Photoplethysmography to detect circulatory arrest: A study in patients with induced cardiac arrest

Every week in the Netherlands, 300 victims suffer from out-of-hospital cardiac
arrest (OHCA). Early recognition and help are of the utmost importance as
survival chances decrease with 5-10% per minute delay to treatment. Over the
past decade, deployment of lay rescuers initiating cardiopulmonary
resuscitation prior to ambulance arrival has contributed to shortening of
treatment delays and improvement in OHCA survival to up to 23%. Unfortunately,
for cardiac arrest victims without a witness, the chance of survival is dismal.
Automated cardiac arrest detection and activation of the emergency medical
chain may help to further shorten treatment delays and provide early help for
victims of unwitnessed cardiac arrest. In the DETECT project, we aim to develop
a smartwatch/wristband that is able to automatically detect cardiac arrest and
alert the emergency services. This project is supported by the Dutch Heart
Foundation. We aim to use photoplethysmography (PPG) to detect circulatory
arrest non-invasively. This is an optical technique with which blood volume
changes can be measured on the skin. It is well-known for its use to measure
oxygen saturation at the finger tip or to monitor heart rate using a
sportwatch. The current submission concerns the DETECT-1 study. This is a first
study to come to above-described solution for automated cardiac arrest
detection. It is an observational study in patients with short-lasting induced
circulatory arrest during which we collect additional data (PPG signals) using
a wristband. We will study the collected PPG signals in relation to the
presence and absence of circulatory arrest. Subsequently, this data will be
used to construct a PPG-based algorithm for automate cardiac arrest detection
and its performance will be evaluated. It is important to emphasize that the
induced circulatory arrests are regularly performed during clinically-indicated
procedures (defibrillation testing after implantable cardioverter
defibrillator (ICD) implantation and rapid ventricular pacing during valve
placement during transcatheter aortic valve implantation (TAVI)) and are by no
means a study-related intervention. For further information, I refer to Chapter
1 of the study protocol ('Introduction and rationale').

1) To study PPG signal characteristics in relation to circulatory arrest in
patients with induced circulatory arrest.
2) To construct a PPG-based algorithm for detection of circulatory arrest based
on induced circulatory arrest data in patients.
3) To study the performance (sensitivity and specificity) of the developed
PPG-based algorithm for detection of circulatory arrest in patients with
induced circulatory arrests.

The DETECT-1 is a Dutch prospective multicenter observational cohort study
performed in a hospital setting.

Given the observational nature of this study and the non-invasive comfortable
wristband used for PPG signal recording, the medical risk associated with study
participation is negligible. For ICD/VT ablation patients, an arterial line
will be inserted in the radial artery to measure blood pressure invasively
during the procedure. Complications following radial artery cannulation include
permanent ischemic damage (0.09%), temporary occlusion (20%), sepsis (0.13%),
local infection (0.72%), pseudoaneurysm (0.09%), hematoma (14.4%), bleeding
0.53%. Generally temporary occlusion of the radial artery has no serious
sequelae. Permanent occlusion appears to be rare (0.09%) (14). In the majority
of studies on which these incidences are based, the arterial line was inserted
for longer term invasive monitoring (e.g. during intensive care unit stay). It
is expected that complication rates will be lower with planned insertion of an
arterial line under optimal circumstances and only for monitoring during the
procedure.