Abstract
Background
Delirium is a common complication in patients at the intensive care unit (ICU) and is associated with prolonged ICU-stay and hospitalization and with increased morbidity. The impact of ICU-delirium on long-term survival is not clearly understood.
Methods
This retrospective single center observational study was conducted at the Institute of Intensive Care Medicine at the University Hospital Zurich, Switzerland. All adult ICU-survivors over a four-year period were screened for eligibility. ICU-delirium was defined based on the Intensive Care Delirium Screening Checklist (ICDSC), together with the coded diagnosis F05 in the International Classification of Diseases (ICD-2019). ICU-survivors who developed delirium during their ICU stay (group D) were compared with ICU-survivors who did not (group ND). Survival was evaluated according to data from hospital electronic health records up to four years from ICU-discharge. The survival analysis was reported using Kaplan-Meier curves and absolute risk differences (ARD). A multivariable logistic regression model was fitted with long-term survival at four years after ICU-discharge as outcome of interest, including several clinical conditions and interventions associated with long-term survival for ICU patients. For subgroup analysis, ICU-survivors were grouped based on age at the time of admission (45–54, 55–64, ≥ 65 years), and on relevant clinical conditions.
Results
A total of 9’604 patients fulfilled the inclusion criteria, of them 22.6% (n = 2’171) developed ICU-delirium. Overall, patients in the group D had a significantly lower probability of survival than patients in the group ND (p < 0.0001, ARD = 11.8%). In the multivariable analysis, ICU-delirium was confirmed as independently associated with long-term survival. After grouping for age categories, patients between 55 and 64 years of age in the group D were less likely to survive than patients in the group ND at every time point analyzed, up to four years after ICU discharge (p < 0.001, ARD = 7.3%). This difference was even more significant in the comparison between patients over 65 years (p < 0.0001, ARD 11.1%). No significant difference was observed in the other age groups.
Conclusions
In the study population, ICU-delirium was independently associated with a reduced long-term survival. Patients who developed ICU-delirium had a reduced survival up to four years after ICU discharge and this association was particularly evident in patients above 55 years of age.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12883-025-04025-7.
Keywords: Encephalopathy, Delirium, ICU, Long-term survival
Background
Delirium represents a common complication in patients treated at the intensive care unit (ICU), affecting approximately one third of the critically ill adults, with a prevalence that can reach the 80% among the mechanically ventilated patients [1, 2]. It is associated with prolonged ICU-, hospital-length of stay (LOS) [3, 4] and duration of mechanical ventilation [5], increased disability [6, 7], reduced long-term cognitive performance [8], as well as increased health care costs [9].
Available data on ICU-delirium and its association to short-term survival are discordant [3, 5, 10–13]. Considering the association between ICU-delirium and long-term survival, data are even more scare and not conclusive. A few studies reported an increased mortality in patients who developed delirium at 6-months [14], 1-year [15], and − in a population of only post-operative patients −even at 5-years [16], compared to patients who did not. On the contrary, in two prospective cohort studies no difference in 1-year mortality has been observed among patients who developed delirium and patients who did not [17, 18].
These discrepancies in the available data could be explained with methodological issues, such as the small sample size of the analyzed populations, the heterogeneity in the definition of delirium, different timing in the assessment of the possible association between delirium and mortality, as well as possible inclusion bias.
Our retrospective observational study focuses on the association between ICU-delirium and long-term survival. To do that, we applied a standardized definition of delirium based both on regular scoring and clinical judgment to a large population of ICU-patients and investigated survival at different time point: 1-year, 2-year, 3-year and up to 4-year following ICU-discharge.
Methods
This retrospective single center observational study was conducted at the Institute of Intensive Care Medicine at the University Hospital Zurich, Switzerland. All patients admitted to the ICU between January, 1st 2019 and December, 31st 2023 were screened for eligibility. Inclusion criteria were: (1) age ≥ 18 years old, (2) clinical diagnosis based on clinical judgment and coded according to the International Classification of Diseases (ICD-10, version 2019). Exclusion criteria were (1) patients’ documented refusal to have their data analyzed for research projects; (2) no documented Intensive Care Delirium Screening Checklist (ICDSC) [19]; (3) ambiguous delirium status; (4) death at the ICU.
Data were obtained from the hospital electronic health records (KISIM-TM, Cistec, Zurich, Switzerland) and from the ICU Patient Data Management System (PDMS, MetaVision, iMDsoft, Israel). Collected data were: Demographics, including age and sex; SAPS II scores [20]; ICU- and hospital-length of stay (ICU-LOS, H-LOS); survival at different time points (at one, two, three and four years) after ICU-discharge. Moreover, to analyze the influence of other possible risk factors affecting long term survival in ICU patients, the following conditions or interventions, selected in the light of the available literature, were analyzed: diabetes [21], atrial fibrillation/flutter [22], acute respiratory failure and chronic obstructive pulmonary disease (COPD) [23], dementia [24], malignancies [25], hours of mechanical ventilation [26], need for renal replacement therapy [27] or need for hemodynamic mechanical support [28]. The main diagnosis for ICU admission were presented after grouping according to ICD-10 codes.
