Even after the return of spontaneous circulation (ROSC) in patients with out-of-hospital cardiac arrest (OHCA), comprehensive post-cardiac arrest care is critical to mitigating ongoing ischemia–reperfusion injury [1, 2, 3]. Maintaining arterial pressure (AP) above a certain threshold is essential to ensure adequate organ and tissue perfusion. In the post-ROSC phase, mean AP (MAP) plays a key role in supporting cerebral perfusion and has been associated with overall patient prognosis [4, 5]. Adequate MAP is crucial for preserving oxygen delivery to the brain and other vital organs, minimizing secondary ischemic damage, and facilitating neurological recovery.

To ensure sufficient tissue perfusion, preserve renal function (urine output), and stabilize metabolic processes (e.g., lactate clearance), current international guidelines recommend maintaining MAP at $\ge$65 mmHg [4, 5]. However, several studies have suggested that targeting a MAP above this threshold after ROSC may be associated with improved neurological outcomes [6, 7, 8, 9]. By contrast, a randomized clinical trial found no significant difference in clinical outcomes between patients managed with a MAP of 65–75 mmHg and those managed with 80–100 mmHg, using 80 mmHg as the comparative threshold [10]. Notably, that study excluded patients with unwitnessed arrests, non-shockable rhythms, and non-cardiac etiologies [10], limiting its generalizability to the broader OHCA population. As a result, it is likely that the study cohort primarily comprised patients with milder disease severity, potentially contributing to the reported favorable prognosis ($>$60%) [10]. Additionally, while previous studies have assessed average MAP during the observation period [6, 7, 8, 9], none have investigated the relationship between the duration of exposure to elevated MAP and neurological outcomes following ROSC.

The present study quantified the duration of MAP $>$80 mmHg within the first 48 hours after ROSC and assessed its association with neurological outcome groups in OHCA survivors. Furthermore, using the revised post-Cardiac Arrest Syndrome for Therapeutic hypothermia score (rCAST) as a measure of illness severity [11, 12], we evaluated whether this association varied according to the severity of injury following cardiac arrest.

This study demonstrated that the duration of MAP $>$80 mmHg during the 25–48 and 0–48 hour intervals, but not during the initial 0–24 hours, was associated with good neurological outcomes at 6 months following ROSC. Notably, among patients classified in the moderate severity category based on rCAST scores, the association between MAP $>$80 mmHg duration and good neurological outcomes was observed during both the 25–48 and 0–48 hour periods.

Previous research has reported that cerebral autoregulation is often impaired in patients with post-cardiac arrest, with a rightward shift in the lower limit of autoregulation. Specifically, patients resuscitated from cardiac arrest exhibited a significantly higher threshold for maintaining cerebral perfusion (114 mmHg) compared to healthy controls (76 mmHg) [17]. These findings suggest that a higher-than-normal MAP may be required to ensure adequate cerebral perfusion in this population. Additional studies using brain tissue regional oxygen saturation identified mean optimal MAP thresholds exceeding 76 mmHg [18] and 89 mmHg [19]. Kilgannon et al. [7] further reported a threshold effect, wherein MAP values above 70 mmHg were associated with favorable neurological outcomes. Similarly, an observational study identified the MAP range of 76–86 mmHg as optimal for maximizing survival in cardiac arrest survivors [8], while another study linked a MAP $>$90 mmHg during the first 6 hours after ROSC to both improved survival and better neurological outcomes [9].

In our study, the threshold of MAP $>$80 mmHg was selected based on prior randomized controlled trials, including those by Jakkula et al. [10] (MAP 65–75 mmHg vs. 80–100 mmHg) and Kjaergaard et al. [20] (MAP 63 mmHg vs. 77 mmHg), which commonly used approximately 80 mmHg as the higher target range for post-cardiac arrest care. Although this threshold does not align with the clinical definition of hypertension, it was applied in this study to represent relatively elevated perfusion pressures. The observed association between sustained MAP $>$80 mmHg and favorable neurological outcomes may be attributable to its role in preserving cerebral microvascular perfusion. The no-reflow phenomenon—wherein microvascular obstruction persists despite restoration of large-vessel flow—has been described in both cardiac and cerebral ischemia. Kloner et al. [21] proposed that microvascular impairment resulting from endothelial swelling, pericyte constriction, and interstitial edema may hinder capillary perfusion after ischemia–reperfusion injury. Maintaining optimal MAP levels may therefore help sustain perfusion across damaged microvascular beds, potentially reducing secondary brain injury and promoting neurological recovery. Our findings support this mechanism, demonstrating that longer durations of MAP $>$80 mmHg were associated with good neurological outcomes.

