The effect of preoperative cognitive impairment and type of vascular surgery procedure on postoperative delirium with associated cost implications

Transcription

1 From the Society for Vascular Surgery The effect of preoperative cognitive impairment and type of vascular surgery procedure on postoperative delirium with associated cost implications Rima Styra, MD, MEd, a,b Elisabeth Larsen, MSc, a Michelle A. Dimas, MSc, a Dorina Baston, MN, a Jeanne Elgie-Watson, MA, a Linda Flockhart, MHS, a and Thomas F. Lindsay, MDCM, MSc, c,d Toronto, Ontario, Canada ABSTRACT Objective: Postoperative delirium (POD) has a high prevalence among vascular surgery patients, increasing morbidity, mortality, and length of stay. We prospectively studied preoperative risk factors for delirium that can be assessed by the surgical team to identify high-risk patients and assessed its impact on hospital costs. Methods: There were 173 elective vascular surgery patients assessed preoperatively for cognitive function using the Montreal Cognitive Assessment (MoCA) and the Confusion Assessment Method for POD, which was verified by chart and clinical review. Demographic information, medications, and a history of substance abuse, psychiatric disorders, and previous delirium were prospectively recorded. An accompanying retrospective chart review of an additional 434 (elective and emergency) vascular surgery patients provided supplemental cost information related to sitter use and prolonged hospitalization secondary to three factors: delirium alone, dementia alone, and delirium and dementia. Results: Prospective screening of 173 patients (73.4% male; age, years) identified that 119 (68.8%) had MoCA scores <24, indicating cognitive impairment, with 7.5% having severe impairment (dementia). Patients who underwent amputation had significantly (P <.000) lower MoCA scores (17 of 30) compared with open surgery and endovascular aneurysm repair patients (23.7 of 30). The incidence of delirium was 11.6% in the elective cohort. Regression analysis identified predictors of delirium to be type of surgical procedure, including lower limb amputation (odds ratio [OR], 16.67; 95% confidence interval [CI], ; P <.000) and open aortic repair (OR, 5.33; 95% CI, ; P <.000); cognitive variables (dementia: OR, 5.63; 95% CI, ; P <.001); MoCA scores #15, indicating moderate to severe impairment (OR, 6.13; 95% CI, ; P ¼.02); and previous delirium (OR, 2.98; 95% CI, ; P ¼.03). Retrospective review (N ¼ 434) identified differences in sitter needs for patients with both delirium and dementia (mean, 13.6 days), delirium alone (mean, 3.9 days), or dementia alone (mean, <1 day [17.7 hours]). Fifteen patients required >200 hours (8.3 days), accounting for 69.7% of sitter costs for the surgical unit; 43.7% of costs were accounted for by patients with pre-existing cognitive impairment. Conclusions: POD is predicted by type of vascular surgery procedure, impaired cognition (MoCA), and previous delirium. Costs and morbidity related to delirium are greatest for those with impaired cognitive burden. Preoperative MoCA screening can identify those at highest risk, allowing procedure modification and informed care. (J Vasc Surg 2018;-:1-9.) Keywords: Preoperative; Cognition; Delirium; Vascular surgery; MoCA Delirium is a complex and challenging neuropsychiatric disorder that is commonly seen in patients who are hospitalized and occurs more frequently in the elderly. Postoperative delirium (POD) is defined as a disturbance of consciousness accompanied by impaired attention or ability to focus that cannot be explained by a pre-existing or evolving neurocognitive disorder. It is a serious complication for surgical patients. 1 Most deliriums are considered to be reversible, yet it is known that delirium increases morbidity and mortality and exacts a high personal toll on the patient and the patient s family. 2,3 Delirium is associated with increased length of stay in the hospital and associated increased health care costs. 4 The duration of delirium has been found to be associated with length of stay even after adjustments for variables such as illness severity and age. 5,6 Delirium rates tend to vary in hospitalized populations. Intensive care unit (ICU) populations have delirium From the Peter Munk Cardiac Center, a and Division of Vascular Surgery, Peter Munk Cardiac Center, c University Health Network; and the Department of Psychiatry, b and Department of Surgery, d University of Toronto. This work was funded by the Peter Munk Cardiac Center Innovation Fund. Author conflict of interest: none. Presented at the 2018 Vascular Annual Meeting of the Society for Vascular Surgery, Boston, Mass, June 20-23, Correspondence: Rima Styra, MD, MEd, University Health Network (TGH site), 200 Elizabeth St, Ste 8EN-219, Toronto, Ontario M5G 2C4, Canada ( rima.styra@uhn.ca). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest Copyright Ó 2018 by the Society for Vascular Surgery. Published by Elsevier Inc. 1

