Phased Chest and Abdominal Compression-Decompression Versus Conventional Cardiopulmonary Resuscitation in Out-of-Hospital Cardiac Arrest
Background Several methods have been developed to improve the efficacy of mechanical resuscitation, because organ perfusion achieved with conventional manual resuscitation is often insufficient. In animal studies, phased chest and abdominal compression-decompression resuscitation by use of the Lifestick device has resulted in a better outcome compared with that of conventional resuscitation. In end-of-life patients, an increased coronary perfusion pressure was achieved. The aim of the present study was to determine the feasibility, safety, and efficacy of the Lifestick compared with conventional resuscitation in patients with sudden nontraumatic out-of-hospital cardiac arrest.
Methods and Results The crews of 4 mobile intensive care units, staffed by an emergency physician and a paramedic, were trained to use the device. Fifty patients were randomized by sealed envelopes to either Lifestick (n=24) or conventional (n=26) resuscitation. No differences were found regarding demographic and logistical conditions between the groups. Nineteen of the patients (73%) with conventional resuscitation had ventricular fibrillation, 13 of whom survived to hospital admission (no survivals with other arrhythmias) and 7 were discharged. In contrast, in the Lifestick-CPR group, only 9 patients had ventricular fibrillation (38%; P=<0.02; OR, 2.5; 95% CI, 0.6 to 10.6). Four of these 9 patients and 5 of 15 patients with other arrhythmias survived to hospital admission, but none survived to hospital discharge. Autopsy in a subgroup of patients who died at the scene revealed less injuries with Lifestick than with conventional resuscitation.
Conclusion Lifestick resuscitation is feasible and safe and may be advantageous in patients with asystole or pulseless electric activity.
Received March 14, 2001; revision received May 22, 2001; accepted May 31, 2001.
The principal target of mechanical resuscitation measures is to raise vital organ blood flow. Apparently, conventional external cardiac compression does not always lead to a perfusion pressure sufficient to maintain blood flow as has been shown in the studies performed by Paradis et al.1 Several methods have been developed to improve the efficacy of mechanical resuscitation, eg, vest cardiopulmonary resuscitation,2 interposed abdominal compression,3 and the active compression-decompression technique (ACD-CPR) by use of the Cardio-Pump device.4,5 Although results from animal studies were promising, none of the new methods proved to be convincingly superior compared with conventional cardiopulmonary resuscitation (CPR) in randomized studies. Only in 1 study by Plaisance et al.5 did ACD-CPR result in a significantly higher survival rate compared with that of conventional CPR in an emergency medical system with surprisingly long response intervals.
Recently a new method has been developed that combines ACD-CPR and interposed abdominal compression techniques.6 This phased chest and abdominal compression-decompression–CPR (Lifestick-CPR) has been shown to result in increased coronary perfusion pressure and total cerebral blood flow, higher end-tidal carbon dioxide, and better survival in animals.6,7 In a series of patients with prolonged unsuccessful resuscitation efforts, the new technique has also led to a significant increase of coronary perfusion pressure.8 The aim of the present study was to determine feasibility, safety, and efficacy of phased chest and abdominal compression-decompression compared with conventional CPR in patients with nontraumatic out-of-hospital cardiac arrest.
The Berlin emergency medical services (EMS) are organized by the fire department as a 2-tiered system. The first tier consists of paramedic-staffed ambulances at fire stations. The paramedics are trained to perform standard CPR, including bag mask ventilation, and to use semiautomated defibrillators. The second tier consists of hospital-based mobile intensive care units (MICUs), each staffed by a paramedic and an emergency physician.
In cases of a life-threatening situation (eg, cases of cardiac arrest), the nearest MICU is alarmed parallel to the paramedic ambulance (so-called rendezvous system). If the MICU next to the scene is not available during the daytime, a rescue helicopter (also staffed by an emergency physician and a paramedic) is alerted to guarantee an alarm-scene interval of ≤10 minutes for the emergency physician in ≥90% of alarms.
The Lifestick consists of a rigid frame attached to 2 adhesive pads (Figure 1). The smaller pad (20×17 cm) has to be placed on the mid-sternum, the larger pad (37×25 cm) on the epigastrium. The pads are fixed to the Lifestick before placement on the patient. For this study, the Lifestick resuscitator was used in a 15:2 compression-ventilation ratio, at 60 cycles per minute. A metronome-driven 240° thoracic-abdominal phase shift (waltz-timing) was applied, which has been shown to result in optimal hemodynamic response.9 The abdominal compression force was limited to 18 to 28 kg and controlled by a colored LED display on the top of the frame. The display for the chest forces can be switched from a low (28 to 45 kg) setting to a medium (41 to 63 kg) or a high (54 to 82 kg) setting to achieve the target compression depth of 4 to 5 cm.
