Analysis of the prevalence and risk factors of post-intensive-care syndrome and post-sepsis syndrome in survivors of sepsis

Abstract

Objective

In the intensive care unit (ICU), sepsis is a leading cause of mortality, and survivors frequently experience serious long-term sequelae. In addition to identifying risk variables influencing post-sepsis syndrome (PSS) and post-intensive care syndrome (PICS), this study sought to present comprehensive data on the three-month prognostic and functional outcomes of sepsis patients in the ICU—a demographic frequently overlooked in long-term recovery discourse.

Methods

A prospective cohort study was conducted in the intensive care units of Jining Medical University Affiliated Hospital between April 2024 and July 2025, enrolling patients diagnosed with sepsis or septic shock. Follow-up assessments were conducted one and three months post-discharge via telephone or outpatient visits. The frequency of each dimension in PICS and PSS was investigated, alongside patients’ health-related quality of life. Risk factors for the two syndromes were analyzed using multivariable logistic regression and stepwise multiple linear regression.

Results

Sepsis survivors included in the one- and three-month follow-ups totaled 147 and 132, respectively. PICS and PSS had overall respective prevalence rates of 93.2–98.6% and 61.4–77.6%. Multivariable analyses revealed that body mass index (BMI), red blood cell (RBC) count, partial pressure of oxygen (PO₂), and globulin levels were independent predictors for PSS. For PICS severity, Glasgow Coma Scale (GCS) score, venous thromboembolism (VTE) risk score, bicarbonate (HCO₃^-) levels, and specific infection sites were identified as independent predictors. Sepsis survivors’ quality of life improves over the three months following discharge; however, PICS and PSS maintain a high prevalence, with a notable symptom overlap between the two syndromes. Clinicians must be mindful of specific in-hospital risk factors to tailor post-discharge care.

Conclusion

This study highlights the profound and persistent risks for ICU sepsis survivors. The results indicate that early multidisciplinary intervention is necessary, which may potentially reduce long-term sequelae and improve recovery trajectories.

Keywords

follow-up studies intensive care units post-intensive care syndrome morbidity post-sepsis syndrome sepsis

Introduction

Sepsis remains a formidable systemic inflammatory response triggered by infection, exacting a massive global toll.¹ Septic shock, its most severe form, is marked by severe cellular, metabolic, and circulatory abnormalities that significantly raise the chance of death.² Multiple organ failure and substantial mortality rates are observed in intensive care units (ICUs) patients with sepsis and septic shock.³ Recent epidemiological analyses have demonstrated an increase in the prevalence of sepsis and septic shock. In 2017, there were approximately 49 million sepsis cases and 11 million sepsis-related deaths globally.⁴ Furthermore, the advances in critical care medicine and sepsis care induced by the Surviving Sepsis Campaign guidelines have improved survival among patients with sepsis.⁵ As sepsis survival has improved over time, the population of patients who live beyond the acute phase has expanded progressively.^6,7 Many survivors have significant comorbidities that may be life limiting or significantly impair quality of life.⁵ Hence, mapping the long-term trajectories of these survivors is not merely an academic exercise—it is a clinical imperative.

As more sepsis survivors return to the community, the burden of illness is shifting from acute care hospitals to chronic long-term care settings.⁸ They could encounter restrictions and long-term consequences that significantly affect their health-related quality of life (HRQoL), as well as their physical, mental, and cognitive well-being.⁹ This is often referred to as the “post-sepsis syndrome” (PSS).¹⁰ Studies have shown that survivors of critical illness, including sepsis, have similarly poor quality of life.¹¹ Consequently, recent studies have investigated the prevalence of Post-intensive care syndrome (PICS) in sepsis survivors.^12,13 PICS can bring a heavy burden to patients’ lives and damage their long-term prognosis.^12,14 Both PICS and PSS involve cognitive, psychological, and physical impairments, resulting in significant overlap in clinical manifestations. However, it remains unclear whether these are truly distinct pathologies or merely contextual manifestations of a universal post-critical illness dysfunction.^15,16 Therefore, we wish to explore their relationship. Furthermore, the existing literature addresses the cognitive, emotional, and physical impacts experienced by sepsis survivors.¹⁷ There is a significant gap in the exploration of potential in-hospital risk factors associated with an altered PICS and PSS. Not enough research has been done on certain in-hospital factors that influence these long-term results.

To address this critical gap, this study aims to investigate the prevalence and characteristics of PICS and PSS in survivors of sepsis and septic shock three months after hospital discharge, and to identify independent in-hospital predictors of these outcomes to inform strategies for improving long-term recovery.

Materials and methods

Study design

This prospective longitudinal cohort study was conducted in accordance with the Declaration of Helsinki (as revised in 2024), and its reporting follows the STROBE guidelines.¹⁸ Ethical approval was obtained from the local ethics commission on 22 November 2023 (approval number: 2023-11-C032), and the study was registered prior to publication (Chinese Clinical Trial Registry, CTR2500112184). All potential participants or their legal next of kin were informed about the study using a consent form. They were assured that their free will to participate (or to permit participation on the patient’s behalf) would be respected, that the patients would remain anonymous, and that refusal or withdrawal during the study would not be disadvantageous.

Patients and setting

Patients were consecutively enrolled from the ICUs of Jining Medical University Affiliated Hospital between April 2024 and July 2025. Those with a confirmed diagnosis of sepsis or septic shock according to the Sepsis-3 criteria were enrolled within 24 hours of ICU admission. Sepsis was defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, operationalized as an acute increase in the Sequential Organ Failure Assessment (SOFA) score of 2 points or more from baseline in the presence of suspected or documented infection.¹ Septic shock was defined as a subset of sepsis with persistent hypotension requiring vasopressor therapy to maintain mean arterial pressure ≥ 65 mmHg and having a serum lactate level > 2 mmol/L despite adequate fluid resuscitation.^1,2

Written informed consent was obtained from the patient or their legal next of kin prior to any study procedures. In many cases, consent was obtained from the legal next of kin because the study subjects were in a state of critical illness, such as sepsis or septic shock, and lacked the decisional capacity to provide written informed consent themselves at the time of enrollment (due to sedation or sepsis-associated encephalopathy). When patients regained decisional capacity, they were approached to provide their own consent to continue participation, ensuring full respect for their autonomy. Patients were then followed up for three months after discharge.

