DOS 2017

160

· DOS Abstracts

Intra- and inter-observer variability in computed

tomography assessment of gaps after primary ce-

mentless total hip arthroplasty

Maartje Belt, Omar Muharemovic, Bjørn Gliese, Hendrik Husted, Kirill Gromov,

Anders Troelsen

Dept. of Orthopaedic Surgery, Copenhagen University Hospital, Hvidovre

Background:

Progressive radiolucency is often used as a measure of the per-

formance of the cup and bearing surface in total hip arthroplasty (THA). Tradi-

tionally, radiolucency is assessed on plain x-rays, but CT scans can accurately

assess gaps around the acetabular component. However, it is time consuming.

Therefore, a faster measurement protocol was developed to measure total gap

dimensions on postoperative CT scans.

Purpose / Aim of Study:

The aim was to validate the measurement protocol in

terms of accuracy, and intra- and inter-observer variability.

Materials and Methods:

Patients (N=41) receiving a primary cementless THA

between July 2015 and March 2016 at one hospital that were enrolled in a RCT

were included in this study. Post-operative CT scans were analyzed for gaps

around the acetabular cup. The protocol is a manual segmentation in axial view

at 0%, 12.5%, 25%, 37.5%, 50%, 67.5%, 75%, 87.5%, and 100% of the gap.

The slices in between were interpolated by the software. Two observers indi-

vidually measured the gap volume, and it was compared to full segmentation.

Findings / Results:

In 95% of the subjects a gap was found, volume ranging

from 0.18 to 7.33 ml (median 2.05 ml). The ICC for intra-observer variability

was 0.988 and 0.997. The inter-observer ICC was 0.962. Using linear mixed

model, a difference of 0.42 ml (p=0.017) was observed in gap volume between

the observers. There was no statistically significant difference observed in vol-

ume between the protocol and assessment by full segmentation.

Conclusions:

The measurement method is accurate in assessing the gap vol-

ume when compared to full segmentation of CT scans. The intra-observer vari-

ability is good, and the inter-observer variability is within the levels of expected

precision. As bearing surfaces improve and the osteolytic potential decreases,

assessment of gaps by CT-scan is a useful tool.

No conflicts of interest reported

112.