Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1050
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-09-10T06:06:05.872652Z
status: published
lens: g_space
sector: health
headline: Advancing preoperative planning technology in total joint arthroplasty with a real-time machine learning calculator: 1-year mortality risk in a value-based care era
dek: Background Technological innovation in total joint arthroplasty (TJA) has largely focused on intraoperative precision through robotics, navigation, and implant design, while preoperative decision-making remains comparatively underdeveloped. Accurate estimation of patient-specific risk is central to surgical indications, yet existing tools provide limited resolution for consequential outcomes such as 1-year mortality. Methods A machine learning model was developed using the TriNetX Research Network. Patients unde…
gain_title: An XGBoost model using 43 routine preoperative variables estimated 1-year mortality after total knee and hip arthroplasty with AUROC 0.761 and stable calibration, stratifying patients so the top 5% had 6.2-fold higher mortality than baseline.
problem_title: (none)
trace_subject: (none)
gain_reading: An XGBoost model using 43 routine preoperative variables estimated 1-year mortality after total knee and hip arthroplasty with AUROC 0.761 and stable calibration, stratifying patients so the top 5% had 6.2-fold higher mortality than baseline.
gain_evidence: Area Under the Receiver Operating Characteristic (AUROC) of 0.761 (fold-level range: 0.744-0.787, standard deviation [SD]: 0.016) | One-year mortality following primary THA and TKA can be estimated with discriminative accuracy using routinely available preoperative variables | enabling individualized preoperative risk quantification at the point of care
problem_reading: (none)
problem_evidence: (none)
quick_read: In a study published September 9, 2026, investigators built a machine learning calculator to estimate 1-year mortality after primary total knee and hip arthroplasty using 43 routinely available preoperative variables from the TriNetX Research Network. On internal validation the model achieved AUROC 0.761 with Brier score 0.006, and stratified risk monotonically from 0.574% overall to 3.57% in the top 5% of predicted risk.

The work matters because preoperative decision-making in joint replacement has lagged behind intraoperative precision tools, and individualized mortality estimation could inform surgical indications and value-based care discussions. Uncertainty remains because results are limited to internal validation with modest overprediction in the highest decile, and the authors note that prospective deployment depends on future external validations.
limitation: Model evaluated only on internal validation without external validation, and calibration showed modest overprediction in the highest risk decile, indicating need for further validation before clinical deployment.
tag: Evidence-backed gain
key_points: Study used TriNetX Research Network patients undergoing TKA or THA with verifiable 1-year follow-up; observed mortality was 0.574% (n=1,344). | Model trained with nested cross-validation and isotonic calibration; Brier score 0.006 and average precision 0.037 on internal validation. | Risk stratification showed monotonic increase: highest quintile 1.488% mortality, highest decile 2.50%, top 5% 3.57%. | Top predictors by SHAP were age (mean |SHAP| = 0.407), male sex (0.218), serum creatinine (0.162), hemoglobin (0.153), and BMI (0.145).
rundown: Researchers developed an XGBoost model incorporating 43 preoperative variables from the TriNetX Research Network for patients undergoing total knee or hip arthroplasty with verifiable 1-year follow-up. Training used nested cross-validation with isotonic calibration, and feature contributions were assessed using SHapley Additive exPlanations (SHAP).

Performance included a sensitivity-anchored threshold of 0.235% that captured 90.1% of all deaths and a Youden-optimal threshold of 0.767% that achieved 83.9% specificity. Calibration remained stable across risk deciles with mean absolute error below 0.15% for D1-D9, while the web calculator was derived from machine-learning-derived inputs for point-of-care use.
sources:
- peer_reviewed | Arthroplasty | https://doi.org/10.1186/s42836-026-00428-0 | 2026-09-09
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

1c194d3bbf06d04f1dc3fb5e2f0fd610dfc90ab27a505a96e86d3cd2af6aeed6

previous

0000000000000000000000000000000000000000000000000000000000000000