To define delirium and stratify the population accordingly, two criteria were considered: ICDSC score and clinical diagnosis. The ICDSC assessment is routinely performed three times a day by trained ICU-nurse staff and is considered positive if ≥ 4. If confirmed from the clinical judgment of the treating physician, a diagnosis of delirium is reported in the electronic medical records, coded according to ICD-10 as diagnosis F05 (delirium, not induced by alcohol and other psychoactive substances). ICU-Delirium patients (group D) were defined based on the above two complementary criteria: (1) at least one positive ICDSC assessment, and (2) the coded diagnosis F05. Likewise, non-delirious ICU patients (group ND) were defined as: (1) no positive ICDSC assessment, and (2) no coded F05 diagnosis. The ICU-delirium status of patients that fulfilled only one criterion was considered ambiguous, leading to exclusion of these patients.
The survival analysis was performed based on data from the electronic medical records. In the Canton Zurich, death certificates from the Civil Registry Department of the Municipal Office are automatically transferred into the patient’s electronic record regardless of whether the death occurred in the hospital or not. The population of ICU survivors was arbitrary grouped in four age categories, also taking into account the age distribution in the population: 45–54 years, 55–64 years, and ≥ 65 years. The cut-off for older adult patients was set at 65 years, being the ‘retirement age’ according to the Swiss federal social insurance system.
Statistical analysis
All statistical analyses were performed using the Scientific Python Development Environment Spyder IDE (Python 3.9.7 64-bit), and a p-value ≤ 0.05 was considered statistically significant. Descriptive statistics are reported as counts/percentages, mean ± standard deviation, or as median including the interquartile range, as appropriate. All continuous data were tested for normality using Shapiro–Wilk’s test. Data not normally distributed were compared using the Mann–Whitney test. Numerical variables with normal distribution were compared using independent sample t test. Ordinal variables or numerical variables with not normal distribution were compared using Mann–Whitney-Wilcoxon test. Categorical variables were compared with chi-squared test. A multivariable logistic regression analysis was performed with covariates selected based on clinical relevance and prior evidence. The logistic regression model was fitted using the maximum likelihood estimation method, with the exposure variable and all covariates included in the model simultaneously. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. For the significant covariates, survival analysis was conducted using Kaplan-Meier curves, with significance between groups being assessed with the log-rank test. The Absolute Risk Difference (ARD) was calculated at specific time points by subtracting the survival probability of the delirium group from that of the non-delirium group.
Reporting of the study results adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [29].
Results
Overall, 16’557 patients were screened for eligibility. Of them 6’953 were excluded. Reasons for exclusion are listed in Fig. 1. In particular, 2’151 patients (11.9%) were excluded due to the missing coded diagnosis F05, and 465 patients (2.6%) due to the absence of a positive ICDSC score (Fig. 1).
Fig. 1.
Flow chart. Inclusion and exclusion criteria. study population flow chart. ICU: intensive care unit, ICDSC: Intensive Care Delirium Screening Checklis,; SAPS II: Simplified Acute Physiology Score II, ICD: International Classification of Diseases
The study population included 9’604 patients (females n = 3’250 (33.8%), median age: 63 years [IQR: 51, 72], median ICU-LOS: 1.8 days [IQR: 0.9–5.0]) (Table 1).
Table 1.
