Related Factors and Treatment of Postoperative Delirium in Old Adult Patients: An Integrative Review

“Postoperative delirium” is defined as delirium occurring in the hospital up to one week after a procedure or before discharge (whichever occurs first) that meets the DSM-5 diagnostic criteria. Objectives: To describe the risk factors related to this pathology and identify effective non-pharmacological forms of treatment. An integrative review of the available literature was performed. The search results considered included all quantitative studies published between 2011 and 2019 in both English and Spanish. A total of 117 studies were selected. Advanced age was identified as the principal risk factor for postoperative delirium. Nursing interventions appear to be the key to preventing or reducing the seriousness of delirium after an anaesthetic episode. The aetiology of postoperative delirium remains unknown, and no treatment exists to eliminate this pathology. The role of nursing staff is fundamental in the prevention, diagnosis, and management of the pathology.


Introduction
"Postoperative delirium" is defined as delirium that occurs in the hospital up to one week after the procedure or before discharge (whichever occurs first) that meets the DSM-5 diagnostic criteria. Factors to consider in postoperative delirium include previous drug use, alcohol use, dementia or previous delirium, persistent effects of drugs, metabolic abnormalities, hypoxia, electrolyte imbalance, and infection [1]. The condition is characterized by changes in cognitive function (memory impairment, disorientation, agitation, and/or confused speech) and in the level of consciousness (alternating periods of alert and periods of delirium) [2]. Onset is sudden, fluctuating, and reversible; it cannot be explained by pre-existing neurocognitive disorders or as a sign that the patient is developing such a disorder, and it manifests shortly after surgery (within days or hours) [2][3][4].
Emergence delirium shares certain similarities to postoperative delirium, and although it does not have a clear definition, it shares some risk factors with postoperative delirium: both can be triggered by noxious stimuli during the perioperative and postoperative periods, but they are different phenomena [5]. Surgical patients may experience emergence delirium initially, which later progresses to postoperative delirium. While emergence delirium is more prevalent in healthy pediatric patients and young adults, postoperative delirium usually affects older patients with multiple comorbidities. The two phenotypes are often referred to as synonymous in the literature, but it is important to differentiate them, the highest incidence, 50%, with the hyperactive and hypoactive forms having incidences of 10-30% and 20-40%, respectively.
Over the years, a series of postoperative delirium assessment scales have been developed for use by nursing staff in anaesthesia recovery rooms and intensive care units in the immediate postoperative period. Among these, several stand out, including the Confusion Assessment Method (CAM) [16,19,42]; the Confusion Assessment Method-Intensive Care Unit (CAM-ICU) [50,[60][61][62]; the Nursing Delirium Screening Scale (NuDESC) [24,60,61], and the Intensive Care Delirium Screening Checklist (ICDSC) [27,63]. This list is not exhaustive and other scales do exist, such as the Delirium Rating Scale Revised-98 [27].
Since nursing staff are usually the first to observe changes in the behaviour of patients, it is very important that these professionals are able to identify postoperative delirium early so that the pathology can be treated and complications avoided in the short-and long-term [64].
It must be emphasized that there are no existing treatments, either pharmacological or otherwise, that can completely eliminate the risk of postoperative delirium. Nevertheless, the literature does identify certain pharmaceuticals and interventions that alleviate it. Interventions most often used by nursing professionals include placing clocks and calendars in highly visible locations to help the patient regain spatio-temporal orientation, encouraging family members to visit, designing non-drug-based schedules to promote good-quality sleep, clear differentiation between night and day, elimination of nuisance noise at night, allowing the use of hearing aids and glasses to avoid audio-visual deficit, monitoring liquid intake to ensure an optimal water-electrolyte balance, avoiding malnutrition, and the evaluation and active treatment of pain [17,65].
In terms of research questions and objectives, this review integrates theory and empirical evidence concerning postoperative delirium in the context of postoperative recovery (Post Anaesthesia Care Unit) and intensive care units in order to establish a framework to support nursing interventions aimed at reducing the incidence of this condition. Because postoperative delirium has a higher incidence in the old adult population, this study focuses on this age group.
To this end, we propose the following objectives related to postoperative delirium: in the first instance, we aim to determine the major factors associated with its onset and development and, in the second instance, we aim to outline the principal treatment methods, distinguishing nursing interventions from pharmacological interventions while also determining which drugs have been proven to be useful in the treatment of postoperative delirium and which have not, be studied in the context of post-anaesthesia care units and intensive care units.

