Sepsis is characterized by excessive systemic inflammatory and dysfunctional infection responses, which results in possible organ dysfunction and shock ([[NoAuthor]]; Kuipers et al., 2014; Bashar et al., 2020). Although progress has been made in the diagnosis and treatment of sepsis, it remains a life-threatening issue with high morbidity and mortality worldwide (Vincent et al., 2014; Angus et al., 2015; Rudd et al., 2020), especially for those in intensive care units (Baykara et al., 2018). Sepsis is one of the leading causes of death among critically ill patients admitted to intensive care unit (ICU) (Gaborit et al., 2021). Clinicians have tried multiple measures, including anti-infection treatment, organ protection, and fluid resuscitation to improve the survival rate of sepsis patients (Burke et al., 2019; Zhong et al., 2020). However, there is still no specific treatment for sepsis patients.

In recent years, accumulated evidence has shown that ondansetron (OND) might have broader pharmacological effects, especially in the anti-inflammation aspect. For instance, researchers found that OND attenuated pancreatic injury in the cerulein-induced acute pancreatitis model (Tsukamoto et al., 2017), the severity of dextran sulfate sodium salt-induced colitis (Utsumi et al., 2016) and targeted peritoneal macrophages as an anti-inflammatory agent (Maehara et al., 2015). These studies suggested that OND might have potential in treating inflammatory diseases. Some indicators reflecting the level of systemic inflammation, including neutrophil to lymphocyte ratio (NLR) and platelet to lymphocyte ratio (PLR), were highly correlated with mortality in septic patients (Kaushik et al., 2018; Shen et al., 2019; Huang et al., 2020). Interestingly, researchers found that OND pre-treatment decreased the mortality of ICU patients on mechanical ventilation, which could be explained by regulation of NLR (Zhou et al., 2022).

It has been reported that the use of OND may be associated with a potential prolongation of the cardiac QT interval (Charbit et al., 2008). However, it is important to note that when ondansetron is administered orally at therapeutic doses, the likelihood of it causing a clinically significant prolongation of the QT interval is very low, and the risk of inducing arrhythmias is even lower (Freedman et al., 2014). Furthermore, the use of OND during pregnancy has not been found to be significantly linked to an increased risk of adverse fetal outcomes, including conditions like cardiac malformations, oral clefts, or congenital malformations overall (Pasternak et al., 2013; Huybrechts et al., 2020). Therefore, it is appealing for us to investigate whether OND medication during ICU stay plays a protective role in septic patients or not.

In the present study, we aimed to explore the effects of OND exposure during ICU stay on clinical outcomes of septic patients, including in-hospital, 28-day, 90-day mortality rates and length of stay in hospital and ICU. Our research might provide a new treatment option for patients with sepsis in ICU to forecast and improve their prognosis.

In the present study, we explored the possible association between OND use during ICU stay with sepsis and the clinical outcomes in critically ill patients. We found that the use of OND might relate to lower in-hospital mortality of sepsis patients and improved long-term prognosis, evident by lower 28-day and 90-day mortalities. Combining with multivariate logistic or Cox regression analyses and propensity score matching, the potential confounders were balanced for a more robust conclusion. Taken together, this study indicated that OND exposure could be a protective factor in critically ill patients with sepsis. Therefore, our study revealed the unrecognized and promising role of OND in sepsis patients, which might be a novel potential treatment strategy for these high-risk patients.

Sepsis or septic shock was defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection (Singer et al., 2016), characterized by an inadequate systemic immune response to an initial stimulus. There is no doubt that sepsis is a major health issue worldwide, and it is estimated that 5.3 million people die of this disease every year (Vincent et al., 2006; Kaukonen et al., 2014; Fleischmann et al., 2016). To make matters worse, accumulated evidence revealed that the number of sepsis patients is rising rapidly (Martin et al., 2003; Lagu et al., 2012; Shankar-Hari et al., 2017). Hence, it is essential for us to diagnose and treat sepsis patients at an early stage.

In current medical practice, the treatment of sepsis is still limited, which can be summarized into three aspects, including infection control, hemodynamic management and modulation of the host response (Vincent, 2022). Antibiotics were necessary for nearly every patient at an early stage even when no particular microorganism was isolated from blood (Sakr et al., 2018; Lakbar et al., 2022). Regarding hemodynamic management, fluid administration and vasopressin, such as noradrenaline or dobutamine, were also required in some situations for an initial mean arterial pressure (MAP) of 65 mmHg was recommended (Evans et al., 2021).Immunomodulatory therapy is gradually gaining traction in sepsis, because early and over-activation of the immune inflammatory response is a major cause of infectious death (Liu et al., 2022). Low-dose of GM-CSF could improve oxygenation index in septic patients with respiratory dysfunction but had no positive effect on 30-day mortality (Orozco et al., 2006). Similarly, a phase-II clinical study showed that granulocyte-macrophage colony-stimulating factor therapy decreased antibiotics treatment time and infectious complications but didn’t improve the in-hospital mortality rate of sepsis patients (Presneill et al., 2002). Combination therapy of ulinastatin and Tα1 reduce 28d and 90d mortality rates and organ damages in sepsis patients while the effect of the medication alone use was not satisfactory (Han et al., 2015; Liu et al., 2017). This evidence suggest that immune regulation might be a potential treatment option for sepsis patients.

