Calibration of the PREdiction of DELIRium in ICu Patients (PRE-DELIRIC) Score in a Cohort of Critically Ill Patients: A Retrospective Cohort Study : Dimensions of Critical Care Nursing

Secondary Logo

Journal Logo

Advanced Search
Clinical DIMENSION

Calibration of the PREdiction of DELIRium in ICu Patients (PRE-DELIRIC) Score in a Cohort of Critically Ill Patients

A Retrospective Cohort Study

Gravante, Francesco RN, MSN; Giannarelli, Diana BSc; Pucci, Antonello RN, MSN; Pisani, Luigi MD, PhD; Latina, Roberto RN, MSN, PhD

Author Information

Francesco Gravante, RN, MSN, is a nurse in the Intensive Care Unit at the Department of Anesthesiology, Local Health Authority of Caserta, Caserta, Italy.

Diana Giannarelli, BSc, is a biostatistician in the Biostatistics and Bioinformatic Unit, IRCCS Regina Elena, National Cancer Institute, Regina Elena, Rome, Italy.

Antonello Pucci, RN, MSN, is a nurse in the Trauma Intensive Care Unit at AO S. Camillo-Forlanini Hospital, Rome, Italy.

Luigi Pisani, MD, PhD, is a physician and researcher in the intensive care unit at Miulli Regional Hospital, Acquaviva delle Fonti, Italy.

Roberto Latina, RN, MSN, PhD, is an assistant professor in Nursing Science at the Department of Health Promotion Science, Maternal and Infant Care, Internal Medicine, and Medical Specialties, School of Medicine and Surgery, University of Palermo, Palermo, Italy.

Author contributions: All authors contributed to the realization of the project. Wrote the manuscript: Francesco Gravante, Roberto Latina and Luigi Pisani; concept: Francesco Gravante, Antonello Pucci; data collection: Antonello Pucci; interpretation and data analysis: Diana Giannarelli and Francesco Gravante; critical review: Roberto Latina, Diana Giannarelli, Luigi Pisani. All authors approved the final version.

The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article.

Address correspondence and reprint requests to: Francesco Gravante, RN, MSN, Intensive Care Unit, San Giuseppe Moscati Hospital, Via Antonio Gramsci, 1, 81031, Aversa (CE), Italy ([email protected]).

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.dccnjournal.com).

Dimensions of Critical Care Nursing 42(4):p 187-195, 7/8 2023. | DOI: 10.1097/DCC.0000000000000586

Abstract

Background 

To predict delirium in intensive care unit (ICU) patients, the Prediction of Delirium in ICU Patients (PRE-DELIRIC) score may be used. This model may help nurses to predict delirium in high-risk ICU patients.

Objectives 

The aims of this study were to externally validate the PRE-DELIRIC model and to identify predictive factors and outcomes for ICU delirium.

Method 

All patients underwent delirium risk assessment by the PRE-DELIRIC model at admission. We used the Intensive Care Delirium Screening Check List to identify patients with delirium. The receiver operating characteristic curve measured discrimination capacity among patients with or without ICU delirium. Calibration ability was determined by slope and intercept.

Results 

The prevalence of ICU delirium was 55.8%. Discrimination capacity (Intensive Care Delirium Screening Check List score ≥4) expressed by the area under the receiver operating characteristic curve was 0.81 (95% confidence interval, 0.75-0.88), whereas sensitivity was 91.3% and specificity was 64.4%. The best cut-off was 27%, obtained by the max Youden index. Calibration of the model was adequate, with a slope of 1.03 and intercept of 8.14. The onset of ICU delirium was associated with an increase in ICU length of stay (P < .0001), higher ICU mortality (P = .008), increased duration of mechanical ventilation (P < .0001), and more prolonged respiratory weaning (P < .0001) compared with patients without delirium.

Discussion 

The PRE-DELIRIC score is a sensitive measure that may be useful in early detection of patients at high risk for developing delirium. The baseline PRE-DELIRIC score could be useful to trigger use of standardized protocols, including nonpharmacologic interventions.