Materials and Methods
An integrative review was completed, as it was felt that this was the most appropriate method for summarizing and analysing both empirical and theoretical literature in a systematic fashion such that the results would have direct applicability to clinical practice.
The bibliographic search and selection of articles were completed in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). To assess the methodological quality of the included articles, the assessment guidelines established in the John Hopkins Nursing Evidence-Based Practice Model were used [66].

Search Strategy
The literature search was completed between the months of May and June 2021. The scientific databases used for this search were PubMed, Cochrane, Google Scholar, SciELO, CUIDEN, and CINAHL.
We used search terms from the Medical Subject Headings (MeSH) and the Descriptores de Ciencias de la Salud (DeCS), in both English and Spanish. These included "delirium" (D003693), and "Emergence delirium" (D000071257); the related search terms "Postoperative Delirium", "Anaesthesia delirium", and "surgery delirium"; additionally, their corresponding Spanish translations.
For the purposes of the search, related terms were linked using the Boolean operator "OR", while in order to link different concepts the Boolean operator "AND" was used.
As an example of a search strategy, an example of a PubMed search is shown. The filters used were "All types of articles", "Full text", "Since 2011", "Human", "Not pediatric", and "Not children", and the search term was "postoperative delirium".
Another example is a Cochrane search that was performed with the limits "all content types", "Cochrane Library Publication date between 2011 and 2019", "CENTRAL Trials only Original publication year between 2011 and 2019", and "Variations of search words".
A search in Google Scholar was performed with the filters "Articles", Specific Interval "2011-2019", "Include patents", and "Include citations" A search in SciELO was performed with the filters "Language: Spanish", "Language: English", and "Year of publication: 2011-2018".
A search in CUIDEN was performed with the filter "years between 2011 and 2019". The last example was a search performed in CINAHL with the filters "Search mode: Boolean/phrase, apply related words", "Full text", "Abstract available", "English language", "Human", "Publication subset: all", "Publication type: all", "Topics of interest: all", "Language: all", "Exclude medline records", " Randomized controlled trials", "Sex: all", "Age groups: all adult", and "Full text in PDF".

Selection Criteria
Articles selected for this study complied with the following criteria: all articles were published between 2011 and 2021 with the aim of selecting only the most recent articles on postoperative delirium. The reason for starting the search in 2011 was because there were two high-quality reviews on the subject covering articles published up until 2010 [65,66]. Articles published in English and Spanish were included. All quantitative studies were included, and full-text versions were obtained; randomized clinical studies, systematic reviews, integrative studies, and bibliographic reviews were included. Descriptive studies with the goal of synthesizing and giving an improved picture of the literature published to date were also included. The population included in our study comprised old adult patients (aged over 65 years old, because this is the age group associated with a higher incidence of delirium) of both sexes requiring admittance to a recovery or intensive care unit immediately after surgery. Regarding the results, the articles selected combined information about associated factors, treatment (pharmacological as well as non-pharmacological), nursing interventions to reduce the incidence of postoperative delirium, and methods used for diagnosis.
All articles that did not meet these criteria were excluded.

Article Selection and Data Abstraction
Initially, in order to minimize the risk of bias, the literature search was completed by two independent reviewers using the aforementioned databases, and the search criteria were outlined. Once the duplicates had been eliminated, the first selection of articles was completed through an independent analysis of article titles.
Following on from this, a second review was completed. This involved reading the abstracts and keywords of the articles selected by the initial two reviewers. In this way, the final selection of articles was composed. Any discrepancies were discussed and a solution found by consensus between reviewers.
Once the full texts of the articles had been obtained, articles were put through a third and final selection process in which articles to be included in this review were chosen.
Finally, once the final selection of articles for inclusion in this review was determined, the following data were extracted from each article: date of publication, title of article, type of study, population, type of intervention, intervention/variables of results, results, influencing factors, influencing pharmaceuticals, pharmaceuticals that reduced the incidence of the condition, non-pharmacological interventions, and diagnostic tests/scales used. This was conducted with the aim of synthesizing all of the information gathered and facilitating its management.
Given that the studies reviewed were very heterogeneous and used many different interventions and forms of evaluation, it was not possible to perform a meta-analysis. Figure 1 shows a PRISMA flow chart of the results obtained from the search process. In the initial search, a total of 50,658 results were obtained. In total, 181 articles fulfilled the selection criteria, and full-text versions of these were obtained. After a full analysis, 45 articles were excluded for the following reasons: 24 were not related to postoperative delirium, 5 were related to delirium types other than postoperative delirium, 7 studies were not complete at the time of access, 1 was a qualitative study, 2 articles concerned case studies of particular patients, 1 article involved a simulation rather than a study on humans, and 5 articles were reviews of other articles. lution found by consensus between reviewers.

