European Journal of Cardiovascular Medicine

Sepsis is a life-threatening organ dysfunction caused by the dysregulation of a host’s response to infections (1) which is further complicated by an altered metabolic state.

The following now forms part of the modified Sepsis 3 clinical criteria (2) for sepsis:

1. Suspected infection; and
2. Acute organ dysfunction, which is indicated by an increase of at least two points from baseline in the sequential (or sepsis-related) organ failure assessment (SOFA) score.”

“The SOFA score, which includes six organ systems, is a 24-point evaluation of organ dysfunction (renal, cardiovascular, pulmonary, hepatic, neurologic and hematologic)” (2).

“Septic shock (2) is a subset of sepsis in which circulatory and cellular/metabolic abnormalities are profound enough to increase mortality risk. The criteria proposed for septic shock include:

1. a) Suspected infection
2. b) Need for vasopressor to elevate mean arterial pressure to more than or equal to 65mmHg
3. c) Serum lactate concentration more than 2 mmol/L after adequate fluid resuscitation.”

Sepsis stands as a powerful obstacle in critical care, persisting as principal reason for death among extremely sick patients, inspite of notable advancements in supportive care and antibiotic treatment. (3)

The release of “brain natriuretic peptide (BNP)” and “NT-proBNP”, the inactive product cleaved off BNP into the bloodstream is triggered by atrial and ventricular wall stretch, indicative of increased cardiac chamber pressure or volume overload, as well as myocardial ischemia, a consequence of insufficient blood supply to the myocardium. BNP is quickly removed from the bloodstream due to its brief half-life of about twenty minutes. In contrast, NT-proBNP has a half life which is longer, ranging from one to two hours, allowing it to persist in the bloodstream for a more extended period.(4)

Beyond heart failure, BNPs have been shown to be reliable indicators of death in a variety of clinical contexts. They show predictive ability in patients undergoing non-cardiac surgery, acute coronary syndromes, pulmonary embolisms, and stable coronary disease. Even with their proven usefulness, research on BNPs' potential to predict death in sepsis patients is still underway.(5)Among the plethora of biomarkers scrutinized, biomarkers like procalcitonin, lactate and C-reactive protein, the role of natriuretic peptides, particularly NT-proBNP, has emerged as an area of significant interest. (6)

This study's objective is to thoroughly look at NT-proBNP's prognostic significance as a critical indicator of in-hospital severity among individuals suffering from sepsis. The study also endeavors to delve into several secondary objectives of significant relevance such as comparison between NT-proBNP and other established biomarkers employed in the context of sepsis. (7)

Objective

The objective of this study is to evaluate the efficacy of Levofloxacin as an adjunct to standard antitubercular chemotherapy (ATT) in improving treatment outcomes for pulmonary tuberculosis (PTB) patients with Type II diabetes.

The current study was a hospital-based prospective cross sectional study that was carried out from March 2023 to February 2024on 100 sepsis patients who had been admitted to the department of general medicine at Silchar Medical College and Hospital. The study was approved by the Silchar Medical College institutional ethical committee, and the ethical clearance number was SMC/ETHICS/17th Feb/2023/32. Informed written consent was obtained from the patient in their own language prior to data collection.

The inclusion criteria were:

• Age >18years
• Suspected infection
• Acute organ dysfunction defined as increase by 2 or more SOFA points

The exclusion criteria were:

• Patients with pre existing reduction of left ventricular functionor Dilated cardiomyopathy on echocardiography
• Pre existing acute or chronic cor pulmonale, pulmonary arterial hypertension, valve disease on echocardiography
• Pre existing chronic renal failure with eGFR <60ml/min per 1.73 m2, Albuminuria with structural abnormalities on ultrasonography
• Acute coronary ischemia
• Patients with severe burns
• Patients on cancer chemotherapy
• Pre existing Cirrhosis of liver on ultrasonography
• Patients with BMI more than 30 kg/m2

Using blood that was obtained right after admission, NT- proBNP levels were assessed. We performed additional laboratory tests including white blood cell count, bilirubin fractions, creatinine level. As the definitive diagnosis, we established the probable cause of suspected sepsis and provided microorganism results based on blood culture data.

Inthecalculationof the SOFA score the variables required are:

• Plateletcount, Bilirubin, Creatinine, PaO2/FiO2 ratio, Mean Arterial Pressure and GlasgowComaScalescore; where, FiO2=0.21+(0.04*Oxygen flow in litres)

Completebloodcountwasmeasuredbyflowcytometryusingan automated analyzer SYSMEX XN 550. Biochemical tests like Kidney and liver function tests were done using VITRIOS 5600 Machine. Serum “NT-proBNP” levels were measured by the VITROS 5600 autoanalyzer using immunometric method. Positive results for NT-proBNP is defined as levels of 300 pg/ml or higher.

The clinical outcome was mortality at 28 days.

Statistical analysis

A predetermined proforma was used for collecting thedata.CalculationsandanalysisweredoneusingtheSocial Statistics Website, Tables and graphs weremade using Microsoft Word and Excel.Descriptive statistics, such as means, frequencies, percentages and standard deviations were used to analyze the data. Figures, charts, and tables were utilized for data presentation.Clinical and demographic characteristics of “survivors” and “non-survivors” was compared using the relevant statistical tests such as t-test and chi-square test.Using “receiver operating characteristic (ROC)” curve analysis, the cut-off values for “NT-proBNP” predicting death was determined.After accounting for possible confounders, independent mortality predictors were established using logistic regression.

Analphalevelof5%wastaken,i.e.,apvalueoflessthan0.05wasconsideredsignificant.

Table 1: Outcomes by age group

32.86% of patients in the 0–60 age group died. More patients i.e., 56.67% died in the age range of 61 to 100 years. With a corresponding p-value of 0.045, the age groups and outcomes were found to be statistically significantly associated indicating that people over 60 have a considerably higher chance of dying than patients under 60.

Table 2: Comparison of Alive and Dead with Sex.

Females constitute a larger proportion among those who are "Alive" compared to those who are "Dead." The chi-squared test statistic for the comparison of "Alive" and "Dead" individuals with respect to sex is 4.102. The p-value associated with this test is 0.043 indicating that there is a significant statistical difference in the distribution of survival status across different sexes.

Table 3: Outcome Distribution.

The outcome distribution indicates that out of the total cases, 64% were classified as "Alive" and 36% as "Dead". The p-value for the outcome distribution (Alive vs. Dead) is approximately 0.0051 which indicates that there is a significant statistical difference.

Table 4:  Comparison of Alive and Dead with MAP

Among individuals who are dead, the mean MAP is 70.65 mmHg, with a standard deviation of 20.71. The z-test statistic comparing the mean MAP between alive and dead individuals is 3.06, with a p-value of 0.0022. This indicates a significant statistical difference in mean MAP between the alive and dead groups.

             Table 5: Comparison of Alive and Dead Among Study Variables (Total Count, Platelet Count)

Total Count: p-value of 0.2339 suggests there is no significant statistical difference in the total count between alive and dead individuals.

PLATELET COUNT: p-value of 0.5273, indicating no significant statistical difference in platelet counts between alive and dead individuals.

Table 6: Comparison of Alive and Dead Among Various Study Variables (GCS, PaO2/FiO2, RBS, T. Bilirubin, S. Creatinine, Na, K, PCT, NT proBNP, SOFA Score)

GCS (Glasgow Coma Scale): Alive individuals have a mean GCS score of 14.25 with a standard deviation of 1.41. Deceased individuals have a mean GCS score of 9.33 with a standard deviation of 3.67. The z-test statistic is 7.73 with a p-value (p < 0.0001). This suggests a significant difference in GCS scores between alive and dead individuals.

PaO2/FiO2: Alive individuals have a mean PaO2/FiO2 ratio of 379.91 with a standard deviation of 86.69. Deceased individuals have a mean PaO2/FiO2 ratio of 308.46 with a standard deviation of 93.36. The z-test statistic is 3.77 with a p-value of 0.0002, indicating a significant difference in the PaO2/FiO2 ratio between alive and dead individuals.

