Validation of cardiogenic shock phenotypes in a mixed cardiac intensive care unit population
Bibliographic record
Abstract
BACKGROUND: Proposed phenotypes have recently been identified in cardiogenic shock (CS) populations using unsupervised machine learning clustering methods. We sought to validate these phenotypes in a mixed cardiac intensive care unit (CICU) population of patients with CS. METHODS: We included Mayo Clinic CICU patients admitted from 2007 to 2018 with CS. Agnostic K means clustering was used to assign patients to three clusters based on admission values of estimated glomerular filtration rate, bicarbonate, alanine aminotransferase, lactate, platelets, and white blood cell count. In-hospital mortality and 1-year mortality were analyzed using logistic regression and Cox proportional-hazards models, respectively. RESULTS: We included 1498 CS patients with a mean age of 67.8 ± 13.9 years, and 37.1% were females. The acute coronary syndrome was present in 57.3%, and cardiac arrest was present in 34.0%. Patients were assigned to clusters as follows: Cluster 1 (noncongested), 603 (40.2%); Cluster 2 (cardiorenal), 452 (30.2%); and Cluster 3 (hemometabolic), 443 (29.6%). Clinical, laboratory, and echocardiographic characteristics differed across clusters, with the greatest illness severity in Cluster 3. Cluster assignment was associated with in-hospital mortality across subgroups. In-hospital mortality was higher in Cluster 3 (adjusted odds ratio [OR]: 2.6 vs. Cluster 1 and adjusted OR: 2.0 vs. Cluster 2, both p < 0.001). Adjusted 1-year mortality was incrementally higher in Cluster 3 versus Cluster 2 versus Cluster 1 (all p < 0.01). CONCLUSIONS: We observed similar phenotypes in CICU patients with CS as previously reported, identifying a gradient in both in-hospital and 1-year mortality by cluster. Identifying these clinical phenotypes can improve mortality risk stratification for CS patients beyond standard measures.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 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".