Increased femoral head size reduces the rate of dislocation after total hip arthroplasty (THA). With the introduction of highly crosslinked polyethylene (HXLPE) liners in THA there has been a trend towards using larger size femoral heads in relatively smaller cup sizes, theoretically increasing the risk of liner fracture, wear, or aseptic loosening. Short to medium follow-up studies have not demonstrated a negative effect of using thinner HXLPE liners. However, there is concern that these thinner liners may prematurely fail in the long-term, especially in those with thinner liners. The aim of this study was to evaluate the long-term survival and revision rates of HXLPE liners in primary THA, as well as the effect of liner thickness on these outcomes. We hypothesized that there would be no significant differences between the different liner thicknesses.

We performed a retrospective database analysis from a single center of all primary total hip replacements using HXLPE liners from 2010 and earlier, including all femoral head sizes. All procedures were performed by fellowship trained arthroplasty surgeons. Patient characteristics, implant details including liner thickness, death, and revisions (all causes) were recorded. Patients were grouped for analysis for each millimeter of PE thickness (e.g. 4.0-4.9mm, 5.0-5.9mm). Kaplan-Meier survival estimates were estimated with all-cause and aseptic revisions as the endpoints.

A total of 2354 patients (2584 hips) were included (mean age 64.3 years, min-max 19-96). Mean BMI was 29.0 and 47.6% was female. Mean follow-up was 13.2 years (range 11.0-18.8). Liner thickness varied from 4.9 to 12.7 mm. Seven patients had a liner thickness <5.0mm and 859 had a liner thickness of <6.0mm. Head sizes were 28mm (n=85, 3.3%), 32mm (n=1214, 47.0%), 36mm (n=1176, 45.5%), and 40mm (n=109, 4.2%), and 98.4% were metal heads. There were 101 revisions, and in 78 of these cases the liner was revised. Reason for revision was instability/dislocation (n=34), pseudotumor/aseptic lymphocyte-dominant vasculitis associated lesion (n=18), fracture (n=17), early loosening (n=11), infection (n=7), aseptic loosening (n=4), and other (n=10). When grouped by liner thickness, there were no significant differences between the groups when looking at all-cause revision (p=0.112) or aseptic revision (p=0.116).

In our cohort, there were no significant differences in all-cause or aseptic revisions between any of the liner thickness groups at long-term follow-up. Our results indicate that using thinner HXPE liners to maximize femoral head size in THA does not lead to increased complications or liner failures at medium to long term follow-up. As such, orthopedic surgeons can consider the use of larger heads at the cost of liner thickness a safe practice to reduce the risk of dislocation after THA when using HXLPE liners.

The benefit of using acetabular screws in primary total hip arthroplasty (THA) has been questioned in recent years. The disadvantages of using screws include increased operative time, risk of injury to surrounding neurovascular structures and metal ware breakage. Recent large registry studies have reported that screws do not confer a protective effect against acetabular loosening or the presence of osteolysis. Other studies have even described an increased risk of aseptic acetabular loosening with the selective use of screws. We report findings from a multicentre cohort study.

This large cohort study compared clinical outcomes between primary acetabular components that were inserted with and without screws. Independent variables included the presence (or absence) of screws, the total number of screws used and the cumulative screw length (CSL). Outcome measures included all-cause revision, acetabular component revision and acetabular component loosening. Statistical software (Stata/IC 13.1 for Mac [64-bit Intel]) was used to conduct all statistical analyses. A p-value < 0 .05 taken to be significant.

There were 4,583 THAs performed in total. Screws were used in 15.9% (n=733). At a mean follow-up of 5.2 years, the all-cause revision rate in the screw cohort was 1.5% compared to 0.83% in the no screw cohort (p=0.085). There was no difference in acetabular component revision rates for screws (3/733, 0.41%) versus no screws (12/3,850, 0.31%) (p=0.439). The rate of acetabular loosening noted during the time of revision surgery was significantly higher when screws were used in the index procedure (2/733, 0.2%) compared to the no screw cohort (1/3,850, 0.02%) (p=0.017). There was no difference in outcomes when stratifying by the number of screws used or the cumulative screw length.

Primary acetabular components do not require screws for fixation. All cause revision rates and acetabular component revision rates are comparable for the screw and the no screw cohorts. The rate of acetabular component loosening, as observed during revision surgery, is significantly higher when screws are used in the index total hip replacement.