The Association Between Signs of Life and Outcomes in Out‐of‐Hospital Cardiac Arrest: A Systematic Review and Meta‐Analysis

ABSTRACT

Aim

Out‐of‐hospital cardiac arrest (OHCA) carries a poor prognosis despite advances in cardiopulmonary resuscitation (CPR). Traditional prognostic factors have limited accuracy, and “signs of life,” including gasping, pupillary light reflex (PLR), and CPR‐induced consciousness (CPRIC), have been proposed as prognostic markers. We conducted a systematic review and meta‐analysis to evaluate the association between signs of life and outcomes in OHCA.

Methods

PubMed, MEDLINE, and Web of Science were systematically searched. Eligible studies included adult patients with OHCA of presumed medical origin reporting survival or neurological outcomes in relation to signs of life. Study quality was assessed using the Newcastle–Ottawa Scale. Pooled odds ratios with 95% confidence intervals were calculated using a random‐effects model.

Results

Eight cohort studies, including 4,862 patients, were analyzed. All studies were of good methodological quality. The presence of signs of life was associated with higher survival and favorable neurological outcomes within 30 days after cardiac arrest (CA). Similar associations were observed for longer‐term survival, although evidence for longer‐term neurological outcomes was limited.

Conclusion

Signs of life during CA or CPR were consistently associated with improved survival and favorable neurological outcomes in patients with OHCA, both within 30 days and beyond 90 days. These findings suggest that signs of life may have prognostic value; however, their applicability may vary depending on clinical context. Systematic documentation of signs of life may be considered in standardized reporting frameworks.

Signs of life during cardiac arrest or CPR were consistently associated with improved survival and favorable neurological outcomes within 30 days and beyond. These findings underscore their prognostic value and support systematic documentation in standardized reporting frameworks.

1. Introduction

Out‐of‐hospital cardiac arrest (OHCA) remains a major public health challenge worldwide, with persistently high mortality [1]. Despite advances in cardiopulmonary resuscitation (CPR), only 5%–10% of patients survive to hospital discharge with favorable neurological outcomes. Traditional prognostic factors—such as initial cardiac rhythm, witnessed arrest, bystander CPR, and no‐flow or low‐flow time—provide limited predictive accuracy [2]. Consequently, many patients undergo resuscitation without meaningful neurological recovery, highlighting the need for more accurate prognostic markers applicable during ongoing CPR.

Recently, “signs of life” have gained attention as potential prognostic indicators during resuscitation [3, 4, 5]. Signs of life include gasping, pupillary light reflex (PLR), and spontaneous body movement or CPR‐induced consciousness (CPRIC) observed during cardiac arrest or CPR. These findings are thought to reflect preserved cerebral perfusion generated by chest compressions and partial brain function. Previous studies have shown that gasping is associated with markedly improved survival, and among patients undergoing extracorporeal CPR (ECPR), the presence of signs of life strongly predicts favorable neurological outcomes [3, 4, 5, 6, 7]. However, prior systematic reviews have focused exclusively on gasping [6, 7] and did not evaluate other clinically relevant manifestations such as PLR or CPRIC. As a result, the overall prognostic significance of signs of life across multiple physiological domains remains unclear.

In addition, definitions of signs of life and the timing of their assessment vary across studies, and their implications for ECPR eligibility and for patients with initial non‐shockable rhythms remain debated. Therefore, we conducted a systematic review and meta‐analysis to evaluate whether the presence of signs of life during CPR is associated with improved survival and favorable neurological outcomes in adult patients with OHCA.

2. Materials and Methods

This systematic review was conducted in accordance with the PRISMA 2020 statement (Supporting Information S1) [8] and was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD420251025806) (Supporting Information 2).

2.1. Clinical Question

The review addressed the following question based on the Population, Exposure, Comparator, and Outcome framework: In adult patients with OHCA, is the presence of “signs of life” during CA or CPR associated with higher survival or favorable neurological outcomes compared with their absence?

