Abstract
Background and Objectives: This study analyzed the prognostic impact of mechanical cardiopulmonary resuscitation (CPR) devices in out-of-hospital cardiac arrest (OHCA) patients, in comparison to manual CPR. Materials and Methods: This study was a nationwide population-based observational study in South Korea. Data were retrospectively collected from 142,905 OHCA patients using the South Korean Out-of-Hospital Cardiac Arrest Surveillance database. We included adult OHCA patients who received manual or mechanical CPR in the emergency room. The primary outcome was survival at discharge and the secondary outcome was sustained return of spontaneous circulation (ROSC). Statistical analysis included propensity score matching and multivariate logistic regression. Results: A total of 19,045 manual CPR and 1125 mechanical CPR cases (671 AutoPulseTM vs. 305 ThumperTM vs. 149 LUCASTM) were included. In the matched multivariate analyses, all mechanical CPR devices were associated with a lower ROSC than that of manual CPR. AutoPulseTM was associated with lower survival in the multivariate analysis after matching (aOR with 95% CI: 0.57 (0.33–0.96)), but the other mechanical CPR devices were associated with similar survival to discharge as that of manual CPR. Witnessed arrest was commonly associated with high ROSC, but the use of mechanical CPR devices and cardiac origin arrest were associated with low ROSC. Only target temperature management was the common predictor for high survival. Conclusions: The mechanical CPR devices largely led to similar survival to discharge as that of manual CPR in OHCA patients; however, the in-hospital use of the AutoPulseTM device for mechanical CPR may significantly lower survival compared to manual CPR.
Keywords: cardiopulmonary resuscitation, heart arrest, chest compression, mechanical device
1. Introduction
Through recent bioengineering developments and its expansion to various medical fields, advanced medical equipment are being used to improve the quality of cardiopulmonary resuscitation (CPR). In airway management, video laryngoscopes significantly improve intubating performance, including improving the glottic view [1,2]. Mechanical CPR devices perform automatic chest compression (CC) for cardiac arrest patients [3]. As an alternative to manual CPR, the 2020 American Heart Association (AHA) CPR guidelines outline that mechanical CPR devices are indicated for use in special circumstances, such as for coronary angiography, the implementation of extracorporeal membrane oxygenation, or patient transportation via ambulances or helicopters [4].
Generally, mechanical CPR devices can be classified into three different CC mechanisms [5,6]. First, the AutoPulseTM device (AutoPulse® Resuscitation System Model 100, ZOLL®, Chelmsford, CA, USA) performs automatic CC using a load-distributing band with two arms that connect from the backboard [7]. When the whole rib cage, including the sternum, is compressed with the load-distributing band, blood flows to the heart and the whole body. Second, the ThumperTM device (Thumper Model 1007CCMII Mechanical CPR System, Michigan instruments Inc., Grand Rapids, MI, USA) performs CC using the driving force generated by compressed oxygen with one connecting arm from the backboard [8]. The piston of the ThumperTM device mimics the mechanism of manual CC and can directly compress the heart over the sternum. Third, the LUCASTM device (LUCASTM2 Chest Compression System, JOLIFE AB Inc., Lund, Sweden) uses the driving force generated by electricity to perform CC using two connecting arms and a suction cup; the device can directly compress the sternum and induce active decompression via the suction cup [9].
In a recent systematic review, the AutoPulseTM device led to a higher incidence of pneumothorax and subcutaneous hematoma than manual CPR; however, the LUCASTM device has been shown to be equivalent to manual CPR [10]. The difference in the CC mechanism between these mechanical CPR devices may affect patient outcomes, such as the return of spontaneous circulation (ROSC) or survival. Therefore, our aim was to investigate the prognostic impact of three mechanical CPR devices used in-hospital for out-of-hospital cardiac arrest (OHCA) patient outcomes, compared with manual CPR, using nationwide surveillance data from South Korea.
2. Materials and Methods
2.1. Study Design
This was a retrospective nationwide population-based observational study that used data from the Out-of-Hospital Cardiac Arrest Surveillance (OHCAS) database from the Korea Disease Control and Prevention Agency (KDCA) in South Korea (http://kdca.go.kr/, accessed on 1 March 2021). Over the period from 2012 to 2016, all acute OHCA patients transferred to medical institutions via emergency medical services were included in this study. Approximately 30,000 patients per year and 600 medical institutions were included. KDCA investigator visited the medical institution to review patient medical records and verify several items according to the Utstein Style and Resuscitation Outcome Consortium Project.
2.2. Participants
The study population included adult patients (18 years of age and older) who had been witnessed as OHCA patients between January 2012 and December 2016. The intervention group included patients who received mechanical CPR in the emergency room using one of three CPR devices (AutoPulseTM, ThumperTM, or LUCASTM). The control group included patients who received manual CPR. During pre-hospital CPR, the paramedics performed manual CPR for the patients in both groups. In addition, no patients received CPR using mechanical CPR devices until their arrival to the medical institution. Exclusion criteria were: trauma, patient with ROSC prior to arrival at the medical institution, death on arrival, patients with “do not resuscitate” orders, patients under 18 years or age, and patients who were transferred to another medical institution after emergency room management.
2.3. Outcome Measures
The primary outcome was survival at hospital discharge. The secondary outcome was the sustained ROSC (>20 min) in the emergency room.