The exclusion criteria were as follows: age under 18 years, end-stage disease (malignant tumor with metastases, AIDS, end-stage renal or liver disease), refusal to participate in the study, and incomplete or missing data. Incomplete or missing data was defined as: (1) missing baseline characteristics that could not be obtained from medical records (such patients were excluded at enrollment and are not included in the initial cohort); or (2) loss to follow-up due to incorrect contact information, refusal, or non-response at 1 or 3 months. All patient details have been de-identified to ensure anonymity. No personally identifiable information is disclosed in this manuscript.

Data collection

We recorded the following in-hospital characteristics of patients admitted to the ICU within the first 24 hours including: age, gender, comorbidities, infection sites, SOFA score, and laboratory indicators. These variables represent baseline characteristics measured within the first 24 hours of ICU admission. In addition, the following cumulative ICU data were documented throughout the entire ICU stay: duration of mechanical ventilation, sedative use, analgesic use, and ICU length of stay. Follow-up data were collected using the same questionnaire at two time points: one and three months after discharge. Follow-up assessments at post-discharge included only questionnaires, and no laboratory tests were repeated at these time points. Two experienced graduate nursing students conducted follow-up visits with patients either by phone or in the outpatient department. The mortality assessment included cumulative all-cause mortality at 1 and 3 months after discharge. Patients who died during hospitalization and those who were lost to follow-up after discharge were not included in this part of the analysis.

Outcome measures

The primary outcomes were: the prevalence of PICS and the prevalence of PSS among survivors of sepsis. PICS was measured using the Healthy Aging Brain Care Monitor Self-Report (HABC-M SR), a self-report measure that covers the cognitive, physical, and psychological domains. Higher total scores indicate higher levels of PICS.¹⁹ The HABC-M SR Chinese version, utilized in this study, encompasses 19 items, with 7 items allocated to the cognitive domain, 6 items to the physical domain, and 6 items to the psychological domain. The total scale ranges from 1 to 57, with scores of 21, 18, and 18 indicating the worst possible outcome in the three domains, respectively. In line with the previous research,²⁰ the presence of PICS at each follow-up was defined as a HABC-M SR total score greater than 0 points. Additionally, to investigate factors associated with the severity of PICS symptoms, the continuous HABC-M SR total score was also used as an outcome in linear regression analyses.

The clinical diagnostic methods for PSS and PICS are similar.²¹ Universal scales are usually used to assess whether patients have experienced any new or worsening functional impairments in cognition, physiology or psychology in order to operationalise the presence of PSS for research purposes. The existence of any of the following circumstances was considered to be the occurrence of PSS,²² based on instruments widely validated in the context of long-term outcomes after sepsis and critical illness: (1) physical dysfunction,²³ Barthel index (BI) score < 60, the BI has been validated for functional assessment in critically ill patients at ICU discharge; (2) cognitive impairment,²⁴ as indicated by a Mini-Mental State Examination (MMSE) score < 24; (3) mental dysfunction,²⁴ Hospital Anxiety and Depression Scale (HADS) ≥ 8 points. The MMSE and HADS are well-established instruments widely used in sepsis survivor cohorts to assess cognitive and psychological outcomes.²⁴ HRQoL was evaluated using the Short Form-36 (SF-36) measure.²⁵

Sample size calculation

Based on the number of independent variables in the multiple regression analysis, the adjusted sample size required was 210 patients.

Statistical methods

Normality was tested using the Shapiro-Wilk test. For normally distributed data, means with standard deviations (SD) were used; otherwise, medians with interquartile range (IQR) were reported, and compared using the independent samples t-test (for normally distributed data) or the Mann-Whitney U test (for non-normally distributed data). Categorical variables were expressed as numbers and percentages, and compared using the Chi-square test or Fisher’s exact test. The 95% confidence interval of the proportion index was calculated using the Wilson method. A stepwise multiple linear regression analysis was employed to introduce variables with a P value less than 0.05, thereby obtaining the risk factors of PICS. To determine the independent risk factors of PSS, a multivariable analysis was conducted using a binary logistic regression model for PSS occurrence at 3 months post-discharge, incorporating clinically relevant variables. The odds ratio (OR) with a 95% confidence interval (CI) was used to assess the independent contributions of the significant factors. For comparisons between 1-month and 3-month follow-up in patients who completed both assessments (n = 132), the Wilcoxon signed-rank test was used for continuous variables and McNemar’s test for categorical variables, accounting for the paired nature of the data. Statistical tests were two-sided, and statistical significance was defined as a P value < 0.05. Non-parametric sensitivity analyses were performed to validate the robustness of the PICS findings. Spearman’s rank correlation was used for continuous variables, and the Mann-Whitney U test (for two groups) or Kruskal-Wallis H test (for three or more groups) was used for categorical variables (Supplementary Table S4). The collected data were analyzed using IBM SPSS Statistics version 26.0 and the Free Statistics analytic platform (version 2.2).

Result

Participants’ characteristics

From April 2024 to July 2025, a total of 228 patients who were diagnosed with sepsis or septic shock upon admission were screened. A total of 26 patients were excluded, with the specific reasons including: age less than 18 years (n = 3), expected to be transferred out of the ICU within 24 hours (n = 9), having severe underlying conditions (n = 8), and declined to participate (n = 6). A total of 202 patients were enrolled in the final cohort (Figure 1). Among them, 17 patients (8.4%) died in the hospital, and 185 patients (91.6%) were discharged alive. One month after discharge, 6 patients (3.2%) were lost to follow-up and 32 patients died. Based on this, the cumulative mortality rate at 1 month was 17.9% (32/179; 95% CI, 12.6%–24.2%). At 3 months after discharge, 9 patients (4.9%) were lost to follow-up, and 44 patients died. The cumulative mortality rate at 3 months was 25.0% (44/176, 95% CI: 18.8%–32.1%). The characteristics of the participants are summarized in Table 1, which also presents the baseline characteristics of survivors at the 1-month and 3-month follow-up assessments.

