Radial head replacement for acute complex elbow instability: a long-term comparative cohort study of 2 implant designs

BACKGROUNDRadial head fractures not amenable to reconstruction should be treated by radial head replacement (RHR) when there is associated elbow or forearm instability. There are multiple RHR designs with different philosophies, but 2 of the most commonly used implants include the anatomic press-fit radial head system and the loose-fit metallic spacer. There is little information available specifically comparing the long-term clinical and radiographic outcomes of these 2 systems. The objective of this study was to compare the long-term clinical and radiologic outcomes of 2 RHR designs in the context of complex acute elbow instability.MATERIALS AND METHODSNinety-five patients with an average age of 54 years (range, 21-87 years) underwent an acute RHR (46 press-fit Acumed anatomic and 49 loose-fit Evolve metallic spacer) and were prospectively followed for an average of 61 months (range, 24-157 months). There were 34 terrible triads; 36 isolated RH fractures with medial, lateral, or longitudinal instability; and 25 RH fractures associated with a proximal ulnar fracture. Clinical outcome and disability were evaluated with the Mayo Elbow Performance Score (MEPS), the Oxford Elbow Score, and the Disabilities of the Arm, Shoulder, and Hand (DASH) score. Pain and satisfaction were assessed using a visual analog scale. Radiographic analysis included presence of loosening, bone loss, and overstuffing related to the RHR.RESULTSEight patients with an anatomic RHR (2 with overstuffing, 3 for stiffness, and 3 with loose implants) and 1 patient with a spacer (with stiffness) required implant removal. There were no significant differences between spacer RHR and anatomic RHR in arc of motion (120° vs. 113°, P = .14), pain relief (1 vs. 1.7, P = .135), MEPS (94 vs. 88; P = .07), Oxford Elbow Score (42.3 vs. 42.2, P = .4), or DASH score (12.2 vs. 14.4, P = .5). However, patients with a spacer RHR were significantly more satisfied (9 vs. 7.7; P = .004) than those with an anatomic implant. Radiographically, 19 anatomic implants had significant proximal bone loss and 10 showed complete lucent lines around the stem. Lucent lines were common around the spacer RHR. These radiographic changes were not always related to worse clinical outcomes.CONCLUSIONBoth the anatomic and spacer RHR designs can provide good clinical long-term outcomes. However, patients with a spacer showed a higher degree of satisfaction and those with an anatomic press-fit RHR had a higher revision rate, with