Comparing Effectiveness of Nitrous Oxide, Propofol, and Midazolam-Assisted Lumbar Punctures in Childhood Acute Lymphoblastic Leukemia

Background:

Childhood acute lymphoblastic leukemia (ALL) treatment requires numerous lumbar punctures (LPs) with intrathecal (IT) chemotherapy to prevent and treat central nervous system disease. Historically, LPs in this setting are performed using propofol as a sedative-hypnotic agent, which has a strong record of efficacy and safety in this population. Nonetheless, propofol requires a nil per os (NPO) status and supervision by a healthcare provider who is skilled in airway management. At our institution, LPs at times incorporate inhaled nitrous oxide (N₂O) and, less frequently, anxiolytic doses of midazolam, as sedation agents. Both of these sedation methods are well-established in children undergoing procedures, can be performed regardless of NPO status, and do not necessarily require a provider with airway management skills. But unlike propofol, neither result in conscious sedation. To date, there have been no large scale assessments comparing these sedation methods for the purpose of performing LPs in children.

Objective:

To compare the effectiveness of propofol, N₂O, and midazolam sedation for lumbar punctures in childhood ALL patients.

Design/Method:

Retrospective cohort study of all patients aged 0-31 years with ALL treated between 1/1/2013-12/31/2018 at the Children’s Minnesota Cancer and Blood Disorders Center, including all LPs performed in the clinic setting under either propofol, N₂O, or midazolam sedation. Decisions to provide sedation with an agent other than propofol were at the discretion of the provider. Diagnostic LPs and LPs performed in conjunction with bone marrow biopsies were excluded. Comparisons between propofol, N₂O, and midazolam procedures were made using ANOVA, two-sample T-tests, and χ² tests where appropriate.

Results:

Among 307 pediatric patients with ALL, 3142 therapeutic LPs were undertaken: 52.4% (n = 1645) with N₂O, 41.5% (n = 1304) with propofol alone, and 6.1% (n = 193) with midazolam alone. Midazolam was also given in conjunction with 5.7% (n = 94) of N₂O cases. Demographic and efficacy information is presented in Table 1. Younger age (8.1 vs 8.9 vs 11.3 years, p < .001) was associated with an increased discretionary use of N₂0 use when compared to propofol and midazolam; patients ≤ 6 years of age were particularly more likely to undergo N₂O sedation (57.2% vs 47.0% vs 36.2%, p < .001). Hispanic patients were more likely to receive propofol sedation when compared to White and Black counterparts (49.8% vs 44.2% vs 43.5%, p < .001). Patients with BMI ≥ 90^th percentile were more likely to receive propofol sedation (54.4% vs 32.4% vs 13.4%, p < .001). There was no difference in sedation based on gender (p = .18). Traumatic LPs (RBC ≥ 10 cells/µL) were less frequent under N₂O (26.1% vs 31.2% vs 42.5%, p < .001). There were no differences in successful IT chemotherapy delivery (99.8% vs 99.8% vs 99.0%, p = .08) by sedation type, which amounted to 2 aborted propofol (0.2%), 4 aborted N₂O LPs (0.2%), and 2 aborted midazolam LPs (1.0%).

Conclusion:

Approximately half of children with ALL treated in our program received N₂O sedation when undergoing LPs. While there has been no formal process for sedation selection, discretionary selection tactics appear to have been utilized by providers, resulting in equivalent success rates in IT chemotherapy delivery across sedation groups. Additionally, we found that there may be select group of younger patients with lower BMI percentiles for whom N₂O can effectively replace propofol during LP procedures. Future analyses will focus on safety events by sedation type, assessing encounter costs in lieu of sedation used, and developing a multivariable model for predicting optimal N₂O-assisted sedation candidates.