Patients with sickle cell disease (SS/Sβ0) often utilize the emergency department (ED) for treatment of painful vaso-occlusive crisis and other sequalae of their disease. However there is significant variation in use, with a minority of patients making up the majority of visits. We studied whether objective steady state laboratory parameters might be associated with frequent ED use and whether hydroxyurea use modified this relationship.
Methods: We identified all patients with sickle cell disesase seen at our medical center in 2012. Patients were identified as having sickle cell disease if hemoglobin (Hb) electrophoresis demonstrated sickle hemoglobin (HbS), fetal hemoglobin (HbF) and HbA2 but no Hemoglobin A, C, or other detectable hemoglobinopathies. ED, clinic, and inpatient admissions over the entire year were calculated and ED use was categorized as either 0-1, 2-5, or >6 visits a year. Steady state laboratory tests were defined as those not within a day of an ED visit or a week of a hospital admission. All 2012 steady state parameters retrieved for a given laboratory test for a given patient were averaged. “Active” laboratory tests, defined as those within one day of an ED visit, were averaged separately. HbF and weight were not separated by activity. Data were analyzed for normality; parametric values were assessed by mean and SD, non-parametric values were assessed as medians and interquartile ranges. Parametric and non-parametric bivariate tests of association were used as appropriate.
Results: 432 adult sickle cell patients were identified, ages 18-87yrs; 54% were female. 181 patients had 0-1 ED visits within the year, 143 had 2-5 visits in the year and 96 had >6 visits for a total of 2259 visits. Patients who had >6 visits accounted for 1750 (77%) of the total visits for the year. When steady state labs were examined, high WBC and platelet counts were most strongly associated with frequent admissions. Steady state WBC of >12.0 x10^9/L were significantly more likely to have >6 visits/year (OR 2.6; 95% CI: 1.6-4.2, p=.0004). Platelet counts of >420 x10^9/L were also strongly associated with >6 ED visits (OR 2.6, 95% CI:1.5-4.4, p=.0007). LDH and AST were also shown to correlate significantly with ED (p=0.02 and 0.005 respectively) use while Hb and albumin were negatively associated (p<0.001 and 0.02 respectively). Hydroxyurea scripts were associated with increased ED visits (p<0.001); 38.1% of the population had been given a script for hydroxyurea within the year; patients with frequent ED use were 2.8x more likely to have been given a script for hydroxyurea within the year. These patients had higher MCV and Hb levels and lower, but still elevated, WBC and platelet counts per ED visit cohort stratification, a suggestion that hydroxyurea was being taken, but despite this, had similar significant associations with ED visits as the non-hydroxyurea group. MCV, MCHC, absolute reticulocyte count, weight, and %HbF did not demonstrate an association with ED visit frequency. When “active” parameters were examined, leukocytosis was present in all categories and WBC count could no longer predict a predisposed cohort; Hb level, LDH and AST were also no longer significant. Albumin and platelet count remained associated with ED visits. When older patients >40 yrs (n=124) were compared with <40yrs (n=308 ), high WBC and platelet counts remained significantly associated with high (>6) ED visits in both cohorts.
Conclusions: Our data suggest that baseline WBC and platelet count are strongly associated with frequent ED utilization and may be better predictors than Hb, HbF, and other red cell or hemolytic parameters. Hydroxyurea use was not associated with fewer ED visits but those patients with frequent ED use still have relatively higher WBC and platelet counts than those with less frequent ED use, suggesting that hydroxyurea under-dosing may be an important issue. To what extent objective steady state parameters can identify a population that needs more aggressive baseline care and to what extent maximization of therapy can decrease frequent ED visits needs further study.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.