Baseline characteristics of the study population. (A) Demographics, pre-existing conditions and interventions (B) Major diagnostic categories. D: delirium group, ND: non-delirium group, ICU: Intensive Care Unit, SAPS II: Simplified Acute Physiology Score II, IQR: interquartile range, COPD: chronic obstructive pulmonary disease
| A. Demographics, pre-existing conditions and interventions | ||||
|---|---|---|---|---|
| General population (D & ND) |
Statistical analysis between D & ND groups | |||
| D | ND | • p-value | ||
| N of total Patients | 9’604 | 2’171 | 7’433 | - |
| Females, n (%) | 3’250 (33.8) | 626 (28.8) | 2’624 (35.3) | < 0.0001 |
|
Age, median (IQR) |
63.0 (51.0, 72.0) | 68.0 (58.0, 77.0) | 61.0 (49.0, 71.0) | < 0.0001 |
|
SAPS II, median (IQR) |
35.0 (25.0, 47.0) | • 47.0 (37.0, 59.0) | 32.0 (22.0, 43.0) | < 0.0001 |
|
ICU length of stay [days], median (IQR) |
1.8 (0.9, 5.0) | 8.7 (4.0, 17.0) | 1.0 (0.8, 2.6) | < 0.0001 |
|
Hospital length of stay [days], median (IQR) |
11.3 (7.6, 18.6) | 20.4 (12.8, 32.1) | 10.0 (6.9, 15.0) | < 0.0001 |
| Hospital deaths, n (%) | 222 (2.0) | 96 (4.1) | 126 (1.5) | < 0.0001 |
| Mechanically ventilated, n (%) | 8’322 (86.6) | 1’986 (91.5) | 6’336 (85.2) | 0.006 |
|
Mechanical Ventilation [days], median (IQR) |
0.4 (0.2, 1.02) | 2.1 (0.5, 6.4) | 0.3 (0.2, 0.5) | < 0.0001 |
| Malignancies, n (%) | 1’476 (15.4) | 214 (9.9) | 1’262 (17.0) | < 0.0001 |
| Acute Respiratory Failure, n (%) | 795 (8.3) | 286 (13.2) | 509 (6.8) | < 0.0001 |
| COPD, n (%) | 600 (6.2) | 193 (8.9) | 407 (5.5) | < 0.0001 |
| Atrial Fibrillation / Flutter, n (%) | 1’790 (18.6) | 659 (30.4) | 1’131 (15.2) | < 0.0001 |
| Diabetes, n (%) | 1’488 (15.5) | 443 (20.4) | 1’045 (14.0) | < 0.0001 |
| Dementia, n (%) | 171 (1.8) | 66 (3.0) | 46 (0.6) | < 0.0001 |
| Hemodynamic mechanical support, n (%) | 522 (5.4) | 361 (16.6) | 161 (2.2) | < 0.0001 |
| Renal replacement therapy, n (%) | 293 (3.0) | 204 (9.4) | 89 (1.2) | < 0.0001 |
| B. Diagnostic Categories of ICU Admission | ||||
| Cardiovascular system, n (%) | 3’763 (39.2) | 869 (40.0) | 2’894 (38.9) | 0.474 |
| Oncology & Hematology, n (%) | 1’629 (17.0) | 159 (7.3) | 1’470 (19.8) | < 0.0001 |
| Others, n (%) | 1’271 (13.2) | 277 (12.8) | 989 (13.3) | 0.381 |
| Trauma, n (%) | 1’048 (10.9) | 372 (17.1) | 677 (9.1) | < 0.0001 |
| Neurological system, n (%) | 621 (6.5) | 113 (5.2) | 508 (6.8) | 0.009 |
| Respiratory system, n (%) | 565 (5.9) | 148 (6.8) | 417 (5.6) | 0.041 |
| Digestive system, n (%) | 383 (4.0) | 108 (5.9) | 275 (3.7) | 0.009 |
| Infectious & Parasitic Diseases, n (%) | 323 (3.4) | 125 (5.8) | 198 (2.7) | < 0.0001 |
Most of these patients were admitted to the ICU due to a cardiovascular diagnosis (39.2%, n = 3’763), followed by onco-hematological diagnosis (17.0, n = 1’629), trauma (10.9%, n = 1’048), neurological disorders (6.5%, n = 621), and respiratory conditions (5.9%, n = 565) (Fig. 2).
Fig. 2.
Diagnostic Categories of ICU Admission. Pie chart illustrating the distribution of the study population according to the main diagnosis for ICU admission as coded with the ICD-10
Overall, 2’171 ICU-survivors (22.6%) developed delirium during their ICU stay (D = delirium group), while 7’433 (77.4%) did not (ND = non delirium group). In the D group, females were fewer (28.8% vs. 35.3%, p < 0.001) and patients were older (median age 68 [IQR: 58, 77] vs. 61 [IQR: 49, 71], p < 0.001). The median SAPS II was 35 points [IQR: 25, 47]. Patients in the D group had a higher SAPS II score than those in the ND group (median SAPS II score 47 [IQR: 37, 59] vs. 32 [IQR: 22, 43], p < 0.001).
Considering outcomes, patients in the D group had a longer ICU-, as well as hospital-LOS (median stay: 8.7 days [IQR: 4.0, 17.0] vs. 1.0 days [IQR: 0.8, 2.6], p < 0.001 and 20.4 days [IQR: 12.8, 32.1] vs. 10.0 days [IQR: 6.9, 15.0], p < 0.001 respectively) then patients in the ND group (Table 1). After discharge from the ICU, 2% of the patients (n = 222) died during their hospital stay (D group: n = 96, 4.1%; ND group: n = 126, 1.5%, p < 0.0001).
