Development, validation, and comparative analysis of a simple clinical and laboratory-based risk score system for predicting and stratifying moderate-to-severe AKI after cardiac surgery
Bibliographic record
Abstract
Abstract Background Cardiac Surgery-Associated Acute Kidney Injury (CSA-AKI) is prevalent and detrimental following cardiac surgery, without effective prevention. Early prediction and stratification are critical in informing appropriate care. Recent studies highlighted the predictive value of some routine laboratory markers in addition to traditional clinical variables. However, existing models suffer from complexity and practical difficulties in implementation. Purpose We aim to develop a high-performance preoperative risk score system for predicting moderate-to-severe CSA-AKI, while preserving its simplicity and utility through the integration of the most predictive and routinely accessible clinical and laboratory variables. Methods A total of 14 303 patients undergoing coronary artery bypass grafting surgery at a single institution, with comprehensive perioperative data, were included. Cohort 1 (N=9 416, 2013-2017) and cohort 2 (N=4 887, 2018-2019) were used to develop and validate a risk score system for predicting moderate-to-severe CSA-AKI, respectively, according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines. Bootstrapped stepwise regression was employed to select the robust candidate variables for a logistic regression model (i.e., the full model), which was then simplified to a more parsimonious model. Model performance was evaluated and compared to several established models. Results The overall incidence of moderate-to-severe CSA-AKI was 3.5% in the entire cohort. Two preoperative risk scores were developed: the simple 6-variable ABC2-Surgery2 score (age, a biomarker of NT-proBNP, two clinical status variables of hypertension and preoperative critical state, and two surgery-related variables of combined surgery and on-pump surgery), and the complete 10-variable AB2C3-Surgery4 score (ABC2-Surgery2 plus an additional biomarker of blood urea nitrogen, a clinical status variable of Canadian Cardiovascular Society angina class, and two surgery-related variables of urgent surgery and previous surgery). In the validation cohort, the simple score (area under the curve, AUC 0.765) outperformed the previously established UK-AKI model (AUC 0.720), Mehta score (AUC 0.700), and Ng score (AUC 0.697), while statistically similar to the most complex 19-variable Cleveland Clinic Score (AUC 0.725). The complete score (AUC 0.777) was superior to all four models. The current score system also effectively stratified CSA-AKI with respect to requirements for vasoactive support (measured by cumulative vasoactive-inotropic score), intravenous diuretics, and renal replacement therapy, as well as in-hospital and long-term mortality. Conclusion A simple risk score system incorporating routine clinical and laboratory variables was developed to predict and stratify moderate-to-severe CSA-AKI and outperformed previous complex models. Further external validation is currently underway for generalizability.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".