A predictive model for 28-day mortality after discharge in patients with sepsis associated with cerebrovascular disease

Abstract

BACKGROUND:

The present study investigated the association between cerebrovascular diseases and sepsis, including its occurrence, progression, and impact on mortality. However, there is currently a lack of predictive models for 28-day mortality in patients with cerebrovascular disease associated with sepsis.

OBJECTIVE:

The objective of this study is to examine the mortality rate within 28 days after discharge in this population, while concurrently developing a corresponding predictive model.

METHODS:

The data for this retrospective cohort study were obtained from the MIMIC-IV database. Patients with sepsis and cerebrovascular disease in the ICU were included. Laboratory indicators, vital signs, and demographic data were collected within 24 hours of ICU admission. Mortality rates within 28 days after discharge were calculated based on patient death times. Logistic regression analysis was used to identify potential variables for a predictive model. A nomogram visualized the prediction model. The performance of the model was evaluated using ROC curves, Calibration plots, and DCA.

RESULTS:

The study enrolled a total of 2660 patients diagnosed with cerebrovascular disease complicated by sepsis, consisting of 1434 males (53.91%) with a median age of 70.97 (59.60, 80.73). Among this cohort of patients, a total of 751 fatalities occurred within 28 days following discharge. The multivariate regression analysis revealed that age, creatinine, arterial oxygen partial pressure (Pa O2), arterial carbon dioxide partial pressure (Pa CO2), respiratory rate, white blood cell (WBC) count, Body Mass Index (BMI), and race demonstrated potential predictive variables. The aforementioned model yielded an area under the ROC curve of 0.744, accompanied by a sensitivity of 66.2% and specificity of 71.2%. Furthermore, both calibration plots and DCA demonstrated robust performance in practical applications.

CONCLUSION:

The proposed prediction model allows clinicians to promptly assess the mortality risk in patients with cerebrovascular disease complicated by sepsis within 28 days after discharge, facilitating early intervention strategies. Consequently, clinicians can implement additional advantageous medical interventions for individuals with cerebrovascular disease and sepsis.

Keywords

Predictive model 28-day mortality sepsis cerebrovascular disease

1. Introduction

Cerebrovascular diseases, including stroke, are brain dysfunctions caused by abnormalities in the blood vessels supplying the brain [1, 2]. Endothelial dysfunction is a key factor in the pathogenesis of stroke and sepsis. It not only increases the susceptibility of patients to sepsis, but also weakens the immune system response to pathogens, leading to the so-called stroke-induced immunosuppression syndrome [3, 4, 5, 6]. Sepsis is an abnormal host response to infection, which is characterized by systemic inflammation, immune regulation disorder, abnormal coagulation cascade, endothelial dysfunction and metabolic process disorder, and these pathological processes are also related to cerebrovascular disease [7, 8, 9, 10, 11].

The consequences of sepsis extend beyond the initial infection, as survivors face a significantly heightened risk of cerebrovascular events during the first few weeks following discharge [12]. The long-term mortality rate was significantly higher among stroke patients who developed sepsis in comparison to those who did not [13]. The presence of cerebrovascular disease in sepsis patients was associated with an extended duration of stay both in the intensive care unit and general ward, as well as a heightened mortality rate [14, 15, 16]. The noteworthy fact is that sepsis accounts for a mortality rate of 14% among patients with acute cerebrovascular disease, and this risk can be mitigated through timely intervention [17]. Despite previous studies identifying risk factors associated with mortality in patients with cerebrovascular disease complicated by sepsis, such as the Simplified Acute Physiology Score II for chronic obstructive pulmonary disease, malignancy, ischemia, and heart failure, as well as the National Institutes of Health Stroke Scale score and volume of hemorrhagic stroke [15], a prognostic model to predict mortality in such patients has not yet been established.

Therefore, the primary objective of this study is to investigate the correlation between laboratory indicators and vital signs within 24 hours of admission to the ICU, as well as their association with mortality within 28 days after discharge in patients with cerebrovascular disease complicated by sepsis, while also developing a corresponding predictive model.