The analysis of comorbidities and interventions showed that more patients in the D group were mechanically ventilated (91.5%, n = 1’968 vs. 85.2%, n = 6336, p = 0.006), and for longer time (2.1 days, [IQR: 0.5–6.4] vs. 0.3 days [IQR: 0.2–0.5]). As pre-existing conditions, delirious patients had more acute respiratory failure (13.2% vs. 6.8%, p < 0.001), COPD (8.9% vs. 5.5%, p < 0.001), atrial fibrillation (30.4% vs. 15.2%, p < 0.001), diabetes (20.4% vs. 14.0%. p < 0.001), and dementia (3.0% vs. 0.6%, p < 0.001). On the contrary, patients in the D group had fewer pre-existing malignancies then patients in the ND group (9.9% vs. 17.9%, p < 0.001). Regarding interventions, delirious patients received more hemodynamic mechanical support (16.6% vs. 2.2% p < 0.001) and renal replacement therapy (9.4% vs. 1.2%, p < 0.001) (Table 1-B).
In the multivariable logistic regression model, ICU-delirium was independently associated with reduced survival at 4 years after ICU discharge (OR: 1.41, 95%CI 1.26–1.86, p < 0.001) (Table 2; Fig. 3).
Table 2.
Multivariable logistic regression model. Long-term survival at 4 years was considered as outcome. Y/N: yes/no, M/F: male/female, ICU-LOS: intensive care unit - length of stay, COPD: chronic obstructive pulmonary disease, OR: odds ratio, CI: confidence interval
| Variable | Coefficient | p-value | OR (95% CI) |
|---|---|---|---|
| Acute respiratory failure (Y/N) | -0.44 | < 0.0001 | 0.65 (0.53, 0.79) |
| Hemodynamic mechanical support (Y/N) | -0.13 | 0.547 | 0.88 (0.58, 1.34) |
| Diabetes (Y/N) | -0.05 | 0.496 | 0.95 (0.82, 1.10) |
| Sex (M/F) | -0.04 | 0.510 | 0.96 (0.86, 1.08) |
| Atrial fibrillation/Flutter (Y/N) | -0.01 | 0.851 | 0.99 (0.86, 1.13) |
| ICU-LOS (days) | 0.01 | 0.014 | 1.01 (1.00, 1.01) |
| Mechanical ventilation (days) | 0.02 | 0.012 | 1.02 (1.00, 1.03) |
| Delirium (Y/N) | 0.34 | < 0.0001 | 1.41 (1.23, 1.62) |
| Age (/10 years) | 0.41 | < 0.0001 | 1.50 (1.44, 1.56) |
| COPD (Y/N) | 0.43 | < 0.0001 | 1.53 (1.26, 1.86) |
| Renal replacement therapy (Y/N) | 0.71 | < 0.0001 | 2.03 (1.63, 2.53) |
| Dementia (Y/N) | 0.77 | 0.0002 | 2.16 (1.45, 3.22) |
| Malignancy (Y/N) | 1.51 | < 0.0001 | 4.52 (3.99, 5.13) |
Fig. 3.
Forrest plot with OR and 95% CI for different factors associated with long term survival at 4 years after ICU-discharge. Y/N: yes/no, M/F: male/female, ICU-LOS: intensive care unit - length of stay, COPD: chronic obstructive pulmonary disease, OR: odds ratio, CI: confidence interval
Throughout the observational time window of four years, patients in the D group had a significantly lower probability of survival than patients in the ND group (Fig. 4). After grouping for age categories, among patients between 45 and 54 years, no differences in the survival curves were observed in patients in the D and ND groups (p = 0.607) (Fig. 4-B). Among patients between 55 and 64 years, patients in the D group were less likely to survive than patients in the ND group (p = 0.0003, ARD = 7.3%) (Fig. 4-C). This difference was even more evident among patients older than 65 years (p < 0.0001, ARD = 11.1%) (Fig. 4-D) (Table 2). Moreover, for each condition or intervention with significant contribution in the multivariable model, survival was significantly lower in the D group with the exception of COPD and dementia (Fig. 4E-J).
Fig. 4.
Survival analysis of patients in the ND and D group illustrated by Kaplan-Meier plots. A. Survival analysis for the study population. B.Ppatients between 45 and 54 years. C. Patients between 55 and 64 years. D. Patients aged ≥ 65 years. E. Mechanically ventilated patients. F Patients with malignancies. G Patients with acute respiratory failure. H. Patients with COPD. I. Patients with pre-existing dementia. J. Patients undergoing renal replacement therapy. ND: non-delirium group, D: delirium group, CI: confidence interval, COPD: chronic obstructive pulmonary disease
Table 3.
