Abstract
This study compares two pain intensity metrics, calculated percentage pain reduction (CPPR) and patient-reported percentage pain reduction (PRPPR), in patients undergoing medial branch nerve blocks (MBB) to determine their comparability and agreement.
A retrospective review of prospectively enrolled consecutive MBB patients with axial spine pain from multiple centers. Data was collected via self-report pain diaries and analyzed using linear regression models, concordance correlation coefficients (CCC), and Bland-Altman plots.
One hundred and fifty pain diaries yielded 1,182/1,350 (88%) valid comparisons between CPPR and PRPPR. The CCC was lowest immediately and at 0.5 hours post-injection (0.44 and 0.47) and ranged from 0.58 to 0.62 at later times. Precision, measuring correlation, was poor while accuracy, measuring the degree of deviation from perfect agreement, was greater than0.90. Bland-Altman plots showed PRPPR indicated greater pain relief than CPPR at all times, with bias decreasing as pain relief increased. Agreement was higher at 0% and 100% improvement compared to 50% on Bland-Altman plots.
The CCC ranged from 0.44 to 0.62 between PRPPR and CPPR in patients with chronic axial pain undergoing diagnostic MBB. PRPPR scores showed greater pain relief than CPPR scores, especially over time. Agreement was higher at extremes of pain relief, while partial average pain relief demonstrated less agreement. Clinicians should therefore not assume equivalence between CPPR and PRPPR. More importantly, there is significant variability in how a patient is categorized for MBB success depending on the measure employed.
Accepted manuscripts