The crews of 4 MICUs and the helicopter crew were trained in small groups in application of the Lifestick, by use of a CPR-training mannequin. The initial training was performed with help from and under supervision of the monitoring staff of the manufacturer (Datascope Cardiac Assist Division). Further crew training was continuously repeated under supervision of the emergency physician on duty.
In principal, CPR was performed according to the guidelines of the European Resuscitation Council,10,11 except that patients randomized to the treatment arm had Lifestick-CPR instead of manual CPR. Patients were allocated in a 1:1 ratio to Lifestick-CPR or conventional CPR, by numbered prerandomized sealed envelopes.
The inclusion criteria were age ≥18 years, normothermia, and nontraumatic eye-witnessed arrest of assumed cardiac etiology irrespective of underlying arrhythmia. The estimated time interval of arrest and start of randomized treatment had to be ≤15 minutes, and the duration of advanced cardiac life support (defibrillation and/or pharmacologic intervention) ≤5 minutes. Patients with ventricular fibrillation were included only if up to 3 consecutive countershocks did not result in return of spontaneous circulation (ROSC). Exclusion criteria were pregnancy, any thoracic or abdominal trauma/surgery appearing to be recent, known terminal or end-stage disease, or severe neurologic impairment.
Time course of resuscitation measures (eg, time point of intubation, pharmacotherapy) were taped on the voice recorder of the semiautomated defibrillators used in all resuscitation attempts (Lifepack 300, Physio-Control). These devices also simultaneously record the rhythm of the patient. Lifestick-CPR or conventional CPR was applied for ≥20 minutes after randomization if ROSC was not achieved earlier. For patients who could not be resuscitated, relatives were asked for consent to perform an autopsy. Autopsies were performed by one of us (T.R.), blinded to the applied CPR method. Special attention was given to possible CPR-related injuries.
Study endpoints were safety (adverse events and autopsy results), ROSC ≥5 minutes, hospital admission and 1-hour survival rates, the hospital discharge rate, and 6-month survival with neurologic assessment. The study protocol was approved by the Ethics Committee of the Universitätsklinikum Benjamin Franklin, Free University, Berlin, Germany.
Values of continuous variables are given as median, maximum, and minimum. We used the χ2 test with continuity correction for dichotomous variables and the Wilcoxon test for continuous variables for group comparisons. Statistical significance was considered at P<0.05.
According to the study plan, 50 patients were blindly randomized to conventional CPR (n=26) or Lifestick-CPR (n=24). These patients were selected from a total of 180 adult patients in whom a resuscitation attempt was performed by one of the participating MICUs. The trial profiles and reasons for exclusion are shown in Figure 2.
Table 1 gives the demographic data, response intervals, and information on the situation at the scene. The first arrhythmia registered was derived from the tape recordings of the semiautomated defibrillators. Although we found no differences within the 2 study arms in most of the variables, there were significant differences in the distribution of arrhythmias (P<0.02; OR, 4.5; 95% CI, 1.4 to 15) at first registration in favor of ventricular fibrillation in the conventional CPR group. A similar trend was observed at randomization (P=0.11; OR, 3.3; 95% CI, 0.9 to 11.4).
The response intervals correspond well with the general experience in the Berlin EMS. Other data reveal some minor differences compared with previous results from the Berlin EMS,12 which may all be explained by the fact that we included only patients with an eye-witnessed arrest. The rate of bystander CPR and the proportion of patients found in ventricular fibrillation was somewhat higher in our study than generally observed in Berlin. Also, the median age of patients in this study was lower than that usually seen in resuscitations in Berlin, where it was 69 years.12
In a total of 22 of the 50 patients (44%), ROSC was observed (Table 2). With conventional CPR, 13 of 26 patients (50%) had ROSC compared with 9 of 24 (38%) with Lifestick-CPR (P=0.55; OR, 1.7; 95% CI, 0.5 to 5.2). If related to baseline arrhythmia, however, differences become more distinct. In patients with ventricular fibrillation, ROSC was achieved in 13 of 19 patients (68%) with standard CPR and in 44% (4 of 9) with the new technique (P=0.43; OR, 2.7; 95% CI, 0.5 to 13.9). In contrast, in patients with pulseless electric activity or asystole, ROSC was achieved in 5 of 15 patients (33%) with Lifestick-CPR but in none of the 7 patients with conventional CPR (P=0.23).
