Research Article
Machine Learning Models for Survival and Neurological Outcome Prediction of Out-of-Hospital Cardiac Arrest Patients
Table 1
Characteristics of the patients at baseline.
| | Variables | All patients () |
| | Demographic characteristics | | | Age (years), | | | Sex, male, (%) | 596 (55.6) | | Underlying medical conditions, (%) | | | Heart failure | 161 (15.0) | | Cerebrovascular disease | 248 (23.2) | | Peripheral vascular disease | 37 (3.5) | | Diabetes mellitus | 244 (22.8) | | Chronic obstructive pulmonary disease | 247 (23.1) | | Chronic kidney disease | 232 (21.7) | | Liver cirrhosis | 15 (1.4) | | Malignancy | 146 (13.6) | | Tumor metastasis | 23 (2.1) | | Dementia | 100 (9.3) | | CCI scored ≥3 | 715 (61.8) | | Laboratory data, | | | White blood cell (1,000/μL) | | | Segmented neutrophils (%) | | | Band neutrophils (%) | | | Hemoglobin (g/dL) | | | Creatinine (mg/dL) | | | Alanine aminotransferase (ALT) (U/L) | | | Na (mEq/L) | | | K (mEq/L) | | | Troponin I (ng/mL) | | | pH | | | ED diagnosis, (%) | | | Hypothermia | 5 (0.5) | | Hyperkalemia | 216 (20.2) | | Acidosis | 722 (67.4) | | Acute myocardial infarction | 140 (13.1) | | Pulmonary embolism | 4 (0.4) | | Tension pneumothorax | 3 (0.3) | | Toxin | 30 (2.8) | | Diabetes ketoacidosis | 27 (2.5) | | Medication and intervention | | | Epinephrine use, (%) | 1050 (98.0) | | Epinephrine dose, | | | Sodium bicarbonate use, (%) | 690 (64.4) | | Dopamine use, (%) | 655 (61.2) | | Norepinephrine use, (%) | 212 (19.8) | | Amiodarone use, (%) | 179 (16.7) | | Lidocaine use, (%) | 38 (3.5) | | Calcium use, (%) | 196 (18.3) | | Defibrillation at ED, (%) | 93 (8.7) | | PCI, (%) | 86 (8.0) | | ECMO, (%) | 18 (1.7) | | Outcome, (%) | | | CPC class 1 or 2 | 86 (8.0) | | Survival-to-discharge | 216 (20.2) | | 30-day survival | 249 (23.2) |
|
|
CCI: Charlson comorbidity index; PCI: percutaneous coronary intervention; ECMO: extracorporeal membrane oxygenation; CPC: cerebral performance category.
|
