Skip to main content
Download PDF

Novel cortisol trajectory sub-phenotypes in sepsis

Critical Care volume 28, Article number: 290 (2024) Cite this article

Abstract

Background

Sepsis is a heterogeneous syndrome. This study aimed to identify new sepsis sub-phenotypes using plasma cortisol trajectory.

Methods

This retrospective study included patients with sepsis admitted to the intensive care unit of Zhongshan Hospital Fudan University between March 2020 and July 2022. A group-based cortisol trajectory model was used to classify septic patients into different sub-phenotypes. The clinical characteristics, biomarkers, and outcomes were compared between sub-phenotypes.

Results

A total of 258 patients with sepsis were included, of whom 186 were male. Patients were divided into two trajectory groups: the lower-cortisol group (n = 217) exhibited consistently low and slowly declining cortisol levels, while the higher-cortisol group (n = 41) showed relatively higher levels in comparison. The 28-day mortality (65.9% vs.16.1%, P < 0.001) and 90-day mortality (65.9% vs. 19.8%, P < 0.001) of the higher-cortisol group were significantly higher than the lower-cortisol group. Multivariable Cox regression analysis showed that the trajectory sub-phenotype (HR = 5.292; 95% CI 2.218–12.626; P < 0.001), APACHE II (HR = 1.109; 95% CI 1.030–1.193; P = 0.006), SOFA (HR = 1.161; 95% CI 1.045–1.291; P = 0.006), and IL-1β (HR = 1.001; 95% CI 1.000–1.002; P = 0.007) were independent risk factors for 28-day mortality. Besides, the trajectory sub-phenotype (HR = 4.571; 95% CI 1.980–10.551; P < 0.001), APACHE II (HR = 1.108; 95% CI 1.043–1.177; P = 0.001), SOFA (HR = 1.270; 95% CI 1.130–1.428; P < 0.001), and IL-1β (HR = 1.001; 95% CI 1.000–1.001; P = 0.015) were also independent risk factors for 90-day mortality.

Conclusion

This study identified two novel cortisol trajectory sub-phenotypes in patients with sepsis. The trajectories were associated with mortality, providing new insights into sepsis classification.

Introduction

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection [1]. The course of sepsis is highly heterogeneous due to different causal agents, patient conditions, comorbidities, and treatments, resulting in highly variable patient outcomes with the same treatments in different septic patients [2]. Consequently, identifying patient subgroups precisely is crucial to optimize sepsis treatment and improve prognosis.

Despite significant advances in modern medicine, accurately assessing the severity and effectively predicting the prognosis of sepsis remains a major challenge. Sepsis is a dynamic condition that exhibits different biological responses over minutes to hours [3, 4], and static laboratory tests often fail to recognize sepsis sub-phenotypes that evolve [5, 6]. Group-based trajectory modeling (GBTM) has become an important topic in the field of critical care as a new phenotypic grouping tool that can be used to analyze potential categories using changes in indicators in the time dimension [7,8,9]. Recent studies on sepsis sub-phenotypes utilizing the GBTM approach have incorporated routinely accessible clinical variables, such as inflammation factors [10] and vital signs [7, 9].

Sepsis is associated with an abnormal response of the hypothalamic–pituitary–adrenal (HPA) axis [11]. Cytokines (including IL-1, IL-6, and TNF-α) stimulate the HPA axis, increasing the secretion of corticotropin-releasing hormone (CRH) and adrenocorticotropic hormone (ACTH) [12, 13]. The body can counteract the inflammatory response by activating the HPA axis to increase cortisol levels [14, 15], which has immunosuppressive effects, decreases cytokine concentrations, and maintains vascular tension, catecholamine sensitivity, and endothelial integrity [15]. Nevertheless, with sepsis progression, the typical cortisol secretion and feedback mechanisms of the HPA axis can be disturbed, and critical illness-related corticosteroid insufficiency (CIRCI) may occur, leading to uncontrolled inflammatory responses and disturbed circadian rhythms [12]. Therefore, in patients with sepsis, cortisol levels are highly heterogeneous [14].