Twelve of the 26 patients (46%) in the standard CPR group survived to hospital admission and to ≥1 hour after admission. Seven patients of this group were discharged, and 6 survived for >6 months (P<0.05 compared with the Lifestick-CPR group). One of the patients remained in a coma in a vegetative state. He died after 12 weeks while being cared for in a nursing home. A second patient who was a long-term survivor had a moderate cerebral and overall disability; a third patient had a moderate cerebral disability but a good overall performance. The other 4 long-term survivors were without neurological deficit.
With Lifestick-CPR, 7 patients were admitted to a hospital (4 patients with ventricular fibrillation, 3 with other arrhythmia at baseline registration). All but one of these patients survived for >1 hour after admission, but none regained consciousness or survived to discharge. One patient with asystole at first arrhythmia registration died in irreversible cardiogenic shock shortly after admission to an intensive care unit.
Safety of the Device
There were no reports of severe injuries or resuscitation-related major complications during clinical course in survivors. In 9 of the 14 patients with conventional CPR and 7 of 17 patients with Lifestick-CPR who died at the scene, an autopsy was performed (Table 3). Sternal or rib fractures were found more frequently with conventional CPR, the difference being significant (P<0.05) with rib fractures. With conventional CPR, we observed 1 unusual fracture of the thoracal spine in an elderly female patient with severe osteoporosis. With Lifestick-CPR, minor abdominal lesions (insignificant peritoneal bleedings) were observed in 1 patient, and 1 patient had a minor hematoma in the hilar region.
Practicability of the Lifestick Device
The median time of interruption of cardiac massage necessary to fix the device was 20 seconds (range, 10 to 73 seconds). More than 30 seconds were needed in 4 patients. In one of the patients, fixation of the device was unsatisfactory because of hairy and wet skin, leading to a pause in the cardiac massage of 73 seconds. There were 9 reports of poor and unsatisfactory fixation of the adhesive pads to the skin, leading to instability of the device in 4 patients. No other difficulties with respect to application of the Lifestick were noticed. The ability to operate and deliver the compression force was judged to be good in 14 cases and fair in 10 cases on a 4-step scale from superior to poor. The use of the device was subjectively felt to not be more strenuous than conventional CPR and was well accepted.
An increased venous refill of the thorax is the basic mechanism leading to an increased coronary perfusion pressure and an increased vital organ blood flow with interposed abdominal compression CPR and active compression-decompression CPR.3,4 By combining and extending both techniques—ie, by phased chest and abdominal compression-decompression—a further improvement of organ blood flow could be expected from experiments according to the results of an electronic model.9 With this novel technique, thoracic decompression during abdominal compression leads to an increased venous return to the thorax by negative intrathoracic pressure. Moreover, abdominal decompression during chest compression may lead to an increased blood flow because afterload is decreased.13 In an animal model, a significant increase of coronary perfusion pressure and a better 48-hour outcome were documented.6 Also, in a recent animal study, a significantly elevated cerebral perfusion pressure, total cerebral blood flow, and left ventricular myocardial blood flow were achieved before administration of epinephrine with Lifestick-CPR compared with standard CPR.7 Finally, in a small number of patients with cardiac arrest refractory to conventional CPR, an increase of coronary perfusion pressure was shown8 after initiation of Lifestick-CPR. In an attempt to evaluate the new device in a routine environment, we compared the Lifestick-CPR technique with conventional CPR in the prehospital setting in patients with nontraumatic cardiac arrest. The principal goal of this pilot study was to test feasibility, practicability and safety of the new device.
One major problem with CPR is the mechanical trauma, which in some cases may lead to life-threatening injuries. Depending on the definition and exactness of investigation, the rate of injuries is reported to be between 21% and >65%.14 In a series of 705 autopsies after conventional resuscitation, thoracic complications of different degrees were observed in a total of 42.7% of cases. Rib fractures were found in 31.6%, sternal fractures in 21.1%. In 18.3%, mediastinal hemorrhages were observed. Moreover, abdominal injuries were described in 30.8%, and pulmonary complications in 13%. In nearly 0.5% of the cases, the injuries were categorized as life-threatening.15 With ACD-CPR, the rate of severe thorax injuries was equal or somewhat higher than that of standard CPR.16 In 1 investigation in cadavers, sternal fractures were found in 11 of 37 cases and rib fractures in 27 of 37 cases at autopsy after ACD-CPR,17 predominantly in women. In our study, we generally found rib and sternal fractures in a high rate of autopsies (Table 3) with conventional CPR, which corresponds to literature,15,16,18 and a significantly lower incidence of fractures after Lifestick-CPR. Most importantly, there was no tendency of an increase of abdominal lesions with the new technique. Autopsy was, however, not performed in all nonsurvivors, and moreover, there was no systematic search for injuries in survivors, which may lead to underreporting.
