Skip to main content

Table 1 Characteristics of the included patients

From: Machine learning for early prediction of acute myocardial infarction or death in acute chest pain patients using electrocardiogram and blood tests at presentation

 

Total

30d AMI or death

Neither AMI nor death

p-values***

n (%)

9519 (100.0)

804 (8.4)

8715 (91.6)

 

Female, %

47.3

35.1

48.4

< 0.01

Age, mean (std)

59.1 (18.9)

71.7 (12.9)

57.9 (18.9)

< 0.01

Disease history*

    

Acute myocardial infarction, %

11.1

19.7

10.3

< 0.01

Congestive heart failure, %

8.6

14.9

8.0

< 0.01

Peripheral vascular disease, %

3.7

7.8

3.4

< 0.01

Cerebral vascular accident, %

6.4

8.8

6.2

< 0.01

Dementia, %

0.9

1.2

0.8

0.205

Pulmonary disease, %

11.3

11.3

11.2

0.95

Connective tissue disorder, %

2.9

3.9

2.8

0.095

Liver disease, %

0.2

0.4

0.2

0.426

Diabetes, %

10.7

17.8

10.1

< 0.01

Diabetes complications, %

4.5

9.2

4.1

< 0.01

Renal Disease, %

3.2

6.2

2.9

< 0.01

Cancer, %

8.4

12.7

8.0

< 0.01

Metastatic cancer, %

1.4

2.6

1.3

< 0.01

Severe liver disease, %

0.1

0.1

0.1

0.586

Biomarkers

    

Glucose mmol/L, median (IQR)

6.1 (5.5–7.2)

7.2 (6.1–9)

6.1 (5.5–7.0)

< 0.01

Hb g/L, mean (std)

139.4 (16.4)

138.2 (20.1)

139.5 (16.0)

0.064

Creatinine (μmol/L), median (IQR)

78 (66–92)

85.5 (71–105)

77 (66–91)

< 0.001

hs-cTnT (ng/L), median (IQR)

6 (4–15)

52 (22–159)

6 (4–12)

< 0.001

hs-cTnT sample time, median (IQR)**

30.0 (17.0–49.0)

20.0 (10.0–36.0)

30.0 (17.0–50.0)

< 0.001

  1. n, number; AMI, acute myocardial infarction; std, standard deviation; Hb, hemoglobin; hs-cTnT, high sensitivity cardiac troponin T; IQR, interquartile range
  2. *As recorded up to 5 years prior to study event
  3. **Time from patient presentation to sampling of hs-cTnT
  4. ***p-values for differences between the group with 30d AMI or Death, and the group with neither AMI nor Death