2. Methods

2.1 Study design

The data for this retrospective cohort study were obtained from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. This publicly accessible and comprehensive intensive care database contains de-identified clinical data of substantial volume. By leveraging the MIMIC database, we gained access to vital clinical information such as laboratory indicators and vital signs within the initial 24 hours of ICU admission. This study is an analysis of a public database. Approval from the Institutional Review Board of hospital was not required.

Patients admitted to the ICU for the first time with a diagnosis of cerebrovascular disease and sepsis were included with information from MIMIC-IV database. The inclusion criteria were as follows: (1) Over 18 years of age (2) Initially admitted and subsequently transferred to the ICU, (3) The duration of ICU stay exceeded 24 hours. The exclusion criteria were as follows: (1) Patients who below the age of 18 years, (2) with previous admissions, (3) with malignant tumors, (4) Duration of ICU stay shorter than 24 hours.

Sepsis was diagnosed according to the sepsis-3 criteria [18]. Patients with documented or suspected infection and an acute increase in total Sequential Organ Failure Assessment (SOFA) score of $\geqslant$ 2 points were considered to meet the diagnostic criteria for sepsis.

2.2 Screening of covariates

The MIMIC data was extracted using Navicate 16 with structured query language. Based on the pathophysiological mechanisms involved in the occurrence and progression of sepsis, including systemic inflammation, activation of the coagulation cascade, metabolic dysregulation, and potential tissue damage pathways, we have incorporated inflammatory markers, coagulation indicators, and parameters for evaluating renal function to comprehensively investigate their correlation with prognosis [5, 6]. Demographic information of patients, including age, gender, Body Mass Index (BMI), and race, was obtained. Baseline information also included vital signs such as respiratory rate, systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate. Additionally, covariates for our study were extracted which encompassed white blood cell (WBC) count, Creatine levels, Base excess, arterial oxygen partial pressure (Pa O2) levels, arterial carbon dioxide partial pressure (Pa CO2) levels, Activated Partial Thromboplastin Time (APTT) levels, international normalized ratio (INR) values and mechanical ventilation status within 24 hours of ICU admission. The average value was employed as the outcome for indicators with multiple test results.

2.3 Calculation of mortality

We conducted a mortality analysis within 28 days post-discharge, utilizing patient death data and discharge timestamps extracted from MIMIC-IV. Patients were classified into the survival group if the duration between hospital discharge and death exceeded 28 days or if they remained alive during follow-up. Otherwise, patients were categorized in the death group.

2.4 Statistical analysis

The data cleaning, imputation of missing values, covariate screening, and establishment of prediction models were performed using R version 4.3.0. The variables with missing values less than 30% underwent multiple imputation, and comparative analyses were conducted before and after the imputation process. The logistic regression analysis (Step forward method) was utilized to identify and select potential variables, thereby establishing a predictive model. Variables with a $P$ -value $<$ 0.05 in univariate analysis were included in the multivariate logistic regression to analyze the prognostic value. Nomogram was used to visualize the prediction model. The performance of the model was evaluated using Receiver Operating Characteristic (ROC) curves, Calibration plots, and Decision Curve Analysis (DCA).

The measurement data with a normal distribution were reported as the mean $\pm$ standard deviation (Mean $\pm$ SD). For data with homogeneity of variance, a T test was employed to compare between two groups, while for data with non-homogeneity of variance, Satterthwaite’s T test was utilized. Measurement data that did not follow a normal distribution were described using the median and quartile, and differences between any two groups were compared using the Mann-Whitney test. Count data were presented as frequencies and proportions, and group differences were assessed using the chi-square test. The statistical significance was determined when the p value was less than 0.05.

3. Results

3.1 Clinical characteristics of patients

The data extraction and screening process was conducted in accordance with the workflow depicted in Fig. 1. The data before and after imputation did not exhibit any significant statistical differences ( $P>$ 0.05). (Supplementary Table 1).

Figure 1.

The process of extracting and excluding data.

A total of 10,250 patients diagnosed with cerebrovascular disease complicated by sepsis and admitted to the ICU were included in this study. After excluding 2,700 patients with non-primary admissions, 3,011 patients with malignant tumors, and 1,879 patients with ICU stays less than 24 hours, a final cohort of 2,660 patients was analyzed. Among these individuals, a total of 751 deaths within 28 days after discharge. The baseline characteristics of the enrolled participants are presented in Table 1. The median age of the participants was approximately 70.97 (59.60, 80.73) years old, and males accounted for around half (53.91%) of the cohort population. Mechanical ventilation was required for about two-thirds (66.37%) of the patients while diabetes with complications were present in approximately one-tenth (9.4%) of them.

