Abstract
The purpose of this study is to demonstrate the characteristics, contributing factors and recommended policy changes associated with emergence delirium. Relevant data were extracted from the PAAd Thai database of 2,006 incident reports which were conducted from 1 January to 31 December 2015. Details pertinent to the patient, surgery, anaesthetic and systematic factors were reviewed independently. Seventeen incidents of emergence delirium were recorded. Emergence delirium was common in the following categories: male (70.6%), over 65 years of age (53%), elective surgery (76%) and orthopedic surgery (35%). Physical restraint was required in 53% (9 of 17) of cases and 14 patients (82%) required medical treatment. One patient developed postoperative delirium and required medical treatment. The study led to the following recommendations: Development of a classification of practice guidelines and a screening tool, and training for restraint use.
Introduction
Emergence delirium (ED) is defined as a disturbance of a patient's awareness or attention to his/her environment, with disorientation and perceptual alterations. These symptoms include hallucination, delusion, hypersensitivity to stimuli, restlessness and involuntary physical activity occurring immediately after the postoperative period (Sikich & Lerman 2004). ED can result in several consequences such as physical injury to patients and medical staff, harm to the operative site, removal of the wound dressing, endotracheal tubes or intravenous catheters, or the requirement of more physical restraint and pharmacological treatment (Sikich & Lerman 2004, Lepousé et al 2006).
Emergence delirium can occur in all age groups, it is commonly found in pediatric patients. ED usually occurs immediately after emerging from general anaesthesia and often resolves spontaneously, whereas postoperative delirium (POD) may not relate to emergence delirium and can develop between one and three days after surgery (Silverstein et al 2007). However, a previous study found that ED could be a strong predictor of POD (Sharma et al 2005).
The incidence of ED in adult patients varies between studies, ranging from 3 to 22% (Rose 1996, Lepousé et al 2006, Yu et al 2010, Radtke et al 2010, Kim et al 2015b, Munk et al 2016), while the occurrence is even higher in pediatric patients, varying from 10 to 80% (Cravero et al 2000, Cole et al 2002). Several risk factors of ED include age (Banchs & Lerman 2014); being male (Yu et al 2010, Munk et al 2016); alcohol abuse (Kim et al 2015a); preoperative anxiety (Banchs & Lerman 2014); exposure to specific anaesthetic agents (Banchs & Lerman 2014, Munk et al 2016); specific types of surgery (Lepousé et al 2006, Radtke et al 2010); and postoperative pain (Radtke et al 2010, Yu et al 2010).
The Royal College of Anaesthesiology of Thailand previously conducted two prospective cohort studies into anaesthesia related adverse events in several hospitals across Thailand. These studies were the Thai-Anaesthesia Incident Study in 2005 (Charuluxananan 2005) and the Thai Anaesthesia Incident Monitoring Study in 2007 (Charuluxananan et al 2008). However, both studies were conducted more than 10 years ago and neither of these studies included ED as an adverse anaesthetic outcome.
In 2015, Perioperative Anaesthetic Adverse Events in Thailand (PAAd Thai), a multicenter study among 22 hospitals in Thailand, was conducted. The outcomes of the study included various adverse anaesthetic events and their consequences during the perioperative periods (Charuluxananan et al 2017, Punjasawadwong et al 2017). The overall incidence of suspected ED in all age groups according to the database of the PAAd Thai study was 0.60 per 10,000 (95% CI: 0.34-0.97) (Charuluxananan et al 2017). The authors then explored factors affecting ED more extensively. This study was retrospectively analyzed. The purpose of this study was to determine characteristics, management, outcomes, and contributory factors associated with ED in PAAd Thai Studies. The outcomes of the study were used to make recommendations for policy changes associated with ED.
Methods
A prospective observational study was conducted between 1 January and 31 December 2015 after being approved by each of the institutional ethical committees. This was a multicenter study which included patients receiving anaesthesia in the eight university hospitals and fourteen non-university hospitals across Thailand.
The attending anaesthetists, resident anaesthetists, or nurse anaesthetists of each institution were requested to complete a standardized incident reporting form and clinical details in addition to usual anaesthetic records whenever ED occurred. They were required to complete the forms within a 24 hour postoperative period. Details of data recorded included demographics:
age gender weight height and medical data American Society of Anesthesiologists of Physical Status (ASA PS) co-morbidities anaesthesia profile (anaesthetic techniques, airways, and monitors) surgical profile (types and sites of operation, types of surgical procedures) level and experience of anaesthetists anaesthetic adverse events (types, timing and location of adverse events) information relating to surgical safety checklists immediate outcomes (≤24 hours) and long term outcomes (≤7 days).
