VOL. 97-B, No. 11, NOVEMBER 2015 1539  SHOULDER AND ELBOW The effect of the anatomy of the distal humerus and proximal ulna on the positioning of the components in total elbow arthroplasty H. Lenoir, M. Chammas, J. P. Micallef, C. Lazerges, T. Waitzenegger, B. Coulet From CHRU Montpellier University Hospital, Montpellier, France  H. Lenoir, MD, Orthopaedic Surgeon,  M. Chammas, PhD, Orthopaedic Surgeon,  C. Lazerges, MD, Orthopaedic Surgeon,  T. Waitzenegger, MD, Orthopaedic Surgeon,  B. Coulet, PhD, Orthopaedic Surgeon, Hand and Upper Extremity Surgery Unit CHRU Montpellier University Hospital, 371 Avenue du Doyen Gaston Giraud, Montpellier Cedex 5, 34295, France.  J. P. Micallef, PhD, Research Laboratory, Engineer Movement to Health (M2H), EA 2991, EuroMov, 700 Avenue du Pic Saint Loup - 34090 Montpellier, France. Correspondence should be sent to Dr H. L. Lenoir; e-mail: hubert.lenoir@laposte.net ©2015 The British Editorial Society of Bone & Joint Surgery doi:10.1302/0301-620X.97B11. 36071 $2.00 Bone Joint J 2015;97-B:1539–45. Received 2 March 2015; Accepted after revision 11 June 2015 Determining and accurately restoring the flexion-extension axis of the elbow is essential for functional recovery after total elbow arthroplasty (TEA). We evaluated the effect of morphological features of the elbow on variations of alignment of the components at TEA. Morphological and positioning variables were compared by systematic CT scans of 22 elbows in 21 patients after TEA. There were five men and 16 women, and the mean age was 63 years (38 to 80). The mean follow-up was 22 months (11 to 44). The anterior offset and version of the humeral components were significantly affected by the anterior angulation of the humerus (p = 0.052 and p = 0.004, respectively). The anterior offset and version of the ulnar components were strongly significantly affected by the anterior angulation of the ulna (p < 0.001 and p < 0.001). The closer the anterior angulation of the ulna was to the joint, the lower the ulnar anterior offset (p = 0.030) and version of the ulnar component (p = 0.010). The distance from the joint to the varus angulation also affected the lateral offset of the ulnar component (p = 0.046). Anatomical variations at the distal humerus and proximal ulna affect the alignment of the components at TEA. This is explained by abutment of the stems of the components and is particularly severe when there are substantial deformities or the deformities are close to the joint. Cite this article: Bone Joint J 2015;97-B:1539–45. In the early 1970s, the first fully constrained total elbow arthroplasties (TEAs) were associ- ated with high rates of loosening. 1,2 Inglis and Pellici 1 reported an overall complication rate of 53% and re-operation rate of 22% on 31 patients who underwent 36 TEAs at a mean follow-up of 3.7 years (2 to 5.5). These compli- cations decreased in the early 1980s with the emergence of new designs. 3-7 Nevertheless, loosening remains a common complication. 8,9 Many authors have evoked various patho- physiological hypotheses to explain these failures, 10-17 with malalignment of the com- ponents being prominent. 18-24 Proper func- tioning of the elbow requires a satisfactory flexion-extension axis with normal kinetics and appropriate constraints. 22,23 Soft-tissue imbalance may overload the implant- cement-bone interface and contribute to early loosening. Because of variations in the morphology of the distal humerus and proximal ulna, posi- tioning the components correctly may be diffi- cult. The distal humerus has variable anterior angulation 25,26 and in the proximal ulna, both the anterior 27-29 and varus angulation 27,30-32 can vary. The identification of these variations can require alteration in the positioning of the osteosynthetic implants when treating frac- tures. 27,30,31 No authors have reported the effects of these morphological features on the quality of the positioning of the components in TEA. We aimed to determine whether the anatom- ical characteristics of the distal humerus and proximal ulna affect the positioning of compo- nents in TEA. Patients and Methods Population and therapeutic care. Between October 2008 and January 2012, 32 TEAs were per- formed in 31 patients in our institution using the Discovery elbow system (Biomet, Warsaw, Indiana). One patient died and five were lost to follow-up. For three elbows, surgery was a revision procedure. We excluded four elbows because in three (two patients) bipolar Type III and IV evidence of radiolucency was present and in one elbow, the anatomical landmarks could not be identified. There were thus 21