The aim of this study was to evaluate the ability of a machine-learning algorithm to diagnose prosthetic loosening from preoperative radiographs and to investigate the inputs that might improve its performance. A group of 697 patients underwent a first-time revision of a total hip (THA) or total knee arthroplasty (TKA) at our institution between 2012 and 2018. Preoperative anteroposterior (AP) and lateral radiographs, and historical and comorbidity information were collected from their electronic records. Each patient was defined as having loose or fixed components based on the operation notes. We trained a series of convolutional neural network (CNN) models to predict a diagnosis of loosening at the time of surgery from the preoperative radiographs. We then added historical data about the patients to the best performing model to create a final model and tested it on an independent dataset.Aims
Methods
In arthritis of the varus knee, a high tibial
osteotomy (HTO) redistributes load from the diseased medial compartment
to the unaffected lateral compartment. We report the outcome of 36 patients (33 men and three women)
with 42 varus, arthritic knees who underwent HTO and dynamic correction
using a Garches external fixator until they felt that normal alignment
had been restored. The mean age of the patients was 54.11 years
(34 to 68). Normal alignment was achieved at a mean 5.5 weeks (3
to 10) post-operatively. Radiographs, gait analysis and visual analogue
scores for pain were measured pre- and post-operatively, at one
year and at medium-term follow-up (mean six years; 2 to 10). Failure
was defined as conversion to knee arthroplasty. Pre-operative gait analysis divided the 42 knees into two equal
groups with high (17 patients) or low (19 patients) adductor moments.
After correction, a statistically significant (p <
0.001, At final follow-up, after a mean of 15.9 years (12 to 20), there
was a survivorship of 59% (95% CI 59.6 to 68.9) irrespective of
adductor moment group, with a mean time to conversion to knee arthroplasty
of 9.5 years (3 to 18; 95% confidence interval ± 2.5). HTO remains a useful option in the medium-term for the treatment
of medial compartment osteoarthritis of the knee but does not last
in the long-term. Cite this article:
