We adhered to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) standards to prepare this literature review. ²⁹ Bibliographic searches were conducted using four electronic databases: PubMed, Google Scholar, Cochrane, and IEEE Explore.

Using a set of keywords, computerized bibliographic searches were carried out in PubMed, Google Scholar, Cochrane, and IEEE Explore. The search was limited to English language articles dating from 2010 to 2023. We utilized the following search terms: (“extracorporeal cardiopulmonary resuscitation” OR “ECPR” OR “extracorporeal oxygenation” OR “extracorporeal life support” OR “percutaneous cardiopulmonary support” OR “ECMO” OR “extracorporeal circulation”) AND (“cardiopulmonary resuscitation” OR “CPR” OR “conventional CPR” OR “CCPR”) AND (“out-of-hospital cardiac arrest” OR “in-hospital cardiac arrest” OR “cardiac arrest” OR “heart arrest” OR “OHCA” OR “IHCA”).

Articles were included in this literature review for fulfilling following criteria: •

Empirical primary studies evaluating ECPR in human subjects with either IHCA or OHCA.

Studies were eliminated based on the following exclusion criteria: •

Secondary sources including other reviews, newspaper articles, magazines, conference proceedings, and unpublished data sets.

Each eligible publication’s study characteristics and data were extracted by us in adherence to the PICO framework. ³⁰ These characteristics and data included the first author’s name, the publication year, the country of origin, the study’s design, the location of the arrest (OHCA or IHCA), the number of patients, the sex ratio, and the mean age.

Authors pooled the findings from the included articles, and the results of all the research under consideration were then available for synthesis and analysis. A modified version of the Newcastle-Ottawa Quality Assessment Scale for Cohort Studies was used to rate effectiveness of each study (NOS) ³¹ ( Table 1).

A total of 1,092 pertinent articles were found in the literature search across four electronic databases. The authors eliminated 373 duplicate articles. The remaining articles’ titles and abstracts were scrutinized for the research subject and 338 were excluded as these studies did not meet the research criteria. The remaining 381 articles were reduced to 140 after the inability to retrieve 241, and the remaining 140 were evaluated based on the predetermined eligibility standards. Eventually, only 12 articles met the requirements after being evaluated using the qualifying criteria (Tabulated in study characteristics in Table 2 and Table 3). The other 140 papers were disregarded on the grounds that 26 were published before 2010, 38 were non-English articles, 39 were abstracts and 25 were systematic reviews, letters to the editor, case reports, and magazines. The study selection process is shown in Figure 1.

The NOS Scale was used to assess the quality of the collected studies and considered ten factors: selection, comparability, and outcome and rating each study on a range of 0 to 10. High-quality observational cohort studies constituted those with 7 or more stars; a score of less than four denotes low quality, a score of five to six suggests moderate quality, a score of seven to eight indicates good quality, and a score of nine to 10 denotes extremely good quality. ⁴⁴ Four studies were deemed moderate quality, while eight were judged to be of high quality. No studies were deemed to be of low quality. The quality assessment score is shown in Table 1.

The main objective of the current literature review is to examine the efficacy of ECPR compared to conventional CPR in patients with cardiac arrest across various settings. In doing so, we aim to establish the effectiveness of ECPR as a future intervention for either IHCA or OHCA patients. We also identified the existing gaps that would require further research. Based on the synthesis of included studies, our review showed that employing ECPR in both IHCA and OHCA is associated with improved survival rate and corresponding positive neurological outcomes compared to conventional CPR. Our review entails a total of twelve studies comprising observational and retrospective studies, analyzing a cumulative of 2,897 IHCA and OHCA patients. Six studies examined ECPR in OHCA, while the remaining six explored IHA and OHCA. The outcome measures evaluated across the studies mainly comprised ROSC (return of spontaneous circulation), survival, neurological outcomes, eligibility for ECPR, and time to extra corporeal membrane oxygenation (ECMO). Furthermore, among the 12 studies analyzed, five studies reported an increased long-term neurologically intact survival with the use of ECPR, four of which were OHCA, ^{32 , 34 , 37 , 39} while one evaluated ECPR in IHCA ³⁵ ( Table 3 in Supplemental Material shows the studies’ design, objectives, interventions, outcomes, and results with neurological outcomes.)