2.4. Data Extraction
We extracted covariates to compare and analyze the survival and ROSC of patients in the intervention and control groups. The results were analyzed as follows: (1) Mechanical CPR using AutoPulseTM versus manual CPR; (2) mechanical CPR using ThumperTM versus manual CPR; and (3) mechanical CPR using LUCASTM versus manual CPR. The covariates included prognostic factors that had been reported to be significantly related to the survival and ROSC of arrest patients in previous research [3,4,10]. More specifically, the following covariates were included and analyzed in the univariate analysis: age, sex, location of the cardiac arrest (public vs. non-public), bystander CPR, cause of arrest (cardiac vs. non-cardiac), initial cardiac arrest rhythm during CPR (shockable vs. non-shockable), transport time to the medical institution, percutaneous coronary intervention (PCI), target temperature management (TTM), pacemaker, and extracorporeal membrane oxygenation (ECMO).
2.5. Statistical Analyses
The categorical variables were analyzed using Pearson’s Chi-squared and Fisher’s exact tests. Continuous variables were analyzed using an independent samples t-test for parametric data and Mann–Whitney U-test for non-parametric data. A Shapiro–Wilk test was used to assess data normality. Propensity score matching (PSM) analysis was used to adjust for the covariates and lower their confounding effects for both the control and experimental groups. Since matching cannot be performed when there are missing values, all missing values for each variable were completely omitted prior to matching. PSM model was developed including all variables with no missing value. Matching was performed for each of the three devices separately, and 1:1 was conducted without replacement of variables. Multivariate analysis using logistic regression was additionally performed using all statistically significant covariates from the univariate analysis; any variables with p < 0.05 in the univariate analyses were included in the regression. Logistic regression with backward elimination was performed for all significant factors in univariate analysis. Following the stepwise elimination of factors in the regression, only the factors that optimize the model’s coefficient of determination remained. The p-value criterion for covariate entry was 0.2. All data analyses were performed using R (version 4.1.1, The R foundation for Statistical Computing).
2.6. Ethics Statement
The local ethics committee approved this study (Kangnam Sacred Heart Hospital’s Institutional Review Board No. HKS 2018-10-020); the need for informed consent was waived off due to the study’s retrospective nature and the use of anonymous clinical data. The KCDC approved the use of the data for this study. In addition, the methodology fulfilled the criteria of the Strengthening the Reporting of Observational Studies in Epidemiology checklist [11].
3. Results
3.1. Study Subject Characteristics
In this study, mechanical CPR using three different mechanical CPR devices (AutoPulseTM, ThumperTM, and LUCASTM) was analyzed, in comparison with manual CPR. The specifications of the mechanical CPR devices are shown in Table 1. AutoPulseTM is characterized by CC performed by a compression band, and ThumperTM is characterized by CC performed by oxygen or air pressure. The compression tool in the LUCASTM device can be absorbed in the form of a cup shape, such that it adheres to the patient’s chest surface to induce active decompression.
Table 1.
Comparison of the compression mechanism in three different mechanical CPR devices.
| Classification | AutoPulseTM | ThumperTM | LUCASTM | |
|---|---|---|---|---|
| Specification of device | Backboard | Yes | Yes | Yes |
| Connecting arm | 2 | 1 | 2 | |
| Compression band | Yes | No | No | |
| Compression tool | Band type | Pad type | Suction cup type | |
| Movement of the device during CPR | Possible | Possible | Possible | |
| Driving force | Battery | Oxygen or air | Battery | |
| CC parameters | CC depth | 20% of Chest AP diameter | 5–6 cm | 4–5 cm |
| CC rate | 80 ± 5/min | 100 ± 6/min | 102 ± 2/min | |
| Duty cycle | 50 ± 5% | 50/50 cycle | 50 ± 5% | |
| CC location | Unknown | The lower half of the sternum | The lower half of the sternum | |
| Active decompression | Impossible | Impossible | Possible | |
Acronyms: CPR, cardiopulmonary resuscitation; CC, chest compression; and AP, anteroposterior.
A total of 142,905 OHCA patients were registered in the database, all of which were evaluated for their eligibility for inclusion in this study. After excluding 32,293 patients with missing data, 20,170 patients were included in our analysis (Figure 1). The number of patients in each of the four comparison groups was as follows: 19,045 patients received manual CPR; 671 patients received mechanical CPR with AutoPulseTM; 305 patients received mechanical CPR with ThumperTM; 149 patients received mechanical CPR with LUCASTM.
Figure 1.
Flow diagram for the identification of relevant studies. A star (*) indicates chest compression in emergency room. OHCA, out-of-hospital cardiac arrest; ER, emergency room; DOA, dead on arrival; ROSC, return of spontaneous circulation; DNR, do not resuscitate; and CC, chest compression.
The clinical characteristics of the patients from the unmatched data in the manual and mechanical CPR groups are summarized in Supplementary Table S1. In the unmatched univariate analysis, there was a significant difference between the mechanical and manual CPR groups for some covariates. In the AutoPulseTM vs. manual CPR comparison, the significant covariates were age, arrest rhythm, PCI, and pacemaker (p = 0.016, 0.010, 0.009, and 0.017, respectively). In the ThumperTM vs. manual CPR comparison, the only significant covariate was TTM (p = 0.028). In the LUCASTM vs. manual CPR comparison, the significant covariates were bystander CPR and ECMO (p = 0.005 and <0.001, respectively). Comparing each device of mechanical CPR and manual CPR, sustained ROSC and survival at discharge were significantly higher in manual CPR (p of AutoPulseTM = less than 0.001, 0.028, p of ThumperTM = less than 0.001, 0.011, and p of LUCASTM = 0.003, 0.048, respectively).
