Table 1 Baseline patient characteristics.

From: Machine learning-based prediction of acute coronary syndrome using only the pre-hospital 12-lead electrocardiogram

 

Cohort 1 (n = 745) (training and testing)

Cohort 2 (n = 499) (external validation)

Demographics

 Age in years

59 ± 17

59 ± 16

 Sex (female)

317 (42%)

243 (49%)

Race (Black)

301 (40%)

202 (40%)

Past medical history

 Hypertension

519 (69%)

329 (66%)

 Diabetes mellitus

196 (26%)

132 (26%)

 Old myocardial infarction

205 (27%)

122 (24%)

 Known CAD

248 (33%)

179 (36%)

 Known heart failure

130 (17%)

74 (15%)

Prior PCI/CABG

207 (28%)

124 (25%)

Presenting chief complaint

 Chest pain

665 (89%)

454 (91%)

 Shortness of breathing

250 (34%)

234 (47%)

 Indigestion, nausea, or vomiting

117 (16%)

109 (22%)

 Dizziness or syncope

106 (14%)

79 (16%)

 Palpitation

96 (13%)

62 (12%)

Other atypical symptoms

54 (7%)

37 (7%)

Baseline ECG rhythm

 Normal sinus rhythm

648 (87%)

442 (88%)

 Atrial fibrillation

71 (9%)

46 (9%)

 Pacing

26 (4%)

8 (2%)

 Right bundle branch block

31 (4%)

27 (5%)

 Left bundle branch block

19 (3%)

16 (3%)

Left ventricular hypertrophy

37 (5%)

24 (5%)

Primary study outcome

 Any ACS event

114 (15.3%)

92 (18.4%)

 Prehospital STEMI

31 (4.2%)

18 (3.6%)

NSTE-ACS

83 (11.1%)

74 (14.8%)

Course of hospitalization

 Length of stay (median [IQR])

2.3 [1.0–3.0]

1.2 [0.6-2.5]

 Stress testing with SPECT

180 (24%)

115 (23%)

 Treated by primary PCI/CABG

74 (10%)

65 (13%)

30-day cardiovascular death

33 (4.4%)

24 (4.8%)