Search for a command to run...
Protein C is a key regulator of endogenous anticoagulant pathways through its effects on thrombin generation and endothelial homeostasis. Reduced protein C activity is associated with a hypercoagulable state and enhanced thrombotic risk [1-3]. Atrial fibrillation (AF), beyond an electrical arrhythmia, is increasingly recognized as a manifestation of atrial cardiomyopathy characterized by structural remodeling, endothelial dysfunction, impaired blood flow, and activation of prothrombotic pathways [4-8]. While abnormalities in downstream coagulation markers such as D-dimer have been described in AF [9], the contribution of endogenous anticoagulant pathways to AF-related thrombogenic phenotypes remains incompletely characterized in clinical populations. We therefore investigated the association between protein C activity and AF in patients with acute ischemic stroke (AIS), an enriched clinical setting in which AF-related thrombogenic mechanisms can be systematically evaluated [10]. We analyzed a prospective registry of consecutive patients admitted with AIS between 2016 and 2021 in whom serum protein C activity was measured within 24 h of admission. Patients with unclear AF-related stroke mechanisms were excluded, yielding a final analytic cohort of 1083 patients with complete clinical and laboratory data (Figure S1). AF was defined as an episode lasting ≥ 30 s, documented by medical history or detected during in-hospital cardiac monitoring, including 12-lead electrocardiography and continuous telemetry, in accordance with contemporary guideline recommendations [11], and classified as either previously known or newly detected during hospitalization. Stroke subtype was classified according to the Trial of Org 10 172 in Acute Stroke Treatment (TOAST) criteria [12]. The primary outcome was the presence of AF. Exploratory analyses evaluated the association between protein C activity and cardioembolic stroke among patients with AF. Baseline demographic characteristics, vascular risk factors, stroke severity assessed by the National Institutes of Health Stroke Scale (NIHSS) [13], laboratory parameters, and pre-stroke antithrombotic therapy were collected using standardized protocols. Serum protein C activity was measured using a chromogenic assay, with a laboratory reference range of 65–135 IU/dL. Protein C activity was modeled as a continuous variable, with effect estimates reported per 10 IU/dL decrease for interpretability. Associations with AF were evaluated using multivariable logistic regression, adjusting for age, sex, vascular risk factors, admission stroke severity, and pre-stroke antithrombotic therapy, including oral anticoagulants. Potential non-linear relationships were assessed using restricted cubic spline models. Given the vitamin K–dependent nature of protein C [14], pre-stroke oral anticoagulant use was explicitly included as a covariate, and sensitivity analyses excluding patients receiving vitamin K antagonists (VKA) or direct oral anticoagulants (DOAC) were performed to mitigate potential medication-related effects on protein C activity. Detailed multivariable models and pre-stroke antithrombotic patterns are provided in Tables S1 and S2. Additional methodological details are provided in the Supplementary Methods. Baseline characteristics according to AF status are summarized in Table 1. Among 1083 patients included in the final analytic cohort, 217 (20.04%) had AF. Patients with AF were older and had more severe neurological deficits at presentation. Protein C activity was significantly lower in patients with AF compared with those without AF, with values clustering closer to the lower bound of the laboratory reference range (Figure 1). In unadjusted analyses, lower protein C activity was strongly associated with the presence of AF. This association remained statistically significant after sequential adjustment for demographic characteristics, vascular risk factors, pre-stroke antithrombotic therapy, and admission stroke severity. In the fully adjusted model, each 10 IU/dL decrease in protein C activity was associated with approximately 19% higher odds of AF (adjusted odds ratio 1.19, 95% confidence interval 1.05–1.36). Restricted cubic spline analyses demonstrated a graded inverse association between protein C activity and AF across its continuous range, without evidence of a clear threshold effect (Figure 2). Sensitivity analyses excluding patients receiving oral anticoagulants yielded directionally consistent results with similar effect estimates (Table S3), supporting the robustness of the observed association. Among patients with AF, protein C activity did not significantly differ between those with previously known AF and those newly diagnosed during hospitalization (88.6 ± 24.1 vs. 97.9 ± 18.3; p = 0.18; Table S4). Protein C functions within the endogenous anticoagulant system to limit thrombin generation and preserve endothelial stability [1-3]. Reduced protein C activity permits sustained thrombin signaling, promotes endothelial activation, and enhances platelet reactivity—features compatible with a prothrombotic atrial environment [15-17]. Experimental data further suggest that excessive thrombin activity may engage protease-activated receptor pathways, triggering inflammatory responses and fibrotic remodeling that contribute to atrial electrical and structural vulnerability [15-17]. This interplay between coagulation, inflammation, and fibrosis provides a conceptual framework linking impaired endogenous anticoagulation to progressive atrial remodeling. The graded inverse relationship observed in our spline analyses, without a discrete threshold, aligns with the concept of atrial cardiomyopathy as a continuum of atrial vulnerability in which structural, functional, and hemostatic abnormalities evolve progressively rather than as a binary transition [5, 8, 18]. Beyond structural remodeling, atrial cardiomyopathy encompasses endothelial dysfunction within the left atrium and left atrial appendage, impaired local blood flow, and dysregulation of hemostatic balance [5-8]. In this context, reduced protein C activity may reflect not only systemic anticoagulant impairment but also a localized atrial prothrombotic milieu, consistent with Virchow's triad operating at the atrial level [6]. Such an imbalance between procoagulant and endogenous anticoagulant forces could facilitate microthrombus formation and perpetuate atrial inflammation and fibrosis, thereby