1. Bilirubin (Total Bilirubin):There is no significant statistical difference in mean total bilirubin levels between the two groups (p = 0.6519).

1. CREATININE (Serum Creatinine): The difference in mean serum creatinine levels between alive and dead individuals is marginally significant, with a p-value of 0.0582.

PCT (Procalcitonin): There is a significant difference in mean procalcitonin levels between alive and dead individuals, with a p-value (p < 0.0001).

SOFA SCORE

Figure 1: Comparison of SOFA SCORE between Alive and Dead groups. The green bar portrays the group that survived and the blue bar portrays the group that died. Error bars indicate standard deviation.

The analysis of the “Sequential Organ Failure Assessment (SOFA) scores” reveals significant differences between survivors and non-survivors. For survivors, the average SOFA score was 4.70, with a standard deviation of 2.04. This indicates that the extent of organ dysfunction among survivors was relatively moderate and consistent, as reflected by the low mean and modest variability. The patients who died had a much higher average SOFA score of 9.42, with a standard deviation of 3.79. The statistical significance of these differences was established through a z-test, which produced a value of -6.93, with a corresponding p-value of <0.001 indicating a highly significant result. These findings underscore the critical role of the SOFA score as a predictive tool in clinical settings.

Table 7: Results of the Logistic Regression Analysis

Figure 2: The logistic regression analysis investigates the relationship between SOFA SCORE and patient outcome (Alive or Dead).

The intercept is -4.0505. This value represents the log-odds of the outcome being "Dead" when the SOFA SCORE is zero. While not directly interpretable in a clinical sense, it serves as the baseline for the logistic regression equation. The coefficient for SOFA SCORE is 0.5306. This indicates that for each one-unit increase in SOFA SCORE, the log-odds of the outcome being "Dead" increase by 0.5306. In terms of odds ratio, this can be interpreted as each additional point in the SOFA SCORE increasing the odds of death by approximately 70% (exp(0.5306) ≈ 1.70). The p-value for the SOFA SCORE coefficient is less than 0.001, demonstrating strong statistical significance. This suggests that there is a very low probability that the observed relationship is due to random chance. The pseudo R-squared value of 0.3488 indicates that approximately 34.88% of the variability in the outcome can be explained by the SOFA SCORE. While not an exceptionally high value, it does suggest a moderate explanatory power of the model. The plot visualizes the logistic regression curve (blue line) and the actual data points (red dots). The curve shows the probability of the outcome being "Dead" as a function of the SOFA SCORE. As the SOFA SCORE increases, the probability of death also increases, aligning with the positive coefficient found in the regression model. The logistic regression analysis indicates a significant and positive association between SOFA SCORE and the probability of death. Higher SOFA Scores are associated with increased likelihood of a negative outcome, and the statistical significance and model fit provide confidence in these findings.

NT proBNP

Figure 3: Comparison of NT proBNP between Alive and Dead groups. The green bar portrays the group that survived and the blue bar portrays the group that died. Error bars indicate standard deviation.

The analysis of NT proBNP levels among patients revealed significant differences between those who survived and those who did not. For patients who were alive, the average NT proBNP level was 9005.16 pg/mL, with a standard deviation of 22134.35 pg/mL. In contrast, patients who had died had a markedly higher average NT proBNP level of 32630.83 pg/mL, with a standard deviation of 22357.26 pg/mL. This stark difference in mean values indicates that higher NT proBNP levels are strongly related with mortality. The statistical significance of this difference was confirmed through a z-test, which yielded a value of -5.09. The corresponding p-value was less than 0.001, indicating a highly significant result. In clinical terms, this suggests that NT proBNP is a potential biomarker for predicting patient outcomes, with elevated levels being a critical indicator of higher mortality risk.  These results emphasize the importance of monitoring NT proBNP levels in patients as part of their assessment and management, particularly in critical care settings.