2.2. Data Sources

A systematic literature search of PubMed, MEDLINE, and Web of Science was performed on December 1, 2024, without restrictions on publication date. Search terms included OHCA‐related terms and signs of life (e.g., gasping, pupillary light reflex, CPR‐induced consciousness). The full search strategy is provided in Supporting Information 3.

2.3. Eligibility Criteria

Eligible studies were English‐language, peer‐reviewed human studies involving adults (≥ 18 years) with OHCA of presumed medical origin that reported survival or neurological outcomes according to the presence or absence of signs of life. Signs of life were defined as gasping, PLR, or body movement/CPRIC observed during CA or CPR. For the purpose of this review, signs of life were defined as occurring during CA or ongoing CPR. Breathing, PLR, and body movement observed after the return of spontaneous circulation were not included. Given the variability in reporting across studies, the timing of signs of life during CA and CPR was not always clearly distinguishable across studies and was therefore treated as a unified concept in this analysis. Studies involving pediatric populations, externally caused CA (e.g., trauma, accidental hypothermia, intoxication), animal experiments, case reports, and abstracts without full text were excluded.

2.4. Study Selection and Data Extraction

Study selection was performed using Rayyan [9] by four reviewers working independently in pairs, with disagreements resolved by consensus. Data extraction was conducted independently by two reviewers using a standardized form, collecting information on study design, population, timing, and type of signs of life, and clinical outcomes.

2.5. Quality Assessment

Methodological quality was assessed using the Newcastle–Ottawa Scale (NOS) for cohort studies [10] by two reviewers independently, with disagreements resolved by discussion.

2.6. Outcomes

The primary outcomes were survival and a favorable neurological outcome (Cerebral Performance Category 1–2). Outcomes assessed at hospital discharge or within 28–30 days were defined as outcomes within 30 days, whereas outcomes assessed at 90 days or later were considered longer‐term outcomes.

2.7. Data Synthesis and Analysis

Adjusted odds ratios were extracted when available; otherwise, crude odds ratios were calculated from reported aggregated data. For studies that reported effect estimates using different measures (e.g., risk ratios for mortality), risk ratios were not converted into odds ratios; instead, crude odds ratios for survival were derived directly from available data to ensure consistency in both outcome definition and effect measure across studies. Pooled odds ratios with 95% confidence intervals were estimated using a random‐effects model (inverse variance method). Heterogeneity was assessed using the I² statistic and Cochran's Q test. Sensitivity analyses were performed by excluding individual studies. Publication bias was evaluated by visual inspection of funnel plots. Analyses were conducted using Review Manager (version 5.3) and EZR. Ethical approval was not required because this study was based on previously published data.

3. Results

3.1. Study Selection

A total of 2230 citations were identified, of which 1316 remained after deduplication. Following title and abstract screening, 26 full‐text articles were assessed, and 18 were excluded for predefined reasons. Finally, eight studies were included in the meta‐analysis (Figure 1) [3, 11, 12, 13, 14, 15, 16, 17].

FIGURE 1.

PRISMA chart.

3.2. Study Characteristics

The eight included studies were observational cohort studies published between 2004 and 2022 and conducted in Japan, Italy, Denmark, France, the Czech Republic, and the United States, comprising 4862 patients with OHCA (Table 1). Six studies involved patients treated with ECPR [3, 11, 12, 13, 14, 15]. Most cohorts were multicenter registry‐based studies.

TABLE 1.

Characteristics of included studies.