The current understanding regarding whether the cortisol trajectory in septic patients can be used to identify sepsis sub-phenotypes remains unclear. This study aimed to investigate a novel plasma cortisol trajectory to identify new sepsis sub-phenotypes.

Methods

Study design and patients

This retrospective study included patients diagnosed with sepsis at the intensive care unit (ICU) of Zhongshan Hospital Fudan University between March 2020 and July 2022. Sepsis was diagnosed according to the third international consensus definition of sepsis [16]. The study was approved by the Ethics Committee of Zhongshan Hospital, Fudan University (#B2021-501R). The requirement for informed consent was waived by the committee. The exclusion criteria were as follows: (1) < 18 years old, (2) ICU stay < 72 h, (3) pregnancy, (4) history of corticosteroid treatment or with drugs that affect adrenal function during hospitalization or within 1 year period before admission, or (5) pituitary or adrenal gland abnormalities.

Data collection and outcome

The demographic and clinical characteristics of the patients, including age, sex, acute physiology and chronic health status score II (APACHE II) [17], SOFA score [18], body mass index (BMI), the source of infection at ICU admission, and comorbidities were collected within the first 24 h after admission to the ICU. Laboratory data were collected from the patient charts, including blood routine examination, C-reaction protein (CRP), procalcitonin (PCT), tumor necrosis factor-α (TNF-α), interleukins (ILs), albumin (Alb), creatinine (Cr), alanine transaminase (ALT), aspartate aminotransferase (AST). All biochemical measurements were performed at the Clinical Chemistry Laboratory of the study hospital.

Routine twice-daily cortisol screenings were conducted at 08:00 and 16:00 for all sepsis patients in our ICU to monitor potential variations in cortisol levels and rhythms. To minimize disturbances to patients' sleep, midnight data were not collected. The cortisol levels measured at 08:00 and 16:00 during the first 3 days following sepsis diagnosis were defined as T1 to T6, respectively. Specifically, T1 refers to 08:00 on the first day after sepsis diagnosis, T2 to 16:00 on the same day, T3 to 08:00 on the second day, T4 to 16:00 on the second day, T5 to 08:00 on the third day, and T6 to 16:00 on the third day. All patients were followed for 90 days from ICU admission. The primary outcome of the present study was the 28-day mortality. The secondary outcomes were the 90-day mortality, the duration of mechanical ventilation, and the length of ICU and hospital stays.

Statistical analysis

The GBTM to establish the distinct trajectories of cortisol was conducted using the TRAJ package for Stata 17.0 (StataCorp LP, College Station, TX, USA) [19]. This model [20] delineates the dynamics between cortisol concentrations and time across various trajectories. The calculated likelihoods were termed posterior probabilities of group affiliation. Based on the array of cortisol measurements, the patients were categorized into the trajectory sub-phenotype for which they had the highest probability of group membership. The determination of the ideal count of latent groups within a preferred grouping was influenced by multiple factors [7]. Firstly, enhanced model fit correlates with increased log-likelihood ratios and higher entropy values, the latter maintaining a threshold of no less than 0.7. Secondly, optimal model fit was associated with reduced values of the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and the sample-size corrected BIC [21]. Thirdly, a critical criterion was that each identified group’s average posterior probability (AvePP) had to meet or exceed 70%. Fourthly, each identified category constituted at least 1% of the overall sample volume. Lastly, the ability of the model to be clinically interpretable was an essential consideration.

The statistical analyses were all performed using SPSS 22.0 (IBM, Armonk, NY, USA), R 4.2.1 (Thermo Fisher Scientific, Waltham, MA, USA), or GraphPad Prism 8.0.2 (GraphPad Software Inc., San Diego, CA, USA). Normally distributed continuous variables were described as means ± standard deviations (SDs). Medians and interquartile ranges (IQRs) were used for the continuous variables that were non-normally distributed. Depending on their distributions, between-group comparisons were performed using parametric (Student’s t-test) or non-parametric tests (Mann–Whitney U-test). Categorical variables were described as n (%), with intergroup comparisons conducted using the chi-squared test or Fisher’s exact test. Univariable and multivariable Cox proportional hazard models were constructed to identify the factors associated with 28- and 90-day mortalities. The variables that were significant in the univariable were included in the multivariable Cox regression analyses. The 28- and 90-day cumulative survival probability was examined using the Kaplan–Meier method. Subgroup analysis was conducted for sex, age, BMI, with or without high blood pressure (HBP), diabetes mellitus (DM), and coronary artery disease (CAD). Multiple imputation (MI) based on five imputed data sets was used to address the missing data with SPSS 22.0 (IBM, Armonk, NY, USA). Statistical significance was determined by two-sided P < 0.05.