Number of patients at risk and absolute risk difference (ARD). Number of ICU survivors at risk and absolute risk difference (ARD) A. for the general population, B-D. After grouping by age, E-J.By different clinical conditions or interventions. Survival is assessed up to 4 years after ICU-discharge. N: number, ARD: absolute risk difference, ND: non-delirium group, D: delirium group, COPD: chronic obstructive pulmonary disease
| Years from ICU discharge | p-value Log-rank test | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 1 | 2 | 3 | 4 | |||||||
| N patients | N patients | ARD | N patients | ARD | N patients | ARD | N patients | ARD | |||
| A. General population | D | 2’171 | 1’453 | 0.106 | 1’026 | 0.102 | 671 | 0.117 | 332 | 0.118 | < 0.0001 |
| ND | 7’433 | 5’616 | 3’913 | 2’562 | 1’396 | ||||||
| B. Patients 45 ≤ age ≤ 54 years | D | 216 | 167 | 0.034 | 134 | 0.004 | 83 | 0.003 | 44 | 0.007 | 0.6071 |
| ND | 1’129 | 915 | 659 | 464 | 255 | ||||||
| C. Patients 55 ≤ age ≤ 64 years | D | 448 | 316 | 0.063 | 219 | 0.073 | 143 | 0.082 | 66 | 0.073 | 0.0003 |
| ND | 1’829 | 1’385 | 971 | 622 | 341 | ||||||
| D. Patients ≥ 65 years | D | 1’296 | 807 | 0.116 | 561 | 0.098 | 375 | 0.111 | 183 | 0.111 | < 0.0001 |
| ND | 3’035 | 2’133 | 1416 | 901 | 490 | ||||||
| E. Mechanical ventilation | D | 1’986 | 1’364 | 0.090 | 961 | 0.088 | 633 | 0.099 | 315 | 0.098 | < 0.0001 |
| ND | 6’336 | 4’766 | 3’280 | 2’142 | 1’185 | ||||||
| F. Malignancies | D | 214 | 97 | 0.165 | 49 | 0.138 | 22 | 0.150 | 5 | 0.106 | < 0.0001 |
| ND | 1’262 | 759 | 415 | 216 | 64 | ||||||
| G. Acute respiratory failure | D | 286 | 172 | 0.084 | 97 | 0.106 | 46 | 0.133 | 4 | 0.152 | 0.0002 |
| ND | 509 | 259 | 150 | 87 | 11 | ||||||
| H. COPD | D | 193 | 111 | 0.052 | 63 | 0.037 | 31 | 0.059 | 8 | 0.061 | 0.178 |
| ND | 407 | 265 | 154 | 84 | 19 | ||||||
| I. Dementia | D | 66 | 33 | 0.097 | 21 | 0.134 | 8 | 0.136 | 2 | 0.136 | 0.201 |
| ND | 46 | 24 | 15 | 8 | 3 | ||||||
| J. Renal replacement therapy | D | 361 | 201 | 0.096 | 139 | 0.082 | 89 | 0.126 | 50 | 0.140 | 0.018 |
| ND | 161 | 101 | 62 | 41 | 25 | ||||||
Discussion
Our study investigates a possible association between ICU-delirium and long-term survival in a large population of ICU survivors. As the main finding, we report that ICU survivors over 55 years of age who suffered from delirium during their ICU stay had a lower survival over time up to four years from ICU discharge compared to patients who did not.
To date, only few studies analyzed long-term outcomes related to ICU-delirium, and methodological issues limit these previous works [15–18]. Similar to our findings, Moskowitz et al. observed an increased 5-years mortality in patients with ICU-delirium. However, this study only included patients over 50 years of age admitted to the ICU after elective surgery [16], which limits the generalization of the findings. Pisani et al. reported an association between ICU-delirium duration and 1-year mortality. In this study, the generalization of the finding is limited by including only older adult patients (over 65 years of age) [15]. On the contrary, Wolters et al. could not confirm an association between ICU-delirium and 1-year mortality [17]. The population investigated, however, was small and patients with neurological conditions were excluded due to possible confounding factors. Additionally, in this previous study, delirium was defined only based on a positive score − CAM-ICU − and/or administration of haloperidol [17]. The choice of this definition lacks, in our opinion, clinical judgment. Indeed, there are clinical conditions that could result in a positive CAM-ICU or ICDSC score for delirium without the patient actually having it. More recently, Fiest et al. performed a similar analysis on a larger population and, similarly to the work of Wolters et al., defined delirium based only on a score − in this case the ICDSC ≥ 4. Also in this study, no association was found between ICU-delirium and long-term survival [18].
In our study, we provide evidence for an association between ICU-delirium and long-term survival up to four years.