Intensive care unit (ICU) delirium is an acute disorder characterized by a rapid onset, impaired cognition, and a fluctuating course.1,2 The available literature presents discordant data regarding the prevalence of delirium in the ICU, reported in a range between 26% and 60%,3-5 with a significant increase in patients receiving mechanical ventilation.6 Intensive care unit delirium constitutes an independent predictor of mortality and is associated with prolonged ICU length of stay, increased duration of mechanical ventilation, and respiratory weaning.

In literature, ICU delirium has been classified into 3 types: hypoactive, hyperactive, or mixed.2 Intensive care unit patients with hyperactive delirium were characterized by psychomotor agitation, whereas hypoactive delirium was described as slowing psychomotor functions.1,2 Misdiagnosis and underestimation of the phenomenon lead to higher care costs and long-suffering patients.7 Several risk factors such as low Glasgow Coma Scale (GCS) score, low mean arterial pressure (MAP), high creatinine level, low hematocrit level, low pH, and low Pao2/Fio2 contribute to the onset of ICU delirium.8-12

Early detection of high-risk patients for delirium would facilitate the early use of nonpharmacological interventions to manage and reduce the duration of ICU delirium and its impact on health outcomes.7,13,14 The Guidelines for the Prevention and Management of Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption also suggest using nonpharmacologic interventions to reduce modifiable risk factors for ICU delirium, such as early mobilization and programs for sleep optimization.7

Several predictive models of ICU delirium are available in the literature.14,15 The Prediction of Delirium in ICU Patients (PRE-DELIRIC) score is a 10-factor score that estimates the risk of delirium in the first 24 hours after ICU admission.16 These factors include age, Acute Physiology and Chronic Health Evaluation II (APACHE II) score, admission group, coma, infection, metabolic acidosis, sedatives and morphine, urea concentration, and urgent admission. Since its development in 2012, several authors have validated the PRE-DELIRIC score in several countries,17 yielding good discrimination and calibration ability.5,14 Recently, studies on the Chinese population have reported an area under the receiver operating characteristic curve (AUROC) of 0.81, with similar findings from studies performed across 7 European countries.14,18 In contrast, in the United Kingdom, a lower AUROC of 0.65 is just above the practical use cut-off.14 The higher the AUROC, the better the performance of the PRE-DELIRIC score in distinguishing between patients with and patients without delirium.14,18

Despite the different validation studies of the PRE-DELIRIC in Europe and beyond, we need more data to verify the discrimination and calibration of the PRE-DELIRIC score in the Italian population. We thus performed a retrospective study seeking to externally validate the PRE-DELIRIC score in a cohort of Italian ICU patients and to determine predictive factors and outcomes for ICU delirium. We hypothesized that the PRE-DELIRIC score has a high predictive capacity for ICU delirium defined by international standards.

Prediction Model

The PRE-DELIRIC score is a prediction model that aids health care practitioners in detecting patients at high risk for delirium.

METHODS

Design and Ethics

We conducted a retrospective cohort study using the electronic medical records of consecutive patients admitted to 2 Italian ICUs (general ICU and surgical ICU) between December 2017 and July 2018. The authors wrote this research report according to the Strengthening the Reporting of Observational Epidemiological Studies.19 We obtained ethical approval from the Ethics Committee of Latium Region 1 (no. 1331/CE Lazio 1, 5/11/2020). The research did not influence patient care, and patient data were anonymized. All data were encrypted and were accessible only to the researchers.

Study Population and Setting

Patients were included according to the following inclusion criteria: (a) 18 years or older; (b) both sexes; (c) admitted to the 2 ICUs between December 1, 2017, and July 31, 2018; and (d) stayed in the ICU for ≥24 hours. The following exclusion criteria were applied: (a) certified diagnosis of alteration of cognitive status or delirium at ICU admission; (b) history of cognitive impairment; (c) having an essential sensitive disorder (auditory or visual); and (d) transferred from other ICUs. The surgical ICU comprises 8 beds, whereas the general ICU caters to trauma and has 12 beds. The patients included were trauma, surgical, and medical patients hospitalized at the surgical or general ICU in the same hospital.