On first glance, outcome with Lifestick-CPR seems to be unfavorable compared with that of standard CPR. Because of the small sample size, we cannot exclude the possibility of a type II error, ie, standard CPR is superior to Lifestick-CPR. There is, however, a surprisingly high proportion in the subgroup with asystole and pulseless electric activity who developed ROSC, a result similar to observations with interposed abdominal compression CPR.3 Finally, bias is a special problem in a small study with so many inherent factors that have a decisive influence on the prognosis of the victim of cardiac arrest.
The 2 most important factors with respect to prognosis are the cause of arrest (some of which are absolutely irreversible) and the arrhythmia first registered with the patient, which is in turn also related to the principal underlying cause of arrest. Therefore, when evaluating resuscitation results, autopsy diagnoses with respect to cause of death are of particular interest (Table 4). In fact, in our study we found a significantly unequal distribution of rhythm disturbances in favor of ventricular fibrillation in the standard CPR group, which has the best prognosis of all arrhythmias. As may be expected from this unequal distribution, autopsy revealed a cardiac cause of death (6 patients died from coronary heart disease, 1 patient from dilatative cardiomyopathy) in nearly all autopsies performed in patients with standard CPR. In contrast, in the group with Lifestick-CPR, consisting of 63% of patients with asystole or pulseless electric activity at first registration, only 2 patients died from coronary heart disease, whereas 4 patients had extracardial causes of arrest, which was absolutely irreversible in at least one of them.
With respect to practicability, the Lifestick was well accepted by the rescue personnel. In contrast to the Cardio-Pump,19 Lifestick users did not complain about exhaustion or have the impression of strenuous handling. There were also no reports of discomfort in the back bone or other complaints by the operators. There were a few reports on instability of the device because of unsatisfactory fixation of the pads to the skin, which lead in 1 case to an interruption of cardiac massage of 73 seconds. No problems with respect to defibrillation were reported. Defibrillation was exclusively performed via the adhesive pads used with semiautomated defibrillators.
The metronome giving the thoracic-abdominal phase-shift was found to be useful. In contrast to the abdominal force control, the chest force switch and control light was felt to be troublesome. The operators would have preferred a purely optical control of the necessary compression depth of 4 to 5 cm, because the force needed to achieve this depth has to be individualized depending on the elasticity of the thorax and may not easily be controlled by the 3 force settings employed by the device. This problem while using the device may also explain failure to continuously stay exactly with the 2:15 ratio of ventilation to compression/decompression cycles in 9 cases and the observation of temporary simultaneous compression and ventilation during resuscitation in 3 patients. Both handling errors may lead to a decreased coronary blood flow and consecutively worsen the chances for successful resuscitation.20
We conclude from our results that cardiopulmonary resuscitation by use of the Lifestick after a comprehensive and repeated training is feasible, safe, and easily accepted by users. There are less CPR-related injuries with Lifestick-CPR compared with standard CPR, and it may be advantageous for a subgroup of patients, eg, patients with asystole or pulseless electric activity. Clearly, this promising new technique deserves a large-scale clinical trial for a conclusive evaluation.
This study was supported in a part by an unrestricted grant from Datascope Cardiac Assist Division, Fairfield, NJ.
Sack JB, Kesselbrenner MB, Jarrad A. Interposed abdominal compression-cardiopulmonary resuscitation and resuscitation outcome during asystole and electromechanical dissociation. Circulation. 1992; 86: 1692–1700.
Tang W, Weil MH, Shock R, et al. Phased chest and abdominal compression-decompression: a new option for cardiopulmonary resuscitation. Circulation. 1997; 95: 1335–1340.
Sterz F, Behringer W, Berzianovich A, et al. Active compression-decompression of thorax and abdomen (Lifestick-CPR) in patients with cardiac arrest. Circulation. 1996 (Suppl I); 94: I-9.Abstract.
Kloss T, Puschel K, Wischhusen F, et al. Resuscitation injuries. Anaesth Intensivther Notfallmed. 1983; 18: 199–203.
Kern KB, Hilwig RW, Berg RA, et al. Optimizing ventilation with phased chest and abdominal compression-decompression (Lifestick) CPR. J Am Coll Cardiol. 1999; 33: 369.Abstract.