Table 2 Baseline characteristics and variables 2.

From: Machine learning prediction models for prognosis of critically ill patients after open-heart surgery

Variables

P value of mortality

P value of thrombocytopenia

P value of septic shock

P value of liver dysfunction

Demographics

Age

 < 0.001

 < 0.001

0.007

0.032

Body mass index

0.459

0.025

0.005

0.453

Male

0.226

0.067

0.953

0.536

Vasopressin

 < 0.001

 < 0.001

 < 0.001

 < 0.001

Lab variables

Glucose

 < 0.001

0.008

0.003

0.065

Platelet

 < 0.001

 < 0.001

0.057

 < 0.001

Potassium

0.187

 < 0.001

0.255

 < 0.001

Sodium

0.001

0.064

0.251

0.007

White blood cell

 < 0.001

0.011

 < 0.001

 < 0.001

Bicarbonate

 < 0.001

 < 0.001

 < 0.001

 < 0.001

Hematocrit

 < 0.001

 < 0.001

0.293

0.092

International normalized ratio

 < 0.001

0.005

0.203

0.002

Lactate

 < 0.001

 < 0.001

0.017

 < 0.001

Partial pressure of carbon dioxide

0.37

 < 0.001

0.346

0.393

PH

 < 0.001

0.001

 < 0.001

 < 0.001

Partial pressure of oxygen

 < 0.001

0.001

 < 0.001

 < 0.001

Oxygen saturation

 < 0.001

0.003

0.015

 < 0.001

Base excess

 < 0.001

0.106

0.143

 < 0.001

Total CO2

 < 0.001

 < 0.001

 < 0.001

 < 0.001

Calcium

 < 0.001

0.29

 < 0.001

 < 0.001

Creatinine

 < 0.001

 < 0.001

 < 0.001

 < 0.001

Urea nitrogen

 < 0.001

 < 0.001

 < 0.001

 < 0.001

Anion gap

 < 0.001

0.442

 < 0.001

 < 0.001

Hemoglobin

 < 0.001

 < 0.001

0.374

0.142

Partial thromboplastin time

 < 0.001

 < 0.001

0.437

 < 0.001

Prothrombin time

 < 0.001

 < 0.001

0.344

0.003

Vital signs

Heart rate

 < 0.001

0.097

0.123

0.126

Systolic BP

 < 0.001

0.012

 < 0.001

 < 0.001

Diastolic BP

0.007

0.014

0.007

0.346

Respiratory rate

 < 0.001

 < 0.001

 < 0.001

 < 0.001

Body temperature

 < 0.001

0.071

0.008

0.129

Comorbidities

Coronary heart disease

0.051

 < 0.001

0.073

0.658

Diabetes

0.423

0.901

0.231

0.092

History of stroke

0.283

0.532

 < 0.001

0.343

Urine output

 < 0.001

 < 0.001

 < 0.001

 < 0.001

  1. P value of outcomes presents whether a variable has significant difference between positive samples (ill patients) and negative samples (normal patients).