Figure 1.

Flowchart of the follow-up process for patients.

Table 1.

Baseline characteristics of all enrolled patients and survivors at 1-month and.

Variable	Enrolled patients	1-Month follow-up	3-Month follow-up
Variable	N = 202	N = 147	N = 132
Age (years)	68.3 ± 13.8	67.8 ± 14.3	67.0 ± 14.0
Female Sex, n (%)	80 (39.6)	59 (40.1)	53 (40.2)
BMI (kg/m ² )	22.7 ± 4.9	22.6 ± 4.7	22.6 ± 4.8
Marital status, n (%)
Married	166 (82.2)	120 (81.6)	110 (83.3)
Divorced	5 (2.5)	5 (3.4)	3 (2.3)
Unmarried	5 (2.5)	5 (3.4)	4 (3.0)
Widowed	26 (12.9)	17 (11.6)	15 (11.4)
Medical insurance type, n (%)
Medical insurance for residents	129 (63.9)	102 (69.4)	93 (70.5)
Employee health insurance	70 (34.7)	42 (28.6)	36 (27.3)
Out of pocket	3 (1.5)	3 (2.0)	3 (2.3)
Degree of education, n (%)
Primary school and below	57 (28.2)	43 (29.3)	37 (28.0)
Junior high school	103 (51.0)	77 (52.4)	69 (52.3)
High school	23 (11.4)	15 (10.2)	15 (11.4)
College	15 (7.4)	10 (6.8)	9 (6.8)
Bachelor’s degree or above	4 (2.0)	2 (1.4)	2 (1.5)
Length of the ICU stay (days)	5 (3, 10)	4 (3, 4)	4.5 (3, 9)
Length of hospital stay (days)	5 (3, 10)	11 (7, 18)	11 (8, 18)
Smoking history, n (%)	82 (40.6)	61 (41.5)	57 (43.2)
Alcohol use history, n (%)	72 (35.6)	54 (36.7)	50 (37.9)
Hypertension history, n (%)	89 (44.1)	67 (46.2)	62 (47.0)
Diabetes history, n (%)	54 (26.7)	39 (26.5)	37 (28.0)
Site of infection, n (%)
Abdominal infection	63 (31.2)	55 (37.4)	54 (40.9)
Pulmonary infection	109 (54.0)	73 (49.7)	65 (49.2)
Urinary tract infection	25 (12.4)	15 (10.3)	15 (11.4)
Infection in other sites	40 (19.8)	31 (21.4)	29 (21.5)
APACHE II score	22.4 ± 9.7	21.2 ± 10.1	20.6 ± 9.9
SOFA score	8.0 ± 4.2	7.2 ± 4.0	7.1 ± 3.9
GCS score	11.6 ± 4.5	12.1 ± 4.3	12.3 ± 4.2
VTE score	4.7 ± 2.3	4.4 ± 2.2	4.4 ± 2.2
Nutrition score	3.0 ± 1.4	2.9 ± 1.4	3.1 ± 1.2
Mechanical ventilation, n (%)	90 (44.6)	58 (39.5)	57 (43.2)
Mechanical ventilation duration (hours)	0.0 (0.0, 24.0)	0.0 (0.0, 14)	0.0 (0.0, 10.5)
Sedative use, n (%)	83 (41.1)	56 (38.1)	53 (40.2)
Analgesic use, n (%)	71 (35.1)	48 (32.7)	46 (34.8)
Norepinephrine use, n (%)	98 (48.5)	65 (44.2)	66 (50.0)
RBC count (×10 ¹² /L)	3.8 ± 0.9	3.9 ± 0.9	3.9 ± 0.9
WBC count (×10 ⁹ /L)	10.3 (6.3, 16.1)	9.8 (6.8, 15.3)	9.8 (7.1, 15.2)
Hb (g/L)	112.4 ± 26.9	114.5 ± 26.6	114.6 ± 26.9
RDW-SD (fL)	48.3 ± 9.6	47.0 ± 8.4	46.9 ± 7.5
RDW-CV (%)	14.2 (13.3, 15.6)	14.1 (13.2, 15.3)	14.1 (13.2, 15.3)
Lymphocyte count (×10 ⁹ /L)	0.7 (0.4, 1.1)	0.7 (0.4, 1.1)	0.7 (0.4, 1.1)
Neutrophil percentage (%)	79.8 ± 22.5	80.9 ± 20.8	81.0 ± 21.1
CRP (mg/L)	120.7 (54.1, 209.7)	131.3 (61.8, 232.5)	131.3 (60.8, 231.9)
PCT (ng/mL)	5.9 (0.4, 32.2)	7.5 (0.4, 38.9)	6.9 (0.4, 36.0)
ALB (g/L)	29.4 (25.8, 33.8)	30.2 (26.2, 34.5)	30.1 (26.1, 34.4)
GLB (g/L)	26.4 ± 6.2	26.5 ± 6.3	26.5 ± 5.8
TBil (μmol/L)	17.0 (11.8, 29.8)	17.0 (11.9, 32.4)	16.9 (12.2, 31.9)
Lac (mmol/L)	1.8 (1.0, 2.8)	1.4 (1.0, 2.7)	1.4 (1.0, 2.5)
PO ₂ (mmHg)	95.0 (71.0, 132.0)	95.0 (71.0, 128.0)	95.0 (71.0, 127.2)
HCO ₃ ^- (mmol/L)	21.8 ± 7.0	21.8 ± 7.1	21.9 ± 7.2
BE (mmol/L)	−2.2 (-6.4, 0.9)	−2.2 (−6.2, 0.7)	−2.2 (−6.4, 0.6)
AG (mmol/L)	14.5 ± 6.8	13.9 ± 6.2	13.7 ± 6.0
D-dimer (mg/L)	4.1 (2.1, 7.0)	3.8 (2.1, 6.9)	4.0 (2.0, 7.0)
IL-6 (pg/mL)	209.2 (41.3, 2234.3)	383.0 (50.9, 3357.7)	406.3 (50.5, 3391.7)
cTnT (ng/L)	11.0 (0.1, 66.9)	8.0 (0.0, 50.5)	8.0 (0.0, 52.2)