In our cohort, ICU-survivors who developed delirium during their ICU stay were significantly older than patients who did not. This finding is confirmatory, since increasing age is a known risk factor for delirium [30]. To exclude the association between ICU-delirium and long-term survival being due only to age differences, we conducted the survival analyses grouping the population by age. We found that ICU survivors above 55 years of age are those in whom the effect of ICU-delirium has the greatest impact on survival, and therefore worthy of further investigation − for example − for targeted prevention.
Moreover, in our cohort, patients with ICU-delirium had a higher median SAPS II score than patients who did not. One might therefore assume that patients with delirium have a lower survival also because they have a more severe clinical condition at ICU-admission.
However, two important aspects should be considered: firstly, the SAPS II is a score used to predict mortality in the ICU and not validated to predict long-term survival [20]. In our case, given the selected population of ICU survivors, the score would lose its applicability. Secondly, the SAPS II is inaccurate in patients with delirium. In fact, one of the items of the SAPS II score is the assessment of the Glasgow Coma Scale (GCS). In ICU patients with delirium, the GCS is very often reduced. In this case, the predictive value of the SAPS II is poor [31]. Therefore, to exclude possible confounding factors affecting ICU long-term survival, several pre-existing comorbidities or interventions affecting ICU long term survival were analyzed in the light of the available literature on the topic [21–28] and introduced in the multivariable logistic regression model with survival at 4 years after ICU-discharge as outcome of interest.
The multivariable logistic regression model confirmed that ICU-delirium is independently associated with reduced long-term survival. This effect on the survival is more evident in the first year after ICU discharge but it persists at all the analyzed time points.
Our paper has several strengths: firstly, data on long-term survival for patients with ICU-delirium are scarce and we add evidence on it. Secondly, we decided to perform our survival analysis only on ICU survivors. This choice was justified by the fact that in patients who died at the ICU, a negative delirium score could be due to the extreme severity of the initial clinical condition rather than to an actual absence of delirium. Thirdly, we adopted a standardized definition of delirium based not only on the ICDSC delirium score, but also on the confirmation of the diagnosis through clinical judgement, as ultimately expressed by the ICD-10 coding. The ICDSC score has a pooled sensitivity of 0.83 and a specificity of 0.87 [32, 33]. The two complementary criteria allowed us to improve our specificity excluding those patients who, even presenting a positive ICDSC, were not considered by the clinician to have delirium. Fourthly, we performed the survival analysis grouping the patients by age and according to possible confounding factors that have already been demonstrated to have an association with long term survival after ICU. Lastly, our study is, to the best of our knowledge, the first one to analyze in a large patient population the association of ICU delirium and survival with a long-term follow-up of up to four years.
Some limitations need to be mentioned. Our analysis is in fact limited by the monocentric and retrospective nature of the study, which possibly reduce the generalizability of the results. Moreover, the grouping by age and the selection of comorbidities and interventions in the analysis was based on the available literature and this does not completely exclude a selection bias. We also recognize that in the definition of the delirious patients, the clinical judgment is subjective and, ultimately, a further selection bias cannot be excluded. Furthermore, the difference in the ICU- and hospital-length of stay between groups may rise some concerns leading to possible inclusion bias. However, ICU-LOS was considered in the multivariable logistic regression model and ICU-delirium still remained independently associated with long-term mortality.
Conclusion
In the study population, ICU-delirium was independently associated with reduced long-term survival. Patients who developed ICU-delirium had a reduced survival up to four years after ICU discharge and this association was particularly evident in patients above 55 years of age.
These discrepancies in the available data could be explained with methodological issues, such as the small sample size of the analyzed populations, the heterogeneity in the definition of delirium, different timing in the assessment of the possible association between delirium and mortality, as well as possible inclusion bias.
Our retrospective observational study focuses on the association between ICU-delirium and long-term survival. To do that, we applied a standardized definition of delirium based both on regular scoring and clinical judgment to a large population of ICU-patients and investigated survival at different time point: 1-year, 2-year, 3-year and up to 4-year following ICU-discharge.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
Not applicable.
Abbreviations
- ICU
Intensive Care Unit
- LOS
Length of stay
- ICDSC
Intensive Care Delirium Screening Checklist
- SAPS II
Simplified Acute Physiology Score II
- PDMS
Patient Data Management System
- ICU-LOS
Intensive Care Unit – Length of stay
- H-LOS
Hospital – Length of stay
- ICD
International Classification of Diseases
- IQR
Interquartile range
- ARD
Absolute Risk Difference
- D
Delirium group
- ND
Non-delirium group
- CAM-ICU
Confusion Assessment Method – Intensive Care Unit
- GCS
Glasgow Coma Scale
- COPD
Chronic Obstructive Pulmonary Disease
Author contributions
ID, MAK, JB and GB designed the study. JB and GB supervised the study. ID and GB wrote the initial draft. MAK and JB performed the analysis and created tables and figures. EJ and RS gave critical input to the manuscript. All authors read and approved the final version of the manuscript.
