Characteristics and Risk Factors for Delirium in The Surgical Step-Down Unit

Article information

J Acute Care Surg. 2023;13(3):124-130

Publication date (electronic) : 2023 November 23

doi : https://doi.org/10.17479/jacs.2023.13.3.124

Soyeon Yoon, Jinjoo Kim, Yoon Jung An, Jeonghui Ok, Yooun-Joong Jung

Department of Nursing, Asan Medical Center, Seoul, Korea

^*Corresponding Author: Yooun-Joong Jung, Department of Nursing, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea, Email: joong@amc.seoul.kr

Received 2023 October 18; Revised 2023 November 1; Accepted 2023 November 3.

Abstract

Introduction

Due to aging societies and advances in medical technology, the demand for long-term care is increasing [1]. As a result, the need for intensive care unit (ICU) bed capacity has increased [2]. Moreover, the premature transfer of patients to general wards increases the ICU readmission rate and mortality [3].

Efforts have been made to ensure patient recovery and intensive care of ICU patients to prevent unnecessary prolonged stays in the ICU [4–6]. This effort led to the formation and establishment of the Step-down unit (SDU) in the US which provides recovery and intensive care [7].

The use of SDU beds has been reported in the US and Europe to significantly reduce the rates of rehospitalization and ICU readmission for intensive care patients, and also increases the utilization rate of ICU beds [4,8]. In Korea, SDUs were first set up in 2017 in university hospitals [9]. They offer recovery care for long-term ICU patients and provide postoperative monitoring for high-risk patients with underlying diseases.

While delirium occurred in 7.6% of the patients in an Italian study who were transferred from the ICU to an SDU after acute-phase treatment [10], another study in the US reported that it persisted in the post-ICU period in 47% of the ICU delirium cases [11]. Management of delirium is important because untreated delirium is associated with long-term cognitive impairment, falls, and prolonged hospital stays [12–14].

Nurses play an important role in detecting the presence of delirium because they spend more time than doctors with patients at their bedside and can observe changes in the patients’ conditions [15]. Currently, the Confusion Assessment Method (CAM) is the most widely used screening tool for delirium [16]. However, since the CAM is a 10-item behavioral observation scale, which takes more than 10 minutes to complete, there is a need for a delirium assessment tool that can be used in hospital wards for a rapid delirium check [17].

The short CAM was developed in 2014 based on the CAM algorithm for quickly assessing delirium [18]. This 4-item scale can be completed within 5 minutes, greatly simplifying the delirium assessment [19]. It has been reported that the short CAM is useful in emergency rooms and hospital wards, particularly for elderly patients aged 70 years and older [18].

This study was conducted using the short CAM to identify risk factors for delirium and to provide foundational data for the development of delirium prevention guidelines for patients in the SDU.

Materials and Methods

Results

1. Incidence rate and patterns of delirium

The incidence rate of delirium in the surgical SDU was 15.4% (Figure 1). Delirium occurred most frequently on the day of admission to the surgical SDU, (n = 16, 45.7%), followed by (n = 15, 42.8%) on the 2nd day of admission. The main occurrence of delirium was between 14:30 and 22:30 accounting for 30 cases (85.7%). Regarding the types of delirium, hypoactive delirium was the most common (n = 19, 54.3%), followed by hyperactive delirium (n = 10, 28.6%; Table 1).

Figure 1

Flow diagram of the subject selection.

CAM = confusion assessment method.

Table 1

Characteristics and patterns in patients with delirium (N = 35)

2. Comparison of general characteristics between the delirium and non-delirium groups of surgical SDU patients

When comparing the groups (without delirium and delirium) differences in various aspects were observed (Table 2). In terms of general characteristics, there were distinctions in age (p < 0.001). Regarding disease-specific characteristics, differences were observed in CCI (p = 0.016), hypertension (p = 0.021), stroke (p = 0.001), and APACHE II scores (p = 0.002). In terms of treatment characteristics, disparities were noted for HD (p = 0.006), WBC (p = 0.043), and albumin (p = 0.050). In terms of environmental characteristics, there was a difference in the use of restraints (p < 0.001).

Table 2

Comparison of patient characteristics between delirium and non-delirium groups (n = 227)

3. Risk factors for delirium in patients admitted to a surgical SDU

Based on multivariate logistic regression analysis significant variables including age, CCI, HTN, stroke, APACHE II score, HD, WBC, albumin level, and use of restraint were included. Logistic regression analysis determined that the regression model was valid statistically significant (p < 0.001), with an explanatory power of 41.9% (using Nagelkerke’s coefficient of determination). The classification accuracy was 87.2%. The model’s goodness of fit was assessed using the Hosmer and Lemeshow’s method and indicated that data fitted the model well (p = 0.562). The analysis results indicated that age, HTN, stroke, WBC, and the use of restraint was the most significant factors influencing the occurrence of delirium (Table 3). As age increases, there is a higher association with delirium (95% CI: 1.025–1.15, p = 0.005). Having hypertension increases the odds 3.3 times (95% CI: 1.206–9.463, p = 0.024), and having a history of stroke increases the odds 4.4 times (95% CI: 1.583–12.362). The higher the WBC count, the higher the odds of delirium (95% CI: 1.044–1.246, p = 0.004). The use of restraint increases the odds 22.2 times (95% CI: 4.748–103.644, p < 0.001).

