National patterns and variation in regional anesthesia use for shoulder, hip, knee, and ankle arthroplasty: a National Inpatient Sample analysis

Background Although regional anesthesia (RA) is associated with improved pain control, reduced opioid use, and accelerated recovery after arthroplasty, racial and geographic differences have been identified in hip and knee arthroplasty, with limited data for shoulder and ankle arthroplasty. We examined differences in RA utilization for shoulder, hip, knee, and ankle arthroplasty, based on patient-level and hospital-level characteristics.Methods Using the National Inpatient Sample, we conducted a cross-sectional study of adults who underwent shoulder, hip, knee, and ankle arthroplasties from 2016 to 2020. RA was defined using procedure codes for neuraxial anesthesia and peripheral nerve blocks. Weighted descriptive statistics summarized RA utilization and compared patient-level and hospital-level characteristics using standardized mean differences (SMDs), with ≥0.10 indicating meaningful differences.Results We identified 1132 011 unweighted admissions, representing approximately 5.7 million weighted arthroplasties. RA was used in 7.0% of cases. RA utilization varied by procedure type, with the highest rates in shoulder (11.4%) and ankle arthroplasty (11.4%), followed by knee (9.1%) and hip arthroplasty (2.4%). Variation in RA utilization was observed in ankle arthroplasty based on race (SMD=0.11) and primary payer (SMD=0.12). RA use also varied by hospital size for shoulder (SMD=0.13), hip (SMD=0.29), and knee arthroplasty (SMD=0.22), by hospital location and teaching status for shoulder (SMD=0.17), hip (SMD=0.18), knee (SMD=0.19), and ankle arthroplasty (SMD=0.12), and by geographic region for shoulder (SMD=0.26), hip (SMD=0.45), knee (SMD=0.24), and ankle arthroplasty (SMD=0.17).Conclusions There is notable variation in RA use for inpatient arthroplasty across hospitals and geographic regions, with limited variation based on patient characteristics. These findings highlight the presence of system-level variation in RA use and underscore the need for further research to better understand the drivers of these patterns.