3.2. Matched Univariate Analysis
For the mechanical CPR with AutoPulseTM and manual CPR comparison, 1:1 PSM was applied to the three imbalanced covariates (i.e., age, arrest rhythm, PCI); the results are shown in Table 2. Both groups equally included 671 patients. Patients in the mechanical CPR with AutoPulseTM group showed a higher frequency of witnessed cardiac arrest (62.9 vs. 56.6%, p = 0.022), TTM rates (7.0 vs. 3.3%, p = 0.003), and ECMO (2.8 vs. 1.0%, p = 0.029), compared to manual CPR. Other covariates did not significantly differ across both groups. In terms of outcomes, there was a lower rate of sustained ROSC in the AutoPulseTM group (30.3 vs. 35.8%, p = 0.037), but there was no significant difference between groups in survival at discharge (4.9 vs. 6.3%, p = 0.342).
Table 2.
Comparison of the compression mechanism in three different mechanical CPR devices after propensity score matching.
| Devices | AutoPulseTM | ThumperTM | LUCASTM | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Manual CPR ** N = 671 |
Mechanical CPR ** N = 671 |
p-Value * | Manual CPR ** N = 305 |
Mechanical CPR ** N = 305 |
p-Value * | Manual CPR ** N = 149 |
Mechanical CPR ** N = 149 |
p-Value * | |
| Age, years | 69 (57–78) | 69 (57–78) | 0.951 | 69 (56–78) | 71 (56–78) | 0.790 | 68 (56–79) | 70 (56–78) | 0.705 |
| Male | 436 (65.0%) | 445 (66.3%) | 0.646 | 185 (60.7%) | 205 (67.2%) | 0.109 | 96 (64.4%) | 101 (67.8%) | 0.624 |
| Witnessed | 380 (56.6%) | 422 (62.9%) | 0.022 | 189 (62.0%) | 184 (60.3%) | 0.740 | 94 (63.1%) | 95 (63.8%) | 1.000 |
| Place | 0.888 | 0.048 | 1.000 | ||||||
| Non-public | 545 (81.2%) | 548 (81.7%) | 258 (84.6%) | 238 (78.0%) | 116 (77.9%) | 115 (77.2%) | |||
| Public | 126 (18.8%) | 123 (18.3%) | 47 (15.4%) | 67 (22.0%) | 33 (22.1%) | 34 (22.8%) | |||
| Bystander CPR | 176 (26.2%) | 191 (28.5%) | 0.391 | 69 (22.6%) | 80 (26.2%) | 0.346 | 53 (35.6%) | 53 (35.6%) | |
| Arrest cause | 0.923 | 1.000 | 0.193 | ||||||
| Cardiac | 612 (91.2%) | 614 (91.5%) | 280 (91.8%) | 281 (92.1%) | 134 (89.9%) | 141 (94.6%) | |||
| Non-cardiac | 59 (8.8%) | 57 (8.5%) | 25 (8.2%) | 24 (7.9%) | 15 (10.1%) | 8 (5.4%) | |||
| Arrest rhythm | 1.000 | 0.506 | 0.561 | ||||||
| Non-shockable | 568 (84.6%) | 568 (84.6%) | 260 (85.2%) | 253 (83.0%) | 117 (78.5%) | 122 (81.9%) | |||
| Shockable | 103 (15.4%) | 103 (15.4%) | 45 (14.8%) | 52 (17.0%) | 32 (21.5%) | 27 (18.1%) | |||
| PCI | 10 (1.5%) | 10 (1.5%) | 1.000 | 8 (2.6%) | 6 (2.0%) | 0.787 | 12 (8.1%) | 8 (5.4%) | 0.487 |
| TTM | 22 (3.3%) | 47 (7.0%) | 0.003 | 9 (3.0%) | 9 (3.0%) | 1.000 | 14 (9.4%) | 14 (9.4%) | 1.000 |
| Pacemaker | 3 (0.4%) | 0 (0.0%) | 0.249 | 4 (1.3%) | 3 (1.0%) | 1.000 | 4 (2.7%) | 1 (0.7%) | 0.371 |
| ECMO | 7 (1.0%) | 19 (2.8%) | 0.029 | 6 (2.0%) | 4 (1.3%) | 0.750 | 11 (7.4%) | 11 (7.4%) | 1.000 |
| Sustained ROSC | 240 (35.8%) | 203 (30.3%) | 0.037 | 110 (36.1%) | 62 (20.3%) | <0.001 | 69 (46.3%) | 41 (27.5%) | 0.001 |
| Survival to discharge | 42 (6.3%) | 33 (4.9%) | 0.342 | 14 (4.6%) | 10 (3.3%) | 0.532 | 8 (5.4%) | 4 (2.7%) | 0.377 |
Categorical and continuous variables are represented by a number (%) and median (interquartile range), respectively. Acronyms: CPR, cardiopulmonary resuscitation; PCI, percutaneous coronary intervention; TTM, target temperature management; ECMO, extracorporeal cardiopulmonary support; and ROSC, return of spontaneous circulation. * Calculated by Mann–Whitney test for continuous variables, and Chi-squared or Fisher’s exact test for categorical variables. ** Performed by univariate analysis after 1:1 propensity score matching.
