 (79%) ED visits. Overall admission rate was 179/279 (64%), with 167/179 admitted on their first visit and 12 on their return visit. Subjects with Hgb SS accounted for 147/179 (82%) admissions. Among those discharged who later returned to the ED during the same VOC, the admission rate was 86% (5 returned in 24 hours, 5 in 48 hours, 3 in 72 hours, and one 4 days later). Pre-ED home opioid use was reported in 75% of visits and was associated with increased likelihood of discharge (OR 1.63, CI 0.94–2.84, p=0.082). Duration of VOC before presentation did not significantly affect admission rate and averaged 53.2 hours for admitted patients and 49.7 hours for those discharged (p=0.689). Patients who received IVF in the ED (219/279; 79%) were less likely to be admitted (31% vs 56%, p<0.001). Hgb concentration was increased in 61%, decreased in 36% and unchanged in 3% of patients from steady-state values and was not associated with admission. After receiving 2 doses of morphine, 31 patients were discharged from the ED, while only 5 patients were discharged after receiving 3 or 4 doses of morphine. A departmental VOC protocol was followed for 25.4% of patients, with no impact on admission rate (p=0.290). In this retrospective analysis of a large series of pediatric sickle cell ED visits, patients presented later in their VOC, and admission rate was higher than previously reported. A small proportion of older patients with SCD accounted for most ED visits and hospitalizations. Hgb variation from steady-state was neither clinically significant nor predictive of admission or discharge. In contrast to previous studies, receiving IVF in the ED was associated with a greater likelihood of discharge. Home opioids prior to ED presentation seemed to decrease hospital admission. Adherence to our center's VOC protocol did not appear to influence disposition from the ED.