Results

From March 2020 to July 2022, 3986 patients were admitted to the ICU, of which 457 were diagnosed with sepsis. According to the exclusion criteria, 199 patients were excluded: 11 were < 18 years old, 72 stayed in the ICU < 72 h, 18 were pregnant, 72 had corticosteroid treatment or drugs that affect adrenal function, and 26 had a history of pituitary or adrenal gland abnormalities. Therefore, 258 patients were included in this study. The missing data rate was 1.04% (Supplementary Table 1).

Cortisol trajectories and model adequacy

Two cortisol trajectories were identified using a GBTM approach. The lower-cortisol group included 217 patients (84%), and the higher-cortisol group consisted of 41 patients (16%). The cortisol levels in the lower-cortisol group persisted at a relatively low starting level with a slow and steadily decreasing trend. On the other hand, the cortisol levels in the higher-cortisol group were elevated noticeably at the beginning (> 1500 nmol/L), had sustained high levels at T2-4, and a rapid decrease at T5-6 but remained significantly higher than in the lower-cortisol group (Fig. 1). Model adequacy was evaluated. The entropy value was 0.906 (expected > 0.7 for adequacy). The levels of AIC and BIC were the lowest. The AvePP of the two trajectories were 98% and 92%, respectively (expected > 70% for adequacy). Finally, the ability of the trajectory model was clinically interpretable.

Fig. 1
figure 1

Group-based trajectory modeling of cortisol levels in septic patients. The dotted gray line represents the 95% confidence interval

Full size image

Clinical characteristics and biomarker comparison

No significant differences were observed between the trajectory groups in age, sex, BMI, comorbidity, blood routine, liver or kidney function (all P > 0.05). Compared with the lower-cortisol group, patients in the higher-cortisol group showed significantly higher SOFA scores (P < 0.001) and APACHE II scores (P = 0.001), PCT (P = 0.004), TNF-α (P < 0.001), IL-6 (P = 0.007), IL-1β (P = 0.004), IL-10 (P < 0.001), and IL-8 (P = 0.003) (Table 1).

Table 1 Characteristics of the patients according to the trajectory
Full size table

Mortality and patient outcomes

There were no significant differences in the duration of mechanical ventilation, the length of ICU, or hospital stay between the two groups. The 28-day mortality (65.9% vs. 16.1%, HR = 5.140; 95% CI 2.426–10.890, P < 0.001) and 90-day mortality (65.9% vs. 19.8%, HR = 5.037; 95% CI 2.388–10.620, P < 0.001) of the patients in the higher-cortisol group was significantly higher than in the lower-cortisol group (Table 1 and Fig. 2).

Fig. 2
figure 2

The 28-day survival probability curve (A) and the 90-day survival probability curve (B) of patients between the two sub-phenotypes

Full size image

Independent influence factors of 28-day and 90-day mortality

Univariable analysis revealed that the trajectory sub-phenotype was significantly associated with both 28-day mortality (HR = 6.152; 95% CI 3.698–10.235; P < 0.001) and 90-day mortality (HR = 5.129; 95% CI 3.156–8.335; P < 0.001). Other significant factors for 28-day mortality included APACHE II, SOFA, MV duration, length of hospital stay, IL-1β, and IL-10. Significant factors for 90-day mortality included BMI, APACHE II, SOFA, MV duration, ICU days, and IL-1β (all P < 0.05) (Table 2). Multivariable Cox regression analysis showed that the trajectory sub-phenotype (HR = 5.292; 95% CI 2.218–12.626; P < 0.001), APACHE II (HR = 1.109; 95% CI 1.030–1.193; P = 0.006), SOFA (HR = 1.161; 95% CI 1.045–1.291; P = 0.006), and IL-1β (HR = 1.001; 95% CI 1.000–1.002; P = 0.007) were independent risk factors for 28-day mortality. Besides, the trajectory sub-phenotype (HR = 4.571; 95% CI 1.980–10.551; P < 0.001), APACHE II (HR = 1.108; 95% CI 1.043–1.177; P = 0.001), SOFA (HR = 1.270; 95% CI 1.130–1.428; P < 0.001), and IL-1β (HR = 1.001; 95% CI 1.000–1.001; P = 0.015) were also independent risk factors for 90-day mortality (Table 2).