For the mechanical CPR with ThumperTM and manual CPR comparison, 1:1 PSM was applied to the imbalanced covariate (TTM). Both groups included 305 patients. Patients in the mechanical CPR with ThumperTM group showed a higher rate of arrest in a public location compared to the manual CPR group. Other covariates did not significantly differ across both groups. With respect to outcomes, the ThumperTM group showed a lower rate of sustained ROSC compared with manual CPR (20.3 vs. 36.1%, p < 0.001). There was no significant difference across groups for survival at discharge (3.3 vs. 4.6%, p = 0.532).
For the mechanical CPR with LUCASTM and manual CPR comparison, 1:1 PSM was applied to the two imbalanced covariates (bystander CPR and ECMO). Both groups included 149 patients. There were no imbalanced covariates in either group after PSM. In terms of outcomes, the LUCASTM group showed a lower rate of sustained ROSC compared to manual CPR (27.5 vs. 46.3%, p = 0.001); however, there was no significant difference across groups for survival at discharge (2.7 vs. 5.4%, p = 0.377).
3.3. Matched Multivariate Analysis
Across all three mechanical CPR devices, the use of mechanical CPR devices, witnessed arrest, and cardiac origin arrest were common significant predictors for sustained ROSC (Figure 2 and Table 3). Witnessed arrest was significantly associated with high ROSC, but the use of mechanical CPR devices and cardiac origin arrest were associated with low ROSC. With respect to survival at discharge, TTM was the only significant predictor for high survival across all three types of mechanical CPR devices (Figure 2 and Table 4).
Figure 2.
Forest plot of matched multivariate analysis for three types of mechanical CPR devices. (A) AutoPulseTM for sustained ROSC. (B) AutoPulseTM for survival at discharge. (C) ThumperTM for sustained ROSC. (D) ThumperTM for survival at discharge. (E) LUCASTM for sustained ROSC. (F) LUCASTM for survival at discharge. * was given a statistical significance of p < 0.05. ** was set at p < 0.01 indicating statistical significance.
Table 3.
Matched multivariate analysis for three types of mechanical CPR devices with respect to sustained ROSC.
| AutoPulseTM | ThumperTM | LUCASTM | ||||||
|---|---|---|---|---|---|---|---|---|
| Factors | aOR (95% CI) * | p-Value | Factors | aOR (95% CI) * | p-Value | Factors | aOR (95% CI) * | p-Value |
| Witnessed | 2.05 (1.60–2.62) | <0.001 | Witnessed | 2.38 (1.57–3.60) | <0.001 | Male | 0.60 (0.36–1.02) | 0.060 |
| Arrest cause (cardiac origin) | 0.38 (0.26–0.57) | <0.001 | Arrest cause (cardiac origin) | 0.29 (0.15–0.54) | <0.001 | Witnessed | 2.05 (1.19–3.55) | 0.010 |
| Mechanical CPR devices | 0.74 (0.58–0.93) | 0.011 | Arrest rhythm (shockable) | 1.48 (0.91–2.41) | 0.115 | Arrest cause (cardiac origin) | 0.14 (0.05–0.41) | <0.001 |
| Mechanical CPR devices | 0.43 (0.30–0.63) | <0.001 | Mechanical CPR devices | 0.45 (0.27–0.75) | 0.002 | |||
Acronyms: aOR, adjusted odds ratio; CI, confidence interval; CPR, cardiopulmonary resuscitation; and ROSC, return of spontaneous circulation. * Calculated by multivariate logistic regression (stepwise backward elimination).
Table 4.
Matched multivariate analysis for three types of mechanical CPR device with respect to patient survival at discharge.
| AutoPulseTM | ThumperTM | LUCASTM | ||||||
|---|---|---|---|---|---|---|---|---|
| Factor | aOR (95% CI) * | p-Value | Factor | aOR (95% CI) * | p-Value | Factor | aOR (95% CI) * | p-Value |
| Witnessed | 3.89 (1.95–7.73) | <0.001 | Witnessed | 3.53 (0.90–13.92) | 0.071 | Age, year | 0.96 (0.92–1.00) | 0.033 |
| Arrest cause (cardiac origin) | 0.50 (0.23–1.09) | 0.081 | Arrest rhythm (shockable) | 2.92 (1.09–7.79) | 0.032 | Arrest cause (cardiac) | 5379780.42 (0.00–) | 0.990 |
| Arrest rhythm (shockable) | 2.00 (1.11–3.61) | 0.021 | PCI | 4.24 (0.87–20.59) | 0.072 | TTM | 6.51 (1.65–25.74) | 0.007 |
| PCI | 24.60 (6.26–96.76) | <0.001 | TTM | 6.33 (1.75–22.98) | 0.005 | Pacemaker | 14.20 (1.48–136.61) | 0.021 |
| TTM | 15.98 (8.41–30.37) | <0.001 | Pacemaker | 15.95 (2.33–109.00) | 0.004 | |||
| Mechanical CPR devices | 0.57 (0.33–0.96) | 0.035 | ||||||
| ECMO | 0.03 (0.00–0.34) | 0.004 | ||||||
Acronyms: aOR, adjusted odds ratio; CI, confidence interval; CPR, cardiopulmonary resuscitation; ECMO, extracorporeal cardiopulmonary oxygenation; PCI, percutaneous coronary intervention; and TTM, target temperature management. * Calculated by multivariate logistic regression (stepwise backward elimination).