Characterisation of Articles
Once the full texts of the articles had been obtained, articles were put through a thir and final selection process in which articles to be included in this review were chosen.
Finally, once the final selection of articles for inclusion in this review was determine the following data were extracted from each article: date of publication, title of articl type of study, population, type of intervention, intervention/variables of results, result influencing factors, influencing pharmaceuticals, pharmaceuticals that reduced the inc dence of the condition, non-pharmacological interventions, and diagnostic tests/scal used. This was conducted with the aim of synthesizing all of the information gathere and facilitating its management.
Given that the studies reviewed were very heterogeneous and used many differe interventions and forms of evaluation, it was not possible to perform a meta-analysis. Figure 1 shows a PRISMA flow chart of the results obtained from the search proces In the initial search, a total of 50,658 results were obtained. In total, 181 articles fulfille the selection criteria, and full-text versions of these were obtained. After a full analysis, 4 articles were excluded for the following reasons: 24 were not related to postoperative d lirium, 5 were related to delirium types other than postoperative delirium, 7 studies we not complete at the time of access, 1 was a qualitative study, 2 articles concerned ca studies of particular patients, 1 article involved a simulation rather than a study on hu mans, and 5 articles were reviews of other articles.  The final selection included 122 articles with the following distributions: 4 case-control studies, 22 randomized clinical trials, 2 analytic studies, 1 quasi-experimental study, 12 cohort studies, 7 descriptive studies, 37 observational studies, 1 case series, 2 meta-analyses, 31 bibliographic reviews, 1 integrative review, 10 systematic reviews, and 6 systematic reviews with a meta-analysis included.

Evaluation of Methodological Quality
In order to evaluate the methodological quality of the articles included in our study, we used the evaluation guidelines set out in the John Hopkins Nursing Evidence-Based Practice Model: a level was assigned to the evidence (I, II, III, IV, or V) and a quality rating of either A (high quality), B (good quality), or C (low quality, or having significant defects) was determined based on the study design and quality of the evidence contained in each article [66]. Figure 2 shows a table explaining the levels of evidence and the quality ratings proposed in the John Hopkins Nursing Evidence-Based Practice Model [66]. tematic reviews with a meta-analysis included.

Evaluation of Methodological Quality
In order to evaluate the methodological quality of the articles included in our study, we used the evaluation guidelines set out in the John Hopkins Nursing Evidence-Based Practice Model: a level was assigned to the evidence (I, II, III, IV, or V) and a quality rating of either A (high quality), B (good quality), or C (low quality, or having significant defects) was determined based on the study design and quality of the evidence contained in each article [66]. Figure 2 shows a table explaining the levels of evidence and the quality ratings proposed in the John Hopkins Nursing Evidence-Based Practice Model [66].
.  Table 1 presents the predisposing factors associated with the risk of a patient developing postoperative delirium. Table 2, on the other hand, lists the triggers that affect the patient. Although the pathophysiology of postoperative delirium occurring after an anaesthetic episode is still unknown, many factors were associated with its appearance and development. We divided these factors into two groups, which are shown in the following tables with the items ordered according to their level of evidence and quality (John Hopkins Nursing Evidence-Based Practice [66]) according to the studies from which they were extracted.  Table 1 presents the predisposing factors associated with the risk of a patient developing postoperative delirium. Table 2, on the other hand, lists the triggers that affect the patient. Although the pathophysiology of postoperative delirium occurring after an anaesthetic episode is still unknown, many factors were associated with its appearance and development. We divided these factors into two groups, which are shown in the following tables with the items ordered according to their level of evidence and quality (John Hopkins Nursing Evidence-Based Practice [66]) according to the studies from which they were extracted.
Studies suggest that the most important predisposing factors to consider in the development of postoperative delirium are advanced age, pre-existing cognitive deterioration, pain, elevated ASA score, alcohol abuse, hypoxemia, metabolic disturbances/waterelectrolyte imbalances, vision impairment, and auditory impairment. On the other hand, the use of benzodiazepines, cardiac surgery, heart-lung bypass, the duration of surgery, emergency surgery, orthopaedic surgery, the type of surgery, receiving a major transfusion, and being admitted to ICU were described as factors predisposing an individual to its development.