Table 1

Clinical characteristics of enrolled patients

Variables	Total ( $n=$ 2660)	Survival ( $n=$ 1909)	Death ( $n=$ 751)	Statistic	$P$
Age, Years,	70.97 (59.60, 80.73)	69.59 (57.94, 79.50)	75.24 (63.19, 83.34)	$Z=$ $-$ 7.14	$<$ 0.001
Gender, $n$ (%)				$\chi^{2}=$ 0.46	0.497
Female	1226 (46.09)	872 (45.68)	354 (47.14)
Male	1434 (53.91)	1037 (54.32)	397 (52.86)
BMI, kg/m²	27.60 (24.00, 32.10)	28.00 (24.50, 32.50)	26.60 (22.90, 31.20)	$Z=$ $-$ 4.35	$<$ 0.001
Race, $n$ (%)				$\chi^{2}=$ 75.67	$<$ 0.001
Asian	72 (2.71)	49 (2.57)	23 (3.06)
White	1607 (60.41)	1208 (63.28)	399 (53.13)
Black	247 (9.29)	189 (9.90)	58 (7.72)
Hispanic/Latino	91 (3.42)	67 (3.51)	24 (3.20)
Other	134 (5.04)	108 (5.66)	26 (3.46)
Unknown	509 (19.14)	288 (15.09)	221 (29.43)
SOFA, score	3.00 (2.00, 4.00)	3.00 (2.00, 4.00)	3.00 (2.00, 4.00)	$Z=$ $-$ 4.59	$<$ 0.001
Base excess, mmol/L,	0.00 ( $-$ 3.00, 1.25)	0.00 ( $-$ 2.00, 2.00)	$-$ 1.00 ( $-$ 5.00, 1.00)	$Z=$ $-$ 6.30	$<$ 0.001
Pa O2, mmHg	144.00 (76.00, 289.00)	158.00 (81.00, 327.00)	114.00 (66.00, 212.00)	$Z=$ $-$ 7.63	$<$ 0.001
Pa CO2, mmHg	40.00 (35.00, 45.00)	40.00 (35.00, 45.00)	39.00 (33.00, 45.00)	$Z=$ $-$ 3.91	$<$ 0.001
WBC, K/uL,	11.60 (8.60, 15.10)	11.30 (8.40, 14.70)	12.50 (9.25, 16.32)	$Z=$ $-$ 5.25	$<$ 0.001
Creatinine, mg/dL	1.00 (0.70, 1.40)	0.90 (0.70, 1.30)	1.10 (0.80, 1.80)	$Z=$ $-$ 8.16	$<$ 0.001
INR	1.20 (1.10, 1.50)	1.20 (1.10, 1.40)	1.30 (1.10, 1.50)	$Z=$ $-$ 4.21	$<$ 0.001
Heart Rate, bpm	82.51 (73.61, 92.69)	81.68 (73.11, 91.15)	85.44 (75.15, 96.33)	$Z=$ $-$ 4.94	$<$ 0.001
Respiratory rate, insp/min,	18.76 (16.71, 21.46)	18.40 (16.48, 20.93)	19.79 (17.55, 22.80)	$Z=$ $-$ 8.65	$<$ 0.001
Mechanical ventilation, $n$ (%)				$\chi^{2}=$ 4.66	0.031
NO	785 (33.63)	579 (34.98)	206 (30.34)
YES	1549 (66.37)	1076 (65.02)	473 (69.66)

Abbreviation: Z: Mann-Whitney test; $\chi^{2}$ : Chi-square test; BMI: Body Mass Index, Pa O2: arterial oxygen partial pressure, Pa CO2: arterial carbon dioxide partial pressure, SOFA: Sequential Organ Failure Assessment, INR: international normalized ratio, WBC: white blood cell.