In the multivariate analysis of matched cases, mechanical CPR with AutoPulseTM and cardiac origin arrest showed a low sustained ROSC (aOR with 95% CI: 0.74 (0.58–0.93) and 0.38 (0.26–0.57), respectively). However, witnessed cardiac arrest showed a high sustained ROSC (aOR with 95% CI: 2.05 (1.60–2.62); Table 3). Additionally, the use of AutoPulseTM and ECMO was associated with lower survival at discharge (aOR with 95% CI, 0.57 (0.33–0.96) and 0.03 (0.00–0.34), respectively; Table 4). Conversely, the following factors were associated with higher survival: witnessed cardiac arrest, arrest rhythm, PCI, TTM, and ECMO (aOR with 95% CI: 3.89 (1.95–7.73), 2.00 (1.11–3.61), 24.60 (6.26–96.76), 15.98 (8.41–30.37), and 0.03 (0.00–0.34), respectively).
In the analysis between mechanical CPR with ThumperTM and manual CPR, the use of ThumperTM and cardiac origin arrest were significantly associated with low sustained ROSC (aOR with 95% CI: 0.43 (0.30–0.63) and 0.29 (0.15–0.54), respectively; Table 3); however, witnessed cardiac arrest was significantly associated with high sustained ROSC (aOR with 95% CI: 2.38 (1.57–3.60)). The use of the ThumperTM as a mechanical CPR device was not significantly related to survival at discharge because it was removed from the final regression model (Table 4). Three factors were significant predictors for high survival: shockable arrest rhythm, TTM, and pacemaker (aOR with 95% CI: 2.92 (1.09–7.79), 6.33 (1.75–22.98), and 15.95 (2.33–109.00), respectively).
In the analysis between mechanical CPR with the LUCASTM device and manual CPR, the use of LUCASTM and cardiac origin arrest were significantly associated with low sustained ROSC (aOR with 95% CI: 0.45 (0.27–0.75) and 0.14 (0.05–0.41), respectively; Table 3). Moreover, witnessed arrest was the only significant predictor (aOR with 95% CI: 2.05 (1.19–3.55)). The use of the LUCASTM as a mechanical CPR device was not significantly related to survival at discharge because it was removed from the final regression model (Table 4). However, young age, TTM, and pacemaker were significantly associated with high survival (aOR with 95% CI: 0.96 (0.92–1.00), 6.51 (1.65–25.74), and 14.20 (1.48–136.61), respectively).
4. Discussion
This study is a nationwide population-based observational study. OHCA patients who experienced three different types of mechanical CPR devices used in-hospital were compared to those who received manual CPR; their ROSC and survival rates were analyzed. In the univariate analysis after PSM, all three mechanical CPR devices showed lower sustained ROSC than that of manual CPR. Survival at discharge after the use of these mechanical CPR devices was equivalent to that of patients who received manual CPR. In the multivariate analysis for survival, in-hospital procedures such as TTM, PCI, pacemaker, and ECMO were significant prognostic factors, even in OHCA patients. These results may provide additional scientific evidence for the clinical importance of post-cardiac arrest care in OHCA patients.
There were several studies comparing mechanical CPR in hospitals against manual CPR [12,13,14,15]. Hayashida et al. utilized national data in Japan and showed that mechanical CPR was associated with significantly worse outcome [12]. This study reported the long no flow period caused by transmission to device from manual CPR, as well as the device’s lack of real-time input on depth, rate, recoil, and cycle. Several CPR devices were included in this study, however no subgroup analysis was conducted. AutoPulseTM requires a long time to apply for an arrest patient due to its complex and heavy structure. Other machines, on the other hand, have the potential to reduce the time it takes to apply them to a patient. Furthermore, because the mechanisms for compression depth, speed, and decompression used by different machines differ, manual CPR must be compared independently.
In this study, the enrolled OHCA patients received manual CPR during the pre-hospital stage. All three types of mechanical CPR were in-hospital CPR performed in the emergency room. Each device has its own unique features and characteristics (Table 1). The automatic chest compression device’s compression movement is similar to the mechanism that successfully returns blood to the heart during manual chest compression [5], and each device may be classified according on the compression mechanism’s features [6]. To squeeze the entire ribcage, compress the heart, and increase blood flow, AutoPulseTM uses two load distributing bands attached to the backboard [7]. Since compression depth is only 20% of chest depth, compression depth might differ depending on the chest depth. It is then compressed at an 80-beat-per-minute rate. In ThumperTM, one connecting arm attached to the backboard directly compresses the heart on the sternum in the form of a piston motion, with oxygen as the driving force [8]. Compression is performed to the depth of 5-6 cm and the rate of 100 per minute. LUCASTM utilizes two connecting arms linked to the backboard and a suction cup to press in a piston motion, and the cup can induce active decompression [9]. When the overall chest depth is less than 18.5 cm, chest compressions are conducted at a depth of 5 cm or less.
The mechanical CPR with AutoPulseTM has been known to have advantages in terms of improving the outcomes of cardiac arrest patients, compared to manual CPR. First, the AutoPulseTM increases the diastolic blood pressure of cardiac arrest patients during CC compared to manual CPR, since it can continuously perform CC via band compression [16]. Nevertheless, there was insufficient evidence that this physiologic effect leads to better survival than manual CPR [16]. Second, the AutoPulseTM has an additional advantage in the pre-hospital transport of cardiac arrest patients, since CC can be maintained at an appropriate depth and rate without interruption in an ambulance [17,18]. Despite the advantages of the AutoPulseTM, there are few reports that the AutoPulseTM is superior to manual CPR [19]. In large population studies in Europe and the United States, the survival of patients who received the AutoPulseTM was similar to that of patients who received manual CPR [20,21]. Another previous study additionally reported a 11.7% rate of serious organ injury occurring in mechanical CPR with AutoPulseTM (manual CPR: 6.3%) [4]; organ injury included pneumothorax, liver rupture and emphysema, and fractures of multiple ribs and the sternum were accompanied in 45.6% of patients (manual CPR: 41.3%) [4]. A recent meta-analysis also reported that manual CPR has a lower risk of pneumothorax and hematoma than mechanical CPR with the AutoPulseTM [10]. Our study also suggests that the in-hospital use of the AutoPulseTM for mechanical CPR could significantly lower survival compared to manual CPR.