Principal Methods of Treatment
Currently, prevention is perhaps the most efficient and cost-effective way to treat delirium at the time of initial presentation. The role of nursing personnel is key to the prevention and identification of delirium and in terms of implementing non-pharmacological interventions to treat it.
Although there is still no established "gold standard" for the treatment of postoperative delirium, studies have suggested numerous non-pharmacological and effective pharmacological interventions that can reduce the symptoms of this pathology. We consider that for an intervention to be effective, the reduction in the incidence of postoperative delirium must be at least 5%.
In the following text, we described the many interventions that can be used directly by nursing staff to reduce the symptoms of delirium (Table 3). Finally, we indicated the most effective pharmacological treatments for delirium (Table 4) as well as pharmacological treatments that are ineffective ( Table 5). All items in the three tables were ordered according to the level of evidence and quality of the studies from which they were extracted (John Hopkins Nursing Evidence-Based Practice [66]). Table 3. Nursing interventions to reduce postoperative delirium. Table 5. Ineffective pharmacological methods for the treatment of postoperative delirium.

Nursing Interventions
Nursing interventions that have shown greater levels of efficacy in the research studies included in this study are described below.

•
Physiological: Maintain good nutrition and hydration statuses, manage pain adequately, and remove catheters promptly. • Cognitive: Place clocks and calendars in the patient's room to aid with temporal reorientation, general reorientation, and cognitive stimulation. • Behavioural: Reduce preoperative anxiety, give psychosocial support, show concern and empathy, and listen attentively. • Sensory: Maintain a good level of illumination, facilitate the use of hearing aids at an early stage, facilitate the use of glasses, and avoid excessive noise. • Sleep and environment: Establish a daily routine in order to prevent disruption of the sleep-wake cycle, avoid the administration of medication or the taking of vital signs during the night where possible, and adjust routines in order to ensure uninterrupted sleep.
• Family involvement: Involve the patient's family, avoid changes in personnel, and allow family to be present at meal-times. • Patient safety and skin integrity: Avoid mechanical restraints where possible, and promote early mobility, walking, or directed exercise at least three times daily

Effective Pharmacological Treatments
The drugs that seem to be more effective for reducing postoperative delirium are Haloperidol, Dexmedetomidine, Olanzapine, Quetiapine, Risperidone, and Melatonin

Ineffective Pharmacological Treatments
Benzodiazepines, GABA agonists, Glutamate blockers, Haloperidol, Membrane stabilizers, and Donepezil have shown lower levels of treatment effectiveness.

