Effect of positive end-expiratory pressure during cardiopulmonary resuscitation on short-term survival

Introduction: The effect of positive end-expiratory pressure (PEEP) during intermittent positive pressure ventilation in patients undergoing cardiopulmonary resuscitation (CPR) on outcome is unknown. PEEP administration may increase cardiac output by reducing left ventricular afterload, reversing pulmonary atelectasis and hence lowering pulmonary vascular resistance. The purpose of the current study is to evaluate the effect of PEEP on short-term survival in out-of-hospital cardiac arrest (OHCA) patients. Material: Physician-staffed medical team resuscitated OHCA patients. They were intubated and ventilated using a bag with or without PEEP valve or with a mechanical ventilator with or without PEEP at the discretion of the attending physician. During ventilation, tracheal airway pressures were measuredusing two air-filled catheters connected to a custom-made portable external recording device consisting of two pressure sensors, two amplifiers and a logger. Compression data were recorded using an accelerometer on a Zoll E or X series defibrillator (Zoll, Chelmsford, US). Pressure and compression data were analysed with dedicated LabVIEW software (National Instruments, Austin, Texas, U Data are expressed as mean values (standard deviation; range). Results: Data were collected in 298 patients with a mean age of 65 years (15.6; 18-89), 211 (70.8%) were male. Of all patients, 36.9% achieved ROSC. The initial rhythm was asystole in 54.7%, ventricular fibrillation in 17.8%, pulseless electrical activity in 11.1% and ventricular tachycardia in 0.7%. The mean chest compression depth was 4.9 cm (0.91; 2.02–8.50) and the mean chest compression rate was 115 per minute (12.7; 52-155). Five cm H2O PEEP was used in 169 patients(56.7%). Patients ventilated with 5 cm H20 PEEP had significantly higher ROSC compared to patients without PEEP (48.2% vs. 22.2%, P < 0.001). Conclusion: The use of PEEP during CPR in OHCA patients was associated with higher ROSC. These results need to be confirmed in a multivariate analysis.