Unless otherwise specified, all data represent baseline characteristics measured within 24 hours of ICU admission, as detailed in the Data Collection section. Abbreviations: BMI: Body Mass Index; APACHE II: Acute Physiology and Chronic Health Evaluation II; SOFA: Sequential Organ Failure Assessment; GCS: Glasgow Coma Scale; VTE: Venous Thromboembolism; RBC: Red Blood Cell; WBC: White Blood Cell; Hb: Hemoglobin; RDW-SD: Red Cell Distribution Width-Standard Deviation; RDW-CV: Red Cell Distribution Width-Coefficient of Variation; PO₂, Partial Pressure of Oxygen; CRP: C-Reactive Protein; PCT: Procalcitonin; ALB: Albumin; GLB: Globulin; TBil: Total Bilirubin; Lac: Lactate; HCO₃^-: Bicarbonate; BE: Base Excess; AG: Anion Gap; IL-6: Interleukin-6; cTnT: Cardiac Troponin T.

Primary outcome

Table 2 summarizes the longitudinal changes in PICS, PSS and quality of life outcomes between 1- and 3-month follow-up. The prevalence of PICS declined from 98.6% at 1 month to 93.2% at 3 months (P = 0.039), and the total HABC-M SR score decreased from a median of 21 (11, 46) to 8 (4, 33) (P < 0.001). For PSS, at 1 month, the median MMSE score was 20 (6, 24), BI score was 80 (10, 90), and HADS-Anxiety score was 8 (6, 11), with a PSS prevalence of 77.6%. By 3 months, the median MMSE score increased to 23 (10, 26), BI to 100 (36.3, 100), HADS-Anxiety decreased to 5 (4, 8), and HADS-Depression to 6 (4, 8.8), with the PSS prevalence declining to 61.4% (P = 0.014). Figure 2 shows the prevalence of PSS and functional impairment overlap. At 1 month, 35.3% of survivors experienced all three functional impairments simultaneously. This proportion decreased to 22.7% by the third month. SF-36 values improved over 3 months. All dimensions showed significant improvement over time (all P < 0.05).

Table 2.

Comparison of PICS, PSS and health-related quality of life between 1 and 3 months after discharge follow-up among sepsis survivors.

Item	1 month (n = 147)	3 month (n = 132)	P*
Prevalence of PICS among survivors, n (%)	145 (98.6)	123 (93.2)	0.039
PICS total score, median (IQR)	21 (11, 46)	8 (4, 33)	<0.001
Prevalence of Cognitive dysfunction, n (%)	138 (93.8)	104 (79.0)	0.001
Cognitive domain score, median (IQR)	6 (3, 16)	2 (1, 10)	<0.001
Prevalence of Physical dysfunction, n (%)	140 (95.2)	106 (80.3)	0.002
Physical domain score, median (IQR)	8 (4, 17)	2 (1, 14)	<0.001
Prevalence of Mental dysfunction (HABC-M SR), n (%)	141 (95.9)	104 (78.8)	<0.001
Mental domain score, median (IQR)	5 (3, 14)	2 (1, 9.5)	<0.001
Prevalence of PSS among survivors, n (%)	114 (77.6)	81 (61.4)	0.014
Prevalence of Cognitive dysfunction, n (%)	97 (66.0)	71 (53.8)	0.033
MMSE score, median (IQR)	20 (6, 24)	23 (10, 26)	0.003
Prevalence of Physical dysfunction, n (%)	59 (40.1)	40 (30.3)	0.024
Barthel Index score, median (IQR)	80 (10, 90)	100 (36.3, 100)	<0.001
Prevalence of Mental dysfunction (HADS), n (%)	87 (59.2)	45 (34.1)	<0.001
HADS-Anxiety score, median (IQR)	8 (6, 11)	5 (4, 8)	<0.001
HADS-Depression score, median (IQR)	7 (6, 12)	6 (4, 8.8)	<0.001
Sleep quality score, median (IQR)	5 (3, 13)	2 (1, 6)	<0.001
Quality of life scores of survivors (SF–36), median (IQR)
PF	45 (10, 55)	60 (20, 70)	<0.001
RP	25 (0, 50)	50 (0, 100)	<0.001
BP	30 (10, 50)	60 (30, 80)	<0.001
GH	50 (5, 70)	65 (35, 85)	<0.001
VT	55 (10, 75)	75 (45, 90)	<0.001
SF	33.3 (11.1, 55.6)	66.7 (25.0, 88.9)	<0.001
RE	33.3 (0, 33.3)	66.7 (0, 91.7)	<0.001
MH	48 (16, 60)	68 (36, 80)	<0.001

*P values were calculated using Wilcoxon signed-rank test for continuous outcomes and McNemar’s test for paired proportions, based on the 132 patients with complete data at the 1- and 3-month follow-up assessments. PICS was defined as a HABC-M SR total score > 0. Abbreviations: PSS, Post-sepsis syndrome; MMSE, Mini-Mental State Examination; HABC-M SR, Healthy Aging Brain Care Monitor Self-Report; HADS, Hospital Anxiety and Depression Scale; PF, Physical Functioning; RP, Role-Physical; BP, Bodily Pain; GH, General Health; VT, Vitality; SF, Social Functioning; RE, Role-Emotional; MH, Mental Health.

Figure 2.

