Clinical outcomes and risk factors for ICU mortality in patients with sepsis following open-heart surgery

Authors

Abstract

AimThis study aimed to identify factors associated with intensive care unit (ICU) mortality in septic patients with a history of open-heart surgery.
MethodsIn this retrospective observational study, adult patients admitted to a tertiary general ICU with a diagnosis of sepsis between 2020 and 2025 and a prior history of open-heart surgery were included. Demographic characteristics, comorbidities, clinical and laboratory parameters at ICU admission, and variables related to previous cardiac surgery were evaluated. Patients were categorized as survivors and non-survivors based on ICU mortality. Univariate and multivariate logistic regression analyses were performed to determine factors associated with mortality.
ResultsA total of 149 patients were included, of whom 89 (59.7%) died during ICU stay. Non-survivors were older and had higher SOFA scores. Congestive heart failure was more frequent among non-survivors. Additionally, admission lactate and creatinine levels were higher, whereas baseline glomerular filtration rate was lower in this group. No significant differences were observed between groups regarding type of surgery or time elapsed since surgery. In multivariate analysis, age, SOFA score, and congestive heart failure were independently associated with ICU mortality, while admission lactate level showed borderline significance.
ConclusionICU mortality in this population appears to be primarily driven by the severity of current illness rather than prior surgical characteristics. Acute physiological status and cardiovascular reserve play a central role in determining outcomes. Mortality in septic patients with a history of open-heart surgery is more closely related to organ dysfunction and comorbidities than to surgical factors. Emphasis on dynamic clinical parameters may improve prognostic assessment.

Keywords

Introduction

Sepsis is a major clinical syndrome characterized by organ dysfunction resulting from a dysregulated host response to infection.¹ It remains one of the leading causes of morbidity and mortality in intensive care units (ICUs) worldwide.^2,3 Clinical outcomes in patients with sepsis are influenced not only by the severity of infection but also by age, burden of comorbidities, and existing organ reserve.³
The cardiovascular system is among the most frequently affected organ systems during sepsis. Sepsis-induced circulatory dysfunction and myocardial impairment are strongly associated with poor prognosis, and these alterations are largely explained by inflammatory and mitochondrial mechanisms.^4,5 Inflammatory mediators, oxidative stress, and mitochondrial dysfunction contribute to impaired myocardial contractility and altered cellular energy metabolism, ultimately leading to sepsis-induced cardiomyopathy (Kakihana et al., 2016; Hollenberg & Singer, 2021).
In patients with a prior history of open-heart surgery, reduced cardiac reserve, coexisting heart failure, and multiple comorbidities may further influence the course of sepsis.^6,7 However, data regarding factors specifically associated with ICU mortality in this patient population remain limited, and existing studies report heterogeneous results in terms of patient populations, sepsis definitions, and evaluation methodologies.^6,8,9 Furthermore, recent evidence suggests that pre-existing cardiac conditions and active cardiac determinants—such as concomitant heart failure—may play a crucial role in clinical outcomes in septic patients.^10,11
The aim of this study was to evaluate clinical outcomes in septic patients with a history of open-heart surgery and to determine the association between clinical and laboratory variables at ICU admission and ICU mortality.