Data Extraction and Prediction Model

Researchers recovered all data from the nurse's electronic medical records (Centricity Critical Care, GE Healthcare). Two independent authors performed the data extraction and analysis, with no missing data recorded for this cohort. Collected data included new predictive factors such as patient baseline characteristics, neurological examination on admission, and GCS score, biochemical markers, MAP, heart rate, respiratory rate, and biochemical markers such as sodium, potassium, creatinine, hematocrit, white blood count, pH, and Pao2/Fio2. The PRE-DELIRIC score was computed on all patients to stratify the risk of delirium according to the 10 prespecified predictive factors: (1) age, (2) APACHE II score, (3) admission group, (4) coma, (5) infection, (6) metabolic acidosis, (7) use of sedatives and (8) morphine, (9) urea concentration, and (10) urgent admission16 (Supplemental Digital Content 1, Supplementary table, https://links.lww.com/DCCN/A126). According to van den Boogaard, we defined the PRE-DELIRIC score as a percentage representing the probability of developing ICU delirium according to 4 risk groups: (a) low risk (a score between 0% and 20%, a sensitivity of 80%, and a specificity of 74%), (b) moderate risk (a score between >20% and 40%, a sensitivity of 62%, and a specificity of 88%), (c) high risk (a score between >40% and 60%, a sensitivity of 46%, and a specificity of 94%), and (d) very high risk (a score >60%, a sensitivity of 30%, and a specificity of 97%). Nurses registered the PRE-DELIRIC score and variable within 24 hours of admission to the ICU.

Study Outcomes

The primary outcome was the predictive validity of the PRE-DELIRIC score for ICU delirium in terms of discrimination and calibration capacity and calculating the cut-off value and recalibration of the PRE-DELIRIC tool. Secondary outcomes included the new predictive factors for the development of ICU delirium, ICU length of stay, ICU mortality, duration of mechanical ventilation, and weaning status. All patients who reported a progressive reduction of mechanical ventilation needs had a positive weaning status.

Delirium Assessment

Delirium was defined using the Italian version of the Intensive Care Delirium Screening Check List (ICDSC).20 The baseline measurements of ICU delirium included the administration of ICDSC by well-trained nurses, regularly every 8 hours. The ICDSC is an appraisal tool composed of 8 items with 2 possibilities: absent (score = 0) or present (score = 1). The items include the evaluation of (1) consciousness, (2) inattention, (3) disorientation, (4) hallucination-delusion-psychosis, (5) psychomotor agitation or retardation, (6) inappropriate speech or mood, (7) sleep-wake cycle disturbance, and (8) fluctuation of symptomatology, according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. The ICDSC score ranged from 0 to 8; a score of ≥4 indicates the presence of delirium (ICDSC positive), a score of 1 to 3 indicates subsyndromal delirium, and a score of 0 no suggests the absence of delirium (ICDSC negative). We considered patients with delirium as having at least 1 positive ICDSC screening during the patients' complete intensive care stay, whereas we defined subsyndromal delirium as having an ICDSC score between 1 and 3.

Data Analysis

To determine the sample size, we considered a delirium prevalence of 30% and a standard error of 3.5%. We estimated a sample size of 165 patients to obtain a 95% confidence interval (CI) of 23% to 37%. Researchers assessed the association between categorical variables through the χ2 test. We evaluated differences between mean values and standard deviation with the Student t test. To determine the diagnostic performance of the PRE-DELIRIC scale, we used the receiver operating characteristic curve; we reported area under the curve with its 95% CI. We used the Youden index to detect the cut-off with the best performance in terms of sensitivity and specificity. To calculate the calibration power, we plotted the observed outcome against the mean predicted outcome within risk groups of patients that were ranked by increasing estimated probability. We used IBM-SPSS version 20.0 statistical software for analysis.

RESULTS

In total, we screened 253 consecutive patients from the 2 ICUs and excluded 88 of them (Figure 1). A total of 165 patients were eligible for analysis. Overall, 107 (65%) of the patients were men, with a median age of 56.7 years (SD, 18.3 years). The mean (SD) APACHE II score was 17.1 (8.6), with a mean (SD) GCS score of 8.3 (5.5). The ICU survival was 86.7% (143), with a mean (SD) ICU length of stay of 13.5 (15.4) days (Table 1).

