Higher Dose of Opioids in the Emergency Department and Earlier Initiation of Oral Opioids after Hospitalization Are Associated with Shorter Length of Stay in Children with Sickle Cell Disease Treated for Acute Pain

There is currently no standardized approach to emergency department (ED) and inpatient treatment of an acute pain event for patients with sickle cell disease (SCD).  In addition, how the initial treatment of a pain event influences the outcome of that pain event is not well described. In surgical patients, aggressive pre-operative and peri-operative pain management is associated with improved pain outcomes post-operatively. Furthermore, in post-operative patients, using oral opioids alone or in combination with intravenous (IV) opioids is shown to provide better pain control, decreased IV opioid consumption, less side effects and higher patient satisfaction. In SCD, it is not known whether more aggressive opioid treatment at presentation to ED care or earlier initiation of oral opioids after hospitalization is associated with improved outcomes. Thus, we sought to determine the impact of the following treatment-related aspects of acute pain events on length of hospital stay: 1) total opioid dose at presentation, 2) time to initiation of first IV opioid at presentation, and 3) time to initiation of first oral opioid after hospitalization. We hypothesized that increased total initial dose of opioids, more rapid initiation of IV opioids in the ED, and earlier initiation of oral opioids after hospitalization would be associated with shorter length of hospital stay.

We conducted a secondary analysis of data from the randomized controlled Magnesium for children in Crisis (MAGiC) trial. The MAGiC trial was a double-blind randomized controlled trial of IV magnesium versus saline placebo added to standard treatment for pediatric pain events. All patients had HbSS or HbSβ⁰ thalassemia to be eligible. The primary outcome, length of stay, was a composite endpoint of actual discharge or 12 hours after administration of last IV opioid, whichever came first.  Patients were defined as having severe disease if a history of acute chest syndrome and/or 3 or more pain hospitalizations in the prior 3 years existed. A priori, all pain events were considered severe as the need for hospitalization indicated failure of outpatient therapy.  Our primary variables of interest were: 1) Total initial opioid dose: total mg/kg/hr of morphine equivalents administered between first opioid in ED through start of study drug, 2) Time to first IV opioid: total hours between ED arrival and first IV opioid, 3) Time to first oral opioid: total hours between ED arrival and start of first oral opioid after hospitalization. Spearman correlations were used to determine the association between the above variables and length of stay and between length of stay and rehospitalization within 7 days to determine the consequence of earlier discharge.

A total of 204 patients were enrolled at 8 sites. Mean (SD) age was 13.6 (4.7) years, 51.5% were female, 88.2% had severe disease and 60.3% were taking hydroxyurea.  The primary MAGiC trial analysis found IV magnesium had no impact on length of stay. The mean total initial dose of opioid was 0.043 (0.029) mg/kg/hr that was administered over a median of 8.5 (IQR 6.4–13.1) hours. Higher total initial opioid dose was correlated with shorter length of stay (r=0.34, p<0.01). The median time to first IV opioid was 1.02 (IQR 0.65–1.33) hours. There was no association between the time to first IV opioid and length of stay (r=0.006, p=0.93).  The median time to time to first oral opioid was 28.3 (IQR 12.2-61.1) hours. Earlier initiation of oral opioids was strongly associated with shorter length of stay (r=0.61, p<0.01). Neither shorter length of stay (r=-0.08, p=0.28) nor earlier initiation of oral opioids (r=-0.02, p=0.82) was associated with rehospitalization within 7 days.

Earlier initiation of oral opioids was strongly associated with shorter length of stay without associated increased rehospitalizations. Higher doses of opioids during the initial management of a pain event also contributed to shorter length of stay, however, more rapid initiation of IV opioids did not alter length of stay. These data suggest that higher initial doses of opioids in the ED and earlier introduction of oral opioids after hospitalization may improve outcomes of pain events. Prospective trials designed to evaluate the impact of initial ED care on outcomes of hospitalizations for pain events and trials aimed at standardizing inpatient treatment for pain events are needed and could ultimately lead to improved outcomes for pain in children with SCD.