Another factor of AutoPulseTM to consider is compression rate. CPR guidelines recommend a depth of compression of 100–120 per minute for high-quality CPR [4]. The compression rate of AutoPulseTM is only 80 beats per minute, which is insufficient for high-quality CPR. Even when compared to other machines (Thumper and LUCAS are 100 ± 6 per minute and 102 ± 2 per minute, respectively), it is set low. This should be considered a variable that can influence the outcome as compared to manual CPR.
During mechanical CPR with ThumperTM, the high energy compressed by air is converted into energy that the piston can use to compress the chest, leading to an increase in cardiac output [22]. However, the high piston energy and one connecting arm of ThumperTM may also cause serious organ damage and induce a change in the CC point or non-vertical CC. Lin et al. showed no significant difference in early survival for OHCA patients performing mechanical CPR with ThumperTM versus manual CPR in the hospital [13].
The LUCASTM device can perform high-quality CPR by maintaining a consistent compression point using two connecting arms and by inducing active decompression via the suction cup. Previous studies have reported that LUCASTM can maintain a higher cardiac output, higher carotid blood flow, and higher cardiac perfusion pressure than manual CPR [23,24]. In a large population randomized trial of OHCA patients, such as the LINC and PARAMEDIC study, mechanical CPR with LUCASTM demonstrated the same sustained ROSC and survival rate, compared to manual pre-hospital CPR by paramedics [25,26]. In a systematic review and meta-analysis comparing LUCASTM and manual CPR, LUCASTM produced the same ROSC and survival as manual CPR [8,21,27]. Another advantage of LUCASTM is its capability of continuous CC during transport for emergency procedures, such as PCI [17,28]. In addition, mechanical CPR with LUCASTM can significantly reduce the interruption time during CC, compared to manual CPR [29]. Nonetheless, some autopsy studies have shown that LUCASTM can cause microfractures of the sternum or multiple ribs, which were not identified by post-mortem computed tomography [30]. In a multi-center study comparing manual CPR with mechanical CPR using LUCASTM, the LUCASTM group showed a similar frequency of sternal fracture (LUCASTM: 58.3% vs. manual CPR: 54.2%, p = 0.555) and a higher frequency of rib fractures than that of manual CPR (LUCASTM: 78.8% vs. manual CPR: 64.6%, p = 0.021) [31,32].
This study has several limitations. First, this study included a wide representation of the population of South Korea, but these results may be different in studies of another race or country. Second, regardless of our efforts to adjust for confounding factors using PSM and multivariate analysis, the data should be carefully interpretated due to the selection bias inherent with the nature of an observational study. Third, we only analyzed short-term survival provided by the OHCAS database, since this database did not provide long-term survival metrics; therefore, the effect of mechanical CPR may differ for long-term survival. Fourth, the sample size of the mechanical CPR group was not as large as that of the manual CPR group; even though PSM was applied to resolve the imbalanced sample sizes between groups, the effect of mechanical CPR should be reassessed in future large population studies.
5. Conclusions
The investigated mechanical CPR devices mostly led to equal survival as that of manual CPR among OHCA patients; however, the in-hospital use of the AutoPulseTM for mechanical CPR may significantly lower survival compared to manual CPR.
Acknowledgments
The data of this research was provided by Out-of-Hospital Cardiac Arrest Surveillance, 2012, Korea Centers for Disease Control and Prevention.
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/medicina58030353/s1. Table S1: Unmatched univariate analysis for three types of mechanical CPR devices.
Author Contributions
Conceptualization, W.K. and I.-Y.K.; methodology, W.K., C.A., S.M. and J.L. (Juncheol Lee); investigation, W.K., J.K., J.L. (Jongshill Lee) and D.-G.S.; resources, W.K., H.S., Y.C. and Y.L.; data curation, W.K., C.A., H.-Y.C. and J.-G.K.; writing—original draft preparation, W.K. and C.A.; writing—review and editing, all authors; supervision, I.-Y.K. All authors have read and agreed to the published version of the manuscript.
Funding
This research was supported by the Bio & Medical Technology Development Program of the National Research Foundation (NRF) funded by the Ministry of Science & ICT(2021M3E5D2A01022397).
Institutional Review Board Statement
The study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Institutional Review Board (Institutional Review Board of the Kangnam Sacred Heart Hospital (IRB No. 2018-10-020)).
Informed Consent Statement
Patient consent was waived due to the retrospective nature of the study.
Data Availability Statement
The datasets generated during the current study are available from the corresponding author on reasonable request.
Conflicts of Interest
The authors declare no conflict of interest.