All incident report forms were verified by the individual site manager and then sent to the data management unit. The incident reports of ED were reviewed by the group of delirium reviewers. Three peer reviewers independently evaluated all completed record forms and provided their assessment pertaining to causes of ED, incident-related factors including:
patient factors (age, gender, ASA PS, co-morbidities) anaesthetic factors (type of anaesthetic, airway devices, monitors) surgical factors (type and sites of operation) and systematic factors (elective or emergency operation, working day or out-of-hours surgery, level or experience of anaesthesia providers, participation in surgical safety checklists).
Preventive strategies, contributing factors, incident minimizing factors and recommended policy changes were also described. If there were any conflicts in opinion, a consultative agreement was made in order to achieve a consensus.
Details of ED incidents were reviewed and levels of sedation and agitation were classified using the Richmond Agitation Sedation Scale (RASS) (Sessler et al 2002). The RASS scores were described as follows:
−5 unarousable −4 deep sedation −3 moderate sedation −2 light sedation −1 drowsy 0 alert and calm +1 restless +2 agitated +3 very agitated +4 combative.
ED was suspected when a RASS score became ≥+1 at any time in the post anaesthesia care unit (PACU) or in the ward. Descriptive statistics were analyzed using STATA software (version 11.0, College station, TX, USA).
Results
Of the 333,219 patients undergoing anaesthesia in all participating institutions, 2,206 incident reports were reported in this PAAd Thai study. Seventeen incident reports of suspected emergence delirium were retrieved.
Patient characteristics are presented in Table 1. ED occurred most frequently in patients aged over 65 years (53%), males (70.6%), and at ASA PS III (53%). The most frequent surgery related incidence included elective surgery (14 of 17, 82.3%), operating during the working day (14 of 17, 82.3%), orthopedic surgery (6 of 17, 35.3%), and urologic surgery (4 of 17, 23.5%). Fifty-three percent of anaesthesics (9 of 17) were delivered by resident anaesthetists and 17.6% (3 of 17) were performed by nurse anaesthetists. Ninety-four percent of patients received general anaesthesia and required endotracheal intubation (12 of 17, 75%).
Patient characteristic
Abbreviations BMI – body mass index, ASA PS – American Society of Anaesthesiologists Physical Status, NRS – numeric rating scale
One patient receiving spinal anaesthesia experienced ED. Eight out of 17 patients (47%) developed ED during the emergence period in the operating room and the remainder of ED incidences occurred in the PACU. Five patients had a score of RASS +1 (29.4%), 2 patients of RASS +2 (11.7%), and 9 patients RASS +3 (53%).
Details of all patients with ED are described in Table 2. Nine patients (53%) required physical restraint and fourteen patients (82.3%) required medical treatment. Opioid (41.2%) alone, and a combination of opioid with midazolam (23.5%) or propofol (23.5%) were frequently used for relieving clinical symptoms of ED. No patients were injured or pulled out their intravenous catheter. All symptoms of ED disappeared before patients were discharged from PACU. Long term consequences of ED included postoperative delirium in one patient (8 days), this patient requiring psychiatrist consultation and medical treatment. He completely recovered before hospital discharge.
Details of emergence delirium events (n=17)
Abbreviations GA – general anesthesia, ET – endotracheal tube, LMA – laryngeal mask airway, N2O – nitrous oxide, CVA – cerebrovascular accident, ESRD – end stage renal disease, CAD – coronary artery disease
The most common factors related to ED were a systematic factor (inexperience of anaesthesia care providers (14 of 17, 82,3%)) and age related patient factors (advanced age (11 of 17, 64.7%) and being a pediatric patient (2 of 17, 11.7%)) (Table 3). Inexperience (14 of 17, 82.3%) was the main contributory factor.
Factors related to emergence delirium
ED was considered for preventability (5 of 17, 29.4%) and partial preventability (6 of 17, 35.3%). Incident minimizing factors include increased vigilance (13 of 17, 76.5%) and encountering prior experience (10 of 17, 58.8%). Recommended policy changes were additional training (14 of 17, 82.3%), the creation of guideline practices about anaesthesia management in patients at high risk of ED (9 of 17, 52.9%), and quality assurance activities (9 of 17, 52.9%) (Table 4).
Contributory factors, incident minimizing factors and recommend policy changes
Discussion
This study was the first, multicenter prospective study reporting frequency, characteristics, related factors, contributing factors and corrective strategies for ED in Thailand.