                                Figure 4: “Receiver Operating Characteristic (ROC) curve” for NT proBNP versus OUTCOME.

The above-displayed “Receiver Operating Characteristic (ROC) curve” assesses how well NT proBNP levels diagnose by forecasting the result (Alive or Dead). Plotting the curve shows the relationship between sensitivity (recall) and specificity, with the False Positive Rate (FPR) on the x-axis and the True Positive Rate (TPR) on the y-axis.

Key statistical values from the ROC analysis include:

“Area Under the Curve (AUC)”: This ROC curve's AUC is roughly 0.84. The test's overall capability to distinguish between positive (Dead) and negative (Alive) outcomes is gauged by the AUC. Whereas an AUC of 1.0 denotes perfect discrimination, an AUC of 0.5 indicates no discriminative capacity (equal to random guessing). Consequently, a good degree of discriminative ability of NT proBNP levels in predicting the result is shown by an AUC of 0.84.

True Positive Ratio (TPR): The TPR, sometimes referred to as sensitivity, quantifies the percentage of true positives (Dead) that the test accurately identifies. Better sensitivity is indicated by a greater TPR.

False Positive Rate (FPR): This is a measurement of the percentage of true negatives (Alive) that the test erroneously reports as positive. The specificity is higher with a lower FPR.

The ROC curve position above the diagonal line (which represents a test with no discriminative ability) signifies that the NT proBNP test is better than random guessing in predicting outcomes. The steeper the curve towards the top-left corner, the higher the sensitivity and specificity of the test. In this case, the NT proBNP levels show a significant ability to distinguish between patients who are likely to be Alive or Dead, suggesting that it is a valuable biomarker for this purpose. The ROC curve and the AUC value of 0.84 demonstrate that NT proBNP levels have good predictive power for patient outcomes, with a balanced trade-off between sensitivity and specificity. This makes NT proBNP a useful marker in clinical settings for predicting patient prognosis.

Table 8:  Mean and standard deviation of hospital stay based on the outcome

The resulting p-value was approximately 0.14, indicating that there is no statistically significant difference in the hospital stay durations between the two outcome groups.

The current study was undertaken to evaluate the prognostic significance of “N-Terminal Pro Brain Natriuretic Peptide (NT Pro BNP)” as in-hospital severity indicator in patients with sepsis.

Sex distribution with Age group

The study found that patients over 60 are more likely to experience fatal outcomes compared to those aged 60 and below. According to Katarzyna Kotfis et al. (2019)⁽⁸⁾hospital mortality rose with age, with patients over 80 years old about twice as likely to die compared to those under 50.

Sex Distribution and Comparison of Alive and Dead with Sex

This study found that females make up a larger proportion in "Alive" individuals. The chi-squared test statistic for this comparison is 4.102, with a p-value of 0.043, indicating a statistically significant difference in survival status distribution across different sexes.

According to Nosheen Nasir et al. (2015)⁽⁹⁾, in comparison to females, males exhibited a greater death rate.

According to Bernhard Wernly et al. (2021)⁽¹⁰⁾, no significant sex-specific variations in mortality were found in the sensitivity analyses.

Outcome Distribution

The study found that 64% of cases were classified as "Alive" and 36% as "Dead", indicating a majority of cases had positive outcomes. The p-value for the outcome distribution (Alive vs. Dead) is approximately 0.0051, indicating a statistically significant difference.

Mortality differed by region, as demonstrated by Michael Bauer et al. (2020)⁽¹¹⁾, who found that the septic shock mortality at the end of 30 days was 33.7% in North America, 26.4% in Australia and 32.5% in Europe.
According to Li Weng et al. (2018)⁽¹²⁾, China's sepsis-related death rate was 66.7.

Comparison between Alive and Dead with MAP

The study analyzed the “mean arterial pressure (MAP)” of alive and dead individuals. The mean MAP of alive individuals was found to be 83.33 mmHg and of dead was 70.65. The z-test statistic showed a statistically significant difference in MAP between alive and dead individuals.

