The aim of this study was to investigate the incidence and risk factors of postoperative delirium (PD) in elderly patients after general or gastrointestinal surgery.
Societies worldwide are rapidly aging, and the number of surgeries in elderly patients has been increasing. PD, which adversely influences postoperative course, has thus become more common.
The Surgery and Anesthesia Network Group of the National Hospital Organization in Japan conducted this retrospective cohort study of patients older than 70 years of age who underwent general or gastrointestinal surgery.
A total of 219 patients from 9 participating institutes underwent surgery between July 2013 and August 2014. We excluded 2 patients who died within 2 weeks after surgery. Of the remaining 217 cases, 31 (14.3%) developed PD. These patients were older (80 versus 76 years, P = 0.013), more likely to be male (74.2 versus 54.8%, P = 0.039), and had higher American Society of Anesthesia Physical Status scores than those without PD. Emergency surgery was more common than elective surgery in the PD group (41.9 versus 10.2%, P < 0.0001). Multivariate analysis showed that male gender (odds ratio, 3.31; 95% confidence interval, 1.32 to 9.39; P = 0.0098) and emergency surgery (odds ratio, 7.47; 95% confidence interval, 2.79 to 20.83; P < 0.0001) were independent risk factors of PD.
The incidence of PD was high in male patients and those undergoing emergency surgery. Effective interventions in these groups will be necessary to improve treatment outcomes in elderly patients.
Postoperative delirium (PD) is an acute, fluctuating disorder of attention and cognition.1 It is one of the most common serious postoperative complications in elderly patients, with a reported incidence of 9.0% to 13.2% in gastrointestinal and general surgery,2–5 33% in colorectal emergency surgery,6 9.1% to 28.6% in orthopedic surgery,7–9 and 23.0% and 32.0% in cardiovascular surgery.10,11 In addition to being subjectively uncomfortable, PD is associated with accidental removal of drains or catheters, worsened treatment compliance, exhaustion of medical staff, and medical accidents such as falls.
PD is also associated with impaired treatment outcomes, including impaired activities of daily living,7 prolonged hospitalization2,3,5 and intensive care unit stays,2 elevated morbidity2,3,6 and mortality,9,12,13 and increased hospital costs.2
Several studies have identified risk factors of PD, including advanced age, cognitive impairment, preoperative nutritional status, alcohol abuse, increased comorbidity, impaired American Society Anesthesiologists Physical Status (ASA-PS) score, exposure to sedating medications, high-risk surgical procedures, emergency surgery, general anesthesia, elevated intraoperative blood pressure, poorly controlled postoperative pain, and major postoperative complications.2,4,8,11,14–18
However, PD remains frequently unrecognized and poorly understood by surgeons, anesthesiologists, and surgical ward staff members in general hospitals.19 Because many countries face aging populations and the demand for surgical services for elderly patients is growing significantly,20 it is critical to be able to detect risk factors of PD in this age group and to develop optimized interventions for PD. To this end, this multicenter retrospective cohort study was conducted by the National Hospital Organization Surgery and Anesthesia Network Group (PD cohort study) in Japan. The study sought to determine the prevalence and risk factors of PD among elderly patients undergoing general or gastrointestinal surgery based on cohort data on patient backgrounds and surgery and anesthesia parameters.
Study design and participants in the PD cohort study
The PD cohort study protocol was designed by the Surgery and Anesthesia Network Group of the National Hospital Organization in Japan. The study protocol was approved by the institutional review board of each participating hospital and was registered in the University Hospital Medical Information Network (UMIN R000022185). This retrospective observational study analyzed data from the clinical records of all elderly patients older than 70 years of age who underwent general or gastrointestinal surgery during the latest 1 month after study protocol approval, from July 2013 to August 2014. Data were collected on patient background, surgery and anesthesia factors, and details regarding PD. Patient background data included age; gender; history of cerebrovascular disease, dementia, and depression; daily alcohol consumption; current smoking; and ASA-PS. Anesthesia data included type of anesthesia, which anesthetics were used, and whether or not bispectral index (BIS) monitoring was performed. Surgical factors included disease; type and approach of surgery; operative time and blood loss; insertion of urinary catheters, venous catheters, and drains; and total number of drains and catheters. Patients were divided into 2 groups: those who developed PD (PD group) and those who did not (nondelirium group). The 2 groups were compared in terms of patient background, surgery and anesthesia factors, and postoperative clinical course. Using data from the PD cohort study database, we analyzed the risk factors of PD and explored the optimal targets for prevention of PD in elderly patients undergoing general or gastrointestinal surgery.