Materials and Methods

Study Design and SettingThis study was designed as a retrospective observational study conducted in the general intensive care unit (ICU) of a tertiary training and research hospital. Adult patients who were admitted to the ICU with a diagnosis of sepsis between January 2020 and December 2025 and had a history of open-heart surgery were included. Clinical, laboratory, and ICU follow-up data were retrospectively obtained from the hospital’s electronic medical record system.
Study Population, Inclusion and Exclusion CriteriaAdult patients admitted to the ICU with a diagnosis of sepsis between January 2020 and December 2025 and with a prior history of open-heart surgery were included. Sepsis diagnosis was limited to cases confirmed based on clinical and laboratory findings during ICU stay.
Patients in whom sepsis diagnosis could not be confirmed, those with missing clinical or laboratory data, or those with insufficient records in the hospital database were excluded. Additionally, patients admitted to the ICU due to trauma, major surgical complications, or non-sepsis-related causes, as well as those receiving immunosuppressive therapy or with hematological malignancies, were excluded. In cases of multiple ICU admissions for the same patient, only the first admission was considered for analysis.
Study Groups and OutcomesPatients meeting the inclusion and exclusion criteria were divided into two groups according to ICU mortality: patients who died during ICU stay were classified as non-survivors, while those discharged or transferred alive were classified as survivors.
The primary outcome was ICU mortality. Secondary outcomes included ICU-related clinical outcomes such as 24-hour lactate level, presence of septic shock, requirement for invasive mechanical ventilation, vasopressor use, renal replacement therapy, acute kidney injury, ICU readmission, and length of ICU stay.
Data Collection and VariablesData were retrospectively collected from the hospital’s electronic database. Variables included demographic characteristics, comorbidities, details of prior cardiac surgery, clinical and laboratory parameters at ICU admission, and ICU-related clinical variables.
Demographic data included age, sex, and body mass index. Comorbidity burden was assessed using the Charlson Comorbidity Index. Recorded comorbidities included diabetes mellitus, hypertension, chronic kidney disease, congestive heart failure, atrial fibrillation, history of stroke, chronic obstructive pulmonary disease, and malignancy.
Data related to previous cardiac surgery included type of surgery (coronary artery bypass grafting,[CABG] combined CABG and valve surgery, and isolated valve surgery) and time elapsed since surgery.
Clinical and laboratory parameters assessed at ICU admission included lactate and creatinine levels, baseline estimated glomerular filtration rate (eGFR), ejection fraction, and severity scores such as SOFA and APACHE II.
ICU-related clinical variables included 24-hour lactate level, presence of septic shock, need for invasive mechanical ventilation, vasopressor use, renal replacement therapy, acute kidney injury, and ICU readmission.
Ethical ApprovalThe study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of Bursa Yüksek Ihtisas Training and Research Hospital (March 25, 2026; decision no: 2024-TBEK 2026/03-11). The requirement for informed consent was waived due to the retrospective design and the use of anonymized patient data.
Statistical AnalysisThe distribution of continuous variables was assessed using histograms, Q–Q plots, and the Shapiro–Wilk test. Continuous variables were expressed as median (interquartile range), while categorical variables were presented as number (percentage).
For comparisons between groups, the Mann–Whitney U test was used for continuous variables, and the Pearson chi-square test or Fisher’s exact test was used for categorical variables, as appropriate.
To identify factors associated with ICU mortality, univariate and multivariate logistic regression analyses were performed. Variables considered clinically relevant and available at ICU admission were included in the multivariate model. Variables developing during ICU stay were excluded to reduce the risk of reverse causality and overfitting.
To preserve model parsimony and prevent overfitting, variable selection was restricted based on the number of events; therefore, some variables found significant in univariate analysis were not included in the multivariate model. Additionally, to avoid collinearity, highly correlated variables were not entered into the same model; specifically, SOFA and APACHE II scores, as well as creatinine and baseline eGFR, were not included simultaneously.
The final multivariate model included age, SOFA score, admission lactate level, congestive heart failure, and time elapsed since surgery. Model calibration was assessed using the Hosmer–Lemeshow test, and discrimination was evaluated using the area under the receiver operating characteristic curve (AUC). All statistical analyses were performed using a two-sided approach, and a p-value <0.05 was considered statistically significant.
Reporting GuidelinesThis study was reported in accordance with the STROBE guideline.

