Infection induced-sepsis is prevalent in Intensive Care Units (ICU) and gives rise to numerous complications. Notably, it significantly impacts the heart. Sepsis-induced cardiac dysfunction extend beyond systolic and diastolic anomalies, encompassing cardiac rhythm disturbances [1]. While prior research has identified sepsis as a predisposing factor for arrhythmias, the emphasis has largely been on atrial arrhythmias [2]. Notably, malignant ventricular arrhythmias (VAs)—encompassing ventricular tachycardia (VT) and ventricular fibrillation (VF)—are of particular concern due to their potential to induce acute heart failure and result in sudden cardiac death (SCD) [3, 4]. Studies indicate that patients with sepsis are vulnerable to suffering from VAs [3]. These irregular heart rhythms can arise from factors such as electrolyte imbalances, reduced afterload, ventricular dysfunction, elevated catecholamines, and chronotropic dysregulation [5, 6].

While the relationship between sepsis and malignant VAs has been well investigated, the impact of sex differences is less explored. Previous studies have underscored sex-specific variations in the epidemiology, underlying mechanisms, clinical presentations, and outcomes of arrhythmias [7, 8, 9]. However, a significant portion of the evidence shaping clinical decisions in the field of cardiology is derived from studies with a conspicuous lack of female representation. Moreover, the sex-specific distribution of diseases, clinical risk elements, and outcomes of VAs in septic patients have been scarcely addressed. Gaining insights into how sex modulates the risk and prognosis of arrhythmias in these patients could lay the groundwork for personalized therapeutic approaches. With this in mind, we aimed to conduct a thorough assessment of sex disparities in the epidemiology, risk factors, and outcomes of VAs in septic patients.

In this large-scale cohort study, we observed sex disparities in VAs incidence and risk factors. With age as the metric, men had a higher VAs risk, particularly in older age groups. The multivariate model showed a stronger link between NICM, CHF, and VAs in men. Furthermore, the occurrence of VAs during hospitalization was associated with a nearly 2-fold increased risk of in-hospital mortality in septic patients, and this risk was consistent across both sexes.

Despite the underrepresentation of female patients in VAs-focused randomized controlled trials, existing research underscores distinct lifetime risks of VAs and SCD between the sexes [7]. This underrepresentation in clinical trials may be linked to the lower incidence of coronary artery disease and the frequency of VAs in women compared to men [7]. In our study, we found that 4.2% of hospitalized sepsis patients experienced VAs. Notably, men had a significantly higher incidence of VAs than women. Furthermore, being male was found to be a strong predictor of increased VAs risk in sepsis. Additionally, Santangeli et al. [17] reported that women with heart failure (HF) exhibit a lower rate of receiving appropriate implantable cardioverter-defibrillator (ICD) shocks in comparison to men. The underlying mechanisms that lead to VAs in sepsis remain incompletely understood. Several factors, such as electrolyte disturbance, inflammation, oxidative stress, cardiomyocyte apoptosis, exotoxins, endotoxins and ischemic heart disease are believed to play pivotal roles [18, 19, 20, 21]. The sex disparities observed might be attributed to the effects of sex hormones on Ca${}^{2+}$ currents [22, 23].

Our findings reveal that for sepsis patients, the cumulative incidence of VAs remains particularly low up until the age of 50. After this age, there’s a marked increase in incidence for men, whereas for women, this increase is observed after 60 years. Interestingly, we observed a decade-long delay in the onset of VAs for women compared to men, a pattern consistent with prior studies [14]. This decade-long difference in VAs onset between sexes might be attributed to protective hormonal factors in women that diminish post-menopause. As patients approach the age of 95, the incidence rates for both sexes tend to stabilize. Previous studies have established that increasing age is a potent predictor for the onset of VAs, a relationship that was also evident in our current research [5, 24]. Furthermore, the role of CHF as a risk factor for VAs has been previously established, which aligns with our study results [14, 25]. Notably, we observed that CHF has a more pronounced impact on men than on women. Potential mechanisms underlying the occurrence of VAs in CHF patients encompass structural and mechanical alterations in the ventricles, ventricular metabolic abnormalities, electrophysiological changes, neurohormonal imbalances, and the use of vasoactive agents [26, 27]. Moreover, the relationship between CHF and VAs is bidirectional, potentially creating a vicious cycle that exacerbates CHF progression and further heightens susceptibility to VAs [28]. Future prospective studies should further validate these sex differences and elucidate their underlying mechanisms. Risk factors such as NICM, which includes hypertrophic cardiomyopathy (HCM) and NICM, play a crucial role in evaluating patient susceptibility to for VAs and SCD [29, 30]. Our study clearly showcased sex disparities. However, contrasting findings were reported in the Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy (MADIT-CRT) [31]. The authors indicated that there was no statistically significant difference in the incidence of VAs and SCD between males and females within NICM (p = 0.063) [31]. Interestingly, the cumulative incidence of VAs was lower in females compared to males, a pattern consistent with our study [31]. It’s noteworthy to mention that in this research, female participants constituted only 35% of the total cohort. It’s important to highlight that other investigations into sex differences in VAs incidence are constrained by a limited number of female participants.