Footnotes
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Jiang J., Ma D., Li B., Yue Y., Xue F. Video laryngoscopy does not improve the intubation outcomes in emergency and critical patients—A systematic review and meta-analysis of randomized controlled trials. Crit. Care. 2017;21:288. doi: 10.1186/s13054-017-1885-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Savino P.B., Reichelderfer S., Mercer M.P., Wang R.C., Sporer K.A. Direct versus video laryngoscopy for prehospital intubation: A systematic review and meta-analysis. Acad. Emerg. Med. 2017;24:1018–1026. doi: 10.1111/acem.13193. [DOI] [PubMed] [Google Scholar]
- 3.Gates S., Quinn T., Deakin C.D., Blair L., Couper K., Perkins G.D. Mechanical chest compression for out of hospital cardiac arrest: Systematic review and meta-analysis. Resuscitation. 2015;94:91–97. doi: 10.1016/j.resuscitation.2015.07.002. [DOI] [PubMed] [Google Scholar]
- 4.Panchal A.R., Bartos J.A., Cabañas J.G., Donnino M.W., Drennan I.R., Hirsch K.G., Kudenchuk P.J., Kurz M.C., Lavonas E.J., Morley P.T., et al. Circulation-Part 3: Adult basic and advanced life support: 2020 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2020;142:S366–S468. doi: 10.1161/CIR.0000000000000916. [DOI] [PubMed] [Google Scholar]
- 5.Ewy G.A. The mechanism of blood flow during chest compressions for cardiac arrest is probably influenced by the patient’s chest configuration. Acute Med. Surg. 2018;5:236–240. doi: 10.1002/ams2.336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ewy G.A., Zuercher M. Role of manual and mechanical chest compressions during resuscitation efforts throughout cardiac arrest. Future Cardiol. 2013;9:863–873. doi: 10.2217/fca.13.70. [DOI] [PubMed] [Google Scholar]
- 7.Zoll Medical AutoPulse Resuscitation System. [(accessed on 11 January 2022)]. Available online: https://www.zoll.com/products/automated-cpr/autopulse-for-ems.
- 8.Michigan Instruments Thumper. [(accessed on 11 January 2022)]. Available online: https://www.michiganinstruments.com.
- 9.Contol P. LUCAS Chest Compression System. [(accessed on 11 January 2022)]. Available online: https://www.lucas-cpr.com.
- 10.Khan S.U., Lone A.N., Talluri S., Khan M.Z., Khan M.U., Kaluski E. Efficacy and safety of mechanical versus manual compression in cardiac arrest—A Bayesian network meta-analysis. Resuscitation. 2018;130:182–188. doi: 10.1016/j.resuscitation.2018.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P. STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. Int. J. Surg. 2014;12:1495–1499. doi: 10.1016/j.ijsu.2014.07.013. [DOI] [PubMed] [Google Scholar]
- 12.Hayashida K., Tagami T., Fukuda T., Suzuki M., Yonemoto N., Kondo Y., Ogasawara T., Sakurai A., Tahara Y., Nagao K., et al. Mechanical Cardiopulmonary Resuscitation and Hospital Survival Among Adult Patients With Nontraumatic Out-of-Hospital Cardiac Arrest Attending the Emergency Department: A Prospective, Multicenter, Observational Study in Japan (SOS-KANTO [Survey of Survivors after Out-of-Hospital Cardiac Arrest in Kanto Area] 2012 Study) J. Am. Heart Assoc. 2017;6:e007420. doi: 10.1161/JAHA.117.007420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lin C.K., Huang M.C., Feng Y.T., Jeng W.H., Chung T.C., Lau Y.W., Cheng K.I. Effectiveness of Mechanical Chest Compression for Out-of-Hospital Cardiac Arrest Patients in an Emergency Department. J. Chin. Med. Assoc. 2015;78:360–363. doi: 10.1016/j.jcma.2015.01.005. [DOI] [PubMed] [Google Scholar]
- 14.Kim H.T., Kim J.G., Jang Y.S., Kang G.H., Kim W., Choi H.Y., Jun G.S. Comparison of in-Hospital Use of Mechanical Chest Compression Devices for Out-of-Hospital Cardiac Arrest Patients: AUTOPULSE vs LUCAS. Medicine. 2019;98:e17881. doi: 10.1097/MD.0000000000017881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ogawa Y., Shiozaki T., Hirose T., Ohnishi M., Nakamori Y., Ogura H., Shimazu T. Load-Distributing-Band Cardiopulmonary Resuscitation for Out-of-Hospital Cardiac Arrest Increases Regional Cerebral Oxygenation: A Single-Center Prospective Pilot Study. Scand. J. Trauma. Resusc. Emerg. Med. 2015;23:99. doi: 10.1186/s13049-015-0182-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Duchateau F.X., Gueye P., Curac S., Tubach F., Broche C., Plaisance P., Payen D., Mantz J., Ricard-Hibon A. Effect of the AutoPulse automated band chest compression device on hemodynamics in out-of-hospital cardiac arrest resuscitation. Intensive Care Med. 2010;36:1256–1260. doi: 10.1007/s00134-010-1784-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gässler H., Ventzke M.M., Lampl L., Helm M. Transport with ongoing resuscitation: A comparison between manual and mechanical compression. Emerg. Med. J. 2013;30:589–592. doi: 10.1136/emermed-2012-201142. [DOI] [PubMed] [Google Scholar]