First author, Publication year	Study designs	Study period/Region	Patient number	Timing of signs of life	Assessed by	Types of signs of life			Witnessed arrest, %	Bystander CPR, %	Shockable rhythm, %	CA‐Cause of Cardiac, %
						Gasping	PLR	CPRIC
Aoki et al. [11], 2024	MC, cohort	2019–2021/Japan	332	Prior to admission	EMS	Yes	No	No	79.2	47.8	53.9	87.6
Scquizzato et al. [12], 2024	SC, cohort	2013–2022/Italy	307	Before or during CPR	Unknown	Details not provided			95.0	66.0	66.0	79.2
Bunya et al. [13], 2024	MC, cohort	2013–2019/Japan	1395	At hospital arrival	Hospital	Yes	Yes	Yes	76.1	53.4	73.0	86.6
Nara et al. [14], 2023	MC, cohort	2008–2011/Japan	352	During EMS transport and/or on hospital arrival	EMS and Hospital	Yes	Insufficient	No	75.0	43.2	100.0	100.0
Mørk et al. [15], 2021	MC, cohort	2011–2020/Denmark	259	During CPR	Pre‐hospital EMS logistic systems	Details not provided			86.0	95.0	67.0	73.0
Debaty et al. [3], 2021	MC, cohort	2009–2017/France	434	Before or throughout CPR	Pre hospital care providers	Yes	Yes	Yes	93.0	86.0	71.0	86.0
Knor et al. [16], 2018	EMS registry data	2013–2014/Czech Republic	565	EMS arrival	EMS	Yes	No	No	76.8	70.4	27.4	100.0
Bobrow et al. [17], 2008	EMS registry data	2004–2008/USA	1218	Found by bystanders or, on or after EMS arrival at the scene	Emergency medical dispatchers and EMS	Yes	No	No	100.0	39.5	43.3	100.0

Note: MC: Multicenter, SC: Single‐center, EMS: Emergency medical services, PLR: Pupillary light reflex, CPRIC: Cardiopulmonary resuscitation‐induced consciousness, CA: Cardiac arrest.

Definitions, timing, and assessors of “signs of life” varied across studies (Table 1). Signs of life included gasping, PLR, and CPRIC, assessed by emergency medical services (EMS) personnel, hospital staff, or both. Detailed stratified outcome data according to timing or assessor were generally unavailable, precluding subgroup analyses.

3.3. Quality Assessment

All studies were rated as having good methodological quality, with NOS scores ranging from 7 to 9 (Table 2). However, residual confounding, time‐dependent bias (e.g., duration of resuscitation), and selection bias related to ECPR eligibility may not be fully captured by the Newcastle–Ottawa Scale.

TABLE 2.

Newcastle‐Ottawa scale assessment for the included studies.

First author, year	Selection				Comparability	Outcome			Total score
	Representativeness of the cohort	Selection of the non‐exposed cohort	Ascertainment of exposure	Outcome of interest not present at the start of the study	Of cohort based on design or analysis (max: ++)	Assessment of outcome	Follow‐up, adequate	Adequacy of follow ‐up of cohorts
Aoki et al., 2024 [11]	−	+	+	+	++	+	+	+	8
Scquizzato et al., 2024 [12]	−	+	+	+	+	+	+	+	7
Bunya et al., 2024 [13]	−	+	+	+	++	+	+	+	8
Nara et al., 2023 [14]	−	+	+	+	++	+	+	+	8
Mørk et al., 2021 [15]	−	+	+	+	+	+	+	+	7
Debaty et al., 2021 [3]	−	+	+	+	++	+	+	+	8
Knor et al., 2018 [16]	+	+	+	+	+	+	+	+	8
Bobrow et al., 2008 [17]	+	+	+	+	++	+	+	+	9

Note: For the Newcastle–Ottawa Scale assessment, studies were downgraded primarily in the “Representativeness” and “Comparability” domains. ECPR‐focused studies were judged less representative of the general OHCA population because they included only a highly selected subgroup eligible for extracorporeal support. In the Comparability domain, scoring was based specifically on whether the association between signs of life and clinical outcomes was analyzed using multivariable adjustment. In particular, studies were considered to have higher comparability if they adjusted for key prognostic factors such as age, initial rhythm, witnessed arrest, and resuscitation characteristics. Studies evaluating this association only with univariate analyses were assigned one point (“+”), whereas those performing multivariable analyses adjusting for key prognostic confounders were assigned two points (“++”). These criteria account for the variation in NOS ratings across the included studies.