Results

The patient selection process is presented in Figure 1. Of the 635 patients admitted to the ICU with suspected sepsis or septic shock, 149 who met the inclusion and exclusion criteria constituted the final study cohort. Among these patients, 60 (40.3%) were discharged alive from the ICU, while 89 (59.7%) died during ICU stay .
Demographic, clinical, and laboratory characteristics according to survival status are presented in Table 1. Non-survivors were significantly older (p<0.001) and had higher SOFA and APACHE II scores, as well as higher Charlson Comorbidity Index values (all p≤0.002).
Regarding comorbidities, congestive heart failure (p=0.021) and malignancy (p=0.004) were more frequent in the non-survivor group, whereas no significant differences were observed for other comorbid conditions (all p>0.05).
In terms of laboratory parameters at admission, non-survivors had significantly higher lactate and creatinine levels and lower baseline glomerular filtration rates (all p≤0.008). However, there was no significant difference in ejection fraction between the groups (p=0.317).
No significant differences were found between the groups regarding prior cardiac surgery-related variables, including type of surgery and time elapsed since surgery (p=0.152 and p=0.567, respectively).
With respect to ICU course, the 24-hour lactate level was higher in non-survivors (p=0.002), and septic shock, invasive mechanical ventilation requirement, and vasopressor use were more frequent in this group (all p<0.001). In contrast, no significant differences were observed in renal replacement therapy or acute kidney injury (p=0.529 and p=0.102, respectively).
ICU readmission was more frequent in survivors (p=0.015), whereas ICU length of stay was longer in non-survivors (p=0.031) .
Logistic regression analyses were performed to evaluate factors associated with ICU mortality (Table 2). In univariate analysis, age (OR: 1.071, p=0.001), SOFA score (OR: 1.381, p<0.001), and congestive heart failure (OR: 2.537, p=0.005) were significantly associated with mortality. Additionally, admission lactate (OR: 1.708, p=0.004) and creatinine levels (OR: 1.532, p=0.007) were positively associated with mortality, whereas baseline glomerular filtration rate showed an inverse association (OR: 0.986, p=0.013).
In multivariate analysis, age (OR: 1.062, p=0.010), SOFA score (OR: 1.296, p=0.002), and congestive heart failure (OR: 3.321, p=0.004) remained independently associated with ICU mortality, while admission lactate level showed borderline significance (OR: 1.472, p=0.055). Time elapsed since cardiac surgery was not independently associated with mortality (p=0.690).
The multivariate model demonstrated acceptable calibration and discrimination (Hosmer–Lemeshow p=0.178, AUC=0.792) .

Discussion

In this study, factors associated with ICU mortality were evaluated in septic patients with a history of open-heart surgery. The main findings were that non-survivors were older, had higher SOFA and APACHE II scores reflecting disease severity, and presented with higher lactate and creatinine levels and lower baseline glomerular filtration rates at ICU admission. In multivariate analysis, age, SOFA score, and congestive heart failure were independently associated with ICU mortality, while admission lactate level showed borderline significance. In contrast, type of surgery and time elapsed since surgery were not significantly associated with mortality. Furthermore, the acceptable calibration and discrimination of the multivariate model support the analytical reliability of the identified independent predictors .
Sepsis remains a leading cause of mortality and multiple organ dysfunction in ICUs. In our study, non-survivors were older and had a higher comorbidity burden, as well as higher SOFA and APACHE II scores. This finding is expected in retrospective observational studies and reflects the tendency of mortality-related risk factors to cluster in more severely ill patients. Therefore, multivariate analysis provides a more accurate assessment of independent effects. The identification of age as an independent risk factor is consistent with the literature, suggesting that advanced age contributes to poor outcomes through reduced physiological reserve, altered immune response, and increased comorbidity burden.^3,12 However, findings from different patient populations indicate that the impact of age on mortality may vary depending on clinical context, and some studies emphasize that age alone should be interpreted alongside other clinical parameters.¹³
The higher prevalence of malignancy in the non-survivor group may be related to impaired immune response and increased clinical frailty. However, since hematological malignancies were excluded, the malignancy variable in this study predominantly reflects solid tumors and did not emerge as an independent predictor in multivariate analysis.
The identification of congestive heart failure as an independent risk factor highlights the importance of cardiovascular reserve in this specific cohort. Reduced cardiac reserve may limit the ability to adapt to systemic inflammatory response and hemodynamic instability during sepsis, thereby impairing organ perfusion. Sepsis-induced circulatory dysfunction and myocardial impairment are known to be associated with poor prognosis.^4,5,10 Additionally, previous studies suggest that pre-existing cardiac function and heart failure may influence both short- and long-term outcomes in septic patients.^11,14 Recent evidence further indicates that baseline cardiac status and cardiovascular determinants such as heart failure are closely linked to clinical outcomes, while the burden of organ dysfunction remains a key determinant of prognosis.^11,15 Notably, pre-existing cardiac dysfunction has been associated with increased need for organ support and higher mortality during sepsis, although different types of cardiac dysfunction may not always correlate with initial sepsis severity.¹¹ This suggests that not only structural abnormalities but also functional reserve and hemodynamic response capacity play a critical role in determining prognosis. The lack of a significant difference in ejection fraction between groups further supports the notion that static cardiac parameters measured at a single time point may be insufficient to reflect clinical outcomes, and that dynamic physiological responses may be more informative.
Higher lactate and creatinine levels and lower baseline glomerular filtration rates at ICU admission in non-survivors indicate more pronounced circulatory impairment and renal involvement. These findings reflect the critical role of systemic hypoperfusion, microcirculatory dysfunction, and impaired organ perfusion in determining clinical outcomes in sepsis. Lactate remains an important indirect marker of hypoperfusion and metabolic stress and may reflect not only tissue hypoxia but also complex metabolic disturbances.¹ Mikkelsen et al. demonstrated that elevated lactate levels are associated with mortality independently of overt shock or organ failure.¹⁶ In sepsis, systemic inflammatory response, microvascular dysfunction, and myocardial impairment together contribute to impaired organ perfusion and the development of multiple organ failure.^4,5 In this context, although admission lactate was significant in univariate analysis, it showed only borderline significance in the multivariate model, suggesting that its effect may overlap with other determinants such as organ dysfunction, age, and cardiovascular comorbidities. Similarly, the loss of significance of renal function parameters in the multivariate model suggests that these variables should be interpreted as part of a broader pathophysiological process reflecting overall sepsis severity.
No significant differences were observed between groups regarding prior cardiac surgery-related variables, including type of surgery and time elapsed since surgery. This finding suggests that acute physiological derangement and current cardiovascular reserve at the time of sepsis may be more influential than remote surgical characteristics. Indeed, the clinical course of sepsis in cardiothoracic surgical patients is shaped not only by the surgical procedure itself but also by the interaction between systemic inflammatory response and comorbid conditions.¹⁵ Therefore, classifications based solely on surgical type may be insufficient to predict prognosis.
Taken together, a history of open-heart surgery alone does not constitute a homogeneous risk profile; rather, mortality risk appears to be primarily driven by current organ dysfunction, cardiovascular reserve, and comorbidities. These findings suggest that risk assessment in this patient population should focus more on dynamic clinical parameters and indicators of organ dysfunction rather than static surgical characteristics .

