Intraarticular Botulinum Toxin A for Refractory Painful Total Knee Arthroplasty: A Randomized Controlled Trial

↵* Data from 49 patients with 49 TKA, i.e., only one TKA per patient.

↵** Comorbidity assessed using self-reported validated comorbidity index⁵¹.

↵† Many patients were taking multiple medications.

↵†† Included capsaicin and medications used for treatment of neuropathic pain, including gabapentin, sertraline, amitryptiline, trazodone, fluoxetine, venlafaxine, topiramate. IA: intraarticular; BoNT/A: botulinum toxin A; VAS: visual analog scale; WOMAC: Western Ontario McMaster University Arthritis Index.

↵* The difference between botulinum toxin and placebo groups was statistically significant at 2 months (p = 0.025). The difference between botulinum toxin and placebo groups was also statistically significant when all 3 timepoints (2, 3 and 4 months) were included (p = 0.019), using generalized estimating equations analyses.

↵* Comparison of IA-BoNT/A and placebo for all efficacy timepoints up to 4 months using GEE, adjusting for respective baseline scores.

↵† Sensitivity analyses that included bilateral total knee arthroplasty (TKA) did not change the significance for any variable listed above (data not shown), except SF-36 pain, for which p value was 0.14 when all 60 TKA were included.

↵†† Only the SF-36 bodily pain subscale score was significantly different between groups; no other SF-36 subscale scores were significantly different between groups at 2 months. WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index; SF-36: Short-form 36; NA: not applicable since these variables were not assessed at these telephone followup visits.