3.4. Outcomes

3.4.1. Outcomes Within 30 Days

Signs of life were associated with improved survival within 30 days (pooled OR: 3.58, 95% CI: 2.93–4.36; I² = 0%) (Figure 2) [11, 12, 13, 15, 17]. A sensitivity analysis excluding the study with different effect measures [15] showed consistent results (Supplementary Figure S1). Favorable neurological outcomes within 30 days were also more frequent in patients with signs of life (pooled OR: 5.62, 95% CI: 4.09–7.71) [3, 11, 13], with no heterogeneity observed.

FIGURE 2.

Association of signs of life and outcomes within 30 days. Figure 2A shows the forest plot for survival within 30 days, while Figure 2B shows the forest plot for favorable neurological outcomes within 30 days.

3.4.2. Longer‐Term Outcomes

Signs of life were associated with improved longer‐term survival (pooled OR: 3.50, 95% CI: 2.43–5.03; I² = 0%) (Figure 3) [14, 16]. For longer‐term neurological outcomes, only one study was available [14], which showed an association between signs of life and favorable outcomes (OR: 7.01, 95% CI: 2.10–23.40). This result should be interpreted with caution due to the limited evidence.

FIGURE 3.

Association of signs of life and longer‐term outcomes. Figure 3A shows the forest plot for longer‐term survival, while Figure 3B shows the forest plot for longer‐term favorable neurological outcomes.

3.4.3. Subgroup Analyses

In studies without ECPR, consistent associations between signs of life and survival were observed (Figure 4) [16, 17]. Four ECPR cohorts reporting outcomes stratified by initial rhythm consistently demonstrated better outcomes among patients with non‐shockable rhythms who exhibited signs of life, although pooled estimates were not feasible [3, 11, 13, 15].

FIGURE 4.

Association of “signs of life” and outcomes without ECPR studies.

3.4.4. Publication Bias

Visual inspection of the funnel plot did not suggest substantial publication bias; however, publication bias could not be reliably assessed due to the small number of included studies (Figure 5).

FIGURE 5.

Funnel plot to assess publication bias.

4. Discussion

This systematic review and meta‐analysis demonstrated that the presence of “signs of life” during CA or ongoing CPR was associated with improved survival and favorable neurological outcomes in patients with OHCA. This association was consistently observed both within 30 days and in the longer term. Although the association was consistent, the results should be interpreted with caution, given the observational design and potential sources of bias.

Our findings are consistent with previous systematic reviews focusing on gasping, which reported a strong association between gasping and improved outcomes in OHCA [6, 7]. Some studies included in these reviews were not captured in our analysis because we restricted eligibility to cardiac arrest of presumed medical origin [18, 19, 20, 21, 22, 23, 24]. Whereas earlier reviews focused solely on gasping, our analysis integrated multiple forms of “signs of life,” including PLR and CPRIC. This broader approach suggests that their prognostic value is not limited to respiratory activity but may reflect preserved neurological responsiveness across several domains, offering more direct insight into cerebral viability during resuscitation.

A clinical scenario may further clarify this point. In severe ischemic heart disease, such as myocardial infarction involving the left main coronary artery, chest compressions may generate sufficient systemic and cerebral perfusion, while coronary perfusion remains inadequate, making return of spontaneous circulation unattainable. In this setting, the presence of signs of life indicates preserved cerebral responsiveness to resuscitation. Initiation of ECPR followed by definitive cardiac intervention may therefore lead to favorable outcomes. Thus, recognition of signs of life during CPR implies partial preservation of brain function and may reflect preserved physiological responsiveness during resuscitation. These findings were associated with improved outcomes; however, this association may be influenced by treatment selection bias, and causality cannot be inferred. In particular, in ECPR settings, the presence of signs of life may have influenced treatment decisions, leading to confounding by indication. In selected settings, signs of life may also serve as one of several factors considered when evaluating candidacy for advanced resuscitation strategies such as ECPR [3, 4, 11, 12, 13, 14, 15, 25].