reinforcing arrhythmogenic substrate [15-17]. From a clinical perspective, circulating protein C activity may therefore serve as an integrative biomarker capturing both hemostatic and atrial disease burden, potentially aiding in the identification of patients with heightened atrial vulnerability, including those with subclinical or paroxysmal AF. Such biomarker-based risk stratification may complement existing rhythm-based approaches by providing a biological context to atrial vulnerability, particularly in patients without overt AF. Our findings extend prior work that has largely focused on downstream coagulation markers [9] by highlighting endogenous anticoagulant capacity as a potential component of AF-related thrombogenic phenotypes. Notably, the association between protein C activity and AF persisted after adjustment for stroke severity, suggesting that the observed relationship is not solely explained by greater neurological injury among patients with AF. Moreover, consistent results after excluding patients receiving oral anticoagulants mitigate concerns regarding medication-related suppression of protein C activity [14, 19]. Nevertheless, reverse causality cannot be fully excluded, as AF-related hemodynamic disturbance or acute systemic inflammation during ischemic stroke may also influence circulating protein C levels [20]. Protein C activity may also be influenced by acute-phase responses during ischemic stroke, which could partially contribute to observed associations despite adjustment for stroke severity. In exploratory analyses restricted to patients with AF, protein C activity was not independently associated with cardioembolic stroke after multivariable adjustment (Table S5). Although these findings do not establish a direct link between protein C activity and embolic subtype, they are compatible with a model in which impaired endogenous anticoagulation reflects an underlying thrombogenic atrial substrate rather than the mere presence of overt arrhythmia [7, 8]. These findings should be interpreted cautiously and warrant confirmation in larger cohorts, given limited statistical power within AF subgroups and the multifactorial nature of embolic risk. Several limitations warrant consideration. This was an observational, single-center study, and residual confounding cannot be fully excluded despite comprehensive multivariable adjustment. Protein C activity was measured at a single time point during the acute phase of stroke, precluding assessment of longitudinal variability or baseline levels. Although adjustment for stroke severity may have partially mitigated acute-phase influences, residual effects cannot be excluded. AF ascertainment relied on documented medical history and in-hospital cardiac monitoring; therefore, paroxysmal or subclinical AF may have been missed. Conditioning on an ischemic stroke cohort may also introduce selection bias, potentially limiting generalizability to broader populations. Finally, the temporal relationship between reduced protein C activity and AF cannot be established in this cross-sectional analysis, and reverse causality remains possible. Prospective studies incorporating serial biomarker measurements and extended rhythm monitoring are warranted to clarify temporality and causality. In conclusion, reduced protein C activity was independently associated with the presence of AF in patients with AIS in a graded, threshold-free manner. These findings support impaired endogenous anticoagulation as a component of an AF-related thrombogenic atrial phenotype and underscore the relevance of the coagulation–arrhythmia interface in atrial disease biology. Circulating protein C activity may represent an integrative biomarker reflecting both hemostatic imbalance and atrial vulnerability, with potential implications for identifying patients at risk of occult or progressive atrial disease. Longitudinal studies incorporating serial biomarker assessment, extended rhythm monitoring, and complementary mechanistic approaches are warranted to clarify temporality, causality, and the clinical utility of endogenous anticoagulant markers in atrial cardiomyopathy. We thank our investigators of CRCS from (Prof. Dae-IL Chang, Prof. Joung-Ho Rha, Prof. Keun-Sik Hong, Prof. Hee-Joon Bae, Prof. Young-Seok Lee, Prof. Ju-Hun Lee, Prof. Sung Il Sohn, Prof. Jong-Moo Park, Prof. Soo Joo Lee, Prof. Dong-Eog Kim, Prof. Jae-Kwan Cha, Prof. Eung-Gyu Kim, Prof. Kyung Bok Lee, Prof. Young Bae Lee, Prof. Tai Hwan Park, Prof. Jun Lee, Prof. Man-Seok Park, Prof. Jay Chol Choi, Prof. Jun Hong Lee, Prof. Chulho Kim, Prof. Dong-Ick Shin, Prof. Hyun Young Kim, Prof. Jee-Hyun Kwon, Prof. Hye-Yeon Choi, Prof. Hahn Young Kim, Prof. Kyung Yoon Eah, Prof. Sang Won Han, Prof. Hyung-Geun Oh, Prof. Young-Jae Kim, Prof. Byoung-Soo Shin, Prof. Chang Hun Kim, and Prof. Chi Kyung Kim) who provided data that greatly assisted the research, although they may not agree with all of the interpretations/conclusions of this paper. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Science and ICT (Information and Communication Technology) (NRF-2022R1F1A1074643) and was supported by Soonchunhyang University. The study protocol was approved by the institutional review board of Kangdong Sacred Heart Hospital (IRB no. 2019-05-004). Informed consent was obtained from all participants or, when necessary, from their legally authorized representatives, such as next of kin. The authors declare no conflicts of interest. The data underlying this article cannot be shared publicly due to ethical and privacy restrictions related to patient confidentiality. De-identified data may be made available from the corresponding author upon reasonable request and with approval from the institutional review board. Figure S1: Flow diagram of patient selection. Table S1: Sequential multivariable logistic regression models for the association between protein C activity and atrial fibrillation. Table S2: Distribution of pre-stroke antithrombotic therapy according to atrial fibrillation status. Table S3: Sensitivity analysis of the association between protein C activity and atrial fibrillation after exclusion of patients receiving oral anticoagulants. Table S4: Protein C Activity According to Known vs Newly Diagnosed Atrial Fibrillation Among Patients With AF. Table S5: Exploratory analysis of protein C activity and cardioembolic stroke among patients with atrial fibrillation (n = 217). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.