RESUMEN
Stress shielding (SS) around press-fit radial head arthroplasty (RHA) was recently reported as a cause of a new type of proximal radial neck resorption (PRNR). Very few studies have analyzed this phenomenon. No comprehensive classification is currently available. We thus decided to clinically and radiographically analyze 97 patients who underwent a press-fit RHA and who were followed up for a mean period of 72 months (range: 2-14 years). PRNR in the four quadrants of the radial neck was assessed. We designed a novel SS classification based on (1) the degree of resorption of the length of the radial neck and (2) the number of neck quadrants involved on the axial plane. The mean PRNR (mPRNR) was calculated as the mean resorption in the four quadrants. mPRNR was classified as mild (<3 mm), moderate (3 to 6 mm), and severe (>6 mm). Eighty-four percent of the patients presented PRNR. mPRNR was mild in 33% of the patients, moderate in 54%, and severe in 13%. In total, 6% of the patients with mild mPRNR displayed resorption in one quadrant, 18% displayed resorption in two quadrants, 4% displayed resorption in three quadrants, and 72% displayed resorption in four quadrants. All four quadrants were always involved in moderate or severe mPRNR, with no significant differences being detected between quadrants (p = 0.568). mPRNR has no apparent effect on the clinical results, complications, or RHA survival in the medium term. However, longer-term studies are needed to determine the effects of varying degrees of PRNR on implant failure.
RESUMEN
Aims: The aim of this study was to analyze how proximal radial neck resorption (PRNR) starts and progresses radiologically in two types of press-fit radial head arthroplasties (RHAs), and to investigate its clinical relevance. Methods: A total of 97 patients with RHA were analyzed: 56 received a bipolar RHA (Group 1) while 41 received an anatomical implant (Group 2). Radiographs were performed postoperatively and after three, six, nine, and 12 weeks, six, nine, 12, 18, and 24 months, and annually thereafter. PRNR was measured in all radiographs in the four radial neck quadrants. The Mayo Elbow Performance Score (MEPS), the abbreviated version of the Disabilities of the Arm, Shoulder, and Hand questionnaire (QuickDASH), and the patient-assessed American Shoulder and Elbow Surgeons score - Elbow (pASES-E) were used for the clinical assessment. Radiological signs of implant loosening were investigated. Results: The mean follow-up was six years (2 to 14). PRNR started after a mean of 7.5 weeks (SD 2.1) and progressed significantly during the first two years, by the end of which the bone resorption stabilized. PRNR was detected in 81% (n = 45) of patients in Group 1 and 88% (n = 36) in Group 2. The final mean PRNR was 3.0 mm (SD 2.3) in Group 1 and 3.7 mm (SD 2.5) in Group 2. The mean MEPS, QuickDASH, and pASES-E were 95.9 (SD 11.5), 4.4 (SD 9.2), and 94.8 (SD 10.9) in Group 1 and 92.2 (SD 16.2), 9.9 (SD 21.5), and 90.8 (SD 15) in Group 2, respectively. No significant differences were observed between groups in the clinical and radiological outcomes. No correlations were found between PRNR and the clinical results. Conclusion: PRNR after press-fit RHA is a common radiological finding that develops in the first 24 months before stabilizing definitively. PRNR does not affect the clinical results or implant survival in the mid term.
