Routine circumferential patellar denervation during total knee arthroplasty did not consistently eliminate neural structures at the lateral patellar border: A prospective randomised controlled trial with histological outcomes

Abstract Purpose Circumferential patellar denervation is widely performed during total knee arthroplasty under the assumption that electrocautery ablates nociceptive neural structures at the lateral patellar border. However, the microscopic effect of this procedure on neural substrates remains poorly defined. This study aimed to determine (1) whether histological characteristics differ between non‐denervated and denervated lateral patellar borders, and (2) which neural elements are affected by routine circumferential denervation. Methods In this prospective randomised controlled trial with histological outcomes, 19 patients undergoing primary total knee arthroplasty were allocated to either a Facetectomy‐First Group ( n = 10) or a denervation‐first group ( n = 9). In the Facetectomy‐First Group, lateral patellar facetectomy was performed prior to electrocautery; in the denervation‐first group, circumferential monopolar electrocautery (~ 4 mm depth) was performed before facetectomy. Resected specimens were processed using standard histological techniques (Hematoxylin and Eosin, Masson's trichrome). Tissue compartments and neural structures, including nerve fibres and free nerve endings, were analysed semi‐quantitatively by a blinded highly experienced musculoskeletal pathologist. Results All specimens demonstrated a consistent layered architecture with loose connective tissue present in 100% of cases. Nerve fibres were identified in connective tissue in all specimens and in most adipose layers, irrespective of surgical sequence. nerve fibres were detected in multiple tissue compartments in both groups. The mean number of nerve fibres groups per specimen was higher in the Facetectomy‐First Group (1.6 ± 0.7) than in the Denervation‐First Group (1.1 ± 0.3), without reaching statistical significance ( p = 0.07). Although focal thermal alterations were observed in denervated specimens, morphologically identifiable nerve fibres persisted. Conclusions Routine circumferential patellar denervation did not consistently eliminate neural structures at the lateral patellar border, and no significant histological differences were observed between the Facetectomy‐First and Denervation‐First Groups. These findings suggest that standard electrocautery‐based denervation may not reliably disrupt morphologically identifiable neural structures in this region. The functional implications of these findings remain uncertain and were not assessed in this study. Further investigations integrating advanced neural characterisation and clinical correlation are required to better define the biological and clinical impact of this procedure. Level of Evidence Level III.