To the Editor:
Cave et al. 1recently reviewed the many techniques available to support circulation during cardiopulmonary resuscitation (CPR): active compression-decompression CPR; phased thoracic-abdominal compression-decompression CPR with a handheld device, impedance threshold device, or mechanical piston device; and load-distributing band CPR or vest CPR. Among all the commercially available devices globally, two devices are currently available in France: LUCAS (Lund University Cardiopulmonary Assist System; Jolife AB, Lund, Sweden), which is a gas- or electric-powered piston device that produces a consistent chest compression rate and depth,2and the automated LifeBand® (AutoPulse; ZOLL Medical Corporation, Chelmsford, MA), which is a load-distributing, broad compression band that is applied across the entire anterior chest.3These devices can be placed rapidly and used easily. In addition, they do not require extra staff to perform resuscitation.4However, the only randomized study published concerning the use of a chest-compression device showed decreased survival.3Furthermore, a recent meta-analysis concluded that there is insufficient evidence from human randomized controlled trials to conclude that mechanical chest compression during CPR for cardiac arrest is associated with either benefit or harm.5Current American Heart Association6guidelines for CPR do not recommend its immediate application in the case of cardiac arrest. Equivalent French guidelines recommend use of this device only in the case of refractory cardiac arrest.7These devices are only intended to be used by properly trained personnel1—not by a witness of a cardiac arrest, either in or out of the hospital setting, contrary to recommendations for the use of semiautomatic defibrillators. We report here the intraoperative use of a device providing consistent external mechanical cardiac compression for a patient with hypovolemic cardiac arrest. The decision to apply this device was the result of limited human resources.
A 66-yr-old man with a history of paroxystic atrial fibrillation treated by β-blocker and flecainide acetate was admitted to the emergency department for thoracic pain associated with bowel disturbance for 2 days. He had refused to undergo a colectomy on the left side for colic diverticulosis 1 month before.
On arrival, the patient's heart rate was 83 beats/min; blood pressure, 77/61 mmHg; respiratory rate, 18 breaths/min; hemoglobin, 7.6 g/L−1. He promptly developed respiratory failure, with 80% oxygen saturation. He was treated with continuous intravenous epinephrine, volume loading, and tracheal intubation. A femoral artery catheter was inserted to monitor blood pressure.
Abdominal computed tomography scanning was conducted to identify the source of abdominal bleeding. Computed tomography revealed a hemoperitoneum that was the result of splenic rupture.
Cardiac arrest occurred during patient transfer to the operating room. External manual compression was started immediately. The chest compression device was quickly placed and activated, allowing hospital staff to pursue resuscitation and initiate surgery. Automated chest compressions lasted 40 min. During this time, the patient was in asystole when the device was periodically stopped for a few seconds. The no-flow period (i.e ., duration without chest compression) was 0 min; the low-flow period (i.e ., duration with chest compression but without spontaneous cardiac activity) was 40 min. During this time, end-tidal carbon dioxide stayed near 15 mmHg. Systolic arterial blood pressure was 70–90 mmHg. Bispectral index monitoring was not possible concurrent with the use of the chest compression device; the quality of signal being low likely as a result of interference.
The patient received 3 L cell-saving recuperation, six packs of red blood cells, four fresh frozen plasma units, 2 g CaCl2, and 250 ml HCO3−,8.4%, as well as several bolus doses of epinephrine added to continuous infusion. The surgeon performed a splenectomy by a bilateral subcostal incision while the chest compression device was in use. After 40 min, the patient recovered a spontaneous rhythm and the chest compression device was stopped. The epinephrine infusion dose was progressively decreased until it was discontinued after 35 min, when arterial blood pressure reached 110/80 mmHg. Bispectral index was 0 when the chest compression device was stopped; however, this measure increased to 28 within 15 min. Anesthesia was then started with sufentanil and sevoflurane.
The patient was then transferred to the intensive care unit. He was cooled to 34°C for the first 24 h. When sedative drugs were stopped, the patient woke with no neurologic deficit. He was extubated the next day. Pathology results showed a spontaneous spleen rupture from lymphomatic proliferation related to acute lymphoplasmacytic splenic lymphoma. The patient was discharged on postoperative day 16 with a Glasgow Outcome Scale score of 1 (i.e ., no neurologic deficit).
The current report describes a case of successful intraoperative resuscitation using an automated external cardiac compression device. Two studies have reported that CPR quality during out-of-hospital8or in-hospital9cardiac arrest did not meet guideline recommendations. Wik et al. 8reported that adequate CPR was not delivered half of the time. Abella et al. 9calculated that no-flow occurred during 24% of the first 5 min of CPR. One of the explanations for this poor performance is provider fatigue after manual chest compressions lasting a few minutes.10Another element that may explain the poor performance of manual CPR is limited staffing resources, as was true in our case (i.e. , the incident occurred outside working hours). One anesthesiologist (V.D-N.) and one nurse anesthetist were present during the incident described. The automated chest compression device allowed these two hospital staff members to conduct blood verification, set up the cell-saver device, take blood samples, and manage anesthesia and vasoactive drugs. Despite body movement, the surgeon could perform the surgery without difficulty, as has already been described11when the device is used in the catheterization laboratory—particularly during percutaneous coronary intervention, reducing the risk of injury by an instrument.
In conclusion, in some settings (e.g ., limited human resources), an automated device for mechanical chest compression can represent an extra and particularly useful “pair of hands,” without interfering with surgical procedures.
*Hôpital Foch, Suresnes, France. m.fischler@hopital-foch.org