According to Martin W. Dünser et al. 2009⁽¹³⁾,the prognosis for sepsis patients with MAP values greater than or equal to 60 mmHg may be the same as that of patients with higher MAP levels during the first twenty-four hours of intensive care unit treatment.

Total Leukocyte Count and Platelet Count

The study found no significant difference in the total leukocyte count between alive and deceased individuals. The study also found no significant difference in platelet counts between alive and dead individuals.

Emily Rimmer MD, MSc et al. 2022⁽¹⁴⁾found a significant correlation between rising WBC trends and greater mortality. Tobias Schupp et al., 2022⁽¹⁵⁾found elevated risk of death at the end of 30 days when they had thrombocytopenia.

GCS

p value <0.001 indicating a significant difference in GCS scores between alive and dead individuals.

In their study, Ying Wu et al. 2021⁽¹⁶⁾ found that non-survivors had a lower GCS.
The GCS score is a significant predictor of disease prognosis, according to Qiang Lai et al. 2023⁽¹⁷⁾.

Oxygenation index (PaO2/FiO2 ratio)

The study found a significant statistical difference in PaO2/FiO2 ratio between alive and dead individuals, with a p value 0.0002 indicating a significant difference.

Hongying Bi et al. 2023⁽¹⁸⁾ found that in sepsis patients, a higher risk of death after 28 days was associated with either a high or low PaO2/FiO2.

Total bilirubin

The study found no significant difference in total bilirubin levels between alive and dead individuals, as indicated by a non-significant p-value (p = 0.6519).

According to Ying Wu et al. 2021⁽¹⁶⁾, non-survivors had higher bilirubin levels in their study.
According to Jayshil J. Patel, MD, et al. 2013⁽¹⁹⁾,patients with septic shock and severe sepsis who have excessive serum bilirubin levels within 72 hours of arrival are at increased risk of dying.

Serum creatinine

The mean serum creatinine levels between alive and dead individuals show a marginally significant difference, with a p-value of 0.0582.

Patients with a single mild episode of Acute kidney injury have much poorer long-term survival  than terminally ill patients without Acute kidney injury, according to research by Adam Linder et al. (2014)⁽²⁰⁾.

Procalcitonin

Procalcitonin levels showed a significant difference between alive and dead groups (p < 0.0001).

Procalcitonin values > or =7 ng/ml recorded during ICU admission have been found to be predictive of short-term mortality by Saransh Jain et al. 2014⁽²¹⁾.

PCT cut-off concentrations can help predict the course of sepsis and is especially useful in selecting patients who might get advantage from ICU admission, according to E.J. Giamarellos-Bourboulis et al. 2011⁽²²⁾.

Mean and standard deviation of hospital stay based on the outcome

A p-value of approximately 0.14, suggests that the numerical difference in hospital stay durations between alive and dead patients may be due to random chance.

According to Richard Y. Kim MD et al. 2018⁽²³⁾, there is no connection between mortality and emergency department duration of stay.

A study by Zhongheng Zhang et al. (2018)⁽²⁴⁾ demonstrates that among patients with sepsis needing ICU care, a longer stay in the emergency room is independently linked to a higher mortality.

SOFA Score

The “Sequential Organ Failure Assessment (SOFA) scores” reveal significant differences between “survivors” and “non-survivors” p-value of <0.001.

By using a univariate logistic regression model, Flavio Lopes Ferreira, MD et al. (2001)⁽²⁵⁾ discovered that SOFA is a reliable predictor of outcome in the early days following ICU admission.

Bodin Khwannimit et al. (2018)⁽²⁶⁾ reported that compared to qSOFA and SIRS, the SOFA score for hospital mortality was substantially greater.

“NT proBNP”

The study of “NT proBNP” levels in patients showed significant differences between those who survived and those who died p value <0.001.

According to a study by Malik Benmachiche et al. (2018)⁽²⁷⁾, patients with high NT-proBNP levels have a greater chance of mortality in the hospital and requiring longer length of stay. According to research done in by Martina Brueckmann et al.2005⁽²⁸⁾, when patients are admitted with severe sepsis, “NT-proBNP” has been shown to be a useful laboratory measure for predicting survival.

The study highlights the critical role of “NT-proBNP” as a prognostic marker in sepsis. Elevated “NT-proBNP” levels were significantly associated with higher death rate, underscoring its utility in assessing the severity of sepsis and guiding clinical decisions. The findings also reaffirm the importance of comprehensive clinical assessments, including “SOFA scores” and other physiological and laboratory parameters, in managing septic patients. Early identification and timely intervention based on these markers can improve patient outcomes and reduce sepsis-related mortality..

1. Shankar-Hari M, Phillips G, Levy M, Seymour C, Liu V, Deutschman C, et al. Developing a New Definition and Assessing New Clinical Criteria for Septic Shock: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):775-87.
2. Harrison's Principles of Internal Medicine, 21e Loscalzo J, Fauci A, Kasper D, Hauser S, Longo D, Jameson J. Loscalzo J, & Fauci A, & Kasper D, & Hauser S, & Longo D, & Jameson J(Eds.),Eds. Joseph Loscalzo, et al.
3. Nedeva C. Inflammation and cell death of the innate and adaptive immune system during sepsis. Biomolecules. 2021 Jul 10;11(7):1011.
4. Lewis, M.. (2017). Paleopathology of Children: Identification of Pathological Conditions in the Human Skeletal Remains of Non-Adults.
5. Maalouf R, Bailey S. A review on B-type natriuretic peptide monitoring: assays and biosensors. Heart failure reviews. 2016 Sep; 21:567-78.
6. Kerkelä R, Ulvila J, Magga J. Natriuretic peptides in the regulation of cardiovascular physiology and metabolic events. Journal of the American Heart Association. 2015 Oct 27;4(10): e002423.
7. Aljabali AA, Obeid MA, El-Tanani M, Mishra V, Mishra Y, Tambuwala MM. Precision epidemiology at the nexus of mathematics and nanotechnology: Unraveling the dance of viral dynamics. Gene. 2024 May 5; 905:148174.
8. Kotfis K, Wittebole X, Jaschinski U, Solé-Violán J, Kashyap R, Leone M, Nanchal R, Fontes LE, Sakr Y, Vincent JL, ICON Investigators. A worldwide perspective of sepsis epidemiology and survival according to age: Observational data from the ICON audit. Journal of critical care. 2019 Jun 1;51:122-32.
9. Nasir N, Jamil B, Siddiqui S, Talat N, Khan FA, Hussain R. Mortality in Sepsis and its relationship with Gender. Pak J Med Sci. 2015 Sep-Oct;31(5):1201-6. doi: 10.12669/pjms.315.6925. PMID: 26649014; PMCID: PMC4641283.
10. Wernly B, Bruno RR, Mamandipoor B, Jung C, Osmani V. Sex-specific outcomes and management in critically ill septic patients. European Journal of Internal Medicine. 2021 Jan 1;83:74-7.
11. Bauer M, Gerlach H, Vogelmann T, Preissing F, Stiefel J, Adam D. Mortality in sepsis and septic shock in Europe, North America and Australia between 2009 and 2019—results from a systematic review and meta-analysis. Critical Care. 2020 Dec;24:1-9.
12. Weng L, Zeng XY, Yin P, Wang LJ, Wang CY, Jiang W, Zhou MG, Du B. Sepsis-related mortality in China: a descriptive analysis. Intensive care medicine. 2018 Jul;44:1071-80.
13. Dünser MW, Takala J, Ulmer H, Mayr VD, Luckner G, Jochberger S, Daudel F, Lepper P, Hasibeder WR, Jakob SM. Arterial blood pressure during early sepsis and outcome. Intensive care medicine. 2009 Jul;35:1225-33.
