A Cost Comparison Between Propofol and Midazolam Use for Targeting Moderate Sedation During Curvilinear Endobronchial Ultrasound Bronchoscopy

Abstract

SESSION TITLE: Rapid Session: Interventional Procedures SESSION TYPE: Original Investigation Slide PRESENTED ON: Sunday, October 29, 2017 at 03:15 PM - 04:15 PM PURPOSE: Propofol is an effective option for providing moderate sedation during bronchoscopy. However, a cost assessment of its use with endobronchial ultrasound (EBUS)-guided bronchoscopy has not been performed. Our objective was to analyze differences in cost of performing curvilinear-array EBUS-guided sampling under moderate sedation, based on type of sedative delivery. METHODS: This was a retrospective analysis of prospectively collected data. We assessed three sedative delivery methods for all curvilinear EBUS bronchoscopies performed at our institution under moderate sedation over a two year period. Procedures involving radial probe (peripheral) EBUS were excluded. Procedural sedation was administered by a qualified nurse, under the direction of the bronchoscopist. The sedation regimens were propofol by continuous infusion only (PC, n=54), propofol by bolus induction followed by maintenance continuous infusion (PBC, n=44), and midazolam by bolus dosing (M, n=31). Supplemental fentanyl was given per the same protocol for all subjects. Procedure-related and sedation characteristics were assessed. A cost comparison was performed based on differences in labor, drug, and resource utilization between the groups. RESULTS: Patient and procedural characteristics were similar in all three groups, including the breadth of sampling achieved, the specific diagnostic yield, and complications observed. However, total sedation time (sum of induction, maintenance, and recovery) was significantly longer in the midazolam group than in either propofol group (M=80.4 minutes ±26.7 vs PC=67.5 ±22.6 and PBC=66.7 ±19.7; p=0.01). This was primarily driven by the recovery phase. The midazolam group exhibited more agitation-related procedure interruptions, and required more fentanyl (M=125.9 mcg ±54.4 vs PC=90.3 ±35.0 and PBC=99.7 ±39.0; p=0.002). Calculating relevant labor and drug utilization expenditures at our institution revealed that midazolam use resulted in an additional per-procedure cost of $7.42 compared to propofol given by continuous infusion (PC), and $8.02 more than propofol delivery using bolus induction followed by continuous maintenance infusion (PBC). Assessment of the most conservative estimate of mean sedation time differences between the midazolam and propofol groups still resulted in essentially equivalent costs. CONCLUSIONS: Using propofol rather than midazolam for targeting moderate sedation during EBUS bronchoscopy may lead to labor cost savings resulting from shorter sedation time. Our analysis did not demonstrate any differences in procedural characteristics, specific diagnostic yield, or adverse events between the sedative delivery methods. Therefore, we assume that any ‘downstream’ costs of required interventions, follow-up testing, and disease management would also be similar. CLINICAL IMPLICATIONS: Bronchoscopist-directed moderate sedation using propofol infusion for EBUS bronchoscopy may be a cost-efficient alternative to midazolam. DISCLOSURE: The following authors have nothing to disclose: Ara Chrissian, Destry Washburn, Avi Cohen No Product/Research Disclosure Information