Patients who have experienced an out-of-hospital cardiac arrest without return of spontaneous circulation may be conveyed directly to cardiac arrest centres with a mechanical cardiopulmonary resuscitation (mCPR) device in use; these devices may also be used peri-procedurally for complications causing cardiac arrest during percutaneous coronary intervention (PCI).1–3 Performing high-quality cardiopulmonary resuscitation with manual chest compressions during simultaneous PCI is extremely challenging and mortality rates remain high.4
European Resuscitation Council and Joint British Societies guidelines advocate mCPR if high-quality manual chest compressions are not practical or compromise provider safety within the catheter lab.5,6 These mCPR devices generate higher intra-arrest coronary and cerebral perfusion pressure than manual approaches.7–9 Each device has a different profile that may overlay the heart (load distributing bands, cover plates, compression pads, cardiac pump mechanism etc), creating imaging challenges for operators to fully visualise the coronary anatomy during PCI.2,10
We sought to evaluate optimal fluoroscopic angles for visualising coronary arteries using commercially available mCPR devices in the UK (AutoPulse [Zoll Circulation, legacy model], LUCAS [Stryker Medical], Corpuls [GS Elektromedizinische Geräte G Stemple], Easy Pulse [Schiller, Precision Medical]).
We used a manikin equipped with a radio-opaque, 3D-printed heart model of a standard coronary tree. We put each mCPR device in situ and took a set of eight standard fluoroscopic projections – three cranial views, three caudal views and straight antero-posterior (AP) and straight left-anterior-oblique (LAO) views (Figure 1 ). If there was obvious mCPR device overlay, further angulation was added to visualise the desired coronary artery. We deemed there to be significant device overlay if most of the target coronary artery was obscured by the device.
For each coronary artery and mCPR device, 15 participating interventional cardiologists were asked to select the fluoroscopic projection they felt provided the best overall visualisation of the target vessel. Responses were collected using a structured Google Forms survey. The optimal projection was defined as that receiving the highest number of responses (the majority opinion). No weighting or scoring system was applied. In cases where responses were closely split, the projection with the numerically highest preference was used.
Our results demonstrated significant device overlay in four or more views with the AutoPulse and Easy Pulse devices, while Corpuls and LUCAS consistently provided superior coronary visualisation with minimal overlay.
The optimal fluoroscopic projections for each artery are shown in Table 1 and tended to follow usual practice, but with more extreme angulation requirements. The most preferred device among the interventionists surveyed was the Corpuls, followed by the LUCAS.
Our work has several important implications. First, some interventional cardiologists may reconsider which projection they choose for left-sided catheter engagement. With all devices except the Corpuls, there is significant device overlay in the straight AP projection and the next best angle that operators may consider would be LAO. Second, AutoPulse and Easy Pulse devices exhibited significant coronary artery overlay in almost all projections and perhaps should be avoided in catheterisation labs where mCPR is deployed before PCI. Third, interventional cardiologists can use Table 1 as a quick reference to demonstrate an optimal projection for a particular coronary artery.
Limitations include this being a small-scale, qualitative study that is open to bias. The heart model used did not have the ability to sub-select individual arteries and the whole coronary tree was always visible. The angiographic projections were taken with the device stationary and there can be large movement artefacts or patient factors such as extremes of height or weight that are not accounted for.
This work is hypothesis-generating and highlights the challenges associated with different mCPR devices while performing PCI and may prove a useful, practical guide to interventional cardiologists within the catheterisation lab.