F1
Figure 1:
Flowchart of patients included.
TABLE 1 - Description of Patient Characteristics and Outcomes (n = 165)
Clinical Characteristics and Outcomes Mean (SD) n (%)
Characteristics
 Male sex 107 (65.0)
 Age (years) 56.7 (18.3)
 APACHE II 17.1 (8.6)
 GCS 8.3 (5.5)
 Temperature 36.5 (0.8)
 MAP 83.4 (16.9)
 Respiratory rate 12 (2)
 Heart rate 82 (19)
Biochemical markers
 pH 7.4 (0.1)
 Pao 2/Fio 2 101 (36)
 Sodium 142 (4.9)
 Potassium 4.0 (0.6)
 Creatinine 1.2 (1.1)
 Hematocrit 30.7 (8.1)
 WBC 11.4 (6.1)
Outcomes
 ICU length of stay 13.5 (15.4)
 Days of mechanical ventilation 10.4 (12.8)
 Mortality 22 (13.3)
 Weaning respiratory
  Positive 65 (39.4)
  Negative 53 (32.1)
  No mechanical ventilation 47 (28.5)
Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; GCS, Glasgow Coma Scale; MAP, mean arterial pressure; WBC, white blood count; ICU, intensive care unit.

Discrimination Capacity of PRE-DELIRIC

The prevalence of delirium was 55.8% (92), with a high discrimination property of the PRE-DELIRIC score for patients with and without ICU delirium (AUROC, 0.81; 95% CI, 0.75-0.88) (Figure 2). The best cut-off was a PRE-DELIRIC score of 27% (Table 2), which had a sensitivity of 91.3% and a specificity of 64.4% for detecting ICU delirium onset. This result was the most accurate prediction of delirium onset. A high variation in the probability of having a diagnosis of ICU delirium emerged in patients with a positive ICDSC test (likelihood positive ratio, 2.56), whereas there was a low variation in the probability of having a diagnosis of ICU delirium in patients with a negative test (ICDSC negative) (likelihood negative ratio, 0.4) (Table 2).

F2
Figure 2:
Receiver operating characteristic curves of the prediction model (AUROC, 0.81; 95% CI, 0.75-0.88).
TABLE 2 - Sensitivity, Specificity, True-Positive, and False-Positive of PRE-DELIRIC Score
Cut-off Sensitivity (%) Specificity (%) Positive Likelihood Ratio Negative Likelihood Ratio
20% 97.8 54.8 2.16 0.04
25% 92.4 63.0 2.50 0.12
27% (Youden index)a 91.3 64.4 2.56 0.14
30% 88.0 65.8 2.57 0.18
35% 82.6 68.5 2.62 0.25
40% 81.5 72.6 2.98 0.25
45% 75.0 72.6 2.74 0.34
50% 64.1 75.3 2.60 0.48
Abbreviation: PRE-DELIRIC, Prediction of Delirium in ICU Patients.
aBest cut-off.

The sensitivity, specificity, and positive and negative likelihood ratios for the other 7 points, chosen conventionally, are reported in Table 2. The slope of the discriminating property of the model is 0.54, and the AUROC of 0.81, which is associated with a positive likelihood ratio of 2.56, confirms the high discriminative power of the scale in patients with ICU delirium in the critically ill Italian population.

Calibration of the PRE-DELIRIC

The PRE-DELIRIC model yielded good calibration. The event detected was slightly higher than predicted, but once we added an 8% probability increment, we observed a similar result between the event predicted and observed (slope, 1.033; intercept, 8.140). Discrimination capacity was very high (AUROC, 0.81; slope, 0.543). The calibration slope and intercept show a direct relationship between the expected ICU delirium trend and the observed outcome frequency for each risk group (Figure 3).

F3
Figure 3:
Calibration plot of the pooled data, with a calibration of 1.033 and intercept of 8.143.

Risk Factors and Outcomes

In Table 3, we report the variables that define neurological function associated with ICU delirium. Patients who developed ICU delirium scored very low on the GCS, unlike patients who did not develop ICU delirium (P < .0001). The risk factors associated with onset of ICU delirium were low MAP value (P < .0001), low pH value (P < .0001), low Pao2/Fio2 (P = .007), low hematocrit value (P < .0001), and high creatinine value (P = .03) (Table 3).

TABLE 3 - Comparison of Patients With Delirium Versus Without Delirium
Clinical Characteristics and Outcomes Patient With Delirium (n = 92) Patient Without Delirium (n = 73) P
Mean SD Mean SD
Baseline characteristicsa
 Age (years) 57.7 18.9 57.6 17.7 .97
 GCS 6.0 4.5 11.3 5.3 <.0001
 Temperature 36.5 0.8 36.3 0.8 .13
 MAP 79.1 14.9 88.6 17.9 <.0001
 Respiratory rate 11.6 2.2 12.0 2.4 .21
 Heart rate 83.3 18.1 81.2 20.0 .49
Baseline biochemical markersa
 pH 7.34 0.1 7.39 0.1 <.0001
 Pao 2/Fio 2 94.3 24.1 110.5 45.7 .007
 Sodium 142.4 4.5 141.0 5.3 .08
 Potassium 4.0 0.7 4.0 0.5 .88
 Creatinine 1.3 1.4 1.0 0.5 .03
 Hematocrit 28.5 6.2 33.4 9.3 <.0001
 WBC 11.5 6.4 11.4 5.6 .96
Outcomes
 ICU length of stay 17.2 15.7 8.9 13.7 <.0001
 Days of mechanical ventilation 13.8 12.8 6.2 11.6 <.0001
 Mortality n % n % .008
  Survivor (n = 143) 74 51.7 69 48.3
  Not a survivor (n = 22) 18 81.5 4 18.2
 Weaning respiratory n % n % <.0001
  Positive (n = 65) 45 69.2 20 30.8
  Negative (n = 53) 42 79.2 11 20.8
  No mechanical ventilation (n = 47) 5 10.6 42 89.4
Abbreviations: GCS, Glasgow Coma Scale; MAP, mean arterial pressure; WBC, white blood count; ICU, intensive care unit.
aBaseline measures.

Patients with delirium had increased ICU length of stay and duration of mechanical ventilation (P < .0001), with a worsening of the weaning respiratory process (n = 42 [79.2%] vs n = 11 [20.8%]; P < .0001). The presence of ICU delirium was also associated with increased mortality during hospitalization in ICU (P < .008) (Table 3).

Sensitivity and Specificity of the PRE-DELIRIC

The PRE-DELIRIC score showed an excellent discriminative capacity for a patient with or without ICU delirium, with an AUROC of 0.81 (95% CI, 0.75-0.88), sensitivity of 91.3%, and specificity of 64.4%, with adequate calibration of the model (slope, 1.033; intercept, 8.140).

DISCUSSION

The primary findings of this study can be summarized as follows: (1) the PRE-DELIRIC score retains a good discriminative capacity and adequate calibration; (2) low GCS score, low MAP, low hematocrit, low Pao2/Fio2, pH value, and high creatinine value were associated with the development of ICU delirium; and (3) ICU delirium is associated with patient-centered outcomes in critically ill patients.

Discrimination Capacity of PRE-DELIRIC

According to the max Youden index, the best cut-off in our cohort of critically ill patients was 27%. In a Chinese population, the cut-off is different, and Liang and colleagues5 only obtained an excellent discrimination ability of PRE-DELIRIC in high-risk patients for values >49%. The validation findings of our analysis were comparable with the result obtained from the external validation carried out in China.5 In 2014, van den Boogaard and colleagues recalibrated the scale in 8 intensive care units in 5 European countries and Australia. The authors observed a variable result on the performance scale among different countries.14 The lower value of AUROC was obtained in Prescot UK (AUROC, 0.65), whereas in Australia, the figure was similar to our findings (AUROC, 0.81).14 In our sample, the discriminative power of PRE-DELIRIC in critically ill patients shows a good performance in a moderate-risk group of ICU-onset delirium.

Although validation findings are independent of the prevalence of ICU delirium,21,22 the different validation results between the countries may have several explanations. They are attributable to differences in risk factors between populations and varying analgosedation protocols applied in clinical practice. Alcohol withdrawal also impacts the onset of ICU delirium, although it is seldom considered an important precipitating risk factor of ICU delirium, particularly in high-drinking countries.3,23 The heterogeneity of the results may also be related to systematic errors in the sample studies.24 The significant differences in delirium predictors between the countries motivated the external validation of the prediction model in the Italian ICU population.

Calibration of the PRE-DELIRIC

Calibration is the extent to which predictions, on average, agree with the overall results found in the sample. A calibration slope of 1 and an intercept of 0 show an excellent calibration ability.25 In our study, a good calibration emerged between the prediction of ICU delirium and the frequency of the event itself. This finding agrees with the calibrations carried out in other countries (calibration slope, 0.894, intercept, −0.178; calibration slope, 1.09, intercept, 0.08)5,14 and in the parent model study (slope, 0.93, intercept, −0.29).16 In our sample, we detected a slight difference between predicted and observed events (slope, 1.03; intercept, 8.1). This difference may be attributable to the use of coefficient calculated in the Italian population (data not shown). The PRE-DELIRIC model has a good calibration and discrimination in a delirious patient. In clinical practice, nurses can use the prediction model in the first hours after ICU admission.

Risk Factors and Outcomes

Recently, some studies have reported a strong association between high and low GCS scores and the onset of delirium.8,10,26 This study's mean/median GCS value was compatible with severe neurological impairment (mean, 8.3). The literature suggests some factors, theories, and mechanisms leading to delirium, which are associated with an alteration in neurotransmitter synthesis, function, and availability that mediates the complex behavioral and cognitive changes observed in delirium.27 Mean arterial pressure was considered an essential index of the perfusion brain.28 The difference between MAP and intracranial pressure determines cerebral perfusion pressure, which represents the net pressure gradient that provides the proportion of valuable oxygen to the survival of brain tissue. The literature reported that low MAP was a risk factor for ICU-onset delirium,8 like our results. Hypoxia is an independent risk factor for acute brain injury that can lead to long-term cognitive impairment.9,29,30 In our cohort, patients with a low pH value and Pao2/Fio2 were at higher risk of ICU delirium. A recent review reported a low hematocrit level as a risk factor for ICU delirium.12 Low levels of hemoglobin, related to trauma and reduced hematocrit, can significantly reduce the cerebral brain's oxygen supply and lead to ICU delirium.12 Elevated serum creatinine values also characterized our patient cohort with ICU delirium. Acute kidney injury may facilitate inflammatory processes of the cerebral cortex and hippocampus, leading to histologic changes and locomotor dysfunction.31,32

The strong association between ICU delirium and ICU mortality, length of stay, duration of mechanical ventilation, and respiratory weaning mirrors the results of a meta-analysis.33 The risk ratio for mortality among patients with ICU delirium was 2.2 (95% CI, 1.8-2.7; P < .001) compared with those without the syndrome. The same study underlined how patients with longer ICU stay and undergoing mechanical ventilation are at the highest risk of developing delirium.33 On the same line, Jeon and colleagues34 found that intubated patients with ICU delirium have more frequently prolonged weaning (odds ratio, 2.3; 95% CI, 1.3-4.2) compared with those without ICU delirium.

Implications for Nursing and Clinical Practice

Managing critically ill patients with ICU delirium requires early detection of the risk group. A predictive model could help health care professionals identify patients most at risk early. When the predictive model fails the prediction, the bedside nurses work through a continuous 24-hour observation and play a decisive role in early delirium assessment, detection, prevention, and treatments in agreement with the ICU physician.

The early identification of high-risk patients would allow the implementation of standardized protocols based on nonpharmacological interventions such as early mobility and exercise, family engagement and empowerment, and sleep-wake management.13,35 Early detection of high-risk patients may reduce adverse outcomes. This approach offers excellent reliability when the instrument calibration considers the adaptation of a specific population. Therefore, this predictive model could be used in daily practice to improve the quality of intensive care and foster the culture of ICU delirium prevention among ICU teams. Nurses can use PRE-DELIRIC score for early prediction of high-risk group patients, at ICU admission, with ICU-onset delirium and implement no pharmacologic intervention to manage ICU patients.

Strengths and Limitations of the Study

This study has several limitations. The small sample size limits the generalization of results, whereas the retrospective design did not allow for all potential confounders to be included in the analysis. A strength was the systematic method of detecting ICU delirium once a shift by identifying patients with ICU delirium regardless of the fluctuating course of the syndrome.

CONCLUSION

The PRE-DELIRIC model shows good stability and accuracy for the discrimination and calibration properties. These findings foster the closer daily assessment for delirium based on the PRE-DELIRIC score at baseline, by nurses in clinical practice, especially in implementing strategies to prevent delirium.