Table 3

Multiple logistic regression analysis of risk factors associated with delirium

Discussion

This retrospective study was conducted with patients admitted to the Surgical SDU using the short CAM as an assessment tool for delirium in order to identify risk factors for delirium. This analysis, spanning a period of 6 months, revealed a delirium incidence rate of 15.4%. In the 2014 Sullivan study [18], the delirium incidence rate in older adults (≥ 70 years) using the abbreviated CAM ranged from 36% to 68%, which was lower than the mean rate calculated in this study. Direct comparison between the 2 studies is not possible due to contextual differences, including the fact that this study was conducted in a single surgical ward in Korea. In this current study, delirium occurred in the majority of cases on the first and second days following admission to the SDU, which is consistent with the results of similar studies [22,23]. In terms of the time of day, 85.7% of delirium cases occurred between 14:30 and 22:30, which may be attributed to the “sundown syndrome” characterized by a sudden surge of neurotic symptoms such as agitation, confusion, and anxiety late in the afternoon and into the evening [24].

Among the different types of delirium, hypoactive delirium was observed in 54.3% of the patients. Previous studies have reported that hypoactive patients are less likely to be identified as delirious, as they tend to comply with treatment [25] and thus escape the attention of the medical staff [26]. This study showed that a high proportion of hypoactive delirium, was detected suggesting that rapid identification through assessments was beneficial. Delayed delirium diagnosis, negatively impacts mortality and prognosis. Therefore, early detection of delirium through delirium assessment proves to be a crucial factor in delirium management.

The factors influencing the occurrence of delirium were determined to be age, hypertension, stroke, WBC count, and use of restraints. Age was a significant risk factor for the incidence of delirium, which is in line with previous studies that report a high incidence of delirium among elderly patients aged 65 years [27] or 75 years or older [28]. In this current study, the mean age of the delirium group was 10 years higher than that of the non-delirium group (75.2 vs. 65.2%), supporting the finding that age is a risk factor for delirium. This may be explained by the fact that as one gets older, one is more affected by and less able to adapt to changes in physical and mental functions, and is more prone to functional decline [29]. Previous studies have reported that the incidence of delirium is higher when the patient has hypertension as a comorbidity [30,31]. Hypertension was identified as a risk factor in this current study. In general, patients with hypertension as an underlying disease are subjected to sustained vascular damage, leading to cerebral perfusion hypoxia, which may trigger cognitive impairment and delirium [32]. Stroke was also identified as a risk factor in this current study, which is consistent with the findings of previous studies that reported reduced brain perfusion induced by stroke may cause spatial and emotional disturbances, thus increasing the risk of delirium [33,34]. An elevated neutrophil count [35] increases the incidence of infection, which sends signals that induce systemic fever and stress responses to the brain, adversely affecting behavior [36] and cognitive function, which is thought to be responsible for the increased risk for delirium [37]. The use of restraint increased the incidence rate of delirium 32 times in the ICU [38] and several studies have demonstrated that the use of restraint is a cause of delirium [39–41]. In this current study, physically restrained patients showed a 22-fold higher incidence of delirium. The mechanism by which the use of restraint causes delirium is not yet clear, but one possible explanation is that its use can cause tracheal extubation, threaten the patient’s dignity, and trigger negative emotions such as anxiety and anger which may lead to delirium [38,40]. On the other hand, the use of restraint is a measure necessary to keep delirium patients, who undergo sudden changes in their state of consciousness and behavior, from dangerous actions and accidents, such as unplanned extubation or falls [38]. Thus, it may be an important intervention for delirium prevention that surgical SDU nurses pay attention to the emotional changes in physically restrained patients, re-evaluate the necessity of using the restraint, and make efforts and formulate strategies to gradually reduce the time required to apply the restraint.

This study has several potential limitations. The most significant one is its retrospective nature, utilizing EMR, which may hinder the generalizability of the research results. Additionally, the relatively low incidence of delirium is noteworthy. Previous studies have shown a wide range of incidence of delirium in ICU populations, ranging from 30% to 80%, primarily in studies targeting mechanically ventilated patients using CAM-ICU [10–12]. This study is limited by the specific characteristics of one surgical ICU population, the use of the short CAM diagnostic tool limited to patients capable of communication, and potential variability in the handling of the short CAM assessment tool due to nursing staff shifts. Furthermore, the study does not include the relationship with the use of sedatives and antipsychotic drugs which are factors associated with the risk of delirium.

Conclusion

This study is significant as the first attempt to assess delirium using the short CAM tool in patients admitted to the surgical SDU of 3rd-tier hospital in South Korea and to investigate the relationship between the occurrence of delirium and various risk factors. Notable clinical features include the predominance of hypoactive delirium, higher occurrence in the afternoon and evening, and the majority of instances of delirium occurring within 1 or 2 days of admission to the surgical SDU. Consideration of patient-related risk factors such as age, hypertension, and stroke, a disease-related risk factor such as elevated WBC, and an environmental risk factor such as sedative use emphasizes the multifaceted nature of delirium. This underscores the importance of implementing standardized delirium screening protocols, fostering interprofessional collaboration, reducing the frequency of restraint application, and improving the ward environment. Additionally, interventions such as establishing a family support system and minimizing restraint application can be included to reduce the occurrence and impact of delirium. Proactive measures that address the multifaceted nature of delirium and improvements in the ward environment are crucial for optimizing the prevention and management of delirium.

Data Availability

All relevant data are included in this manuscript.

Article information Continued