Skip to main content

Table 2 Association with the outcome: unadjusted and adjusted hazard ratios

From: NMR-based metabolomics identifies patients at high risk of death within two years after acute myocardial infarction in the AMI-Florence II cohort

 

Hazard ratio (univariate)

p-value

Hazard ratio (multivariate)

p-value

Age

 68–79

4.63

2.24×10−04

2.09

1.45×10−01

 > 79

12.86

8.90×10−11

3.87

8.33×10−03

Male sex

0.64

1.35×10−02

0.96

8.72×10−01

Hypertension

1.34

1.47×10−01

0.53

1.98×10−02

Dyslipidemia

0.55

7.16×10−03

0.38

7.91×10−03

Smoking habits

 Yes

0.43

1.67×10−03

1.25

5.04×10−01

 Ex-smokers

1.13

7.95×10−01

4.14×10-08

9.95×10−01

CAD

0.41

2.00×10−03

0.99

9.86×10−01

Previous CABG

1.16

7.10×10−01

1.92

2.61×10−01

Previous PCI

1.28

2.69×10−01

1.66

9.64×10−02

Heart failure

3.67

1.07×10−07

1.81

8.98×10−02

Atrial fibrillation

2.37

7.03×10−04

1.30

4.52×10−01

Cerebrovascular disease

3.17

4.10×10−07

1.96

2.67×10−02

Diabetes

1.99

1.83×10−04

1.05

8.58×10−01

Creatinine (>1.2 mg/dL)

2.89

8.70×10−09

1.29

3.37×10−01

Killip class

 II

3.43

4.89×10−09

1.77

4.26×10−02

 III

4.87

9.12×10−10

3.31

4.76×10−04

ACS classification

 STEMI

0.49

1.34×10−03

0.72

3.00×10−01

NOESY RF risk score

 ≥ 0.454

6.45

2.16×10−16

3.71

2.36×10−05

GRACE score#

 ≥ 170

6.05

3.76×10−06

–

–

NOESY RF + GRACE#

 ≥ 7.7

9.33

2.16×10−16

–

–

  1. Correlation with the outcome for prognostic features and RF risk score in the full dataset, using univariate and multivariate analysis. Age split into tertiles. In the multivariate, hazard ratios of all the variable were included together in the analysis
  2. CAD coronary artery diseases, CABG coronary artery bypass grafting, PCI percutaneous coronary intervention, ACS acute coronary syndrome, STEMI ST-segment elevation myocardial infarction, NOESY nuclear Overhauser effect spectroscopy, RF random forest, GRACE Global Registry of Acute Coronary Events
  3. #These variables were not included in the multivariate analysis due to the strong co-linearity between GRACE score and the other clinical variables [66]