The crucial factors influencing the development and outcomes of postoperative delirium in proximal femur fractures

Abstract Objective The aim of this study is to find the incidence, risks, and reasons for prolonged length of hospital stay, short, long‐term mortality, and the factors contributing to mortality of postoperative delirium in proximal femoral fractures. Methods The data for the study was obtained from National Hip Fracture Database (NHFD) and internal hospital computer systems (Medway, ICE, Clinic letters) between January 2018 and December 2019. One hundred seventy‐five patients were found have developed postoperative delirium. The outcomes measured were postoperative anemia, lower respiratory tract infection, urinary tract infection, acute kidney injury, urinary retention, cardiac event and stroke, alcohol or drug withdrawal, length of hospital stay, and 30 day and 1 year mortality. Results The patients who developed delirium were 68 (38.9%) with American Society of Anesthesiologists (ASA) grade 4 and 94 (22.3%) without delirium (p < 0.05). The average length of stay after developing postoperative delirium was 19.69 days compared to 17.4 days for patients without delirium. The mortality at 30 days and 1 year was 10.9% and 37% in patients who had postoperative delirium compared to 2.1% and 2.8% to those without delirium, respectively. Conclusion Postoperative delirium is three times more common in hip fractures. Early detection and timely management are crucial in the improvement of functional outcomes and mortality.

sometimes predisposes to the development of dementia in hip fractures. 7 The incidence of postoperative delirium is three times more common in hip fracture surgery compared with patients undergoing surgery for other indications. 8 A better understanding of the underlying pathophysiology and the identifiable causes of postoperative delirium is vital in its prevention and management. 9 The National Institute of Health and Clinical Excellence (NICE), in 2010, have focused on the importance of diagnosis, prevention, and management of postoperative delirium in patients with hip fractures. 10 The current Best Practice Tariff (BPT) by NHS England has included a delirium assessment using the 4AT screening tool during the admission as one of the criteria. 11 A multidisciplinary approach to identify those at risk and effective management of delirium in patients with hip fractures has been highlighted in the study by Chuan et al in 2020. 12 The aim of this study is to find the incidence, identifiable causes, reasons for prolonged length of hospital stay, calculate short and long-term mortality, and the factors contributing to mortality of postoperative delirium in patients with proximal femoral fractures.

| ME THODS
The data for the study was obtained from National Hip Fracture Database (NHFD) and internal hospital computer systems (Medway, ICE, Clinic letters). The study period was between January 2018 and December 2019. A total of 598 patients were admitted during the study period. The study was approved by the audit department of our hospital trust. This study used data from a public database and did not require ethical approval and patient consent.
After screening, 175 patients were found have developed postoperative delirium. The diagnosis of delirium was made using the 4AT scoring system (Figure 1). Patients aged 65 years or above F I G U R E 1 The 4AT assessment tool for delirium who sustained a proximal femur fracture and underwent surgery and scored ≥4 in the 4AT assessment were included in the study.
Patients with pathological fractures were excluded from the study.
All the patients admitted with proximal femur fracture were prospectively included in the database. Patient demographics including age, sex, fracture pattern, American Society of Anesthesiologists

| Statistical analysis
The statistical analysis was done using SPSS version 22. Comparison between patients who developed postoperative delirium and those without delirium were compared using chi-squared, Student's t test and Mann-Whitney U test for categorical, and parametric and nonparametric data sets, respectively. Logistic regression analysis was used to find the causes for postoperative delirium and also determining the factors contributing to mortality associated with delirium. A p value of <0.05 was considered significant.

| RE SULTS
The total number of patients during the study period admitted with proximal femur fractures was 598, after applying the inclusion and  Table 1.
In our study, the identifiable factors which caused postopera-

| DISCUSS ION
Surgery for proximal femur fractures in elderly and frail patients can often lead to altered disturbance in their mental status. It is usually manifested as dementia, depression, and delirium. 13 The reported incidence of postoperative delirium varies from 9.5% to 61.3%. [14][15][16] In our study, patients with fractures of the hip the incidence of postoperative delirium was 29.3%. But in a study by Edelstein et al, they noted an incidence as low as 5.1%. 17 The reason for such a wide variation in the range of incidence of postoperative delirium may be be- There is strong association of pre-operative AMT score and the development of postoperative delirium. A one point increase of AMT score can lead to a fall in the odds of postoperative delirium by 0.6 fold. 18 Yang et al in their meta-analysis found that age and preoperative dementia to be a risk factor for the development of delirium. 19 Another study by Kim et al also made similar observations. 20 In the 2018 report by the NHFD, they suggested that an AMT score of less than 8 is a significant predictor of postoperative delirium in hip fractures using the 4AT diagnostic criteria. 21 19,20,22 These findings were in contrast to our observation where no significant differences were noted between delirium and non-delirium.
In the 2018 NHFD report, it was suggested that there is an increased incidence of postoperative delirium in patients who had general anesthesia for the surgical repair of hip fractures. 21 Anh et al, in their retrospective study, found that by administering regional anesthesia for this cohort of patients the incidence of postoperative delirium was reduced by 2.5 times. 23 But our study has