There were no significant differences in gender between the survival group and the death group. The proportion of patients requiring mechanical ventilation was higher in the death group compared to the survival group (69.66% vs 65.02%, $P=$ 0.031). Additionally, the age of patients in the death group was significantly higher than that of those in the survival group ( $P<$ 0.001). The levels of Base excess, PaO2, and PaCO2 were significantly lower in the death group ( $P<$ 0.001). Furthermore, white blood cell count, INR, and creatinine levels were significantly higher in the death group compared to the survival group ( $P<$ 0.001). Heart rate and respiratory rate were also significantly higher in the death group than those observed in the survival group ( $P<$ 0.001).

3.2 Identification and validation of candidate predictors

The potential predictors examined in this study encompassed Age, INR, Creatinine, Po2, PCo2, Base excess, Heart rate, Respiratory rate, SOFA score, WBC, BMI, and Race ( $P<$ 0.05). Following adjustment for confounding factors of covariates through multivariate regression analysis, the variables that retained statistical significance were age, creatinine, Pa O2, Pa CO2, respiratory rate, WBC count BMI and Race. These variables will be integrated into the final prediction model (Table 2). The specific calculation methodology of this prediction model is delineated in Fig. 2.

Table 2
Regressive analysis for screening potential predictors

Variables	Multivariate analysis
	$\beta$	OR (95%CI)	$P$
Age Years	0.03	1.03 (1.02–1.04)	$<$ 0.001
BMI, kg/m²	$-$ 0.04	0.96 (0.95–0.98)	$<$ 0.001
Race
Unknown	Reference
White	$-$ 0.86	0.45 (0.34–0.60)	$<$ 0.001
Black	$-$ 0.88	0.44 (0.27–0.71)	$<$ 0.001
Hispanic/Latino	$-$ 0.52	0.59 (0.30–1.13)	0.173
Asian	$-$ 1.24	0.61 (0.28–1.33)	0.214
Other	$-$ 2.38	0.48 (0.26–0.88)	0.017
Pa O2, mmHg	$-$ 0.01	0.99 (0.99–0.99)	$<$ 0.001
Pa CO2, mmHg	$-$ 0.02	0.98 (0.97–0.99)	0.008
WBC, K/uL	0.03	1.03 (1.01–1.05)	$<$ 0.001
Creatinine, mg/dL,	0.20	1.23 (1.12–1.34)	$<$ 0.001
Respiratory rate, insp/min	0.09	1.09 (1.06–1.13)	$<$ 0.001

Abbreviation: OR: odds ratio, CI: confdence interval, Pa O2: arterial oxygen partial pressure, Pa CO2: arterial carbon dioxide partial pressure, WBC: white blood cell.

Figure 2.

The nomogram of the prediction model.

3.3 Assessment of prediction models

The ROC, calibration, and DCA curves of the prediction model are presented in Fig. 3A, B, and C respectively. The area under the ROC curve for the aforementioned model was determined to be 0.744, with a sensitivity of 66.2% and specificity of 71.2%. The correction plot demonstrated a robust agreement between predicted and actual values (H-L test, $P>$ 0.05). The DCA curve also demonstrates the model’s significant practical value.

Figure 3.

Evaluation of effectiveness of predictive models. A: The ROC curve was used to evaluate the performance of the model; B: The calibration curve was used to evaluate the consistency of the prediction model; C: The DCA curve was used to evaluate the clinical utility of the prediction model.

4. Discussion

The demographic data, vital signs parameters, and laboratory parameters within 24 hours of ICU admission were analyzed for patients diagnosed with cerebrovascular disease complicated by sepsis in the MIMIC-IV database. Additionally, the associations between these factors and mortality within 28 days after discharge were investigated. The key predictors including age, creatinine levels, Pa O2 and Pa CO2 values along with respiratory rate measurements as well as WBC count and BMI data were carefully chosen for integration into our predictive model. The potential application value of the prediction model was confirmed through rigorous analysis using ROC curve, calibration plots, and DCA.