Although different manifestations of “signs of life” may not carry identical prognostic weight, most included studies did not report outcomes separately for individual signs. Gasping was assessed in most studies, whereas PLR and CPRIC were each evaluated in only two reports, precluding formal comparisons. Therefore, these manifestations were treated as a unified concept in this analysis; however, the observed associations may be largely driven by gasping. In addition, different types of signs of life may reflect distinct physiological states and may be associated with different outcomes. Future prospective studies are needed to clarify these differences. Similarly, although four ECPR‐related studies reported outcomes stratified by initial rhythm, heterogeneity in outcome definitions and effect measures prevented a rhythm‐specific meta‐analysis [3, 11, 13, 15]. Nonetheless, all four studies consistently suggested that patients with non‐shockable rhythms who exhibited signs of life had better outcomes than those without. Signs of life may also inform real‐time resuscitation decisions. Prehospital recognition by EMS personnel may prompt early preparation for advanced interventions, including ECPR, while signs observed during ongoing resuscitation may support decisions to continue aggressive efforts or delay termination when neurological responsiveness appears preserved. Together, these findings suggest that signs of life may serve as integrative markers linking physiological viability, prognostic assessment, and clinical decision‐making during resuscitation.

Another important consideration is that signs of life are not systematically recorded in current OHCA reporting frameworks, such as the Utstein style [26]. Although Utstein templates capture key prognostic variables, they do not include the presence or absence of signs of life. Previous reports focusing on gasping have highlighted the need for systematic documentation, and this principle should be extended to all forms of signs of life [20]. While a simple binary variable indicating the presence or absence of any signs of life may be considered for standardized reporting, different manifestations of signs of life, such as gasping, PLR, and CPRIC, may not have equivalent prognostic significance and may be associated with different outcomes. Future prospective studies with standardized data collection are needed to clarify these differences. Across the included studies, the definition, timing, and personnel involved in assessing signs of life also varied substantially, with assessments performed by EMS personnel or hospital staff at different time points during resuscitation. Such variability likely contributed to heterogeneity in reported associations and limited comparability across studies. Standardizing definitions and assessment protocols for signs of life within reporting frameworks would facilitate consistent data collection and improve prognostic evaluation.

This review has several limitations. First, the literature search was restricted to selected databases (PubMed, MEDLINE, and Web of Science) and English‐language publications, and did not include Embase, which may have resulted in missed studies. Second, all included studies were observational, making residual confounding and selection bias unavoidable, particularly in ECPR settings where treatment decisions may have been influenced by the presence of signs of life. In addition, although the Newcastle–Ottawa Scale was used to assess study quality, this tool may not fully account for time‐dependent bias or confounding related to resuscitation duration and treatment selection. Third, a substantial proportion of the included studies involved ECPR‐treated populations, which may limit the generalizability of the findings. The association between signs of life and outcomes may differ between ECPR‐treated and non‐ECPR populations due to differences in patient selection and treatment pathways. Finally, the limited number of eligible studies precluded detailed subgroup or meta‐regression analyses. Accordingly, the overall certainty of evidence was considered low, reflecting the observational nature of the included studies and the presence of potential sources of bias. Future prospective studies with standardized assessment and reporting are needed to clarify the prognostic and operational role of signs of life in OHCA.

5. Conclusion

In this systematic review and meta‐analysis, the presence of signs of life during CA or CPR was consistently associated with improved survival and favorable neurological outcomes in patients with OHCA, both within 30 days and beyond 90 days. These findings suggest that signs of life may have prognostic value in patients with OHCA; however, their applicability may vary depending on clinical context and system‐level factors. Systematic documentation of signs of life may be considered in standardized reporting frameworks.

Funding

This work was supported by a Grant‐in‐Aid for Scientific Research (KAKENHI, Grant No. 23K08480) from the Japan Society for the Promotion of Science. The funding body had no role in the design of the study, data collection, analysis, interpretation of data, or writing the manuscript.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Sensitivity analysis for survival within 30 days, excluding the study with different effect measures.

Supporting Information: 1 PRISMA Checklist.

Supporting Information: 2 PROSPERO registration record for this systematic review (International Prospective Register of Systematic Reviews).

Supporting Information: 3 Full search strategy.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.