Comparative Performance of Prediction Models for Contrast-Associated Acute Kidney Injury After Percutaneous Coronary Intervention
Bibliographic record
Abstract
BACKGROUND: Identifying patients at increased risk of contrast-associated acute kidney injury (CA-AKI) can help target risk mitigation strategies toward these individuals during percutaneous coronary intervention. Illuminating which risk models best stratify risk is an important foundation for such quality improvement efforts. METHODS AND RESULTS: Seven previously published risk prediction models for CA-AKI and 3 models for kidney injury requiring dialysis were validated using 2 definitions for CA-AKI (the Kidney Disease: Improving Global Outcomes definition of ≥0.3 mg/dL within 48 hours or ≥50% increase in serum creatinine from baseline within 7 days and the historical definition of ≥0.5 mg/dL or ≥25% increase in serum creatinine from baseline within 48 hours), and AKI requiring dialysis within 30 days of percutaneous coronary intervention. Model performance was compared based on discrimination, calibration, and categorical net reclassification index before and after model recalibration. Among 7888 patients who underwent percutaneous coronary intervention in Alberta Canada, CA-AKI occurred in 330 patients (4.2%) when CA-AKI was defined using the Kidney Disease: Improving Global Outcomes definition and 571 (7.3%) when using the historical definition. CA-AKI requiring dialysis occurred in 42 (0.6%) patients. When validated using the Kidney Disease: Improving Global Outcomes definition for CA-AKI, the 2 most recently published models for CA-AKI showed better discrimination (C statistics, 0.75-0.76) than older models (C statistics, 0.61-0.68). C statistics of models for kidney injury requiring dialysis ranged from 0.70 to 0.86. The calibration of all models for CA-AKI deviated from ideal, and the proportion of patients classified into different risk categories for CA-AKI differed substantially for the 2 most recent models. Recalibration significantly improved risk stratification of patients into clinical risk categories for some models. CONCLUSIONS: Recent prediction models for CA-AKI show better discrimination compared with older models; however, model recalibration should be examined in external cohorts to improve the accuracy of predictions, particularly if predicted risk strata are used to guide management approaches.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".