Table 2 Cox univariate and multivariate regression of 28- and 90-day mortality
Full size table

Subgroup analysis

The subgroup analyses included all 258 patients revealed that the cortisol trajectory sub-phenotype remained a significant risk factor for 28- and 90-day mortality after stratification by sex, age, BMI, and HBP (all P < 0.05) (Fig. 3). A gender subgroup analysis was performed to mitigate concerns regarding generalizability due to gender bias. In the male and female subgroups, the cortisol trajectory subtype grouping still exhibited statistically significant differences in HR for 28- and 90-day mortality, unaffected by gender (Fig. 3).

Fig. 3
figure 3

Forest plots of subgroup analysis for 28-day mortality (A) and 90-day mortality (B) of the cortisol trajectory group. BMI, Body mass index; HBP, High blood pressure; DM, Diabetes mellitus; CAD, Coronary artery disease

Full size image

Discussion

The study identified two distinct sepsis sub-phenotypes based on plasma cortisol trajectory, revealing that patients in the higher-cortisol group had significantly higher 28-day and 90-day mortality rates compared to those in the lower-cortisol group. Cortisol trajectory sub-phenotype was identified as an independent predictor of mortality, highlighting their potential utility in guiding treatment strategies in sepsis management. This study is the first to use GBTM to identify two distinct cortisol trajectory sub-phenotypes in septic patients. This finding provides new insights into the dynamic role of cortisol in sepsis.

Previous studies have shown that critically ill patients, especially those with sepsis, have a disruption of the normal organization of the HPA axis, which may result in an excessive increase in plasma cortisol and correlate with disease severity [12, 22]. In addition, increased blood bile acid concentrations in critically ill patients may inhibit cortisol metabolizing enzymes, leading to increased cortisol [23]. A downregulation of glucocorticoid receptors (GR) activity can also result in glucocorticoid resistance, which may cause an increase in sepsis mortality [24]. Cortisol rhythms are disrupted, and circadian rhythms are lost in patients with sepsis [14]. Still, the exact rhythmic trajectory is unknown, and few studies were reported about cortisol trajectory sub-phenotypes in sepsis. The literature suggests that the need for longitudinal data trajectories may be more helpful in effectively differentiating sub-phenotypes rather than values at individual time points for the more complex factors in patients with sepsis [7, 25].

In the present study, the higher-cortisol group was associated with a significantly higher 28- and 90-day mortality compared with the lower-cortisol group. The present study also showed that the trajectory group, APACHE II score, SOFA score, and IL-1β were independent risk factors for 28- and 90-day mortality. The high mortality is consistent with the high inflammatory state that leads to excessive activation of the necroinflammatory cell death pathways in multiple organs [26, 27] or may be due to unresolved inflammation or immune suppression [28, 29]. Meanwhile, cytokines involved in immune dysregulation play an important role in patients with sepsis, among which IL-1β acts as a pro-inflammatory cytokine, causing organ destruction in sepsis and activating inflammatory vesicles (including NLRP3 in macrophages and endothelial cells) [30]. In addition, macrophage-produced IL-1β destabilizes the vasculature, which further induces inflammatory progression and tissue damage [31]. It is also possible that cytokines directly affect the adrenal glands and regulate cortisol secretion [32, 33], which may be responsible for persistently high cortisol levels. Further understanding of the pathophysiology and biological mechanisms underlying these sub-phenotypes could be crucial for developing the next generation of management strategies for sepsis [34]. Notably, in our study's higher-cortisol group, no deaths were observed between 28 and 90 days of follow-up. The absence of mortality events after day 28 in the higher-cortisol group may be attributed to the following reasons. First, the patients in the higher-cortisol group were generally more critically ill, resulting in a higher risk of mortality within the initial 28 days, with a mortality rate of 65.9%. Consequently, the most critically ill patients tended to succumb early, leading to no additional deaths between days 28 and 90 in this group [35,36,37]. Second, the small sample size could be involved, as mentioned in the limitations.