Definition of PD
PD is an acute decline in attention and cognition after surgery.17 In reference to the criteria of the Diagnostic and Statistical Manual Mental Disorder 4th edition,21 it was diagnosed in the current study if patients presented with at least 1 of the following 3 fluctuating symptoms within 7 days after surgery: (1) abnormal behavior, (2) one or more hallucinatory episodes, and (3) oversedation. PD was diagnosed by attending physicians or study investigators at each institute based on patient medical records.
We expressed continuous variables as medians (range). Fisher's exact test was used to compare binary variables, and the Mann-Whitney U test was used to compare continuous variables. Univariate and multivariate analyses were carried out using logistic regression. P < 0.05 was considered statistically significant. Statistical analysis was performed using JMP software (SAS Institute, Cary, North Carolina).
A total of 219 elderly patients older than 70 years of age underwent general or gastrointestinal surgery from July 2013 to August 2014 at the 9 institutions enrolled in this PD cohort study. We excluded 2 patients who died within 2 weeks after surgery (1 patient died on the 2nd postoperative day, and the other died on the 12th postoperative day after emergency surgery for perforation peritonitis). The remaining 217 patients were analyzed (Fig. 1). Their median age was 77 (70 to 96) years old, and there were 125 males and 92 females. The numbers (percentage) of patients with a history of cerebrovascular disease, dementia, or depression were 38 (17.5%), 8 (3.7%), and 6 (2.8%), respectively. In terms of surgical factors, 126 patients underwent surgery for malignant disease (esophageal, 2; gastric, 26; colorectal, 67; hepato-biliary-pancreatic, 18; breast, 10; and others, 3), and 91 patients had benign disease (inguinal hernia, 33; gallbladder stone, 27; appendicitis, 8; peritonitis, 4; ileus, 4; and others, 15). There were 185 elective and 32 emergency surgeries. Open surgery was performed in 121 patients, and a laparoscopic approach was used in 96. Regarding anesthesia factors, the ASA-PS was 1 to 2 in 171 patients and 3 to 5 in 46 patients; 193 patients received general anesthesia (intravenous anesthesia, 111; inhalation anesthesia, 82), and 133 patients were monitored by BIS during anesthesia.
Of the 217 evaluable patients, 31 (14.3%) developed PD; details are summarized in Table 1. Observed symptoms were abnormal behavior in 26 patients, 1 or more hallucinatory episodes in 4 patients, and oversedation in 1 patient. The median onset of PD was postoperative day 1 (0 to 7). The median recovery duration was 1 day in 9 patients, 2 to 3 days in 11 patients, 4 to 6 days in 6 patients, and more than 7 days in 5 patients. Eleven patients who developed PD received typical antipsychotic drugs, 4 patients received atypical antipsychotic drugs, and 16 patients did not receive any pharmacologic treatment for PD.
Table 2 presents perioperative patient data with a focus on the development of PD. PD patients were older (80 versus 76 years, P = 0.013) and were more likely to be male (74.2 versus 54.8%, P = 0.039) compared with those without PD (nondelirium). More PD patients underwent emergency surgery (41.9 versus 10.2%, P < 0.0001) and had ASA-PS scores worse than 3 (35.5% versus 18.8%, P = 0.046). There were no significant differences between the 2 groups in anesthesia type (general anesthesia versus other), intravenous or inhalation anesthesia during general anesthesia, disease type, surgical approach, operative time, operative blood loss, or number of drains and catheters. Of the 14 patients who did not undergo perioperative urinary catheter insertion, none developed PD.
In terms of postoperative complications and clinical course, there was no significant difference in overall or specific complication rates between the PD and nondelirium patients (Table 3); however, PD patients were more likely to require intensive care unit treatment (35.5% versus 15.0%, P = 0.011) and had a longer postoperative hospital stay (18 versus 10 days, P = 0.0006).
Table 4 shows the risk factors of PD in the 217 evaluable patients. Age, gender, type of surgery, and ASA-PS were extracted from the analysis of patient background and surgery and anesthesia factors in PD and nondelirium patients, as shown in Table 2. In the univariate analysis, age greater than 80 years, male gender, emergency surgery, and ASA-PS score worse than 3 were significantly correlated with the incidence of PD. In the multivariate analysis, male gender (odds ratio [OR], 3.31; 95% confidential interval [CI], 1.32 to 9.39; P = 0.0098) and emergency surgery (OR, 7.47; 95% CI, 2.79 to 20.83; P < 0.0001) were the only independent risk factors associated with the incidence of PD. The incidences of male (n = 14) and female (n = 18) patients who underwent emergency surgery and male (n = 111) and female (n = 74) patients who underwent elective surgery were 50.0%, 33.3%, 14.4%, and 2.7%, respectively.
PD is a common and deleterious postoperative complication1 that is becoming more prevalent in rapidly aging societies. Identifying patients at higher risk of PD is important to improve outcomes in elderly patients undergoing surgery.
In this cohort study, we focused on the incidence of PD and its risk factors after general or gastrointestinal surgery in elderly patients older than 70 years of age. Of 217 evaluable patients, 31 (14.3%) developed PD. The independent risk factors identified by multivariate analysis were emergency surgery and male gender, findings that are consistent with previous reports.11,22
In our series, male patients were significantly more likely to be current smokers and to consume alcohol daily and were less likely to undergo general anesthesia and BIS monitoring during anesthesia (data not shown). Other biological and sociocultural differences between genders, such as brain function, pain tolerance, and adaptability, might have influenced the results.23–25
Intervention to prevent PD in elderly patients undergoing general or gastrointestinal surgery should be targeted at males and at patients requiring emergency surgery. Several studies, including a randomized controlled trial and a meta-analysis, evaluated various approaches to the complex task of predicting and treating PD in elderly patients undergoing surgery.26,27 Pharmacologic prevention of PD remains controversial; whereas most medications evaluated were ineffective, including haloperidol,28,29 donepezil hydrochloride,30 tryptophan,31 and rivastigmine.32 several trials of prophylactic antipsychotic administration showed efficacy in reducing the incidence of PD in several surgical settings.1,26,27,33 Multicomponent intervention, BIS-guided anesthesia, and dexmedetomidine sedation were reported to be useful in preventing PD.27,34,35 However, preventive interventions for PD are difficult to conduct in patients undergoing emergency surgery.
The primary limitation of our study was its retrospective cohort design. The accurate diagnosis of PD is difficult for surgeons, anesthesiologists, and the medical staff of surgical wards, and PD is often unrecognized and easily overlooked.1,19 PD should be diagnosed by psychiatrists using established diagnostic criteria or instruments such as the Diagnostic and Statistical Manual Mental Disorder 4th edition,21 the International Classification of Diseases-10,36 and the Confusion Assessment Method.37 Although this is not always possible in clinical practice, it should be emphasized, along with assessment of preoperative cognitive function, in prospective interventional studies for PD.
In conclusion, this study demonstrated that elderly patients undergoing general or gastrointestinal surgery were at higher risk of PD if they were male or undergoing emergency surgery. These subpopulations are the optimal targets of interventions such as multicomponent programs to prevent PD and improve treatment outcomes.
This study was performed by the Surgery and Anesthesia Network Group of the National Hospital Organization that includes the following institutions: Kumamoto Medical Center, Kure Medical Center, Kyoto Medical Center, Mito Medical Center, Nagoya Medical Center, Osaka Minami Medical Center, Osaka National Hospital, Sendai Medical Center, and Ureshino Medical Center.