Acknowledgments

The authors thank Surui Liang, PhD, for her critical advice on statistical analysis.

References

1. Guze SB. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV). AJP. 1995;152(8):1228–1228. doi:10.1176/ajp.152.8.1228.
2. Arumugam S, El-Menyar A, Al-Hassani A, et al. Delirium in the intensive care unit. J Emerg Trauma Shock. 2017;10(1):37–46. doi:10.4103/0974-2700.199520.
3. Gravante F, Giannarelli D, Pucci A, et al. Prevalence and risk factors of delirium in the intensive care unit: an observational study. Nurs Crit Care. 2021;26(3):156–165. doi:10.1111/nicc.12526.
4. Kanova M, Sklienka P, Kula R, Burda M, Janoutova J. Incidence and risk factors for delirium development in ICU patients—a prospective observational study. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2017;161(2):187–196. doi:10.5507/bp.2017.004.
5. Liang S, Chau JPC, Lo SHS, Bai L, Yao L, Choi KC. Validation of Prediction of Delirium in ICU Patients (PRE-DELIRIC) among patients in intensive care units: a retrospective cohort study. Nurs Crit Care. 2021;26(3):176–182. doi:10.1111/nicc.12550.
6. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA. 2001;286(21):2703–2710. doi:10.1001/jama.286.21.2703.
7. Devlin JW, Skrobik Y, Gélinas C, et al. Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit Care Med. 2018;46(9):e825–e873. doi:10.1097/CCM.0000000000003299.
8. Brown CH. Delirium in the cardiac surgical ICU. Curr Opin Anaesthesiol. 2014;27(2):117–122. doi:10.1097/ACO.0000000000000061.
9. Gong F, Ai Y, Zhang L, Peng Q, Zhou Q, Gui C. Relationship between PaO2/FiO2 and delirium in intensive care: a cross-sectional study. J Intensive Med. 2022;3(1):73–78. Published 2022 Oct 1. doi:10.1016/j.jointm.2022.08.002.
10. Rasheed AM, Amirah M, Abdallah M, Awajeh AM, Parameaswari PJ, Al Harthy A. Delirium incidence and risk factors in adult critically ill patients in Saudi Arabia. J Emerg Trauma Shock. 2019;12(1):30–34. doi:10.4103/JETS.JETS_91_18.
11. Yazici AB, Yazici E, Erol A. Delirium and high creatine kinase and myoglobin levels related to synthetic cannabinoid withdrawal. Case Rep Med. 2017;2017:1–6. doi:10.1155/2017/3894749.
12. Zhang HJ, Ma XH, Ye JB, Liu CZ, Zhou ZY. Systematic review and meta-analysis of risk factor for postoperative delirium following spinal surgery. J Orthop Surg Res. 2020;15(1):509. doi:10.1186/s13018-020-02035-4.
13. Blair GJ, Mehmood T, Rudnick M, Kuschner WG, Barr J. Nonpharmacologic and medication minimization strategies for the prevention and treatment of ICU delirium: a narrative review. J Intensive Care Med. 2019;34(3):183–190. doi:10.1177/0885066618771528.
14. van den Boogaard M, Schoonhoven L, Maseda E, et al. Recalibration of the delirium prediction model for ICU patients (PRE-DELIRIC): a multinational observational study. Intensive Care Med. 2014;40(3):361–369. doi:10.1007/s00134-013-3202-7.
15. Cowan SL, Preller J, Goudie RJB. Evaluation of the E-PRE-DELIRIC prediction model for ICU delirium: a retrospective validation in a UK general ICU. Crit Care. 2020;24(1):123. doi:10.1186/s13054-020-2838-2.
16. van den Boogaard M, Pickkers P, Slooter AJ, et al. Development and validation of PRE-DELIRIC (Prediction of Delirium in ICU Patients) delirium prediction model for intensive care patients: observational multicentre study. BMJ. 2012;344:e420. doi:10.1136/bmj.e420.
17. Ho MH, Chen KH, Montayre J, et al. Diagnostic test accuracy meta-analysis of PRE-DELIRIC (Prediction of Delirium in ICU Patients): a delirium prediction model in intensive care practice. Intensive Crit Care Nurs. 2020;57:102784. doi:10.1016/j.iccn.2019.102784.