The HR for the cortisol trajectory sub-phenotype in the present study was significantly higher than the HR of the APACHE II score, which is known to predict the prognosis of critically ill patients and patients with sepsis [38, 39]. It suggests that the cortisol trajectory group may be more valuable for discriminating the differences in the heterogeneity of sepsis and may serve as a reference for the classification of future sepsis treatments. Current sepsis management guidelines [1, 40,41,42] do not provide recommendations for precise treatment based on cortisol levels, underscoring the potential for further research into cortisol trajectories in sepsis. For instance, we hypothesize that patients with a high cortisol trajectory might not require corticosteroid therapy, at least not in the early stages, whereas those with a low cortisol trajectory may benefit from earlier and more aggressive corticosteroid treatment. This study may offer some initial insights into the development of more precise, individualized corticosteroid therapy strategies for heterogeneous conditions. Still, how the management of the two subtypes could be optimized requires further studies. The subgroup analysis showed that high cortisol levels (i.e., the higher-cortisol group) indicated high mortality regardless of sex, age, BMI, or HBP, suggesting that the cortisol trajectory grouping is a potentially superior sepsis subgroup indicator. Nevertheless, accurately assessing the severity and effectively predicting the prognosis of sepsis remains a major challenge. The significance of the identified sub-phenotypes lies in the potential need to adjust corticosteroid therapy strategies based on different cortisol trajectories to optimize the therapeutic outcomes.

Currently, there are various methods for classifying sepsis trajectory sub-phenotypes. Trajectories based on body temperature (hyperthermic, slow resolvers; hyperthermic, fast resolvers; normothermic; hypothermic) have been shown to distinguish between sepsis sub-phenotypes and are associated with mortality, with the highest mortality observed in hypothermic patients [8, 9, 25]. Bhavani et al. [7] proposed four sepsis sub-phenotypes based on vital sign trajectories, which were associated with different treatment responses. Xu et al. [34] established four trajectories based on SOFA scores: rapidly worsening, delayed worsening, rapidly improving, and delayed improving, all of which were associated with mortality. Additionally, another classification based on the frequency of infections before sepsis was linked to mortality [43]. This study is the first to identify trajectory sub-phenotype in sepsis patients from the perspective of cortisol, providing valuable insights into existing dynamic models. However, a comprehensive comparison of these different trajectories is still lacking, and direct comparisons among studies are challenging due to patient heterogeneity. Future research should aim to concurrently examine multiple trajectories within the same cohort to identify which provides the most accurate associations with patient outcomes.

This study has limitations. Firstly, it was a retrospective single-center study with a limited sample size, resulting in limited generalizability. Secondly, this study only included cortisol measurements twice daily, and the absence of midnight cortisol measurement may have biased the results. Thirdly, due to excluding patients treated with corticosteroids, there is a lack of exploration into the responsiveness to corticosteroid therapy during different cortisol trajectory sub-phenotype patients. Fourthly, patients with ICU stays of less than 72 h were not included, which may limit the generalizability of our conclusions. Therefore, future studies need to be prospective, with more rigorous designs, larger sample sizes, and more comprehensive time rhythm data to validate these results.

In conclusion, this study identified two distinct sepsis sub-phenotypes using plasma cortisol trajectory. The lower-cortisol group exhibited consistently low and slowly declining cortisol levels, while the higher-cortisol group showed relatively higher levels than the lower-cortisol group. A novel cortisol trajectory sub-phenotype was identified as an independent predictor of mortality, providing new insights into sepsis classification and highlighting its potential in treatment strategies for sepsis management.

Availability of data and materials

All data generated or analysed during this study are included in this published article.

