Background
The risk factors for and clinical impact of radiolucent lines (RLLs) in cementless total hip arthroplasty remain contentious. The aim of this work was to describe a method of classification that has clinical significance and to identify risk factors.
Methods
A cohort of 288 subjects with unrevised Corail stems (DePuy Synthes, Warsaw, IN) were reviewed with radiographs and Oxford Hip Scores at 10 years. Based on clinical experience, three groups were defined; those with no RLLs (NoRLLs), those considered to have benign RLLs (BenRLLs), and those considered to have significant RLLs (SigRLLs). SigRLLs were then compared to BenRLLs and NoRLLs to determine the validity of this classification.
Results
One hundred and nine (37.8%) had NoRLLs, 111 (38.5%) had BenRLLs, and 68 (23.6%) had SigRLLs. No significant difference apart from gender was noted between the occurrence of BenRLLs and NoRLLs after multinomial regression analysis, consequently the NoRLLs and BenRLLs groups were combined (NoSigRLLs) and compared to SigRLLs. Non-cross-linked polyethylene (odds ratio = 4.6, P < .001), collarless stem design (odds ratio = 9.4, P < .001), undersizing (odds ratio = 1.2, P = .028), and male sex (odds ratio = 2.1, P = .008) were risk factors for SigRLLs. Regression analysis also revealed that increasing age at operation decreased the likelihood of SigRLLs (P < .001). Patients with SigRLLs had significantly higher pain scores (P = .005) although overall Oxford Hip Scores were not significantly different (P = .364).
Conclusion
The definition of SigRLLs proposed in this study was significantly associated with that of non-cross-linked polyethylene, absence of a collar, undersizing, and higher pain scores.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)