The findings of a meta-analysis conducted on 15 international citation databases reveal an estimated global incidence of 31.5 million cases of sepsis and 19.4 million cases of severe sepsis, with a potential annual mortality rate reaching 5.3 million deaths [19]. Additionally, electroencephalogram abnormalities were observed in 87% of patients with bacteremia. Among them, 70% exhibited diagnosed alterations in consciousness, ranging from mild lethargy to severe coma [20]. The prevalence of cerebrovascular disease and sepsis as public health concerns is associated with significant disability and mortality rates, while the prognosis for patient survival remains unfavorable [21, 22]. Within ICU settings, stroke patients experiencing septic shock exhibit notably high mortality rates [14]. Recent studies have demonstrated the clinical significance of specific inflammatory factors and protein markers in predicting unfavorable prognosis among sepsis patients [23, 24]. In line with this investigation, WBC, as an inflammatory biomarker, was employed as a predictive factor. Furthermore, the presence of vascular disease or chronic kidney disease has been associated with increased 28-day mortality rates in patients with hospital-acquired sepsis [25].

The prediction model revealed that advanced age significantly contributed to increased risks of 28-day mortality. These results align with the findings of our study, highlighting older age as a pivotal determinant for adverse outcomes among patients diagnosed with hospital-acquired sepsis and postoperative sepsis [25, 26]. The observed association may potentially be attributed to the age-related decline in circulatory function, immune response, and overall organ functionality among geriatric individuals. Age independently contributes to an elevated risk of mortality, irrespective of whether cerebrovascular disease or sepsis is present; nevertheless, this impact becomes more prominent when both conditions coexist. The analysis of hospital-based stroke mortality rates in adult populations within the United States revealed a significant association between older age and increased fatality risk for both ischemic and hemorrhagic strokes [27], which is consistent with our analytical findings. The level of creatinine serves as a direct reflection of the patient’s current renal function, with higher levels indicating a more compromised renal filtration function. Elevated creatinine levels can disrupt the body’s water and salt balance, thereby exacerbating circulatory failure and septic shock. Moreover, it has the potential to alter the bioavailability of drugs primarily metabolized by the liver and intestine [28], consequently impacting their clinical efficacy in conditions such as infection and shock. The impact of creatine value on mortality outcome was found to be more statistically significant in this study, and the contribution of the score with elevated creatine value in our nomogram played a larger role in determining the overall outcome. Prior investigations have not only established a robust association between creatine levels and prognostic indicators for sepsis, but they have also showcased its clinical relevance in diagnosing sepsis by employing metabonomics to assess risk factors linked to the onset and advancement of this pathological state [29, 30]. In addition to creatine, the blood urea nitrogen to albumin ratio has also been identified as an independent prognostic indicator for mortality in septic patients [31].

The role of gender as a regulatory factor in the occurrence and progression of sepsis remains controversial [32]. However, this study found no significant correlation between gender and mortality outcomes in patients with cerebrovascular disease complicated by sepsis. The mortality rate of cerebrovascular patients with sepsis within 28 days after discharge may be correlated with a low BMI. This association is likely attributed to malnutrition in patients with a lower BMI, which aligns with previous findings indicating an elevated risk of unfavorable prognosis among elderly individuals suffering from ischemic stroke [33]. Additionally, our findings indicate a significant association between respiratory function and the prognosis of patients with cerebrovascular disease and sepsis, aligning with previous research that has identified respiratory comorbidities as independent risk factors for mortality in postoperative sepsis [27].

The present study represents a potential endeavor to investigate the prediction of mortality within 28 days following discharge in patients with cerebrovascular disease complicated by sepsis. The final prediction model demonstrates significant practical value following verification through analysis of ROC curve, calibration plot, and DCA curve.

Several limitations exist in this study. Firstly, the retrospective design and single-center setting introduce an inherent selection bias. Moreover, the screening of predictors lacks comprehensiveness and fails to include certain indicators for evaluating other organs, potentially leading to inaccuracies in the predictive accuracy of the model. Future prospective studies are warranted to validate these findings.

5. Conclusion

The proposed prediction model enables clinicians to promptly assess the mortality risk in patients with cerebrovascular disease complicated by sepsis within 28 days after discharge, facilitating early intervention strategies. Consequently, clinicians can implement additional medical interventions that are considered advantageous for individuals with cerebrovascular disease and sepsis.

Data availability

All datasets used during the present study are publicly available in the MIMIC-IV v1.0 database (https://mimic. physionet.org).

Ethical approval

This study is an analysis of a public database. Approval from the Institutional Review Board was not required.

Footnotes

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

Yan

Wang

, et al. Association between triglyceride glucose index and risk of cerebrovascular disease: systematic review and meta-analysis. Cardiovasc Diabetol. 2022; 21(1): 226. doi: 10.1186/s12933-022-01664-9.