References

  1. Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, Machado FR, McIntyre L, Ostermann M, Prescott HC, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med. 2021;47(11):1181–247.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Stanski NL, Wong HR. Prognostic and predictive enrichment in sepsis. Nat Rev Nephrol. 2020;16(1):20–31.

    Article  PubMed  Google Scholar 

  3. Cazalis MA, Lepape A, Venet F, Frager F, Mougin B, Vallin H, Paye M, Pachot A, Monneret G. Early and dynamic changes in gene expression in septic shock patients: a genome-wide approach. Intensive Care Med Exp. 2014;2(1):20.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Namas RA, Vodovotz Y. From static to dynamic: a sepsis-specific dynamic model from clinical criteria in polytrauma patients. Ann Transl Med. 2016;4(24):492.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Wong HR, Cvijanovich NZ, Anas N, Allen GL, Thomas NJ, Bigham MT, Weiss SL, Fitzgerald JC, Checchia PA, Meyer K, et al. Endotype transitions during the acute phase of pediatric septic shock reflect changing risk and treatment response. Crit Care Med. 2018;46(3):e242–9.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Sweeney TE, Shidham A, Wong HR, Khatri P. A comprehensive time-course-based multicohort analysis of sepsis and sterile inflammation reveals a robust diagnostic gene set. Sci Transl Med. 2015. https://doi.org/10.1126/scitranslmed.aaa5993.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Bhavani SV, Semler M, Qian ET, Verhoef PA, Robichaux C, Churpek MM, Coopersmith CM. Development and validation of novel sepsis subphenotypes using trajectories of vital signs. Intensive Care Med. 2022;48(11):1582–92.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Bhavani SV, Wolfe KS, Hrusch CL, Greenberg JA, Krishack PA, Lin J, Lecompte-Osorio P, Carey KA, Kress JP, Coopersmith CM, et al. Temperature trajectory subphenotypes correlate with immune responses in patients with sepsis. Crit Care Med. 2020;48(11):1645–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Bhavani SV, Carey KA, Gilbert ER, Afshar M, Verhoef PA, Churpek MM. Identifying novel sepsis subphenotypes using temperature trajectories. Am J Respir Crit Care Med. 2019;200(3):327–35.

    Article  PubMed  PubMed Central  Google Scholar 

  10. You SH, Kweon OJ, Jung SY, Baek MS, Kim WY. Patterns of inflammatory immune responses in patients with septic shock receiving vitamin C, hydrocortisone, and thiamine: clustering analysis in Korea. Acute Critical Care. 2023;38(3):286–97.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Annane D. The role of ACTH and corticosteroids for sepsis and septic shock: an update. Front Endocrinol (Lausanne). 2016;7:70.

    Article  PubMed  Google Scholar 

  12. Annane D, Pastores SM, Arlt W, Balk RA, Beishuizen A, Briegel J, Carcillo J, Christ-Crain M, Cooper MS, Marik PE, et al. Critical illness-related corticosteroid insufficiency (CIRCI): a narrative review from a multispecialty task force of the society of critical care medicine (SCCM) and the European society of intensive care medicine (ESICM). Intensive Care Med. 2017;43(12):1781–92.

    Article  PubMed  Google Scholar 

  13. Turnbull AV, Rivier CL. Regulation of the hypothalamic-pituitary-adrenal axis by cytokines: actions and mechanisms of action. Physiol Rev. 1999;79(1):1–71.

    Article  CAS  PubMed  Google Scholar 

  14. Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med. 2003;348(8):727–34.

    Article  CAS  PubMed  Google Scholar 

  15. Wasyluk W, Wasyluk M, Zwolak A. Sepsis as a pan-endocrine illness—endocrine disorders in septic patients. J Clin Med. 2021;10(10):2075.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche JD, Coopersmith CM, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA. 2016;315(8):801–10.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985;13(10):818–29.

    Article  CAS  PubMed  Google Scholar 

  18. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22(7):707–10.