Outcomes of sepsis survivors 1- and 3-month after discharge. (a), the prevalence post-sepsis syndrome (PSS) at the first-month follow-up. (b), the prevalence post-sepsis syndrome (PSS) at the third-month follow-up. (c), Median (IQR) HABC-M SR Chinese version score for PICS and its three domains among the 1- and 3-month follow-ups. (d), Median (IQR) SF-36 and its eight domains among the 1- and 3-month follow-ups.

Risk factors for PSS

The dependent variable was whether PSS occurred in the third month (a binary variable: yes/no). The variables with statistical significance in the univariate analysis (P < 0.1; Supplementary Table S2) were included in the multivariable logistic regression. In the multivariable analysis, length of stay in the ICU, mechanical ventilation, norepinephrine use, sedative use, bicarbonate (HCO₃^-), base excess (BE), and Glasgow Coma Scale (GCS) score were not significantly associated with PSS occurrence (all P > 0.05). However, BMI (OR 0.864, 95% CI 0.767–0.974, P = 0.017), RBC count (OR 2.606, 95% CI 1.389–4.922, P = 0.003), PO₂ (OR 1.016, 95% CI 1.002–1.030, P = 0.026), and globulin (GLB) (OR 1.106, 95% CI 1.003–1.221, P = 0.044) were independently associated with PSS at 3 months (Table 3).

Table 3.

Multivariable logistic regression analysis for PSS at 3 months.

Variable	Odds ratio	95%CI	P value
Length of stay in the ICU (days)	1.032	(0.960, 1.108)	0.394
BMI (kg/m²)	0.864	(0.767, 0.974)	0.017
Mechanical ventilation	1.862	(0.600, 5.779)	0.282
Norepinephrine use	2.410	(0.703, 8.258)	0.162
Sedative use	1.177	(0.383, 3.623)	0.776
RBC count (×10¹²/L)	2.606	(1.389, 4.922)	0.003
GLB (g/L)	1.106	(1.003, 1.221)	0.044
PO₂ (mmHg)	1.016	(1.002, 1.030)	0.026
HCO₃- (mmol/L)	0.955	(0.864, 1.055)	0.363
BE (mmol/L)	1.101	(0.973, 1.246)	0.128
GCS score	0.863	(0.738, 1.010)	0.066

All laboratory indicators were measured within the first 24 hours of ICU admission. No laboratory tests were performed during the post-discharge follow-up assessments. Abbreviations: PSS, Post-sepsis syndrome; BMI, Body Mass Index; RBC, Red Blood Cell; GLB, Globulin; BE, Base Excess; GCS, Glasgow Coma Scale; PO₂, Partial Pressure of Oxygen; HCO₃^-, Bicarbonate.

Risk factors for PICS

To identify factors associated with the severity of PICS symptoms (as measured by the continuous HABC-M SR score) at 3 months, we performed univariate linear regression analyses (Supplementary Table S3) followed by a stepwise multiple linear regression model. In the univariate linear regression analysis, factors significantly associated with the HABC-M SR score at 3 months included BMI, nutrition score, procalcitonin (PCT), total bilirubin (TBil), HCO₃^-, length of stay in the ICU, GCS score, VTE score, alcohol history, and site of infection (all P < 0.05). These variables, together with age and Acute Physiology and Chronic Health Evaluation II (APACHE II) score, were entered into a stepwise multiple linear regression model to identify independent predictors. The final model (Table 4) revealed that lower GCS score, higher VTE score, higher HCO₃^- level and pulmonary infection were independently associated with higher HABC-M SR score (abdominal infection as reference). The model accounted for 33.2% of the variance (adjusted R²= 0.332, P < 0.001). Non-parametric analyses (Supplementary Table S4) yielded consistent results.

Table 4.

Multiple linear regression analysis of factors independently associated with HABC-M SR score at 3-month follow-up.

Variables	B	SE	Standardized β	t	P value	95% CI
(Constant)	9.678	6.942	—	1.394	0.166	(−4.063, 23.420)
GCS score	−1.339	0.321	−0.311	−4.176	< 0.001	(−1.973, −0.704)
VTE score	1.996	0.635	0.239	3.143	0.002	(0.739, 3.253)
HCO₃- (mmol/L)	0.462	0.200	0.178	2.309	0.023	(0.066, 0.858)
Pulmonary infection	7.103	2.985	0.197	2.380	0.019	(1.194, 13.011)
Infection at other sites	11.025	3.296	0.252	3.345	0.019	(4.565, 17.485)

Variables were selected using a stepwise regression method. Initial candidates included all factors with P < 0.05 in the univariate analysis, alongside age and APACHE II score, which were forced into the model due to their recognized clinical relevance. Model summary: R² = 0.358, Adjusted R² = 0.332, F = 13.700, P < 0.001. For the site of infection, abdominal infection was set as the reference category. All laboratory indicators were measured within the first 24 hours of ICU admission. No laboratory tests were performed during the post-discharge follow-up assessments. Abbreviations: B, unstandardized coefficient; SE, standard error; CI, confidence interval; GCS, Glasgow Coma Scale; VTE, venous thromboembolism risk score.

Discussion

The mortality rate three months after sepsis in this study was similar to rates reported in previous literature. The in-hospital, 30-day, and 90-day mortality rates for patients with sepsis or septic shock were 8.4%, 17.9%, and 25.0%, respectively. Previous epidemiological studies have reported that the 90-day mortality rate for patients with sepsis was between 17.9% and 31.6%.^26,27 These findings suggest that the long-term prognosis after sepsis is not favorable. In addition to survival and discharge, we should pay more attention to patients’ quality of life after discharge.

