Objective The aim of this study was to compare the incidence of airway haemorrhage between participants who received manual cardiopulmonary resuscitation (CPR) and those who had received mechanical CPR using the LUCAS device.

Methods A retrospective cohort study was conducted by means of a medical chart review. All non-traumatic cardiac arrest patients that presented to the ED, from May 2014 to February 2018, were recruited. The groups were stratified according to those who had the majority of CPR performed using the LUCAS and those who had the majority of CPR performed manually. The primary outcome was the proportion of participants with airway haemorrhage, defined as blood observed in the endotracheal tube, pharynx, trachea or mouth, and documented in the doctor or nursing notes. Logistic regression analysis was performed to adjust for confounders.

Results 12 of 54 (22%) participants in the majority LUCAS CPR group had airway haemorrhage, compared with 20 of 215 (9%) participants in the majority manual CPR group, a difference of 13% (95% CI 3% to 26%, p=0.02). The unadjusted odds for developing airway haemorrhage in the majority LUCAS CPR group was 2.8 (95% CI 1.3 to 6.1). After adjusting for confounders, the odds for developing airway haemorrhage in the majority LUCAS CPR group was 2.5 (95% CI 1.1 to 5.7).

Conclusions The LUCAS mechanical CPR device is associated with a higher incidence of airway haemorrhage compared with manual CPR. Limitations in the study design mean this conclusion is not robust.