    Article  CAS  PubMed  Google Scholar 

  19. Chen HY, Elmer J, Zafar SF, Ghanta M, Moura Junior V, Rosenthal ES, Gilmore EJ, Hirsch LJ, Zaveri HP, Sheth KN, et al. Combining transcranial doppler and EEG data to predict delayed cerebral ischemia after subarachnoid hemorrhage. Neurology. 2022;98(5):e459–69.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Bhavani SV, Xiong L, Pius A, Semler M, Qian ET, Verhoef PA, Robichaux C, Coopersmith CM, Churpek MM. Comparison of time series clustering methods for identifying novel subphenotypes of patients with infection. J Am Med Inform Assoc. 2023;30(6):1158–66.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Eriksson J, Nelson D, Holst A, Hellgren E, Friman O, Oldner A. Temporal patterns of organ dysfunction after severe trauma. Crit Care. 2021;25(1):165.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Widmer IE, Puder JJ, Konig C, Pargger H, Zerkowski HR, Girard J, Muller B. Cortisol response in relation to the severity of stress and illness. J Clin Endocrinol Metab. 2005;90(8):4579–86.

    Article  CAS  PubMed  Google Scholar 

  23. Jenniskens M, Langouche L, Vanwijngaerden YM, Mesotten D, Van den Berghe G. Cholestatic liver (dys)function during sepsis and other critical illnesses. Intensive Care Med. 2016;42(1):16–27.

    Article  CAS  PubMed  Google Scholar 

  24. Fowler C, Raoof N, Pastores SM. Sepsis and adrenal insufficiency. J Intensive Care Med. 2023;38(11):987–96.

    Article  PubMed  Google Scholar 

  25. Yehya N, Fitzgerald JC, Hayes K, Zhang D, Bush J, Koterba N, Chen F, Tuluc F, Teachey DT, Balamuth F, et al. Temperature trajectory sub-phenotypes and the immuno-inflammatory response in pediatric sepsis. Shock. 2022;57(5):645–51.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Schenck EJ, Ma KC, Price DR, Nicholson T, Oromendia C, Gentzler ER, Sanchez E, Baron RM, Fredenburgh LE, Huh JW, et al. Circulating cell death biomarker TRAIL is associated with increased organ dysfunction in sepsis. JCI Insight. 2019. https://doi.org/10.1172/jci.insight.127143.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Linkermann A. Death and fire-the concept of necroinflammation. Cell Death Differ. 2019;26(1):1–3.

    Article  PubMed  Google Scholar 

  28. Cao C, Yu M, Chai Y. Pathological alteration and therapeutic implications of sepsis-induced immune cell apoptosis. Cell Death Dis. 2019;10(10):782.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Mira JC, Gentile LF, Mathias BJ, Efron PA, Brakenridge SC, Mohr AM, Moore FA, Moldawer LL. Sepsis pathophysiology, chronic critical illness, and persistent inflammation-immunosuppression and catabolism syndrome. Crit Care Med. 2017;45(2):253–62.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Evavold CL, Kagan JC. Inflammasomes: threat-assessment organelles of the innate immune system. Immunity. 2019;51(4):609–24.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Xiong S, Hong Z, Huang LS, Tsukasaki Y, Nepal S, Di A, Zhong M, Wu W, Ye Z, Gao X, et al. IL-1beta suppression of VE-cadherin transcription underlies sepsis-induced inflammatory lung injury. J Clin Invest. 2020;130(7):3684–98.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Engstrom L, Rosen K, Angel A, Fyrberg A, Mackerlova L, Konsman JP, Engblom D, Blomqvist A. Systemic immune challenge activates an intrinsically regulated local inflammatory circuit in the adrenal gland. Endocrinology. 2008;149(4):1436–50.

    Article  PubMed  Google Scholar 

  33. Kanczkowski W, Sue M, Zacharowski K, Reincke M, Bornstein SR. The role of adrenal gland microenvironment in the HPA axis function and dysfunction during sepsis. Mol Cell Endocrinol. 2015;408:241–8.