Lee

Slutsky

. Sepsis and endothelial permeability. N Engl J Med. 2010; 363: 689-691.

Xing

Luo

, et al. Role of activating the nuclear factor kappa B signaling pathway in the development of septic cardiomyopathy in rats with sepsis. Technol Health Care. 2023; 31(5): 1671-1681. doi: 10.3233/THC-220471.

Irawany

Lolobali

. Neutrophil-to-lymphocyte Ratio as a New Predictor of Sepsis in Critically-ill Stroke Patients Open Access Maced J Med Sci. 2024; 10 Open Access Maced J Med Sci. doi: 10.3889/oamjms.2022.9446.

Yende

Kellum

Talisa

et al. Long-term host immuneresponse trajectories among hospitalized patients with sepsis. JAMANetw Open. 2019; 2: 1-15. doi: 10.1001/jamanetworkopen.2019.868615.

Hao

Liang

, et al. Identification and analysis of genes associated with the severity and prognosis of sepsis. Technol Health Care. 2024; 32(2): 989-996. doi: 10.3233/THC-230363.

Corrales-Medina

Alvarez

Weissfeld

, et al. Association between hospitalization for pneumonia and subsequent risk of cardiovascular disease. Jama-J Am Med Assoc. 2015; 313(3): 264-74. doi: 10.1001/jama.2014.18229.

Lai

Lee

, et al. Susceptible period for cardiovascular complications in patients recovering from sepsis. Can Med Assoc J. 2018; 190(36): E1062-E1069. doi: 10.1503/cmaj.171284.

De Geer

Oscarsson Tibblin

Fredrikson

, et al. No association with cardiac death after sepsis: A nationwide observational cohort study. Acta Anaesthesiol Scand. 2019; 63: 344-351. doi: 10.1111/aas.13280.

10.

Angriman

Rosella

Lawler

, et al. Sepsis hospitalization and risk of subsequent cardiovascular events in adults: a population-based matched cohort study. Intens Care Med. 2022; 48(4): 448-457. doi: 10.1007/s00134-022-06634-z.

11.

Lai

Lee

, et al. Susceptible period for cardiovascular complications in patients recovering from sepsis. Can Med Assoc J. 2018; 190 CAN MED ASSOC J. doi: 10.1503/cmaj.171284.

12.

Tsou

Wang

, et al. Impact of post-sepsis cardiovascular complications on mortality in sepsis survivors: a population-based study. Crit Care. 2019; 23(1): 293. doi: 10.1186/s13054-019-2579-2.

13.

Phua

Lim

, et al. Sepsis, cardiovascular events and short-term mortality risk in critically ill patients. Ann Acad Med Singap. 2024; 51. doi: 10.47102/annals-acadmedsg.202220.

14.

Silva

Miranda

Massaud

, et al. Abstract WP337: Septic Shock is Associated with Poor Clinical Outcomes in Patients with Ischemic Stroke. Stroke. 2024; 48. doi: 10.1161/str.48.suppl_1.wp337.

15.

Wang

Liu

. Construction of a prediction model for risk of early neurological deterioration following intravenous thrombolysis in patients with acute ischemic stroke. Technol Health Care. 2023; 31(6): 2213-2223. doi: 10.3233/THC-230140.

16.

Gonçalves

Kurtz

Turon

, et al. Incidence and impact of sepsis on long-term outcomes after subarachnoid hemorrhage: a prospective observational study. Ann Intensive Care. 2019; 9. doi: 10.1186/s13613-019-0562-3.

17.

Arboix,

Massons,

Oliveres,

, et al. Mortality in the acute phase of cerebrovascular disease: the registry of La Alianza-Hospital Central of Barcelona. Med Clin-Barcelona. 1994; 103. PMID: 7799664.

18.

Singer

Deutschman

Seymour

, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). Jama. 2016; 315(8): 801-10.

19.

Fleischmann

Scherag

Adhikari

, et al. Assessment of Global Incidence and Mortality of Hospital-treated Sepsis. Current Estimates and Limitations. Am J Resp Crit Care. 2016; 193. doi: 10.1164/rccm.201504-0781OC.

20.