Limitations of the study
There were a few limitations of this study. Firstly, although this study was a prospective data collection study, retrospective analysis of ED was conducted. The diagnosis of ED was based on a clinical description of individual patients in the incident report form only. Secondly, most anaesthesia care providers recognize only agitated patients as delirious and most institutions in Thailand do not use standard tools for the diagnosis of ED in routine clinical practice. Therefore, other delirious patients who present with less obvious symptoms, for example drowsiness, will be undiagnosed. Finally, the incidence of ED in this study may be underestimated particularly after surgery during nonofficial times because this was based on voluntary reports.
Tools used
In this study, more than half of the delirious patients were scored as ‘very agitated’ and most of these patients required either physical and/or medical restraints. The incidence of ED in this study was lower than those reported by previous studies (Rose 1996, Lepousé et al 2006, Radtke et al 2010, Yu et al 2010, Kim et al 2015b, Munk et al 2016). However, direct comparison between studies was difficult because there were differences in study design, study population, diagnostic tools, or type of operation. Several tools have been used for assessment of the level of sedation and agitation during the postoperative period. However, previous studies have demonstrated that both the Riker agitation sedation scale and RASS are easy to use and have excellent inter-rater reliability (Lepousé et al 2006). The RASS was selected for assessment of emergence delirium in this study because it could be used for assessment in both adult and pediatric patients (Kerson et al 2016).
Risk factors for ED
Previous studies reported that potential risk factors associated with ED were patient-related, anaesthesia-related and surgical-related factors (Vlajkovic & Sindjelic 2007, Banchs & Lerman 2014, Kanaya 2016).
In these studies, patient-related factors of ED included age and male gender. Elderly (>70 years old) and pediatric patients (2-7 years) could be more susceptible to ED due to a decrease in level of gamma aminobutyric acid and acetylcholine induced by medication or an immaturity of cholinergic receptors (Banchs & Lerman 2014). The prospective cohort studies reported that male patients were at greater risk of ED than female patients (Yu et al 2010, Munk et al 2016). This might be due to a lower pain tolerance in males after surgery compared to females (Yu et al 2010).
Anaesthesia-related factors of ED were volatile anaesthetic agents (Vlajkovic & Sindjelic 2007) and preoperative benzodiazepine (Lepousé et al 2006). Volatile anaesthetic agents, such as sevoflurane or desflurane, could contribute to the occurrence of ED as a result of a number of mechanisms. For example, volatile anaesthetic agents could interfere with the balance between the neuronal synaptic inhibition and excitation in the central nervous system (Yli-Hankala et al 1999).
Another mechanism associated with ED is the delayed recovery of cognitive function compared to other functions of the brain (Vlajkovic & Sindjelic 2007). In addition, although most incidences of ED occurred in elective patients, only three patients (17.6%) received benzodiazepine for premedication.
From other studies related surgical factors for ED included postoperative pain (Yu et al 2010, Kim et al 2015a) and type of surgical procedure (Vlajkovic & Sindjelic 2007). Moderate to severe postoperative pain was a potential risk factor of ED in several studies (Radtke et al 2010, Xará et al 2013) and inadequate pain control was also found to be a contributory factor of ED (Davis et al 1999). Our study also found that almost 30% of patients had acute moderate to severe pain after emergence from anaesthesia, their symptoms improving after receiving an opioid supplement.
The most common surgical procedures related to ED were head and neck surgery (Vopel-Lewis et al 2003), orthopedic surgery (Radtke et al 2010), breast surgery (Lepousé et al 2006), abdominal surgery (Lepousé et al 2006), and urologic surgery (Kim et al 2015a). In our study, orthopedic and urologic surgery accounted for almost 65% of all surgical procedures (10 of 17 patients).
Precipitating factors
Unlike POD, few studies have reported precipitating factors of ED which might occur during anaesthesia. During the review of all anaesthetic and medical records of ED patients, the authors observed that 35% of patients (6 of 17 patients) had prolonged intraoperative hypotension (systolic blood pressure <90 mmHg for 20-30 minutes) without association with any surgical bleeding and required frequent administration of a vasopressor. Such hemodynamic instability could have an influence on pain management during the intraoperative period resulting in severe postoperative pain. Most of these patients were elderly and some patients had concurrent co-morbidities, which might make them susceptible to hemodynamic disturbances such as hypertension or renal disease. ED could reduce cerebral blood flow and increase the risk of POD.