    Article  CAS  PubMed  Google Scholar 

  34. Xu Z, Mao C, Su C, Zhang H, Siempos I, Torres LK, Pan D, Luo Y, Schenck EJ, Wang F. Sepsis subphenotyping based on organ dysfunction trajectory. Crit Care. 2022;26(1):197.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Rivas M, Motes A, Ismail A, Yang S, Sotello D, Arevalo M, Vutthikraivit W, Suchartlikitwong S, Carrasco C, Iwuji K, et al. Characteristics and outcomes of patients with sepsis who had cortisol level measurements or received hydrocortisone during their intensive care unit management: a retrospective single center study. SAGE Open Med. 2023;11:20503121221146908.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Sam S, Corbridge TC, Mokhlesi B, Comellas AP, Molitch ME. Cortisol levels and mortality in severe sepsis. Clin Endocrinol (Oxf). 2004;60(1):29–35.

    Article  CAS  PubMed  Google Scholar 

  37. De Castro R, Ruiz D, Lavin BA, Lamsfus JA, Vazquez L, Montalban C, Marcano G, Sarabia R, Paz-Zulueta M, Blanco C, et al. Cortisol and adrenal androgens as independent predictors of mortality in septic patients. PLoS ONE. 2019;14(4): e0214312.

    Article  PubMed  PubMed Central  Google Scholar 

  38. 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. Crit Care. 2020;24(1):239.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Liu B, Chen YX, Yin Q, Zhao YZ, Li CS. Diagnostic value and prognostic evaluation of Presepsin for sepsis in an emergency department. Crit Care. 2013;17(5):R244.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Egi M, Ogura H, Yatabe T, Atagi K, Inoue S, Iba T, Kakihana Y, Kawasaki T, Kushimoto S, Kuroda Y, et al. The Japanese clinical practice guidelines for management of sepsis and septic shock 2020 (J-SSCG 2020). J Intensive Care. 2021;9(1):53.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Rello J, van Engelen TSR, Alp E, Calandra T, Cattoir V, Kern WV, Netea MG, Nseir S, Opal SM, van de Veerdonk FL, et al. Towards precision medicine in sepsis: a position paper from the European Society of Clinical Microbiology and Infectious Diseases. Clin Microbiol Infect Off Publ Eur Soc Clin Microbiol Infect Dis. 2018;24(12):1264–72.

    CAS  Google Scholar 

  42. Yu C, Yanfen C, Ying D, Bangjiang F, Minghua L, Zhongqiu L, Yiming L, Shinan N, Chuanyun Q, Yingping T, China sepsis/septic shock emergency treatment guidelines. J Clin Emerg Treat 2018.

  43. Lin HY, Hsiao FY, Huang ST, Chen YC, Lin SW, Chen LK. Longitudinal impact of distinct infection trajectories on all-cause mortality of older people in Taiwan: a retrospective, nationwide, population-based study. Lancet Healthy longev. 2023;4(9):e508–16.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

Clinical Research Funding of Zhongshan Hospital Fudan University (ZSLCYJ202339).

Author information

Author notes

  1. Fei Leng and Zhunyong Gu are co-first authors.

Authors and Affiliations

  1. Department of Critical Care Medicine, Zhongshan Hospital, Fudan University, Shanghai, 200032, China

    Fei Leng, Zhunyong Gu, Simeng Pan, Shilong Lin, Xu Wang, Ming Zhong & Jieqiong Song

Authors
  1. Fei Leng
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Zhunyong Gu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Simeng Pan
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Shilong Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Xu Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Ming Zhong
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jieqiong Song
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Leng Fei and Gu Zhunyong carried out the studies, participated in collecting data, and drafted the manuscript. Pan Simeng, Lin Shilong and Wang Xu performed the statistical analysis and participated in its design. Song Jieqiong and Zhong Ming participated in acquisition, analysis, or interpretation of data and draft the manuscript. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Ming Zhong or Jieqiong Song.

Ethics declarations

Ethics approval and consent to participate

All procedures were performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. This study was approved by the Ethics Committee of the Zhongshan Hospital Fudan University [B2021-501R]. The requirement for individual Informed consent was waived by the Ethics Committee of the Zhongshan Hospital Fudan University because of the retrospective nature of the study. The study was carried out in accordance with the applicable guidelines and regulations.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Leng, F., Gu, Z., Pan, S. et al. Novel cortisol trajectory sub-phenotypes in sepsis. Crit Care 28, 290 (2024). https://doi.org/10.1186/s13054-024-05071-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13054-024-05071-2

Keywords

Critical Care

ISSN: 1364-8535