In this prospective observational study, we found that 92.5% of sepsis survivors experienced symptoms of PICS three months after discharge, with 60.9% meeting the criteria for PSS at the same time point, highlighting the substantial and persistent burden faced by this population. The prevalence of PICS declined from 98.6% at 1 month to 92.5% at 3 months (P = 0.039), while PSS prevalence decreased from 77.6% to 60.9% (P = 0.014), indicating gradual but incomplete recovery across both syndromes. Marra et al. found that 36% of subjects were PICS-free at 3 months.²⁸ This result differs substantially from our findings, where almost all participants exhibited some PICS symptoms at 3 months. This discrepancy may be attributable to the use of the HABC-M SR,²⁹ where a score greater than 0 was considered positive.²⁰ Additionally, the survivors were in the early recovery phase at 1 month, which might explain the high prevalence rate. A significant finding of this study is the substantial symptom overlap between PICS and PSS. Compared to non-PICS survivors, those with PICS exhibited significantly higher rates of impairment across all domains at 3 months. Physical dysfunction was the most prevalent manifestation, affecting 80.3% of survivors, followed by cognitive (79.0%) and mental dysfunction (79.0%). Similarly, for PSS, physical dysfunction affected 30.3% of survivors at 3 months, while cognitive and mental dysfunction affected 53.8% and 34.1%, respectively. Despite the high prevalence of symptoms, we observed a significant longitudinal improvement across all assessed dimensions over the three-month period (all P < 0.05). This functional recovery was mirrored by a broad improvement in health-related quality of life (HRQoL). All eight domains of the SF-36 showed statistically significant gains. These findings suggest that the first three months post-discharge represent a critical window of physiological and psychological adaptation.

In the multivariable logistic analysis for PSS, higher baseline partial pressure of oxygen (PO₂), red blood cell (RBC) count, and globulin (GLB) levels were independently associated with increased odds of PSS at 3 months. These associations may reflect underlying disease severity rather than direct causal effects. Research has found that high-oxygen exposure can trigger inflammatory responses and oxidative damage in the lungs.³⁰ Elevated PO₂ may indicate greater respiratory support requirements in patients with more severe lung injury. Elevated RBC may reflect hemoconcentration related to fluid resuscitation or underlying chronic conditions, impairing microcirculatory flow and tissue perfusion,³¹ while higher globulin levels indicate a pro-inflammatory phenotype or chronic inflammation that predisposes to prolonged recovery.³² Intriguingly, we observed that baseline inflammatory markers—specifically interleukin-6 (IL-6), C-reactive protein (CRP), and PCT—tended to be higher in survivors than in non-survivors. While hyperinflammation is conventionally equated with severity, this paradoxical trend aligns with the emerging concept of “immunoparalysis” in sepsis.³³ Higher BMI was associated with significantly lower odds of PSS (OR 0.864, 95% CI 0.767–0.974, P = 0.017), suggesting a protective effect of greater body mass in this population. This finding aligns with the well-described “obesity paradox” in critical illness,³⁴ where higher BMI is paradoxically associated with improved outcomes, possibly due to greater metabolic reserve or nutritional buffers against the catabolic stress of sepsis.³⁵

In the multivariable analysis for PICS severity, lower GCS score, higher VTE score, higher HCO₃^- level, pulmonary infection, and infection at other sites were independently associated with higher HABC-M SR scores. Lower GCS score emerged as the strongest predictor (standardized β = -0.311, P < 0.001), indicating that impaired neurological status at ICU admission has a sustained impact on long-term functional recovery. This finding aligns with previous studies.^36,37 A higher VTE risk score indicates a greater likelihood of developing VTE. Research shows that after an acute VTE episode, many patients experience functional limitations and a decline in quality of life.^38,39 Elevated HCO₃^- at admission most commonly indicates chronic respiratory acidosis with renal compensation.⁴⁰ The identification of infection site as an independent predictor highlights the importance of source-specific factors in post-sepsis outcomes. While BMI, ICU stay duration, and total hospital stay were significant in univariate analyses, they did not retain independent significance in the final model, suggesting their effects may be mediated by the stronger predictors identified.

At the beginning of the twenty-first century, the first accounts of post-sepsis syndrome were documented, detailing persistent physical, mental, cognitive, and medical abnormalities that followed severe sepsis.¹⁰ At present, a growing body of research on sepsis survivorship focuses on PICS.^12,13 Rather than viewing PICS and PSS as distinct syndromes, our findings suggest they may represent two manifestations of a common underlying condition—post-critical illness dysfunction—differing primarily in the population studied rather than in clinical presentation.¹⁵ Our data reveal a striking co-occurrence of symptoms, with 92.5% of survivors affected by PICS and 60.9% by PSS at 3 months, highlighting the multi-dimensional and overlapping nature of these impairments. These data underscore the substantial and persistent burden faced by this population. Rather than viewing them as isolated entities, the co-occurrence of PICS and PSS embodies the multi-dimensional nature of post-sepsis recovery, necessitating integrated clinical surveillance and supportive care tailored to these overlapping impairments. Consequently, future research should focus on developing integrated assessment tools and intervention protocols that simultaneously address the cognitive, physical, and psychological domains common to both syndromes.^16,41

Several limitations were acknowledged. First, the study was single-centered, which would have limited how broadly our results could be applied. Furthermore, 34.7% of participants (n = 70) died during the follow-up period or were lost to follow-up, which could introduce non-response bias and distort our findings. The high cumulative mortality rate observed at 3 months (25.0%) represents a significant competing risk in this population. Because PICS and PSS can only be assessed in living patients, our findings are subject to survivor bias. The observed survival bias—where non-survivors exhibited paradoxically lower inflammatory markers—highlights the complexity of host responses; our findings thus reflect the burden specifically among those capable of mounting an initial immune response. Consequently, the reported prevalence likely represents the survivor burden of sepsis rather than the total impact on the entire initial cohort. Second, it is possible that the instruments used to evaluate PICS differ from those employed in previous research. Third, the three-month follow-up time is rather brief and is unable to cover all PICS symptoms. Thus, a long-term longitudinal study with multiple follow-ups will be needed. Future analyses involving a multicenter cohort with a broader participant demographic are planned.

Conclusion

In this prospective study, we analyzed PICS and PSS conditions and related quality of life dimensions in sepsis survivors and found significant overlap between the two syndromes. Furthermore, we identified risk factors associated with the severity of these syndromes in order to provide a framework for future research and facilitate better recovery outcomes for sepsis survivors.