OND, as a 5-HTR antagonist, had a binding ability to 5-HT₃R 5-HT_1BR, 5- HT_1CR, α-adrenergic receptor and μ-opioid receptor (Kovac, 2016). In recent years, studies found that 5-HT₃R antagonists might have pharmacological effects in many diseases, particularly their anti-inflammatory effect. For instance, tropisetron showed protective effect by mediating 5-HT₃ receptors (Motavallian et al., 2013) in inflammatory bowel disease and granisetron could ameliorate acetic acid-induced colitis (Fakhfouri et al., 2010). Local injection of tropisetron potently relieved inflammation and pain in arthritis, osteoarthritis and tendinopathies (Stratz and Müller, 2000; Stratz et al., 2002). The cytokine storm caused by excessive inflammatory activity is a key issue during the progression of sepsis (Karbian et al., 2020). Researchers have discovered that intervention with a 5-HT3 receptor antagonist can help control the overproduction of pro-inflammatory cytokines involved in the pathogenesis of severe sepsis or septic shock, and reduce serum levels of noradrenaline in sepsis models (Stratz et al., 2002; Setoguchi et al., 2011). Oxidative stress resulting from the release of reactive oxygen species (ROS) contributes to mitochondrial dysfunction, induces cell apoptosis, and worsens the prognosis of sepsis patients (Huet et al., 2011). Therefore, antioxidant therapy is considered a potential strategy to improve sepsis outcomes. An animal study reported that the use of OND could alleviate oxidative stress in rats with cystitis (Zirak et al., 2020). Similarly, OND was found to maintain the pro-oxidant/antioxidant balance in the brain by increasing GSH concentrations and inhibiting MDA in a mouse model (Gupta et al., 2014). Collectively, these findings suggest that OND might mitigate sepsis-related injuries through its anti-inflammatory and antioxidant properties, although the specific molecular mechanisms still require further investigation.

In current clinical practice, ondansetron (OND) is commonly used to prevent and treat nausea and vomiting in critical care settings. However, emerging evidence suggests that OND may also have a role in improving the prognosis of critically ill patients. This highlights the potential for OND to provide benefits beyond its established use for managing nausea and vomiting. As previously reported, OND exposure was associated with lower 90-day mortality for cardiac surgical patients and postoperative AKI (Xiong and Xiong, 2022). Meanwhile, Gray et al. also considered the use of OND is friendly with kidney and was associated with a significance decrease in 90-day mortality in AKI patients compared with other anti-emetics (Gray et al., 2022). They were consistent with our results and indicated that OND might be a potential clinical treatment for critically ill patients to improve their outcomes. Researchers have also discovered that the 5-HT3 receptor antagonist could reduce the mortality of septic mice (Gong et al., 2019). This suggests that drugs targeting the 5-HT3 receptor may have a positive impact on the survival of mice with sepsis. Additionally, the OND medication was associated with longer lengths of stay in ICU in this study, which might be due to the shorter survival time of non-OND user, thereby shortening their ICU LOS. Therefore, it is necessary to conduct randomized controlled trials in the future to confirm whether OND should be included as a routine medication for septic patients. Such trials will help determine the effectiveness and safety of using OND in the management of sepsis and provide more concrete evidence to guide clinical practice.

There are several limitations in the present study. First, this is a single-center retrospective study, although we have adopted various methods to exclude potential migrations, it may still exist. Secondly, we have not subdivided the dosage and initiated time of OND, as well as the duration of treatment time, although they were not specified, they may also act as a factor affecting the clinical outcomes of patients. Furthermore, it’s worth noting that the septic patients in this study were identified solely based on ICD-9 coding, without the use of any other clinical criteria, potentially limiting the generalizability of the study’s findings. Additionally, some clinical indices with more than 30% missing values, including neutrophil and lymphocyte counts, were excluded in this study. However, it cannot be denied that these excluded variables may have had an impact on the study’s outcomes. Despite the extensive adjustment for numerous potential confounding factors, it’s important to acknowledge that there may still be unaccounted variables that could have influenced the results. These variables might include factors such as the type of pathogen responsible for the patient’s infection, the specific type and dosage of antibiotics administered during the patient’s ICU stay, and more. As a result, further randomized controlled trials are necessary in the future to provide a more comprehensive understanding of this subject.

In this study, we found that OND exposure might be associated to lower in-hospital, 28-day, and 90-day mortality rates in critically ill patients with sepsis. This study indicated that OND might help to improve the prognosis of patients with sepsis, which required further studies.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

Ethical approval was not required for the studies involving humans because The data for this article came from the MIMIC-IV database, a public database that is available to the public in anonymous form, with our authors obtaining research consent. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements because The data for this article came from the MIMIC-IV database, a public database that is available to the public in anonymous form, with our authors obtaining research consent.

BY: Conceptualization, Data curation, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. YZ: Conceptualization, Data curation, Formal analysis, Investigation, Validation, Writing – original draft, Writing – review & editing. XZ: Investigation, Software, Validation, Visualization, Writing – original draft, Writing – review & editing. TL: Data curation, Formal analysis, Resources, Validation, Writing – review & editing. KN: Software, Validation, Writing – review & editing. ZW: Software, Validation, Writing – review & editing. XJ: Writing – review & editing, Validation, Investigation. XL: Supervision, Validation, Writing – review & editing. HQ: Conceptualization, Supervision, Validation, Writing – review & editing. CS: Writing – review & editing.