18. Hanison J, Umar S, Acharya K, Conway D. Evaluation of the PRE-DELIRIC delirium prediction tool on a general ICU. Crit Care. 2015;19(suppl 1):P479. doi:10.1186/cc14559.
19. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–349. doi:10.1016/j.jclinepi.2007.11.008.
20. Gian Domenico G, Piergentili F. Cultural and linguistic validation of the Italian version of the Intensive Care Delirium Screening Checklist. Dimens Crit Care Nurs. 2012;31(4):246–251. doi:10.1097/DCC.0b013e318256e0cc.
21. Metz CE. Basic principles of ROC analysis. Semin Nucl Med. 1978;8(4):283–298. doi:10.1016/S0001-2998(78)80014-2.
22. Obuchowski NA. Receiver operating characteristic curves and their use in radiology. Radiology. 2003;229(1):3–8. doi:10.1148/radiol.2291010898.
23. Newman RK, Stobart Gallagher MA, Gomez AE. Alcohol Withdrawal. In: StatPearls. Treasure Island, FL: StatPearls Publishing; 2022.
24. Park SH, Goo JM, Jo CH. Receiver operating characteristic (ROC) curve: practical review for radiologists. Korean J Radiol. 2004;5(1):11–18. doi:10.3348/kjr.2004.5.1.11.
25. Stevens RJ, Poppe KK. Validation of clinical prediction models: what does the “calibration slope” really measure? J Clin Epidemiol. 2020;118:93–99. doi:10.1016/j.jclinepi.2019.09.016.
26. Maneewong J, Maneeton B, Maneeton N, et al. Delirium after a traumatic brain injury: predictors and symptom patterns. Neuropsychiatr Dis Treat. 2017;13:459–465. doi:10.2147/NDT.S128138.
27. Maldonado JR. Neuropathogenesis of delirium: review of current etiologic theories and common pathways. Am J Geriatr Psychiatry. 2013;21(12):1190–1222. doi:10.1016/j.jagp.2013.09.005.
28. Mount CA, Das JM. Cerebral Perfusion Pressure. In: StatPearls. Treasure Island, FL: StatPearls Publishing; 2022.
29. Girard TD, Thompson JL, Pandharipande PP, et al. Clinical phenotypes of delirium during critical illness and severity of subsequent long-term cognitive impairment: a prospective cohort study. Lancet Respir Med. 2018;6(3):213–222. doi:10.1016/S2213-2600(18)30062-6.
30. Hopkins RO, Weaver LK, Pope D, Orme JF, Bigler ED, Larson-Lohr V. Neuropsychological sequelae and impaired health status in survivors of severe acute respiratory distress syndrome. Am J Respir Crit Care Med. 1999;160(1):50–56. doi:10.1164/ajrccm.160.1.9708059.
31. Altmann C, Andres-Hernando A, McMahan RH, et al. Macrophages mediate lung inflammation in a mouse model of ischemic acute kidney injury. Am J Physiol Renal Physiol. 2012;302(4):F421–F432. doi:10.1152/ajprenal.00559.2010.
32. Hoke TS, Douglas IS, Klein CL, et al. Acute renal failure after bilateral nephrectomy is associated with cytokine-mediated pulmonary injury. J Am Soc Nephrol. 2007;18(1):155–164. doi:10.1681/ASN.2006050494.
33. Salluh JI, Wang H, Schneider EB, et al. Outcome of delirium in critically ill patients: systematic review and meta-analysis. BMJ. 2015;350:h2538. doi:10.1136/bmj.h2538.
34. Jeon K, Jeong BH, Ko MG, et al. Impact of delirium on weaning from mechanical ventilation in medical patients. Respirology. 2016;21(2):313–320. doi:10.1111/resp.12673.
35. Marra A, Ely EW, Pandharipande PP, Patel MB. The ABCDEF bundle in critical care. Crit Care Clin. 2017;33(2):225–243. doi:10.1016/j.ccc.2016.12.005.
Keywords:

Calibration; Delirium; ICU; Intensive Care Units, PRE-DELIRIC

Supplemental Digital Content

Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved.