To examine whether natural language processing (NLP) using a clinically based large language model (LLM) could be used to predict patient selection for total hip or total knee arthroplasty (THA/TKA) from routinely available free-text radiology reports. Data pre-processing and analyses were conducted according to the Artificial intelligence to Revolutionize the patient Care pathway in Hip and knEe aRthroplastY (ARCHERY) project protocol. This included use of de-identified Scottish regional clinical data of patients referred for consideration of THA/TKA, held in a secure data environment designed for artificial intelligence (AI) inference. Only preoperative radiology reports were included. NLP algorithms were based on the freely available GatorTron model, a LLM trained on over 82 billion words of de-identified clinical text. Two inference tasks were performed: assessment after model-fine tuning (50 Epochs and three cycles of k-fold cross validation), and external validation.Aims
Methods
The primary aim of this study was to assess whether patients waiting six months or more for a total hip (THA) or knee (KA) arthroplasty had a deterioration in their health-related quality of life (HRQoL). Secondary aims were to assess changes in frailty and the number of patients living in a state considered to be worse than death (WTD), and factors associated with changes in HRQoL and frailty. This cross-sectional study included 326 patients, 150 males (46.0%) and 176 females (54.0%), with a mean age of 68.6 years (SD 9.8) who were randomly selected from waiting lists at four centres and had been waiting for six months or more (median 13 months, interquartile range 10 to 21) for a primary THA (n = 161) or KA (n = 165). The EuroQol five-dimension questionnaire (EQ-5D) and visual analogue scores (EQ-VAS), Rockwood Clinical Frailty Scale (CFS), and 36-Item Short Form Survey subjective change in HRQoL were assessed at the time and recalled for six months earlier. A state that was WTD was defined as an EQ-5D of less than zero.Aims
Methods
The aim of this study was to assess the quality of life of patients on the waiting list for a total hip (THA) or knee arthroplasty (KA) during the COVID-19 pandemic. Secondary aims were to assess whether length of time on the waiting list influenced quality of life and rate of deferral of surgery. During the study period (August and September 2020) 843 patients (THA n = 394, KA n = 449) from ten centres in the UK reported their EuroQol five dimension (EQ-5D) scores and completed a waiting list questionnaire (2020 group). Patient demographic details, procedure, and date when listed were recorded. Patients scoring less than zero for their EQ-5D score were defined to be in a health state “worse than death” (WTD). Data from a retrospective cohort (January 2014 to September 2017) were used as the control group.Aims
Methods
The EuroQol five-dimension (EQ-5D) questionnaire is a widely used multiattribute general health questionnaire where an EQ-5D < 0 defines a state ‘worse than death’ (WTD). The aim of this study was to determine the proportion of patients awaiting total hip arthroplasty (THA) or total knee arthroplasty (TKA) in a health state WTD and to identify associations with this state. Secondary aims were to examine the effect of WTD status on one-year outcomes. A cross-sectional analysis of 2073 patients undergoing 2073 THAs (mean age 67.4 years (Aims
Patients and Methods
Primary arthroplasty may be denied to very elderly patients based upon the perceived outcome and risks associated with surgery. This prospective study compared the outcome, complications, and mortality of total hip (TKR) and total knee replacement (TKR) in a prospectively selected group of patients aged ≥ 80 years with that of a control group aged between 65 and 74 years. There were 171 and 495 THRs and 185 and 492 TKRs performed in the older and control groups, respectively. No significant difference was observed in the mean improvement of Oxford hip and knee scores between the groups at 12 months (0.98, (95% confidence interval (CI) −0.66 to 2.95), p = 0.34 and 1.15 (95% CI −0.65 to 2.94), p = 0.16, respectively). The control group had a significantly (p = 0.02 and p = 0.04, respectively) greater improvement in the physical well being component of their SF-12 score, but the older group was more satisfied with their THR (p = 0.047). The older group had a longer hospital stay for both THR (5.9
