The blood glucose trajectories among non-diabetic patients with total joint arthroplasty: clinical characteristics and predictors.
To classify the blood glucose trajectories using group-based trajectory models (GBTM) and construct a random forest model to identify predictive factors to forecast different blood glucose trajectories in non-diabetic patients with total joint arthroplasty (TJA).
A prospective observational study was carried out from September 2022 to May 2024 in West China Hospital, Sichuan University. Patients without diabetes mellitus aged 18 to 80 years who underwent unilateral elective primary TJA for end-stage osteoarthritis were included in this clinical study. We measured the blood glucose level of TJA patients on preoperative 1 day and postoperative days (PODs) 0 to 2 to identify the subgroups of blood glucose trajectories in TJA patients by GBTM. Meanwhile, we collected the socio-demographic characteristics, disease-related data, results from routine blood tests, and information regarding perioperative drug use to analyze the predictors of different subgroups of blood glucose trajectories in patients with TJA.
Three distinct groups emerged: Group 1-Normal blood glucose, stable (49.5%); Group 2-Postoperative blood glucose slightly increased with minor fluctuations (41.7%); and Group 3-Hyperglycemia with significant fluctuations (8.8%). Those predictors integrated by random forest (RF) model were age, red blood cell (RBC) one day post-surgery, hypertension, diclofenac, intraoperative blood transfusion volume, and Huaxi Emotional-distress Index (HEI). The RF model achieved an overall accuracy rate of 78.3% (95% CI: 73.1-83.0%), with a Kappa coefficient of 0.600.
In this study, we found three blood glucose trajectories subgroups of TJA patients in the perioperative period using GBTM and analyzed their predictors by a RF model. However, at the current classification threshold, the model has limited ability to identify patients with hyperglycaemic trajectories.
A prospective observational study was carried out from September 2022 to May 2024 in West China Hospital, Sichuan University. Patients without diabetes mellitus aged 18 to 80 years who underwent unilateral elective primary TJA for end-stage osteoarthritis were included in this clinical study. We measured the blood glucose level of TJA patients on preoperative 1 day and postoperative days (PODs) 0 to 2 to identify the subgroups of blood glucose trajectories in TJA patients by GBTM. Meanwhile, we collected the socio-demographic characteristics, disease-related data, results from routine blood tests, and information regarding perioperative drug use to analyze the predictors of different subgroups of blood glucose trajectories in patients with TJA.
Three distinct groups emerged: Group 1-Normal blood glucose, stable (49.5%); Group 2-Postoperative blood glucose slightly increased with minor fluctuations (41.7%); and Group 3-Hyperglycemia with significant fluctuations (8.8%). Those predictors integrated by random forest (RF) model were age, red blood cell (RBC) one day post-surgery, hypertension, diclofenac, intraoperative blood transfusion volume, and Huaxi Emotional-distress Index (HEI). The RF model achieved an overall accuracy rate of 78.3% (95% CI: 73.1-83.0%), with a Kappa coefficient of 0.600.
In this study, we found three blood glucose trajectories subgroups of TJA patients in the perioperative period using GBTM and analyzed their predictors by a RF model. However, at the current classification threshold, the model has limited ability to identify patients with hyperglycaemic trajectories.