Limitations

This study has several limitations. First, its single-center retrospective design carries risks of selection bias and incomplete data. Additionally, baseline differences in clinical characteristics between survivor and non-survivor groups may have led to an overrepresentation of more severely ill patients in the mortality group. This is an expected feature of retrospective observational studies and likely reflects underlying disease severity; therefore, multivariate analysis was emphasized for assessing independent effects.
Second, the analysis was limited to ICU outcomes, and in-hospital outcomes after ward transfer or long-term outcomes were not evaluated. This may have limited the comprehensive assessment of prognosis, particularly in a population with a high burden of cardiovascular disease.
Furthermore, the relatively small number of patients in surgical subgroups may have limited the ability to detect the prognostic impact of cardiac surgery type. Similarly, the presence of unmeasured or unrecorded confounding factors (residual confounding) cannot be entirely excluded.
Despite these limitations, the study’s focus on septic patients with a history of open-heart surgery and the evaluation of admission-related predictors of ICU mortality in this specific cohort represent important strengths .

Conclusion

In septic patients with a history of open-heart surgery, advanced age, higher SOFA score, and congestive heart failure were independently associated with ICU mortality. In contrast, type of surgery and time elapsed since surgery were not significantly associated with mortality.
These findings suggest that mortality in this patient population is more closely related to the burden of current organ dysfunction and cardiovascular reserve rather than the history of prior cardiac surgery alone. Although biochemical markers such as lactate reflect clinical severity, they should be interpreted within the broader context of underlying pathophysiological processes.
In conclusion, risk assessment in septic patients with a history of open-heart surgery should prioritize dynamic physiological parameters and indicators of organ dysfunction rather than static surgical characteristics .