Our findings demonstrate an association between the duration of MAP $>$80 mmHg and favorable neurological outcomes specifically during the 25–48 hour interval, but not during the initial 0–24 hours. As illustrated in Fig. 2, both neurological outcome groups exhibited hemodynamic stabilization following the initial resuscitation phase; however, patients in the good outcome group consistently maintained higher MAP values throughout the 48-hour observation period. Notably, during the 25–48 hour interval, the MAP declined in the poor outcome group, leading to a widening divergence between the groups. This growing disparity in MAP may partly account for the differential associations observed between MAP duration and neurological outcomes. Previous studies have identified that hemodynamic instability is typically most pronounced around 6 hours post-ROSC, corresponding to the lowest observed cardiac index and MAP during the early resuscitation phase [22, 23]. While the precise timing of nadir MAP varies slightly across studies, both consistently reported minimal cardiac output at approximately 6 hours, followed by gradual improvement in MAP beyond 24 hours [22, 23]. Concurrently, this period is also characterized by peak hypoperfusion and secondary brain injury [24, 25]. Although moderate hypothermia during TTM may suppress cerebral metabolic demands and attenuate the inflammatory cascade, the rewarming phase can paradoxically increase metabolic requirements [26] and exacerbate inflammation [27], thereby aggravating secondary neurological injury. Thus, the stronger association between MAP and neurological outcomes during the 25–48 hour period may be attributable to evolving hemodynamic stress linked to these inflammatory processes. This timeframe represents a critical convergence of multiple concurrent pathophysiological mechanisms: the persistently elevated autoregulation threshold (requiring MAP substantially above traditional targets for adequate perfusion) [17], the transition from hypothermic cerebral metabolic suppression to progressively increasing metabolic demands [26], and potential inflammatory cascade activation during rewarming [27]. The temporal coincidence of impaired cerebral autoregulation with rising metabolic requirements and inflammatory burden during the 24–48 hour interval provides a pathophysiological basis for the observed association between MAP $>$80 mmHg maintenance and improved neurological outcomes during this specific period. Collectively, these findings underscore the potential importance of sustaining adequate perfusion and systemic recovery beyond the initial resuscitative window.

Our multivariable analysis revealed strong independent associations between male sex (OR: 2.427, 95% CI: 1.220–4.827) and cardiac etiology (OR: 2.495, 95% CI: 1.290–4.825) with good neurological outcomes. The association between male sex and good neurological outcomes reflects multifactorial biological and clinical mechanisms. Recent meta-analyses encompassing over 1.2 million patients demonstrate that males present with shockable rhythms more frequently (39.6% vs. 25.7% in females), with meta-regression analysis showing initial shockable rhythm as a significant predictor of survival outcomes (p 0.001) [28]. This rhythm disparity confers critical prognostic importance, as an initial shockable rhythm is regarded as one of the most important factors associated with survival after OHCA [29]. Furthermore, Bosson et al. [30] demonstrated that beyond favorable arrest characteristics, males receive more aggressive post-resuscitation interventions including coronary angiography (25% vs. 11%), percutaneous coronary intervention (PCI) (14% vs. 5%), and TTM (40% vs. 33%). Importantly, when these differences in treatment were accounted for in our analysis, the survival advantage for males disappeared, with no significant difference in neurological outcomes between males and females (OR: 0.9, 95% CI: 0.8–1.1) [30].

