Impact of a Pediatric Sickle Cell Outpatient Pain Clinic on Health Outcomes

Over a one-year period at our institution, we identified over 350 emergency department (ED) visits for acute sickle cell pain, with 100 of these acute pain events resulting in hospital admission. While the NIH SCD Management guidelines recommend rapid initiation of analgesia, achieving this metric is challenging in a busy ED setting. Our personal challenge in meeting this rapid analgesic recommendation in our ED prompted our decision to develop an outpatient pain clinic aimed to reduce the time to initial opioid administration and decrease the hospital admission rate for acute pain events. In addition to opioids, relaxation training delivered via virtual reality (VR) is a distraction technique shown to alleviate the pain and anxiety associated with acute and chronic pain. We hypothesized that children and adolescents with SCD presenting with acute pain to our outpatient pain clinic would receive quicker pain management and have a lower rate of admission. In addition, we tested the hypothesis that virtual reality delivered relaxation training would be acceptable in the acute pain setting.

Methods. We compared SCD pain outcomes for patients evaluated in the outpatient pain clinic versus the ED at Children's of Alabama. We abstracted the exact times for triage, time the first dose of NSAID and opioid were ordered and time the first NSAID and opioids were administered from the electronic medical records to determine the time to first opioid from triage for all outpatient pain clinic visits and the most recent ED visit. The dose of NSAID (mg) and opioid (mg/kg) was recorded. The administration and timing of repeated opioid dosing, total time receiving care, and hospitalization were recorded. VR was offered to the patients that presented to the outpatient pain clinic but was not offered to any ED patient. We analyzed differences between all encounters in the pain clinic and ED using t-test for continuous variables and Fisher's exact for categorical outcomes. We then performed a matched-pair analysis to account for repeat patients using only the first pain clinic visit and most recent ED visit using a Wilcoxon Signed Rank test to account for possible non-normal distribution in this small study.

Results. We identified 21 patients presenting to the outpatient pain clinic accounting for 27 pain clinic visits. Fifteen patients have a clinical diagnosis of HbSS, 3 with HbSB0 thalassemia, and 6 with HbSC. Only two (7%) of the 27 pain clinic visits resulted in an admission for further pain management as compared to 6 (29%) of the 21 ED visits for pain (p=0.05). Among all visits, patients seen in the pain clinic had a significantly shorter time to first dose of opioid order than patients seen in the ED (33 vs 50 minutes, p=0.05). The time from triage to discharge disposition was non-significantly shorter in the pain clinic (182 vs 232 minutes, p=0.09). In the pain clinic, oral opioids were the initial treatment in 7 of 27 encounters as compared to 1 of the 21 ED visits (p=0.06). VR was administered to 19 of the 27 outpatient pain clinic encounters. The use of VR in the outpatient clinic did not impact time to first opioid order (34 minutes with VR vs 28 minutes without VR, p=0.5). Additionally, VR did not change time to discharge disposition (180 with VR vs 187 minutes without VR, p=0.85). Using a matched pair analysis to compare the first pain clinic encounter to the most recent ED visit (within 6 months of the pain clinic visit) for 13 patients, we identified a significant reduction in total IV morphine milligram equivalents received (5.3 vs 7.8mg, p=0.007) and a non-significant shorter time to first opioid (34 vs 50 minutes, p=0.08) and time to final disposition (177 vs 299 minutes, p=0.18).

Discussion: An outpatient sickle cell pain clinic was associated with a shorter time to first opioid order, lower total IV morphine milligram equivalents administered, and lower admission rate for pain events. Patients were accepting of VR relaxation training techniques and this intervention did not prolong time to first opioid order.