2 2 Styra et al Journal of Vascular Surgery prevalence rates that are extremely high, with reports ranging from 40% to 87%. 7 Delirium rates in postoperative vascular patients have been reported to range from 5% to 39%, 2-4,8-12 with the highest incidence occurring in abdominal aortic surgery, 46% to 52%. 13 The incidence of cardiac POD has yielded similar variability, ranging from 6% to 50.6% across 21 cohort studies, with 33 identified risk factors. 14 Delirium is particularly common during hospitalization when both predisposing and precipitating factors are present. Patient factors, illness comorbidities, and environmental influences have been identified as being relevant risk factors for delirium. Attempts have been made to identify predictive risk factors that include age, type of vascular surgery procedure, depression, smoking, alcohol, and psychoactive drugs. One factor, cognitive impairment, is more common in older patients and has been shown to have a negative impact on postsurgical outcomes because of the increased risk of delirium. 15 Assessment of cognitive impairment requires specific testing using validated scales such as the Montreal Cognitive Assessment (MoCA). 16 The MoCA is a screening instrument for cognitive dysfunction that covers a number of cognitive domains, such as memory, visuospatial/executive function, attention, and abstraction. The scale is scored out of 30, whereby scores below 26 are considered to indicate cognitive impairment. The MoCA has been shown to detect 90% of mild cognitive impairment, and in a group of patients with mild Alzheimer disease, the sensitivity of the MoCA to detect cognitive impairment was 100% and specificity was 87%. It has a high retest reliability (correlation coefficient, 0.92; P <.001) and has been validated in cerebrovascular disease. A strong relationship of impairment on the MoCA independent of age was shown with cognitive impairment of vascular etiology. 17,18 Many of the studies to identify delirium risk factors are retrospective chart reviews, with few prospective studies focused solely on vascular surgery as a specific population and specific risk factors based on significant independent odds ratios (ORs). 11,12,19-21 Cognitive impairment, in particular, has been found to have a negative impact on outcomes in prospective studies. 22,23 The aim of our prospective study was to determine specific, easily identifiable preoperative risk factors for clinicians to use to detect patients at high risk for POD. Those who were seen in the preadmission clinic were recruited at a minimum of 1 week before their scheduled surgery, whereas those who were seen on the ward were recruited in addition to their assessment for vascular surgery. Delirium was assessed postoperatively using the Confusion Assessment Method (CAM) and was confirmed by a psychiatrist. In addition, a retrospective chart review was carried out of a second cohort of ARTICLE HIGHLIGHTS d Type of Research: Prospective and retrospective cohort studies d Take Home Message: Preoperative screening using the Montreal Cognitive Assessment identified postoperative predictors of delirium of 173 vascular surgery patients that included cognitive variables, amputation, open aortic repair, and previous delirium. In a second study of 434 patients, sitter costs were highest in those with delirium and dementia. d Recommendation: The authors recommend preoperative screening using the Montreal Cognitive Assessment to identify high-risk vascular patients for postoperative delirium, allowing potential procedure modification. elective and emergency vascular surgery patients to provide cost information related to sitter use and prolonged hospitalization related to delirium. METHODS The study was approved by the University Health Network Research Ethics Board. It was carried out at a tertiary hospital and consisted of two parts, a prospective study of elective vascular surgery patients and a retrospective chart review. Patients who were to undergo vascular surgery and were fluent in English were recruited either in the preadmission clinic or when they were admitted for assessment to the vascular surgery unit. A convenience sampling method was used, and those who were emergency patients requiring urgent intervention were excluded in the prospective study. Baseline demographic data as well as comorbidities, medications, and specific risk factors that had been identified by literature review 9-12,19-26 were collected by interview of the patients and review of clinical records. Particular attention was directed to any previous diagnosis of dementia, cognitive impairment, or investigations carried out because of cognitive concerns. Cognitive function was assessed by administration of the MoCA by a trained rater. The MoCA is scored out of 30. Patients scoring below 26 are identified as having cognitive impairment, with declining scores indicating more severe difficulty with cognition (major neurocognitive disorder/dementia, mild cognitive impairment). In the cardiovascular population, the optimum cutoff for detecting mild cognitive impairment was identified as MoCA < The MoCA was administered to patients in the preadmission clinic or on the ward. This took place at a minimum of 1 week before scheduled surgery in the preadmission clinic, and those who were seen on the