The association between cardiac etiology and good neurological outcomes in our study is consistent with established pathophysiological mechanisms. Unlike non-cardiac causes that involve prolonged hypoxia before arrest, cardiac arrests result from acute coronary occlusion with immediate circulatory cessation, limiting the extent of anoxic brain injury. The therapeutic reversibility of cardiac etiology is crucial—Dumas et al. [31] demonstrated that successful PCI in patients with significant coronary lesions improved survival from 31% to 51%. This finding is supported by nationwide data from Japan showing five-fold better neurological outcomes in cardiac versus non-cardiac origin OHCA (5.0% vs. 1.2%) among 547,153 patients [32]. These mechanisms—electrically reversible rhythms, shorter ischemic time, and treatable coronary pathology—collectively explain why cardiac etiology emerges as a powerful predictor of favorable neurological outcome, supporting aggressive interventional approaches in these patients.

Determining the optimal MAP target for post-cardiac arrest care remains challenging due to the complex interplay of ischemic brain injury, cerebral metabolic demands, and impaired autoregulation. This complexity may explain why recent randomized controlled trials have not demonstrated significant differences in neurological outcomes between patients managed with higher versus lower MAP thresholds [10, 20, 33]. Notably, these trials enrolled only patients with OHCA of presumed cardiac origin and excluded individuals with non-cardiac etiologies, who have worse prognoses [10, 20, 33]. By contrast, our study encompassed a broader patient population, with lower proportions of cardiac etiologies (54.9% vs. 100%) and shockable rhythms (34.4% vs. 100%, 66.7%, and 84.8%) compared with previous trials [10, 20, 33]. This wider inclusion likely contributed to the higher incidence of poor neurological outcomes observed in our cohort (71.8% vs. 35.0%, 63.2%, and 33.0%) [10, 20, 33]. These strict inclusion criteria in the randomized controlled trials (RCTs)—particularly the exclusion of non-cardiac etiologies and unwitnessed arrests—likely selected patients with relatively preserved cerebral autoregulation. Our broader inclusion criteria captured patients with more heterogeneous pathophysiology, where the therapeutic window for MAP optimization may be more relevant. This explains why the association between MAP duration and outcomes was most apparent in our moderate-severity subgroup, while previous RCTs with more homogeneous populations showed neutral results. Given the heterogeneity of our study population, including variations in cardiac arrest etiology and severity, we utilized the rCAST score to stratify injury severity and better evaluate the impact of MAP on neurological outcomes in OHCA survivors. The rCAST score has been extensively validated as a reliable prognostic tool in diverse OHCA populations. In a single-center U.S. validation study (n = 505), Kim et al. [34] demonstrated that the rCAST score achieved excellent discrimination for predicting poor neurological outcome (area under the curve [AUC]: 0.815; 95% CI: 0.763–0.867) and mortality (AUC: 0.799; 95% CI: 0.751–0.847), significantly outperforming the Pittsburgh Cardiac Arrest Category score for mortality prediction (p = 0.017). Similarly, in the multicenter study involving 658 patients across 24 intensive care units (ICUs), Lascarrou et al. [35] reported that rCAST maintained good performance (AUC: 0.82; 95% CI: 0.78–0.85), though it did not significantly outperform Utstein criteria (p = 0.16). Nevertheless, the authors emphasized the clinical utility of rCAST, noting its ease of calculation and rapid bedside determination as key advantages for routine implementation [35].

Our analysis showed that the association between the duration of MAP $>$80 mmHg and neurological outcomes was most evident during the 25–48 hour period in patients within the moderate severity group, compared with those in the low and high severity groups. These findings suggest that the effect of MAP on neurological outcomes may be modulated by the severity of illness in OHCA survivors. In our cohort, patients classified as having either low or high severity typically exhibited uniform neurological outcomes—either predominantly favorable or unfavorable. In contrast, the moderate severity group included a mixture of outcomes, indicating the presence of a “gray zone” in which the duration of MAP $>$80 mmHg may exert a meaningful influence on neurological prognosis. In the high-severity group, univariate analysis revealed a significant difference in the duration of MAP $>$80 mmHg between the good and poor prognosis groups (41 vs. 30 hours; p = 0.032), but not in multivariate analysis (p = 0.055). This discrepancy likely reflects the limited statistical power due to only eight patients achieving good outcomes in this subgroup, as well as the combined effects of severe neurological impairment that outweighed the benefits of maintaining MAP $>$80 mmHg. This difference in response by severity is consistent with the results of other post-cardiac arrest interventions. Prior observational studies have suggested that post-cardiac arrest injury severity, as measured by scoring systems or electroencephalography, is an important determinant of TTM effectiveness [11, 36, 37]. Notably, mild therapeutic hypothermia (33–34 °C) has been associated with improved neurological outcomes in moderately injured cardiac arrest survivors, whereas its benefits appear diminished in those with either minimal or severe injury [11, 37].

