Background: Sepsis is a life-threatening organ dysfunction, which still represent one major cause of death worldwide. Atrial fibrillation (AF) often occur during sepsis, and imposes additional treatments, longer hospital stay, and an increased risk of mortality and morbidity.
Current knowledge on the relationship between sepsis and AF is limited. Beyond uncertainties on the pathophysiological mechanisms, the epidemiology and natural history of sepsis-related AF is still unknown, both in terms of prevalence of AF, and the magnitude of its effects on the prognosis of sepsis patients.
We aim to estimate the prevalence of AF, and the AF-associated risk of adverse events (all-cause mortality, thromboembolism and major bleeding) among patients with sepsis or septic shock, through a systematic review and meta-analysis of the literature. Furthermore, using advanced meta-regression techniques, we aim to identify factors associated with the onset of sepsis-related AF, and the AF-associated risk of outcomes.
Methods: This study will be performed according to the MOOSE and PRISMA Guidelines. Two investigators will independently perform the study selection process, bias assessment, and data extraction. We will report pooled estimates for 1) the prevalence of sepsis-related AF, and 2) the risk of outcomes according to the presence of AF. Finally, meta-regression analyses will be performed to identify factors associated with increased risk of AF and outcomes.
Discussion: The results of our analysis will provide a benchmark on the epidemiology of sepsis-related AF, with an effective and innovative use of the currently available data. Meta-regressions are expected to identify factors responsible for the onset of sepsis-related AF and worse progonsis, and these findings will provide a solid ground for the design of future studies and the development of specific preventive and treatment strategies, ultimately improving the care of sepsis patients.
We designed this research project to deliver highly significant scientific content, with the use of advanced and innovative approaches:
- Clinical Significance: To date, the epidemiology of sepsis-related AF is still unclear, both in terms of the prevalence of AF, and its impact on clinical outcomes and management of sepsis patients. Most of the current available data are based on small, retrospective cohorts, with reduced generalizability of their findings, and fragmentation of the evidence. A comprehensive and thorough systematic review and meta-analysis is a suitable approach to summarize and pool all the available data, and it will offer a proper and robust estimation of the actual prevalence of AF among sepsis patients, as well as the magnitude of the AF effects on adverse outcomes. This analysis will represent a landmark study in the field, providing a benchmark to design future research and guidance on the management of sepsis-related AF.
- Identification of characteristics associated with the onset of AF and with worse outcomes: With the use of advanced meta-regression techniques - which expands the power of conventional subgroup analyses - we will try to identify those clinical factors associated with higher risk of developing AF during sepsis, and with the risk of adverse outcomes (all-cause mortality, thromboembolism, bleeding) in patients with vs. without AF. These data are urgently needed to identify prevention and therapeutic targets, design goal-directed trials, and inform clinical practice on the factors that predispose to arrhythmia and AF-related adverse events.
- Innovative Methodology: The use of advanced and modern meta-regression techniques, along with robust meta-analytic methods, allows us to provide answers to several unsolved questions. The application of meta-regressions in this clinical scenario has several advantages: first, it allows for the efficient and pragmatic use of the currently available data, avoiding the excess costs of collecting new data, and the long times needed to perform a new study; second, the analysis of data coming from different cohorts will provide more robust estimates, due to the higher number of patients analyzed and the incorporation of studies coming from different geographical settings, thus taking into account potential regional-based differences that may bias the interpretation of the results. The application of meta-analysis and meta-regression to the sepsis-related AF scenario will provide modern and timely answers, with an optimal allocation of resources. This project represents one pratical application of epidemiological meta-analysis, already known to be valuable sources of information that would be otherwise difficult to obtain with cohort studies.
References:
Walkey AJ, McManus D. Am J Respir Crit Care Med. 2017 Jan 15;195(2):152-154
Moosavi V et al., European Heart Journal. 2020 41;2:ehaa946.0363
Harrell, F. E., Jr. Regression modeling strategies. Springer 2015
van Houwelingen HC et al. Stat Med. 2002 Feb 28;21(4):589-624.
Barendregt JJ et al. J Epidemiol Community Health 2013;67:974-978.