12-lead ECG automatic versus clinician read

StudyParticipantDiagnostic tests
Charbit 2006
Study held in France
Setting: recovery room after anaesthesia
Funding :solely from institution/ department
Population: postoperative patients. Prior tests: Unclear or Not stated.
Inclusion criteria: patients admitted to recovery room after anaesthesia (92% general anaesthesia).
Exclusion criteria: cardiac arrhythmias or bundle branch block.
Patient characteristics: age: 45 (16) years; sex: 57% female
Comorbidities: not stated.

Other study comments: Bazett formula: QTcb = QT/(square root of RR); Fridericia formula: QTcf = QT/(cube root of RR)
Index test: standard 12 lead ECG using Pagewriter M1770 (Hewlett Packard); corrected QTc calculated using Bazett or Fridericia formula; time: not stated (n=108)

Reference standard: analysed by one investigator; RR and QT intervals measured in chest lead with maximal T wave amplitude using digitising pad (SummaSketch III Professional); QTc (Bazett or Fridericia) averaged over 3–7 consecutive beats; time not stated (n=108)
for Target Condition/Outcome: prolonged QT interval (over 450ms for women and 440ms for men)
Christov 2001
Study held in Bulgaria and Italy
Setting: Cardiology
Funding :CNR-NATO Individual Fellowship
Population: routine ECGs from department of cardiology.
Prior tests: unclear or not stated.
patients from an annotated atrial flutter-fibrillation database: ECGs collected routinely in cardiology dept; >80% abnormal

Inclusion criteria: patients from an annotated atrial flutter-fibrillation database: ECGs collected routinely in cardiology dept; >80% abnormal.
Exclusion criteria: Intensive noise in V1 signals preventing accurate detection of P-wave onset and T-wave end.

Patient characteristics: age: not stated; sex: not stated
Comorbidities: not stated. Other details: patients from an annotated atrial flutter-fibrillation database: ECGs collected routinely in cardiology dept; >80% abnormal
Index test: algorithm to calculate atrial flutter/fibrillation parameter (mean value of differentiated filtered and rectified signal); threshold of 0.35% as cut-off value; instrument not specified; time: not stated (n=329)

Reference standard: atrial flutter-fibrillation records diagnosed and annotated by a group of cardiologists (no further details); time not stated (n=329)
for Target Condition/Outcome: either atrial flutter or fibrillation versus normal ECG
Denny 2007
Study held in USA
Setting: Hospital several departments (departments of biomedical informatics and medicine)
Funding :National Library of Medicine grants
Population: database of ECGs from all inpatients

all inpatients admitted for 2–30 days from 1999–2003

Inclusion criteria: all inpatients admitted for 2–30 days from 1999–2003.
Exclusion criteria: not stated
Patient characteristics: age: not stated; sex: not stated
Comorbidities: not stated.
Other study comments: database of 44808 ECGs with cardiologist-generated free text impression and machine calculated QT intervals and heart rate
Index test: machine calculated QT intervals and heart rate (automated QT and QTc) from an ECG management system (no further details); time: not stated (n=44808)

Reference standard: ardiologist-generated free text impression (selected from stock phrases or stock phrase edited or typed free text); time not stated (n=44808)
for Target Condition/Outcome: QTc over 450ms versus probable or possible QT prolongation identified by cardiologist
Fatemi 2008
Study held in Iran
Setting: Hospital several departments (Medical Science Research Institute and University hospital; ECGs from cardiac care unit and cardiac emergency ward)
Funding :grants from Mashhad University
Population: database of ECGs from patients in cardiac care unit and cardiac emergency ward. Prior tests: Unclear or Not stated.

Inclusion criteria: patients admitted to CCU and Cardiac Emergency Ward.
Exclusion criteria: not stated.

Patient characteristics: age: not stated; sex: not stated; Unclear/not stated with existing heart disease (diagnoses not given for all patients); patients admitted to CCU and Cardiac Emergency Ward
Comorbidities: not stated.
Index test: 3-channel digital ECG device (GE industry of Germany); time: not stated (n=200)

Reference standard: 4 cardiologists; time not stated (n=200)

for Target Condition/Outcome: ischaemic (acute MI/IHD); arrhythmia (premature atrial/ventricular contractions, atrial fibrillation, paroxysmal supraventricular tachycardia); structural (enlarged atrium, ventricular hypertrophy); conduction (AV/bundle branch/sinoatrial block)
Kaneko 2005
Study held in Japan
Setting: Hospital several departments (several hospitals in Japan)
Funding :not stated
Population: general population plus specific patient group. patients with Brugada syndrome; other ECGs
Definition of TLoC: not TLOC.

Inclusion criteria: patients with Brugada syndrome; other ECGs.
Exclusion criteria: not stated.

Patient characteristics: age: not stated; sex: 25 male + 2 female patients
Comorbidities: not stated

Other study comments: 97 ECGs from 27 patients with Brugada syndrome plus 21,524 other ECGs (10,564 from population health checkups; 9740 from university hospital; 1220 CSE database)
Index test: ST segment abnormalities defined as characteristic of Brugada syndrome (several sets of rules) (n=21621)

Reference standard: “classified by cardiologists”; time not stated (n=21621)

for Target Condition/Outcome: Brugada syndrome (type 1 or 2 or 3) or suspected Brugada type
Taha 2000
Study held in USA
Setting: unclear
Funding: lead author employed by GE Marquette Medical Systems
Population: database of cardiologist-read ECGs; population unclear.

Inclusion criteria: database of 4172 ECGs. Exclusion criteria: not stated.
Patient characteristics: age: not stated; sex: not stated.
Comorbidities: not stated
Other study comments: database of 4172 ECGs; frequency domain measures of QRST-subtracted signlas to differentiate between atrial flutter and fibrillation versus neither of these
Index test: time-based criteria for detecting atrial flutter or fibrillation (12SL MAC-Rhythm, GE Marquette Medical Systems, Milwaukee, WI); time: not stated (n=4172)

Reference standard: expert cardiologist; time not stated (n=4172)
for Target Condition/Outcome: atrial flutter or fibrillation (each correctly classified)

From: Appendix D1, Included studies characteristics

Cover of Transient Loss of Consciousness (‘Blackouts’) Management in Adults and Young People
Transient Loss of Consciousness (‘Blackouts’) Management in Adults and Young People [Internet].
NICE Clinical Guidelines, No. 109.
National Clinical Guideline Centre for Acute and Chronic Conditions (UK).
Copyright © National Clinical Guideline Centre for Acute and Chronic Conditions, 2010.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.