3 The diagnostic tests

3 The diagnostic tests

The assessment compared 5 interventions with 1 comparator.

The interventions

3.1 The lead-I electrocardiogram (ECG) devices were assessed when they were used in addition to 12‑lead ECGs. Clinical experts advised that a 12‑lead ECG would still be used after lead-I ECGs to identify any additional abnormalities, such as left ventricular hypertrophy, which need to be considered when deciding on further treatment. One of the 5 lead‑I ECG devices in the scope, the RhythmPad GP (Cardiocity Ltd), was removed from this guidance after consultation. This was because the company informed NICE that, following a change in the CE mark, the device is no longer intended for detecting atrial fibrillation in people with signs or symptoms using single time point testing in primary care.

imPulse

3.2 imPulse (Plessey Semiconductors Ltd) is a CE-marked lead‑I ECG device, which is provided with downloadable software for data analysis (imPulse Viewer). The software has to be installed on a personal computer or tablet. ECGs are taken by holding the device in both hands and placing each thumb on a separate sensor on the device for a pre-set length of time (from 30 seconds to 10 minutes). Data are transferred to the hardware hosting the analytical software using Bluetooth, with the recorded ECG trace being displayed in real time.

3.3 Once the recording has finished, the generated ECG trace can be saved in the imPulse viewer. Previously recorded ECG traces can also be loaded into this viewer and can be saved as PDFs. The software's atrial fibrillation algorithm analyses the trace and states whether atrial fibrillation is unlikely, possible or probable. For a 'possible' or 'probable' result, the company recommends that the person should have further investigations, and that the algorithm should not be used to definitively diagnose atrial fibrillation.

Kardia Mobile

3.4 Kardia Mobile (AliveCor Ltd) is a CE-marked lead-I ECG device that works with the Kardia app to record and interpret ECGs. A compatible Android or Apple smartphone or tablet is also needed. Two fingers from each hand are placed on the Kardia Mobile to record an ECG, which is sent wirelessly to the device hosting the Kardia app. The default length of recording is 30 seconds, but this can be extended up to 5 minutes. The ECG trace is then automatically sent as an anonymous file to a server in the European Union for storage as an encrypted file.

3.5 The app's algorithm classifies ECG traces as:

  • normal

  • possible atrial fibrillation detected

  • unclassified.

    The instructions for use state that the Kardia app assesses for atrial fibrillation only, and the device will not necessarily detect other cardiac arrhythmias. Any non-atrial fibrillation arrhythmias detected, including sinus tachycardia, are labelled as unclassified. The company states that any ECG labelled as 'possible atrial fibrillation' or 'unclassified' should be reviewed by a cardiologist or qualified clinician. ECG traces recorded by the device can be sent from a smartphone or tablet by email as a PDF attachment and stored in a patient's records.

MyDiagnostick

3.6 MyDiagnostick (MyDiagnostick Medical BV) is a CE-marked handheld lead‑I ECG device that can produce and interpret an ECG trace. The ECG is generated by holding metal electrodes at each end of the device for 1 minute. The device activates automatically when gripped and deactivates automatically when released. A light on the device turns green if no atrial fibrillation is detected, or red if atrial fibrillation is detected. If an error occurs during the reading, the device produces both an audible warning and a visible warning from the light on the device. Up to 140 ECG traces can be stored in the device before it starts to overwrite previous traces.

3.7 MyDiagnostick can be connected to a computer via a USB connection to download the generated ECG trace for review and storage using free software (downloaded from the MyDiagnostick website). The company states that the device automatically interprets ECGs, but that a clinical professional should examine the ECG trace to confirm the diagnosis.

Zenicor-ECG

3.8 Zenicor-ECG (Zenicor Medical Systems AB) is a CE-marked system with 2 components: a lead‑I ECG device (Zenicor-EKG 2) and an online system for analysis and storage (Zenicor-EKG Backend System version 3.2). The online system sends data to a server in the European Union. This can be accessed using a web browser without prior installation of software and requires a user licence. ECGs are taken by placing both thumbs on the device for 30 seconds.

3.9 Once an ECG is taken using Zenicor-EKG 2, the trace can be transferred from the device (using a built-in mobile network modem) to a Zenicor server in Sweden. Here the ECG is analysed using the Zenicor‑EKG Backend System, which includes an automated algorithm. This categorises an ECG into 1 of 12 groups of potential arrhythmias; 1 of which includes atrial fibrillation. The algorithm also reports if the ECG cannot be analysed. The company states that a clinician needs to manually interpret the ECG trace generated by the Zenicor‑ECG to make a final diagnosis of atrial fibrillation. Clinicians can view the analysis using the Zenicor Doctor System user interface via a web browser. The ECG trace is also available via this interface and can be downloaded or printed as a PDF.

The comparator

12-lead ECG after an irregular pulse is detected

3.10 The comparator for this assessment is a 12-lead ECG, used to check for atrial fibrillation after an irregular pulse has been detected by manual pulse palpation. Clinical experts commented that an irregular pulse on manual pulse palpation is not thought to be sufficient to start anticoagulant treatment, so in this diagnostic pathway patients do not have treatment until a 12‑lead ECG confirms atrial fibrillation.

3.11 Clinical experts commented that there can be delays in arranging 12‑lead ECGs after an irregular pulse is detected, which can delay diagnosis of atrial fibrillation, or potentially miss paroxysmal atrial fibrillation because the initial examination did not include an ECG recording. The length of this delay will vary and depends on local arrangements for doing 12‑lead ECGs, for example, if this can be done in primary care or if a referral to secondary care is needed.