Funding
This study was partially funded by a grant from the Iten-Kohaut Foundation in collaboration with the USZ Foundation, awarded to JB.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
The study was approved by the cantonal ethics committee of Zurich (BASEC: 2020–02695). The study complies with the Declaration of Helsinki, the Guidelines on Good Clinical Practice (GCP-Directive) issued by the European Medicines Agency as well as with Swiss law and regulatory authority requirements.
Consent for publication
Informed consent was obtained from the patients or from their relatives whenever possible. The cantonal ethics committee of Zurich granted permission to use the data of all patients who did not object to the use of their data for research purposes.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Ignazio De Trizio and Maria Angeliki Komninou contributed equally to this work.
Jan Bartussek and Giovanna Brandi contributed equally to this work.
References
- 1.Salluh JI, Wang H, Schneider EB, Nagaraja N, Yenokyan G, Damluji A, et al. Outcome of delirium in critically ill patients: systematic review and meta-analysis. BMJ. 2015;350:h2538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pandharipande PP, Ely EW, Arora RC, Balas MC, Boustani MA, La Calle GH, et al. The intensive care delirium research agenda: a multinational, interprofessional perspective. Intensive Care Med. 2017;43(9):1329–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li HC, Yeh TY, Wei YC, Ku SC, Xu YJ, Chen CC, et al. Association of Incident Delirium with short-term mortality in adults with critical illness receiving mechanical ventilation. JAMA Netw Open. 2022;5(10):e2235339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ouimet S, Riker R, Bergeron N, Cossette M, Kavanagh B, Skrobik Y. Subsyndromal delirium in the ICU: evidence for a disease spectrum. Intensive Care Med. 2007;33(6):1007–13. [DOI] [PubMed] [Google Scholar]
- 5.van den Boogaard M, Schoonhoven L, van der Hoeven JG, van Achterberg T, Pickkers P. Incidence and short-term consequences of delirium in critically ill patients: a prospective observational cohort study. Int J Nurs Stud. 2012;49(7):775–83. [DOI] [PubMed] [Google Scholar]
- 6.Brummel NE, Jackson JC, Pandharipande PP, Thompson JL, Shintani AK, Dittus RS, et al. Delirium in the ICU and subsequent long-term disability among survivors of mechanical ventilation. Crit Care Med. 2014;42(2):369–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bulic D, Bennett M, Georgousopoulou EN, Shehabi Y, Pham T, Looi JCL, et al. Cognitive and psychosocial outcomes of mechanically ventilated intensive care patients with and without delirium. Ann Intensive Care. 2020;10(1):104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wilcox ME, Girard TD, Hough CL. Delirium and long term cognition in critically ill patients. BMJ. 2021;373:n1007. [DOI] [PubMed] [Google Scholar]
- 9.Milbrandt EB, Deppen S, Harrison PL, Shintani AK, Speroff T, Stiles RA, et al. Costs associated with delirium in mechanically ventilated patients. Crit Care Med. 2004;32(4):955–62. [DOI] [PubMed] [Google Scholar]
- 10.Al Huraizi AR, Al-Maqbali JS, Al Farsi RS, Al Zeedy K, Al-Saadi T, Al-Hamadani N et al. Delirium and its association with short- and long-term Health outcomes in medically admitted patients: a prospective study. J Clin Med. 2023;12(16). [DOI] [PMC free article] [PubMed]
- 11.Rood PJT, van de Schoor F, van Tertholen K, Pickkers P, van den Boogaard M. Differences in 90-day mortality of delirium subtypes in the intensive care unit: a retrospective cohort study. J Crit Care. 2019;53:120–4. [DOI] [PubMed] [Google Scholar]
- 12.Salluh JI, Soares M, Teles JM, Ceraso D, Raimondi N, Nava VS, et al. Delirium epidemiology in critical care (DECCA): an international study. Crit Care. 2010;14(6):R210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Duprey MS, van den Boogaard M, van der Hoeven JG, Pickkers P, Briesacher BA, Saczynski JS, et al. Association between incident delirium and 28- and 90-day mortality in critically ill adults: a secondary analysis. Crit Care. 2020;24(1):161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrell FE Jr., et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA. 2004;291(14):1753–62. [DOI] [PubMed] [Google Scholar]