- 18.Lyon R.M., Crawford A., Crookston C., Short S., Clegg G.R. The combined use of mechanical CPR and a carry sheet to maintain quality resuscitation in out-of-hospital cardiac arrest patients during extrication and transport. Resuscitation. 2015;93:102–106. doi: 10.1016/j.resuscitation.2015.05.030. [DOI] [PubMed] [Google Scholar]
- 19.Gao C., Chen Y., Peng H., Chen Y., Zhuang Y., Zhou S. Clinical evaluation of the AutoPulse automated chest compression device for out-of-hospital cardiac arrest in the northern district of Shanghai, China. Arch. Med. Sci. 2016;12:563–570. doi: 10.5114/aoms.2016.59930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hallstrom A., Rea T.D., Sayre M.R., Christenson J., Anton A.R., Mosesso V.N., Jr., Ottingham L.V., Olsufka M., Pennington S., White L.J., et al. Manual chest compression vs use of an automated chest compression device during resuscitation following out-of-hospital cardiac arrest: A randomized trial. JAMA. 2006;295:2620–2628. doi: 10.1001/jama.295.22.2620. [DOI] [PubMed] [Google Scholar]
- 21.Wik L., Olsen J.A., Persse D., Sterz F., Lozano M., Jr., Brouwer M.A., Westfall M., Souders C.M., Malzer R., van Grunsven P.M., et al. Manual vs. integrated automatic load-distributing band CPR with equal survival after out of hospital cardiac arrest. The randomized CIRC trial. Resuscitation. 2014;85:741–748. doi: 10.1016/j.resuscitation.2014.03.005. [DOI] [PubMed] [Google Scholar]
- 22.Koster R.W., Beenen L.F., van der Boom E.B., Spijkerboer A.M., Tepaske R., van der Wal A.C., Beesems S.G., Tijssen J.G. Safety of mechanical chest compression devices AutoPulse and LUCAS in cardiac arrest: A randomized clinical trial for non-inferiority. Eur. Heart J. 2017;38:3006–3013. doi: 10.1093/eurheartj/ehx318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Betz A.E., Menegazzi J.J., Logue E.S., Callaway C.W., Wang H.E. A randomized comparison of manual, mechanical and high-impulse chest compression in a porcine model of prolonged ventricular fibrillation. Resuscitation. 2006;69:495–501. doi: 10.1016/j.resuscitation.2005.09.026. [DOI] [PubMed] [Google Scholar]
- 24.Steen S., Liao Q., Pierre L., Paskevicius A., Sjöberg T. Evaluation of LUCAS, a new device for automatic mechanical compression and active decompression resuscitation. Resuscitation. 2002;55:285–299. doi: 10.1016/S0300-9572(02)00271-X. [DOI] [PubMed] [Google Scholar]
- 25.Rubertsson S., Karlsten R. Increased cortical cerebral blood flow with LUCAS; a new device for mechanical chest compressions compared to standard external compressions during experimental cardiopulmonary resuscitation. Resuscitation. 2005;65:357–363. doi: 10.1016/j.resuscitation.2004.12.006. [DOI] [PubMed] [Google Scholar]
- 26.Perkins G.D., Lall R., Quinn T., Deakin C.D., Cooke M.W., Horton J., Lanb S.E., Slowther A.-M., Woollard M., Carson A., et al. Mechanical versus manual chest compression for out-of-hospital cardiac arrest (PARAMEDIC): A pragmatic, cluster randomised controlled trial. Lancet. 2015;385:947–955. doi: 10.1016/S0140-6736(14)61886-9. [DOI] [PubMed] [Google Scholar]
- 27.Rubertsson S., Lindgren E., Smekal D., Östlund O., Silfverstolpe J., Lichtveld R.A., Boomars R., Ahlstedt B., Skoog G., Kastberg R., et al. Mechanical chest compressions and simultaneous defibrillation vs conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest: The LINC randomized trial. JAMA. 2014;311:53–61. doi: 10.1001/jama.2013.282538. [DOI] [PubMed] [Google Scholar]
- 28.Smekal D., Johansson J., Huzevka T., Rubertsson S. A pilot study of mechanical chest compressions with the LUCAS™ device in cardiopulmonary resuscitation. Resuscitation. 2011;82:702–706. doi: 10.1016/j.resuscitation.2011.01.032. [DOI] [PubMed] [Google Scholar]
- 29.Azadi N., Niemann J.T., Thomas J.L. Coronary imaging and intervention during cardiovascular collapse: Use of the LUCAS mechanical CPR device in the cardiac catheterization laboratory. J. Invasive Cardiol. 2012;24:79–83. [PubMed] [Google Scholar]
- 30.Levy M., Yost D., Walker R.G., Scheunemann E., Mendive S.R. A quality improvement initiative to optimize use of a mechanical chest compression device within a high-performance CPR approach to out-of-hospital cardiac arrest resuscitation. Resuscitation. 2015;92:32–37. doi: 10.1016/j.resuscitation.2015.04.005. [DOI] [PubMed] [Google Scholar]
- 31.Oberladstaetter D., Braun P., Freund M.C., Rabl W., Paal P., Baubin M. Autopsy is more sensitive than computed tomography in detection of LUCAS-CPR related non-dislocated chest fractures. Resuscitation. 2012;83:e89–e90. doi: 10.1016/j.resuscitation.2011.12.003. [DOI] [PubMed] [Google Scholar]
- 32.Smekal D., Lindgren E., Sandler H., Johansson J., Rubertsson S. CPR-related injuries after manual or mechanical chest compressions with the LUCAS™ device: A multicentre study of victims after unsuccessful resuscitation. Resuscitation. 2014;85:1708–1712. doi: 10.1016/j.resuscitation.2014.09.017. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during the current study are available from the corresponding author on reasonable request.