Generally, studies enrolling adult IHCA and OHCA patients reported improved and favorable neurological outcomes associated with ECPR. ^{38 , 40} A study by Bartos et al. demonstrated that for CPR durations of less than 60 minutes, ECPR was associated with favorable neurological outcomes and survival despite significantly higher metabolic derangement as compared with CCPR. ³² Our observations are consistent with findings of another meta-analysis which compared ECPR and CCPR for patients with CA and found that ECPR was related to statistically significant improvement, 30-day survival, and neurologic outcomes for patients with IHCA (RR = 1.60, 95% CI = 1.25-2.06 and RR = 2.69, 95% CI = 1.63-4.46). ⁴⁵ The findings revealed no effect on neurological outcomes and survival with ECPR on OHCA patients. ⁴⁵ However, another study comparing ECPR and CCPR revealed that neurological outcome and survival were the same in both groups for OHCA. ⁴⁶ This finding particularly contradicts the outcomes of the majority of the included studies, which suggest favorable neurological outcomes associated with ECPR relative to CCPR. ^{36 , 43}

The disparities in these results can be attributed to selection bias in meta-analyses which are due to inclusion of studies with recruitment periods spanning a decade, a period in which technological advancements could have changed a lot in this field. For instance, these advancements have seen CPR guidelines being changed twice and the emergence of advanced mechanical CPR equipment becoming widely adopted. ⁴⁷ While our literature review evaluated the criteria under which ECPR should be performed, four studies reported the need to evaluate CA patients for eligibility to see if they fit for ECPR therapy. ^{35 , 39 , 41 , 42} These studies report higher survival associated with carefully selected patients after meeting inclusion criteria. ³⁹ These results are significant as they demonstrate the need to make early patient selection conceivable and necessary prior to ECPR therapy. In that regard, Te Avest et al. suggest that one of the factors to consider is the effectiveness of in-hospital ECPR, which appears to be of little importance in rural setups, as potentially ECPR-eligible patients would deteriorate prior to their arrival to the hospital. ⁴² This is imperative since as ECPR awareness grows and pre-hospital mechanical CPR is introduced, it is plausible that not only more ECPR eligible cases will present to the emergency department (ED), but may also predict a rise in ECPR ineligible patients. ⁴⁸

Time to return of spontaneous circulation, ROSC, CPR duration, and extracorporeal membrane oxygenation were evaluated as an effect of ECPR in this literature review and were reported in most of included articles. Kim et al. showed that longer durations of CPR were associated with decreased neurological outcomes in CCPR compared to with ECPR. ³⁶ In contrast, a significant number of patients who were scheduled for ECPR recovered ROSC before or after hospital admission after receiving prolonged CPR. ³⁷ In addition, Patricio et al. showed that the ROSC rates associated with ECPR were significantly higher relative to CCPR (77/80 (96%) vs. 30/80 (38%), (p <0.001)). ³⁸ Additionally, ECPR effectively reduced the time from OHCA to ECMO under an ECPR-specific program implementation. ⁴⁰ ( Table 3 provided as in Supplemental Material shows the study design, objectives, interventions, outcomes, and results with neurological outcomes.)

The majority of the included studies were observational studies and retrospective in design. This literature review demonstrates existence of a research gap with the absence of randomized control trials, which would raise the level and quality of evidence and close knowledge gaps, since the majority of the current knowledge is derived from single-center observations, and the preponderance of the evidence is derived from case series and cohort studies, making it susceptible to publication bias. ⁴⁹ As a result of observational studies’ well-known flaws, it is impossible to draw valid conclusions from primary data due to its high bias risk and potential for producing accurate but false results when combined. This indicates the importance and need for high-quality research that would explore the viability and patient-centered results of employing ECPR in innovative settings, such as via EMS-based or ED-based large, randomized trials, further exploring the usefulness of ECPR for cardiac arrest. Ideally, the design of future research would identify a successful and unified approach to ECPR and develop an algorithm that could be used for both IHCA and OHCA.