Clinical relevance of postoperative radial neck osteolysis after modular press-fit pyrocarbon radial head arthroplasty: a retrospective radiographic-functional correlation study.

Abstract

Radial neck osteolysis is frequently reported after press-fit pyrocarbon radial head arthroplasty (RHA) and has been proposed as a marker of impaired outcome beyond a 3 mm threshold. Whether this association holds at short-to-midterm follow-up with the MoPyC implant is debated. We retrospectively analysed 16 consecutive adults who underwent modular press-fit pyrocarbon RHA (MoPyC, FX Solutions) for unreconstructable radial head fractures between 2010 and 2023. Clinical evaluation comprised the Mayo Elbow Performance Score (MEPS), QuickDASH, visual analogue scale (VAS) for pain and goniometric range of motion. Anteroposterior and lateral radiographs and a systematic final follow-up CT scan were assessed by a single senior orthopaedic surgeon for radial neck osteolysis depth (mm), periprosthetic radiolucency, stress shielding, Delta sign, heterotopic ossification (Brooker grade) and humero-ulnar osteoarthritis. Associations between osteolysis depth and functional scores were analysed using Spearman rank correlation with exact permutation tests (10 000 permutations) and Fisher z 95% confidence intervals. All analyses were exploratory. Median follow-up was 36 months (range 12-110). Mean MEPS was 80.6 ± 11.7, QuickDASH 20.4 ± 7.5 and VAS 2.2 ± 2.3. Mean flexion was 140 ± 9°, extension deficit 16 ± 11°, pronation 74.5 ± 7.2° and supination 70.8 ± 16.9°. Any radial neck osteolysis (> 0 mm) was found in 12/16 patients (75.0%), reached 2 mm in 9 (56.2%) and reached the 3 mm threshold in only 1 patient (6.2%); mean depth was 1.68 ± 1.20 mm (range 0-3.16). Osteolysis depth was not correlated with MEPS (ρ =  + 0.31, 95% CI - 0.22 to + 0.70, p = 0.24), VAS pain (ρ =  - 0.17, 95% CI - 0.61 to + 0.36, p = 0.54) or any range-of-motion variable; a non-significant trend toward lower QuickDASH with deeper osteolysis was observed (ρ =  - 0.48, p = 0.07), opposite to the working hypothesis. Secondary radiographic findings (Brooker ≥ II 43.8%, humero-ulnar osteoarthritis 43.8%, stress shielding 31.2%, Delta sign 25.0%) were not associated with MEPS (all p > 0.15). One patient (6.2%) underwent ulnar nerve neurolysis; no implant was revised. In this small exploratory cohort of modular press-fit pyrocarbon RHA, radial neck osteolysis ≥ 3 mm was rare and no association was found between osteolysis depth and functional outcomes. These findings do not support the use of an isolated 3 mm osteolysis threshold as a clinical warning sign in this implant at short-to-midterm follow-up. Retrospective case series.