Discussion
This integrative review attempted to conceptualize postoperative delirium and the conditions in which it appears to expand the body of knowledge about the factors that influence its appearance and development as well as the therapeutic measures that reduce the severity of the pathology.
As was mentioned, the aetiology of postoperative delirium is unknown [21,22,52,[68][69][70][71], but it has been shown to be related to a large number of factors.
Advanced age appears to be the principal risk factor predisposing a patient to the development of postoperative delirium. This was explored in a study published by Carrera Castro [14], which attributed this surgical risk factor to the deterioration in biological functions and functional capacity associated with aging.
De las Pozas Abril [1] established age as a factor related to the appearance of postoperative delirium, but it was not reported as an independent factor. In contrast, studies presented by Huang et al. [73], Veiga et al. [102], Ying et al. [17], Guo et al. [51], Van Der Sluis et al. [46], and Wang et al. [103] did establish age as an independent factor in the development of postoperative delirium. The results obtained in this review agree with the results of the study published by Bettelli and Neuner, showing that postoperative delirium occurs more frequently in elderly patients. This seems to coincide with the fact that age is a key factor in its development, in addition to the fact that with increasing age, the likelihood of suffering from other alterations that predispose a patient to the development of postoperative delirium is greater [35].
Regarding predisposing factors, the data obtained in this study agree with those obtained in a systematic review carried out by Aitken et al. [104], which suggested that age and previous cognitive decline are key factors in the development of delirium. Previous studies also agree that the occurrence of postoperative delirium can prolong the length of hospital stay [24].
Another factor closely related to the syndrome is postoperative pain. Álvarez-Bastidas et al. [105] found a two-fold increase in the risk of developing delirium in patients reporting medium-severe pain on the Visual Analogue Scale (VAS). Mei et al. [74] showed an association between pain and postoperative delirium in patients undergoing abdominal surgery requiring a general anaesthetic. The American Geriatrics Society [75] recommends that health professionals optimize post-operative pain control in older patients in order to prevent the occurrence of delirium. If possible, this should be conducted using non-opioid analgesics.
Cognitive deterioration and being male are other factors with strong relationships with the occurrence of postoperative delirium. Che-Sheng et al. [23] attributed the higher incidence of the syndrome among males compared with females to the fact that men have higher rates of auditory loss, comorbidities, and emergency hospital admissions.
It is necessary to develop new lines of investigation to establish the physiopathology of this disease and to drive the implementation of preventative measures and directed, scientifically proven treatments [76].
The identification of delirium continues to be a daily challenge in clinical practice. Diagnostic methods need to be improved and, for this, the involvement and training of nursing personnel is key [64].
In their guidelines, published in 2017, The European Society of Anaesthesiology [24] established that there are two highly sensitive scales that can be used to identify and diagnose postoperative delirium: the Nu-DESC and the CAM scales. However, it warns that, if the CAM scale is administered by untrained staff, it is far less accurate. They state that further investigation is necessary to identify the optimal tools for the identification of delirium on recovery wards.
Nursing interventions are principally focused on the prevention of delirium and the reduction in its symptoms after anaesthetic episodes.
Duarte et al. [65] established that nurses are taking on an ever more important role in the prevention, diagnosis, and treatment of delirium, since they are generally the first professionals to identify it. They recommend the implementation of measures and standard protocols within daily clinical practice.
Among the non-pharmacological interventions, most frequently mentioned for the management of postoperative delirium is the reorientation of patients using objects such as clocks and calendars in the room, the maintenance of good nutrition and hydration practices, adequate management of pain, the maintenance of good illumination, the avoidance of excessive noise, facilitation of the use of hearing aids and glasses, and the promotion of early mobility.
A systematic review completed by Artuz et al. [6] analysed 30 studies concerning the nursing care of old adult patients with postoperative delirium and suggested that non-pharmacological measures are more important than pharmacological treatments for managing this condition. A study by Ocádiz-Carrasco et al. [22] supports this result, confirming that, currently, implementing non-pharmacological measures through a multidisciplinary approach is a strategy that is feasible, economic, and more effective than pharmacological treatments.
Various studies have shown that dexmedetomidine is an effective treatment for postoperative delirium [15,16,112], although without clear evidence [72,82,109]. Furthermore, another study suggested that dexmedetomidine is not an effective treatment for postoperative delirium [8], so we cannot conclude that its use is beneficial for the treatment of patients.
Several authors have proposed [11,53,106,110] the use of Haloperidol prophylactically in order to reduce the severity and duration of symptoms. However, due to the inconsistency of results and lack of evidence, the European Society of Anaesthesiology [24] guidelines do not advise its routine use, despite stating that there is evidence that Haloperidol, when used at low doses and as a preventative treatment, does reduce the incidence, severity, and duration of delirium.
Some of the studies reviewed suggest that the use of Benzodiazepines should be reserved for patients who are habitual users of this drug or who have a history of alcohol abuse, given that this category of drugs is, in fact, associated with the appearance of postoperative delirium [20,24,31,36,41,75,111].
Reviews completed by Borozdina et al. [80] and Popp et al. [60] suggest that melatonin should be administered prior to surgery in order to reduce the risk of the patient developing postoperative delirium. On the other hand, Guenther et al. [31] found that melatonin does not reduce the incidence of postoperative delirium in old adult patients who have undergone surgery. Other authors argue that the results of previous studies are too inconsistent to allow any recommendations to be determined regarding its preoperative use [24,72].
The studies included in this integrative review vary in terms of their level of methodological quality, as defined in the John Hopkins Nursing Evidence-Based Practice [66].
The heterogeneity of the studies and the results found highlight the need for more randomized clinical studies to be performed in this area. This would allow us to obtain a better understanding, based on scientific evidence, on postoperative delirium and give us new data that may help to uncover the unknown factors surrounding this condition.

Study Limitations
One of the principal limitations of this study is that, despite the high number of articles reviewed, they were very heterogeneous in nature, and there was a paucity of randomized clinical trials and systematic reviews. Furthermore, the similarity between the terms "postoperative delirium" and "emergency delirium" may have made it difficult to specifically search for articles on this topic.

Clinical Implications
The aetiology of postoperative delirium is not clear. Due to its high incidence and its clinical and economic repercussions, it is essential to train healthcare staff (especially nurses as staff members who are in close contact with the patient) in the recognition and management of postoperative delirium. As a future line of research, we suggest conducting a greater number of randomized clinical trials on the use of nursing interventions in the treatment of postoperative delirium.

Conclusions
The aetiology of postoperative delirium is still unknown, and no treatments exist to eliminate this pathology. Despite there being several different theories concerning the physiopathology of the condition, its causal mechanism is unclear, and there are no specific treatments that appear to efficiently counteract it.
Advanced age appears to be the principal risk factor for developing postoperative delirium, while pain and prior cognitive deterioration seem to be closely related to the development of this pathology. Interventions such as reorientation, pain management, and ensuring the patient receives good nutrition and hydration, in addition to the use of Haloperidol and Dexmedetomidine, seem to decrease the severity of delirium.
The role of nursing staff is fundamental in the prevention, diagnosis, and management of this pathology, given that these professionals are usually the first to notice symptoms.