Emergence delerium has been reported as being an uncommon neurological complication after spinal anaesthesia (Bilotta et al 2013). However, our study reported one patient as experiencing ED after spinal anaesthesia. This patient received continuous propofol infusion and intravenous midazolam during anaesthesia due to severe anxiety. A randomized controlled study found that patients receiving deep sedation during spinal anaesthesia had a higher incidence of postoperative delirium due to sedative-induced alteration in brain activity (Sieber et al 2010).
In this study, most of the patients with suspected ED received anaesthesia administered by resident anaesthetists (9 of 17) and nurse anaesthetists (3 of 17) who had less than four years’ experience in performing anaesthesia (82%). Therefore, inexperience and inadequate knowledge regarding ED were considered as being the main contributory factors for ED.
Quality assurance activities including the Morbidity and Mortality Conference or Critical Incident Reporting are methods which can be used to improve patient safety by learning from the incidents or near misses. Resident anaesthetists should be supervised by consultant anaesthetists while providing anaesthesia for patients at a high risk of ED or whilst taking care of delirious patients during the emergence period. Evidence-based clinical practice guidelines for the anaesthetic management in patients at a high risk of ED should be developed in order to help anaesthetists and trainees in their provision of appropriate preoperative and postoperative management, and enable them to deliver an appropriate depth of anaesthesia. Although the occurrence of ED is difficult to prevent, these guidelines may help reduce the incidence and severity of ED.
Pre- intra- and post operative management
Preoperative assessment should include an identification of potential risk factors of ED, an assessment of level of anxiety, and provide ED education to patients and their family members for high risk patients (Hudek 2009). On the day of surgery, pediatric patients and their parents should arrive at the operating theatre early in order to help them become familiar with the hospital environments and parents should be encouraged to accompany the patients during the induction period (Hudek 2009).
Intraoperative management should avoid any medication or any events that might contribute to ED such as hypoxemia, hypotension or fluctuation in blood pressure particularly in elderly patients or hypertensive patients, and provide appropriate treatment during extensive surgical bleeding (Hirsch et al 2015).
Postoperative care includes adequate pain control, and the assessment of depth and level of sedation and level of ED by using standard tools such as RASS or the Riker Sedation-Agitation Scale during emergence and the postoperative period. If patients get severely agitated, correctable causes of ED such as hypoxemia, severe pain, electrolyte imbalance, anxiety or sleep deprivation should be identified and appropriately treated (Bray et al 2004). Endotracheal tubes or urinary catheters should be removed quickly after surgery if there is no indication.
Non-pharmacologic interventions should be used in order to prevent and treat ED as well as minimize restraint use (Bray et al 2004). These include early mobilization, a reduction of external stimuli, minimization of sleep deprivation, and reduction of the use of invasive monitoring (Bilotta et al 2013). Provision of a calm environment could promote smooth emergence and reduce the level of patient anxiety (Hudek 2009, Banchs & Lerman 2014).
Prevention and treatment
Appropriate medication including analgesic or sedative medications are effective in the prevention and treatment of ED. However, the role of physical restraint is still in debate. For restraint to be used the benefits, such as the prevention of removal of invasive monitoring or endotracheal tubes, should outweigh the risks, such as physical injury or increased agitation. Physical restraint has been reported as being an independent risk factor of agitation or delirium (Hofsø & Coyer 2007).
The British Association of Critical Care Nurses recommends that physical restraints can only be considered if other therapeutic interventions are ineffective. Guidelines or institutional policies for prevention or treatment of ED including non-pharmacologic intervention, chemical restraint and also physical restraint should be developed. In addition, educational and training programmes about all types of restraints should be provided for all staff with focus on restraint techniques, indications and contraindications, and how and when to use restraint with safety during daily clinical practice (Bray et al 2004).
Conclusion
Policy changes regarding strategic planning need to take place, including an improvement in certain quality assurance programmes in local practice in each hospital.
In order to prevent and reduce severity of ED, the study recommends:
implementation of the clinical practice guidelines for anaesthetic management in patients at a high risk of ED frequent assessment of the level of sedation and delirium by using various standard tools, and the provision of additional training programmes regarding restraint techniques for all staff in anaesthesiology departments are recommended
Footnotes
Acknowledgements
This study was a contributory part of the study into Perioperative Anesthetic Adverse Events in Thailand (PAAd Thai) which was financially supported by the Royal College of Anesthesiologists of Thailand, the Faculties of Medicine of Chiang Mai University, Chulalongkorn University (Rachadapisakesompotch fund), Khon Kaen University, Mahidol University (Siriraj Hospital and Ramathibodi Hospital), and Prince of Songkla University, the Health System Research Institute, and the National Research Council of Thailand.
No competing interests declared