BACKGROUND: Collared, fully hydroxyapatite-coated stems are designed to enhance axial and rotational stability through collar-calcar contact, reducing the risk for implant subsidence and periprosthetic fracture. The purpose of this study was to address the uncertainty surrounding the use of cementless stems in Dorr C femora by evaluating the association between collar-calcar gap with periprosthetic fracture and subsidence in patients who underwent THA through the direct anterior approach.
METHODS: A retrospective analysis was performed on 6,107 total hip arthroplasties performed between September 2020 and June 2025. A total of 387 hips, with Dorr C femora, using collared, fully HA-coated stems, were included in this study. All surgeries were performed using the direct anterior approach. Radiographs were reviewed for the collar-calcar gap within 90 days postoperatively. Outcomes included the presence of periprosthetic femoral fracture (PPF) and/or subsidence.
RESULTS: Full collar-calcar contact was observed in 371 hips (95.9%), while gaps were present in 16 hips (4.1%). There were three PPFs that occurred exclusively with patients who had a collar-calcar gap (18.8%) and were not observed in patients who had collar-calcar contact. Subsidence was present in two hips among the contact group (0.5%) and two hips among the gap group (12.5%). The presence of a gap is associated with higher rates of subsidence and periprosthetic fracture with the collar-calcar gap group as compared to the collar-calcar contact group (P = 0.0000282; P = 0.0000584). The absolute number of events is small; however, the effect size is clinically meaningful.
CONCLUSION: The exclusive occurrence of early PPF and subsidence in the gap group underscores the mechanical importance of immediate load transfer through the collar. Our findings support intraoperative assessment of collar-calcar contact, especially in Dorr C bone, and reinforce the rationale for using collared, fully HA-coated, cementless stems when contact can be secured.
Read the article: PubMed · Publisher (DOI)