- 15.Pisani MA, Kong SY, Kasl SV, Murphy TE, Araujo KL, Van Ness PH. Days of delirium are associated with 1-year mortality in an older intensive care unit population. Am J Respir Crit Care Med. 2009;180(11):1092–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Moskowitz EE, Overbey DM, Jones TS, Jones EL, Arcomano TR, Moore JT, et al. Post-operative delirium is associated with increased 5-year mortality. Am J Surg. 2017;214(6):1036–8. [DOI] [PubMed] [Google Scholar]
- 17.Wolters AE, van Dijk D, Pasma W, Cremer OL, Looije MF, de Lange DW, et al. Long-term outcome of delirium during intensive care unit stay in survivors of critical illness: a prospective cohort study. Crit Care. 2014;18(3):R125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fiest KM, Soo A, Hee Lee C, Niven DJ, Ely EW, Doig CJ, et al. Long-term outcomes in ICU patients with delirium: a Population-based Cohort Study. Am J Respir Crit Care Med. 2021;204(4):412–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bergeron N, Dubois MJ, Dumont M, Dial S, Skrobik Y. Intensive care Delirium Screening Checklist: evaluation of a new screening tool. Intensive Care Med. 2001;27(5):859–64. [DOI] [PubMed] [Google Scholar]
- 20.Le Gall JR, Lemeshow S, Saulnier F. A new simplified Acute Physiology score (SAPS II) based on a European/North American multicenter study. JAMA. 1993;270(24):2957–63. [DOI] [PubMed] [Google Scholar]
- 21.Ali Abdelhamid Y, Plummer MP, Finnis ME, Biradar V, Bihari S, Kar P, et al. Long-term mortality of critically ill patients with diabetes who survive admission to the intensive care unit. Crit Care Resusc. 2017;19(4):303–9. [PubMed] [Google Scholar]
- 22.Garside T, Bedford JP, Vollam S, Gerry S, Rajappan K, Watkinson PJ. Increased long-term mortality following new-onset atrial fibrillation in the intensive care unit: a systematic review and meta-analysis. J Crit Care. 2022;72:154161. [DOI] [PubMed] [Google Scholar]
- 23.Doherty Z, Kippen R, Bevan D, Duke G, Williams S, Wilson A, et al. Long-term outcomes of hospital survivors following an ICU stay: a multi-centre retrospective cohort study. PLoS ONE. 2022;17(3):e0266038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zhu B, Chen X, Li W, Zhou D. Effect of Alzheimer Disease on Prognosis of Intensive Care Unit (ICU) patients: a propensity score matching analysis. Med Sci Monit. 2022;28:e935397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lee SY, Huh JW, Hong SB, Lim CM, Ahn JH. Short-term and long-term outcomes of critically ill patients with solid malignancy: a retrospective cohort study. Korean J Intern Med. 2024;39(6):957–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Damuth E, Mitchell JA, Bartock JL, Roberts BW, Trzeciak S. Long-term survival of critically ill patients treated with prolonged mechanical ventilation: a systematic review and meta-analysis. Lancet Respir Med. 2015;3(7):544–53. [DOI] [PubMed] [Google Scholar]
- 27.De Corte W, Dhondt A, Vanholder R, De Waele J, Decruyenaere J, Sergoyne V, et al. Long-term outcome in ICU patients with acute kidney injury treated with renal replacement therapy: a prospective cohort study. Crit Care. 2016;20(1):256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kondo T, Araki T, Imaizumi T, Sumita Y, Nakai M, Tanaka A, et al. Prognosis in patients with cardiogenic shock who received Temporary Mechanical Circulatory support. JACC Asia. 2023;3(1):122–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The strengthening the reporting of Observational studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495–9. [DOI] [PubMed] [Google Scholar]
- 30.Wang CM, Huang HW, Wang YM, He X, Sun XM, Zhou YM, et al. Incidence and risk factors of postoperative delirium in patients admitted to the ICU after elective intracranial surgery: a prospective cohort study. Eur J Anaesthesiol. 2020;37(1):14–24. [DOI] [PubMed] [Google Scholar]
- 31.Kornbluth J, Bhardwaj A. Evaluation of coma: a critical appraisal of popular scoring systems. Neurocrit Care. 2011;14(1):134–43. [DOI] [PubMed] [Google Scholar]
- 32.Chen TJ, Chung YW, Chang HR, Chen PY, Wu CR, Hsieh SH, et al. Diagnostic accuracy of the CAM-ICU and ICDSC in detecting intensive care unit delirium: a bivariate meta-analysis. Int J Nurs Stud. 2021;113:103782. [DOI] [PubMed] [Google Scholar]
- 33.Krewulak KD, Rosgen BK, Ely EW, Stelfox HT, Fiest KM. The CAM-ICU-7 and ICDSC as measures of delirium severity in critically ill adult patients. PLoS ONE. 2020;15(11):e0242378. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.