Szatmári

Végh

Csomós

, et al. Impaired cerebrovascular reactivity in sepsis-associated encephalopathy studied by acetazolamide test. Crit Care. 2010; 14(2): R50. doi: 10.1186/cc8939.

21.

Machado

Cavalcanti

Monteiro

, et al. Predictive Accuracy of the Quick Sepsis-related Organ Failure Assessment Score in Brazil. A Prospective Multicenter Study. Am J Resp Crit Care. 2020; 201(7): 789-798. doi: 10.1164/rccm.201905-0917OC.

22.

Wang

, et al. Diagnostic value of coagulation index and serum inflammatory cytokines in hemorrhagic stroke patients with pulmonary infection in the sequelae stage. Technol Health Care. 2024; 32(3): 1383-1391. doi: 10.3233/THC-230345.

23.

Chen

Yan

, et al. Pentraxin-3 Is a Strong Biomarker of Sepsis Severity Identification and Predictor of 90-Day Mortality in Intensive Care Units via Sepsis 30 Definitions. Diagnostics (Basel). 2021; 11(10): doi: 10.3390/diagnostics11101906.

24.

Alba-Patiño

Vaquer

Barón

, et al. Micro- and nanosensors for detecting blood pathogens and biomarkers at different points of sepsis care. Microchim Acta. 2022; 189. doi: 10.1007/s00604-022-05171-2.

25.

López-Mestanza

Andaluz-Ojeda

Gómez-López

, et al. Clinical factors influencing mortality risk in hospital-acquired sepsis. J Hosp Infect. 2018; 98. doi: 10.1016/j.jhin.2017.08.022.

26.

Thakkar

Weiss

Fitzgerald

, et al. Risk Factors for Mortality in Pediatric Postsurgical versus Medical Severe Sepsis. J Surg Res. 2019; 242. doi: 10.1016/j.jss.2019.04.011.

27.

Neves

Cole

Lee

, et al. Demographic and institutional predictors of stroke hospitalization mortality among adults in the United States. eNeurologicalSci. 2022; 26. doi: 10.1016/j.ensci.2022.100392.

28.

Saito

Tsuchiya

Itoga

, et al. Multiple Blood Culture Sampling, Proper Antimicrobial Choice, and Adequate Dose in Definitive Therapy Supported by the Antimicrobial Stewardship Team Could Decrease 30-Day Sepsis Mortality Rates. Infect Drug Resist. 2024; 17. doi: 10.2147/IDR.S445917.

29.

íIzquierdo-Garcia

Nin

Cardinal-Fernandez

, et al. Identification of novel metabolomic biomarkers in an experimental model of septic acute kidney injury. AM J Physiol-Renal. 2019; 316. doi: 10.1152/ajprenal.00315.2018.

30.

Naveen Kumar,

Gupta,

Kumar,

, et al. Differentiation between sepsis survivors and sepsis non-survivors through blood serum metabolomics: A proton nuclear magnetic resonance spectroscopy (NMR) study. Magn Reson Imaging. 2022; 89. doi: 10.1016/j.mri.2022.02.003.

31.

Han

Cheng

Liao

, et al. Analysis of the Value of the Blood Urea Nitrogen to Albumin Ratio as a Predictor of Mortality in Patients with Sepsis. J Inflamm Res. 2022; 15. doi: 10.2147/JIR.S356893.

32.

Failla

Connelly

. Systematic Review of Gender Differences in Sepsis Management and Outcomes. J Nurs Scholarship. 2017; 49. doi: 10.1111/jnu.12295.

33.

Jiang

Zhao

Bao

, et al. Prognostic value of geriatric nutritional risk index in patients with acute ischemic stroke. Chin J Geriatr Heart Brain Vessel Di. 2023; 25(1): 35-38. doi: 10.3969/j.issn.1009-0126.2023.01.010.

A predictive model for 28-day mortality after discharge in patients with sepsis associated with cerebrovascular disease

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Methods

2.1 Study design

2.2 Screening of covariates

2.3 Calculation of mortality

2.4 Statistical analysis

3. Results

3.1 Clinical characteristics of patients

Table 2 Regressive analysis for screening potential predictors

5. Conclusion

Data availability

Ethical approval

Footnotes

Conflict of interest

References

Table 2
Regressive analysis for screening potential predictors