Declarations

Abbreviations

AKI - Acute kidney injury
APACHE II - Acute Physiology and Chronic Health Evaluation II
BMI - Body mass index
CABG - Coronary artery bypass grafting
COPD - Chronic obstructive pulmonary disease
GFR - Glomerular filtration rate
ICU - Intensive care unit
IMV - Invasive mechanical ventilation
RRT - Renal replacement therapy
SOFA - Sequential Organ Failure Assessment

References

1. Prescott HC, Antonelli M, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock 2026. Crit Care Med. 2026;54(4):725-812. doi:10.1097/ccm.0000000000007075
   Article PubMed Google Scholar
2. Vincent JL, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med. 2006;34(2):344-353. doi:10.1097/01.ccm.0000194725.48928.3a
   Article PubMed Google Scholar
3. Gotts JE, Matthay MA. Sepsis: pathophysiology and clinical management. BMJ. 2016;353. doi:10.1136/bmj.i1585
   Article PubMed Google Scholar
4. Lin Y, Xu Y, Zhang Z. Sepsis-induced myocardial dysfunction: the pathophysiological mechanisms and therapeutic strategies targeting mitochondria. J Clin Med. 2020;9(11):3645.
   PubMed Google Scholar
5. Walley KR. Sepsis-induced myocardial dysfunction. Curr Opin Crit Care. 2018;24(4):292-299. doi:10.1097/mcc.0000000000000507
   Article PubMed Google Scholar
6. Howitt SH, Herring M, Malagon I, McCollum CN, Grant SW. Incidence and outcomes of sepsis after cardiac surgery as defined by the Sepsis-3 guidelines. Br J Anaesth. 2018;120(3):509-516. doi:10.1016/j.bja.2017.10.018
   Article PubMed Google Scholar
7. Napoli F, Aleman R, Zadneulitca N, Navia J, Brozzi NA. Sepsis in cardiothoracic surgery. Surgery. 2024;175(2):556-558. doi:10.1016/j.surg.2023.10.015
   Article PubMed Google Scholar
8. de Oliveira DC, de Oliveira Filho JB, Silva RF, et al. Sepsis in the postoperative period of cardiac surgery: problem description. Arq Bras Cardiol. 2010;94(3):332-336.
   PubMed Google Scholar
9. Karamnov S, Brovman EY, Greco KJ, Urman RD. Risk factors and outcomes associated with sepsis after coronary artery bypass and open heart valve surgeries. Semin Cardiothorac Vasc Anesth. 2018;22(4):359-368. doi:10.1177/1089253218785362
   Article PubMed Google Scholar
10. Aissaoui N, Boissier F, Chew M, Singer M, Vignon P. Sepsis-induced cardiomyopathy. Eur Heart J. 2025;46(34):3339-3353. doi:10.1093/eurheartj/ehaf340
    Article PubMed Google Scholar
11. Iyer S, Kennedy JN, Jentzer JC, Senussi MH, Seymour CW. Cardiac function before sepsis and clinical outcomes. JAMA. 2024;331(17):1496-1499. doi:10.1001/jama.2024.3917
    Article PubMed Google Scholar
12. Alhamyani AH, Khan RM, Nasser AY, et al. Sepsis in aging populations: a review of risk factors, diagnosis, and management. Cureus. 2024;16(12). doi:10.7759/cureus.74973
    Article PubMed Google Scholar
13. Gönen C, Baltacı Özen S, Kayar Çalılı D, Erdem D, Özkocak Turan I. Retrospective evaluation of COVID-19 patients receiving low-dose dexamethasone therapy in the intensive care unit. Anatol J Anesthesiol Intensive Care. 2025;1(1):1-8. doi:10.71350/ajaic.4
    Article PubMed Google Scholar
14. Wu Z, Chen F, Xiao C, et al. Long-term cardiovascular outcomes and risk factors in adult sepsis survivors: a systematic review and meta-analysis. World J Emerg Med. 2025;16(5):423. doi:10.5847/wjem.j.1920-8642.2025.098
    Article PubMed Google Scholar
15. Zhang Q, Xu L, He W, Lai X, Huang X. Survival prediction for heart failure complicated by sepsis: based on machine learning methods. Front Med (Lausanne). 2024;11:1410702. doi:10.3389/fmed.2024.1410702
    Article PubMed Google Scholar
16. Mikkelsen ME, Miltiades AN, Gaieski DF, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med. 2009;37(5):1670-1677. doi:10.1097/ccm.0b013e31819fcf68
    Article PubMed Google Scholar

Additional Information

Publisher’s Note
Bayrakol MP remains neutral with regard to jurisdictional and institutional claims.

Rights and Permissions

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). To view a copy of the license, visit https://creativecommons.org/licenses/by-nc/4.0/

View full license details →

About This Article

Received:

May 5, 2026

Accepted:

June 11, 2026

Published Online:

June 22, 2026