Supplemental material

Supplemental material - Analysis of the prevalence and risk factors of post-intensive-care syndrome and post-sepsis syndrome in survivors of sepsis

Supplemental material for Analysis of the prevalence and risk factors of post-intensive-care syndrome and post-sepsis syndrome in survivors of sepsis by Yuanqing Li, Anhao Liu, Xuehui Zhang, Ningkang Lv, Tianqi Chen, Cuiping Hao and Dongmei Wu in Science Progress.

Footnotes

ORCID iDs

Yuanqing Li

Cuiping Hao

Dongmei Wu

Ethical considerations

This study adhered to the Declaration of Helsinki, was approved by the local ethics committee on 22 November 2023 (approval number: 2023-11-C032), and registered at the Chinese Clinical Trial Registry (CTR2500112184).

Consent to participate

Written informed consent was obtained from all participants or their legal next of kin prior to their participation in the study.

Consent for publication

Written informed consent was obtained from all participants or their legal next of kin prior to their participation in the study. A copy of the signed consent form is available for review upon request.

Author contributions

LYQ, LAH and LNK contributed to data collection and manuscript drafting. LYQ,ZXH and CTQ participated in data analysis, interpretation, and critical revisions of the manuscript. WDM and WCP reviewed and approved the final version of the manuscript. All authors have reviewed and agreed to publish this final draft.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Key R&D Program of Jining City, Shandong Province, China (No. 2023YXNS180, No. 2023YXNS107) and the Wu Jieping Medical Foundation (No.320.6750.2024-07-4). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript, and no additional financial support will be provided.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.*

Supplemental material

Supplemental material for this article is available online.

Appendix

References

Singer

Deutschman

Seymour

, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). Jama 2016; 315: 801–810. https://doi.org/10.1001/jama.2016.0287

Shankar-Hari

Phillips

Levy

, 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: 775–787. https://doi.org/10.1001/jama.2016.0289

Patnaik

Azim

Singh

, et al. Serial Trend of Neutrophil CD64, C-reactive Protein, and Procalcitonin as a Prognostic Marker in Critically Ill Patients with Sepsis/Septic Shock: A Prospective Observational Study from a Tertiary Care ICU. Indian J Crit Care Med 2024; 28: 777–784, 20240731. https://doi.org/10.5005/jp-journals-10071-24777

Rudd

Johnson

Agesa

, et al. Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study. Lancet 2020; 395: 200–211. https://doi.org/10.1016/s0140-6736(19)32989-7

Evans

Rhodes

Alhazzani

, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med 2021; 47: 1181–1247, 20211002. https://doi.org/10.1007/s00134-021-06506-y

Kempker

Martin

. The Changing Epidemiology and Definitions of Sepsis. Clin Chest Med 2016; 37: 165–179, 20160323. https://doi.org/10.1016/j.ccm.2016.01.002

Prescott

Kepreos

Wiitala

, et al. Temporal Changes in the Influence of Hospitals and Regional Healthcare Networks on Severe Sepsis Mortality. Crit Care Med 2015; 43: 1368–1374. https://doi.org/10.1097/ccm.0000000000000970

Huang

Daniels

Lembo

, et al. Life after sepsis: an international survey of survivors to understand the post-sepsis syndrome. Int J Qual Health Care 2019; 31: 191–198. https://doi.org/10.1093/intqhc/mzy137

Rababa

Bani Hamad

Hayajneh

. Sepsis assessment and management in critically Ill adults: A systematic review. PLoS One 2022; 17: e0270711, 20220701. https://doi.org/10.1371/journal.pone.0270711

10.

Mostel

Perl

Marck

, et al. Post-sepsis syndrome - an evolving entity that afflicts survivors of sepsis. Mol Med 2019; 26: 6, 20191231. https://doi.org/10.1186/s10020-019-0132-z

11.

Granja

Dias

Costa-Pereira

, et al. Quality of life of survivors from severe sepsis and septic shock may be similar to that of others who survive critical illness. Crit Care 2004; 8: R91–R98, 20040220. https://doi.org/10.1186/cc2818

12.

Liu

Watanabe

Nakamura

, et al. One-year outcomes in sepsis: a prospective multicenter cohort study in Japan. J Intensive Care 2025; 13: 23, 20250501. https://doi.org/10.1186/s40560-025-00792-0

13.

Inoue

Nakanishi

Sugiyama

, et al. Prevalence and Long-Term Prognosis of Post-Intensive Care Syndrome after Sepsis: A Single-Center Prospective Observational Study. J Clin Med 2022; 11: 5257, 20220906. https://doi.org/10.3390/jcm11185257

14.

Nakanishi

Liu

Kawakami

, et al. Post-Intensive Care Syndrome and Its New Challenges in Coronavirus Disease 2019 (COVID-19) Pandemic: A Review of Recent Advances and Perspectives. J Clin Med 2021; 10: 3870, 20210828. https://doi.org/10.3390/jcm10173870

15.

Fleischmann-Struzek

Joost

FEA

Pletz

, et al. How are Long-Covid, Post-Sepsis-Syndrome and Post-Intensive-Care-Syndrome related? A conceptional approach based on the current research literature. Critical Care 2024; 28: 283. https://doi.org/10.1186/s13054-024-05076-x

16.

Liu

Nakashima

Goto

, et al. Phenotypes of Functional Decline or Recovery in Sepsis ICU Survivors: Insights From a 1-Year Follow-Up Multicenter Cohort Analysis. Crit Care Med 2025; 53: e928–e940, 20250224. https://doi.org/10.1097/ccm.0000000000006621

17.

Taito

Banno

, et al. Rehabilitation for patients with sepsis: A systematic review and meta-analysis. PLoS One 2018; 13(20180726): e0201292. https://doi.org/10.1371/journal.pone.0201292

18.

von

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007; 147: 573–577. https://doi.org/10.7326/0003-4819-147-8-200710160-00010

19.

Gao

Liang

Lyu

, et al. Prevalence of and risk factors analysis for post-intensive care syndrome among survivors of critical care during 3-month longitudinal follow-up. Nurs Crit Care 2025; 30: e13242, 20250114. https://doi.org/10.1111/nicc.13242

20.