Additionally, among sepsis patients who developed VAs, the impact of AF might be more pronounced in females compared to males. There is growing evidence suggesting a mechanistic connection between AF and ventricular tachyarrhythmias [7, 32]. This connection may be attributed to reduced ventricular refractoriness and the occurrence of pro-arrhythmic short-long-short sequences preceding the onset of ventricular tachyarrhythmias in the presence of AF, compared to when the heart is in sinus rhythm [32]. The sex disparity might be related to the fact that female patients often do not receive timely treatment for AF [7]. Furthermore, while our study identified sex differences in the impact of pneumonia on the occurrence of VAs, pneumonia does not appear to be a critical influencing factor for VAs development. A prior study regarding COVID-19 indicated that cardiac arrest and arrhythmias might result from systemic illness rather than being solely a direct consequence of COVID-19 infection [33]. The influence of LOS-ICU and CKD on VAs risk seems limited. For these factors, our study did not observe a significant sex effect on the occurrence of VAs during hospitalization in sepsis patients. However, our study did not incorporate CKD staging, creatinine clearance, or renal replacement therapy in the analysis, so the impact of CKD on VAs should be interpreted with caution.

At present, there remains a debate regarding sex differences in the mortality rate associated with VAs [34]. Our study indicates that the occurrence of VAs in sepsis is correlated with an elevated in-hospital mortality rate and a heightened one-year mortality rate, yet no significant sex disparity was observed. Variations in cardiac regulation, cellular electrophysiology, and the influence of sex hormones might explain the sex-based differences in arrhythmia [7, 34]. Our study unveils the intricate interplay of various risk factors in determining the likelihood of VAs occurrence in sepsis patients. The observed sex disparities in VAs incidence and associated risks underscore the importance of adopting a sex-specific approach in clinical assessments and interventions. Future research should place greater emphasis on potential sex differences within certain cardiovascular diseases.

Our investigation delineates the nuanced interplay between sepsis and VAs, underscoring salient sex-based disparities in both incidence and predisposing factors. Our data reveals an augmented predisposition to VAs in male patients, with determinants such as CHF and NICM exerting a more pronounced influence in this demographic. The etiological underpinnings of these observations are intricate, spanning cellular electrophysiology, cardiac autoregulation, and the modulatory effects of sex hormones. These insights advocate for a sex-centric paradigm in clinical evaluations and therapeutic interventions. Emphasizing this, it becomes crucial to advocate for sophisticated, sex-tailored therapeutic strategies in addressing these challenges.

While this cohort study benefits from a substantial population size, several limitations warrant mention. First, the study’s retrospective design necessitates the need for future prospective investigations. Second, the absence of baseline electrocardiograms in this cohort may have resulted in an underestimation of VA incidence. Third, the data in this article originates from a substantial intensive care medicine cohort, which may limit the generalizability of our findings to other patient populations. Additionally, we did not account for iatrogenic VAs in our analysis despite their relatively frequent occurrence in ICU settings. Given the limitations of our data source, we are unable to account for iatrogenic VAs in our analysis. Studying iatrogenic VAs and their relationship with sex among ICU patients is indeed a valuable topic that merits further research. Lastly, the method used for imputing missing may have influenced our results. In light of these limitations, the conclusions drawn from this study should be approached with prudence.

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Study conception and design: LL, YY and XP. Acquisition of data: LL and XP. Analysis and interpretation of data: LKZ, YLX, ZHZ, ZXZ and ZH. Writing, review, and/or revision of the manuscript: LL, YY, ZXZ, LKZ, ZHZ, YLX, ZH and XP. Study supervision: YY. 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.

Given that our research involved a third party, anonymized, and publicly accessible database with prior institutional review board (IRB) consent, there was no need for additional IRB approval from our side and no additional patient’s informed consent is required.

Not applicable.

This research was funded by Medical and Health Technology Innovation Project of Chinese Academy of Medical Sciences, grant number 2021-CXGC09-1.

The authors declare no conflict of interest.