In patients with low severity, relatively intact cerebral autoregulation may render additional MAP elevation unnecessary. However, individuals with moderate severity are more likely to have impaired autoregulation and borderline cerebral perfusion, placing them within a physiological range where elevated MAP may meaningfully enhance oxygen delivery and mitigate secondary brain injury. Conversely, patients with high severity often have extensive, irreversible brain damage, limiting the potential benefit of elevated MAP. These observations underscore the value of tailoring post-resuscitation hemodynamic management based on individual injury severity. Identifying patients in the moderate injury category who may benefit from higher MAP targets could facilitate more personalized and effective post-cardiac arrest care.

This study had several limitations. First, as a retrospective observational study, it cannot establish causality between MAP duration and neurological outcomes. Second, although the use of a multicenter registry enhances generalizability, approximately 25% of patients were excluded due to missing data, potentially introducing selection bias. Third, MAP was recorded hourly rather than continuously. This intermittent sampling may have failed to capture transient hypotensive episodes between measurements, potentially contributing to secondary brain injury. While continuous monitoring would provide a more granular assessment of hemodynamic stability, our use of hourly data from a large, multicenter registry represents a pragmatic and methodological advance over many prior studies that relied on less frequent, averaged values. Fourth, our analysis did not account for other hemodynamic parameters that may influence cerebral perfusion and outcomes, a notable limitation of this study. Specific data on the dose and duration of vasopressors or inotropes, as well as on net fluid balance, were not included in our analysis. These interventions are significant potential confounders. For instance, higher vasopressor doses may indicate more severe post-cardiac arrest shock, an independent predictor of poor outcomes, while fluid management can affect both MAP and cerebral edema. Although our multivariable models adjusted for overall illness severity using SOFA and rCAST scores, these scores may not fully capture the influence of these specific pharmacologic interventions was not assessed in the current analysis, and the potential for residual confounding remains. Fifth, the primary outcome was assessed through telephone interviews with patients or proxies. This method may be less sensitive than direct examination for detecting subtle cognitive deficits. Proxy reporting is also susceptible to recall bias, potentially overestimating favorable outcomes. While this approach is pragmatic for multicenter studies, such misclassification could influence the observed associations. Finally, cerebral perfusion pressure was not measured, precluding direct assessment of its relationship with MAP.

In this study, duration of MAP $>$80 mmHg during the first 48 hours following ROSC was associated with good neurological outcomes. Further subgroup analysis indicated that this association was significant only during the 25–48 hour interval and was observed exclusively in patients with moderate injury severity, as defined by the rCAST score. No significant associations were found in the low or high severity groups.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further enquiries can be directed to the corresponding author.

DHL, SJR: Conceptualization, Data curation, Investigation, Writing - original draft, Writing - review & editing. BKL: Conceptualization, Formal analysis, Methodology, Project administration, Software, Writing - original draft, Writing - review & editing. YHJ, KWJ: Conceptualization, Data curation, Investigation, Writing - review & editing. HJB, HJK, JSO: Data curation, Resources, Validation, Writing - review & editing. ISC: Data curation, Resources Supervision, Writing - review & editing. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

This study was approved by each participating hospital institutional review board including the Chonnam National University Hospital Institutional Review Board (CNUH-2021-017). Written informed consent was obtained from all patients or their legally authorized representatives, in accordance with national regulations and the principles outlined in the Declaration of Helsinki.

We thank all Korean Hypothermia Network (KORHN) investigators.

This study was supported by a grant (BCRI24052) of Chonnam National University Hospital Biomedical Research Institute. And this research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: RS-2024-00335934).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/RCM42733.