14. Rimmer E, Garland A, Kumar A, Doucette S, Houston BL, Menard CE, Leeies M, Turgeon AF, Mahmud S, Houston DS, Zarychanski R. White blood cell count trajectory and mortality in septic shock: A historical cohort study. Canadian Journal of Anesthesia/Journal canadien d'anesthésie. 2022 Oct;69(10):1230-9.
15. Schupp T, Weidner K, Rusnak J, Jawhar S, Forner J, Dulatahu F, Brück LM, Hoffmann U, Kittel M, Bertsch T, Akin I. Diagnostic and prognostic role of platelets in patients with sepsis and septic shock. Platelets. 2023 Dec 31;34(1):2131753.
16. Wu Y, Huang S, Chang X. Understanding the complexity of sepsis mortality prediction via rule discovery and analysis: a pilot study. BMC medical informatics and decision making. 2021 Dec;21:1-5.
17. Lai Q, Xia Y, Yang W, Zhou Y. Development and validation of a rapid and efficient prognostic scoring system for sepsis based on oxygenation index, lactate and glasgow coma scale. Journal of Inflammation Research. 2023 Dec 31:2955-66.
18. Bi H, Liu X, Chen C, Chen L, Liu X, Zhong J, Tang Y. The PaO2/FiO2 is independently associated with 28-day mortality in patients with sepsis: a retrospective analysis from MIMIC-IV database. BMC Pulmonary Medicine. 2023 May 27;23(1):187.
19. Patel JJ, Taneja A, Niccum D, Kumar G, Jacobs E, Nanchal R. The association of serum bilirubin levels on the outcomes of severe sepsis. Journal of intensive care medicine. 2015 Jan;30(1):23-9.
20. Linder A, Fjell C, Levin A, Walley KR, Russell JA, Boyd JH. Small acute increases in serum creatinine are associated with decreased long-term survival in the critically ill. American journal of respiratory and critical care medicine. 2014 May 1;189(9):1075-81.
21. Jain S, Sinha S, Sharma SK, Samantaray JC, Aggrawal P, Vikram NK, Biswas A, Sood S, Goel M, Das M, Vishnubhatla S. Procalcitonin as a prognostic marker for sepsis: a prospective observational study. BMC research notes. 2014 Dec;7:1-7.
22. Giamarellos-Bourboulis EJ, Tsangaris I, Kanni T, Mouktaroudi M, Pantelidou I, Adamis G, Atmatzidis S, Chrisofos M, Evangelopoulou V, Frantzeskaki F, Giannopoulos P. Procalcitonin as an early indicator of outcome in sepsis: a prospective observational study. Journal of Hospital Infection. 2011 Jan 1;77(1):58-63.
23. Kim RY, Ng AM, Persaud AK, Furmanek SP, Kothari YN, Price JD, Wiemken TL, Saad MA, Guardiola JJ, Cavallazzi RS. Antibiotic timing and outcomes in sepsis. The American journal of the medical sciences. 2018 Jun 1;355(6):524-9.
24. Zhang Z, Bokhari F, Guo Y, Goyal H. Prolonged length of stay in the emergency department and increased risk of hospital mortality in patients with sepsis requiring ICU admission. Emerg
25. Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL. Serial evaluation of the SOFA score to predict outcome in critically ill patients. Jama. 2001 Oct 10;286(14):1754-8.
26. Liu Z, Meng Z, Li Y, Zhao J, Wu S, Gou S, Wu H. Prognostic accuracy of the serum lactate level, the SOFA score and the qSOFA score for mortality among adults with Sepsis. Scandinavian journal of trauma, resuscitation and emergency medicine. 2019 Dec;27:1-0.
27. Benmachiche M, Marques-Vidal P, Waeber G, Mean M. In-hospital mortality is associated with high NT-proBNP level. PLoS One. 2018 Nov 8;13(11):e0207118.
28. Brueckmann M, Huhle G, Lang S, Haase KK, Bertsch T, Weiß C, Kaden JJ, Putensen C, Borggrefe M, Hoffmann U. Prognostic value of plasma N-terminal pro-brain natriuretic peptide in patients with severe sepsis. Circulation. 2005 Jul 26;112(4):527-34.