Liang

Wang

, et al. Screening for post-intensive care syndrome: Validation of the Healthy Aging Brain Care Monitor Self-Report Chinese version. Nurs Crit Care 2024; 29: 824–829, 20230717. https://doi.org/10.1111/nicc.12949

21.

van der Slikke

Beumeler

LFE

Holmqvist

, et al.

Understanding Post-Sepsis Syndrome: How Can Clinicians Help?

Infect Drug Resist 2023; 16: 6493–6511, 20230929. https://doi.org/10.2147/idr.S390947

22.

Yuzhen

Jianhuang

Zhizhao

, et al. Latent class analysis of symptom characteristics in patients with post-sepsis syndrome. Chinese Journal of Emergency and Critical Care 2024; 5: 197–204.

23.

Dos Reis

Figueiredo

Biscaro

RRM

, et al. Psychometric Properties of the Barthel Index Used at Intensive Care Unit Discharge. Am J Crit Care 2022; 31: 65–72. https://doi.org/10.4037/ajcc2022732

24.

Prescott

Angus

. Enhancing Recovery From Sepsis: A Review. Jama 2018; 319: 62–75. https://doi.org/10.1001/jama.2017.17687

25.

Heyland

Hopman

Coo

, et al. Long-term health-related quality of life in survivors of sepsis. Short Form 36: a valid and reliable measure of health-related quality of life. Crit Care Med 2000; 28: 3599–3605. https://doi.org/10.1097/00003246-200011000-00006

26.

Liu

Tao

Xiao

, et al. Low lymphocyte to high-density lipoprotein ratio predicts mortality in sepsis patients. Front Immunol 2023; 14: 1279291, 20231012. https://doi.org/10.3389/fimmu.2023.1279291

27.

Zeng

Zhang

, et al. Prediction of 90-Day Mortality among Sepsis Patients Based on a Nomogram Integrating Diverse Clinical Indices. Biomed Res Int 2021; 2021: 1023513, 20211020. https://doi.org/10.1155/2021/1023513

28.

Marra

Pandharipande

Girard

, et al. Co-Occurrence of Post-Intensive Care Syndrome Problems Among 406 Survivors of Critical Illness. Crit Care Med 2018; 46: 1393–1401. https://doi.org/10.1097/ccm.0000000000003218

29.

Kosilek

Schmidt

Baumeister

, et al. Frequency and risk factors of post-intensive care syndrome components in a multicenter randomized controlled trial of German sepsis survivors. J Crit Care 2021; 65: 268–273, 20210716. https://doi.org/10.1016/j.jcrc.2021.07.006

30.

Zhai

, et al. Hyperoxia but not high tidal volume contributes to ventilator-induced lung injury in healthy mice. BMC Pulmonary Medicine 2023; 23: 354. https://doi.org/10.1186/s12890-023-02626-x

31.

Piagnerelli

Boudjeltia

Vanhaeverbeek

, et al. Red blood cell rheology in sepsis. Intensive Care Med 2003; 29: 1052–1061, 20030612. https://doi.org/10.1007/s00134-003-1783-2

32.

Alagna

Meessen

Bellani

, et al. Higher levels of IgA and IgG at sepsis onset are associated with higher mortality: results from the Albumin Italian Outcome Sepsis (ALBIOS) trial. Ann Intensive Care 2021; 11: 161, 20211126. https://doi.org/10.1186/s13613-021-00952-z

33.

Hotchkiss

Monneret

Payen

. Sepsis-induced immunosuppression: from cellular dysfunctions to immunotherapy. Nat Rev Immunol 2013; 13: 862–874, 20131115. https://doi.org/10.1038/nri3552

34.

Karampela

Chrysanthopoulou

Christodoulatos

, et al. Is There an Obesity Paradox in Critical Illness? Epidemiologic and Metabolic Considerations. Curr Obes Rep 2020; 9: 231–244. https://doi.org/10.1007/s13679-020-00394-x

35.

Jaitovich

Hall

. The flux of energy in critical illness and the obesity paradox. Physiol Rev 2025; 105: 1487–1552, 20250221. https://doi.org/10.1152/physrev.00029.2024

36.

Shen

Zhou

, et al. Associations between Glasgow Coma Scale trajectories and 28-day survival rate in patients with sepsis-associated encephalopathy: insights from longitudinal group trajectory modeling. Front Neurol 2025; 16: 1607946, 20250902. https://doi.org/10.3389/fneur.2025.1607946

37.

Alcamo

Weiss

Fitzgerald

, et al. Outcomes Associated With Timing of Neurologic Dysfunction Onset Relative to Pediatric Sepsis Recognition. Pediatr Crit Care Med 2022; 23: 593–605, 20220503. https://doi.org/10.1097/pcc.0000000000002979

38.

Steiner

Nopp

Weber

, et al. The post-VTE functional status scale for assessment of functional limitations in patients with venous thromboembolism: Construct validity and responsiveness in a prospective cohort study. Thromb Res 2023; 221: 1–6, 20221117. https://doi.org/10.1016/j.thromres.2022.11.006

39.

Monreal

Agnelli

Chuang

, et al. Deep Vein Thrombosis in Europe-Health-Related Quality of Life and Mortality. Clin Appl Thromb Hemost 2019; 25: 1076029619883946. https://doi.org/10.1177/1076029619883946

40.

Bruno

Valenti

. Acid-base disorders in patients with chronic obstructive pulmonary disease: a pathophysiological review. J Biomed Biotechnol 2012; 2012: 915150, 20120201. https://doi.org/10.1155/2012/915150

41.

Torres

JSS

Tamayo-Giraldo

Bejarano-Zuleta

, et al. Sepsis and post-sepsis syndrome: a multisystem challenge requiring comprehensive care and management-a review. Front Med (Lausanne) 2025; 12: 1560737, 20250408. https://doi.org/10.3389/fmed.2025.1560737

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.74 MB