3 Journal of Vascular Surgery Styra et al 3 Volume -, Number - ward were recruited during the time of their assessment for possible vascular surgery. On administration of the MoCA, patients were not given any medications that may affect their cognition or performance on the test. POD was ascertained by using the CAM and the CAM for the ICU (CAM-ICU) ratings and verified by a psychiatrist. The CAM and CAM-ICU are validated tools that are extensively used for diagnosis of delirium in hospital populations of patients. 28,29 An accompanying retrospective study to review the electronic clinical records of patients admitted to the vascular surgery unit as either elective or emergency vascular surgery patients was completed. Emergency patients were classified as those who were admitted through the emergency department. There were 434 charts reviewed. Demographic data, type of vascular surgery procedure, diagnosis of dementia, presence or absence of delirium, and length of time and reason for a sitter were recorded for the period of January 2015 to December Patients were deemed to have a diagnosis of dementia if the diagnosis had been noted in the clinical record and was based on investigations and a physician s assessment, verified by a psychiatrist. Those who required sitters were placed into three categories: patients with delirium, patients with dementia, and patients with delirium and dementia. Statistical analysis. Baseline variables were analyzed using frequencies and descriptives to gather means and standard deviations. Missing values identified visually were recoded and not included in further analyses. Cohorts were created for a number of variables, including age, MoCA scores, type of surgical procedure, and absence or presence of delirium. Predictors for follow-up analyses included age, sex, psychiatric history, alcohol use, psychotropic medication, cognitive enhancers, illicit drugs, opioids, depression, history of cerebrovascular accident and transient ischemic attack, previous delirium, renal disease, MoCA total score, and MoCA domain scores. All assumptions were met before running all statistical analyses. The c 2 tests were conducted for categorical data. If <5% of the data remained in one category, a Fisher exact test was used on contingency tables of 2 2. In instances in which <5% of the data existed on contingency tables greater than 2 2, a Fisher-Freeman-Halton test was used. A 2 (POD and non-pod) 10 (predictive variables) multivariate analysis of variance was analyzed. A univariate analysis of variance approach was used to further investigate with a significance level of P <.05. A multivariate regression analysis was used to identify predictive variables of previous delirium, surgical procedure type, dementia, and MoCA score. Risk factors were formed to determine ORs and a 95% confidence interval, and significance was set at P <.05. The retrospective data were visually inspected for any missing data, with values of missing cases recoded and not included in further analyses. Descriptive data and frequencies were analyzed for those with delirium alone, dementia alone, and delirium with dementia. Independent samples t-tests were analyzed between patient group and incidence of POD, using a significance level of P <.05. All data for both the prospective and retrospective parts of the study were analyzed using statistical software (SPSS version 23; IBM Corp, Armonk, NY). RESULTS A total of 220 patients were originally approached; 184 vascular surgery patients were consented for the prospective study between January 2016 and October One patient was transferred to cardiac surgery and 10 patients either declined surgery or did not require surgery. In total, 173 patients were retained for data analysis. The majority of patients were male (73.4%) and had a mean age of years. MoCA scores <24, which are consistent with some cognitive impairment in cardiovascular patients, were present in 119 patients (68.8%). The incidence of POD was 11.6% (20 patients). POD was associated with reduced MoCA score (P <.01). Delirium rates per age cohort were as follows: 80 to 89 years, 16.6%; 70 to 79 years, 11.8%; 60 to 69 years, 14.9%; and 50 to 59 years, 6.3%. There were no significant differences between age cohorts of $70 years (range, $70 years vs <70 years) and POD. The two age cohorts of <50 years and $90 years did not experience any delirium. The incidence of POD differed by procedure and was most commonly observed after amputation (62.5%) and open aortic surgery (32.0%) compared with endovascular aneurysm repair (EVAR; 9.3%). Further descriptive variables are summarized in Table I. Main effects were identified between group (POD and non-pod) and predictive variables. Univariate analyses did not identify age, heavy alcohol use, substance use, psychiatric history (depression and schizophrenia), renal disease, smoking, previous cerebrovascular accident or transient ischemic attack, opioids, illicit drugs, psychotropic medication, and cognitive enhancers as predictive of POD (Table I). A total list of procedures can be found in Table II. Patients who underwent amputation had significantly (P <.000) lower MoCA scores ( ) compared with open aortic ( ) and EVAR patients ( ). The educational background for each patient was taken into account in the MoCA scoring. A multivariate analysis of variance indicated that reduced MoCA total scores in lower limb amputation were significantly different from those in patients undergoing other vascular surgery procedures (P <.000). Table III lists MoCA scores by cohort. A multivariate logistic regression analysis identified five significant variables for predicting POD. These variables included lower limb amputation (P <.000), open aortic repair (P <.000), pre-existing dementia (P <.004),

4 4 Styra et al Journal of Vascular Surgery Table I. Baseline characteristics and univariate analysis for postoperative delirium (POD) and non-pod patients Data Total sample (N ¼ 173; 100%) POD (n ¼ 20; 11.6%) Non-POD (n ¼ 153; 88.4%) P Sex, male 127 (73.4) 16 (11.7) 121 (88.3).59 Age, years Age $70 years 78 (45.1) 8 (10.3) 70 (89.4).41 Previous delirium 36 (21.8) 8 (22.2) 28 (77.8).03 a Renal disease 39 (22.5) 4 (10.3) 35 (89.7).43 Depression 9 (5.2) 2 (22.2) 7 (77.8).28 Smoking 4 (2.3) 0 (0) 4 (100).61 CVA or TIA 40 (23.1) 8 (20.0) 32 (80.0).06 Opioids 34 (19.6) 4 (11.7) 30 (88.3).58 Illicit drugs 7 (4.0) 1 (14.3) 6 (85.7).58 Psychotropic medication 47 (27.2) 9 (19.1) 38 (80.9).06 Cognitive enhancers 3 (1.7) 1 (33.3) 2 (66.7).31 Schizophrenia 3 (1.7) 0 (0) 3 (100) d Heavy alcohol use 14 (8.0) 2 (14.3) 12 (85.7).26 Total MoCA score (range, 0-30) a MoCA <24 (cognitive impairment) 119 (68.8) 11 (9.2) 108 (90.8).95 MoCA visuospatial/executive (range, 0-5) MoCA memory (range, 0-5) Dementia (pre-existing) 9 (5.1) 4 (44.4) 5 (55.6).00 a Lower limb amputation 8 (4.6) 5 (62.5) 3 (37.5).00 a Open aortic surgery 25 (14.4) 8 (32.0) 17 (68.0).00 a Carotid surgery 20 (11.5) 0 (0) 20 (100) d EVAR 75 (43.4) 7 (9.3) 68 (90.7).29 Femoral/infrainguinal surgery 35 (20.1) 0 (0) 35 (100) d Angioplasty alone 1 (0.57) 0 (0) 1 (100) d Other surgery 9 (5.2) 0 (0) 9 (100) d CVA, Cerebrovascular accident; EVAR, endovascular aneurysm repair; MoCA, Montreal Cognitive Assessment; TIA, transient ischemic attack. Heavy alcohol use is >10 drinks per week for women or >15 drinks per week for men. Illicit drugs consist of marijuana use. Psychotropic medication consists of antipsychotics, antidepressants, anxiolytics, and hypnotics. Lower limb amputation excludes toe surgery. Dementia that was diagnosed previously is noted as pre-existing. Categorical variables are presented as No. (%). Continuous variables are presented as mean 6 standard deviation. a Indicates significance level of P <.01. MoCA scores #15 indicating moderate to severe cognitive impairment (P ¼.024), and previous delirium (P ¼.03). Table IV summarizes findings from the multivariate regression analysis. The receiver operating characteristic curve presented an area under the curve of (95% confidence interval, ; P <.000). The area under the curve shows that the predictive model has the ability to discriminate POD from non-pod patients, with a Youden index of J ¼ to indicate optimum threshold (Fig 1). The charts of 434 patients were reviewed retrospectively for assessment of delirium and its cost implications related to sitter use (44 patients admitted as emergency patients and 390 elective patients). The incidence of delirium in the emergency patients was 45.5% compared with 17.8% in the elective group (P <.01; Table V). There was no significant difference in age between these two groups. Similar to the prospective study, delirium rates were high in both open aortic surgery (33.9%) and amputation (44.1%). EVAR was associated with a lower incidence of delirium (12.5%), similar to the prospective group. Descriptive information for patients regarding the incidence of POD is found in Table V. Supplemental information regarding sitter days was provided for patients with delirium alone, delirium and dementia, and dementia alone. A large variability of sitter days was identified for patients (<1 day to 44.8 days). Patients with dementia and delirium required on average 13.6 days of sitters compared with 3.9 days for delirium alone and <1 day for dementia. Sitter use per patient is provided in Table VI. Twenty-five patients had a pre-existing diagnosis of dementia on admission identified in the medical record. Of these, the incidence of POD was 72%, and this small cohort (n ¼ 18) was responsible for 48% of the sitter costs based on unit records. Those with delirium alone (n ¼ 72) accounted for 51% of the unit s sitter costs; however, the average sitter days required per patient differed

5 Journal of Vascular Surgery Styra et al 5 Volume -, Number - Table II. List of vascular procedures Vascular procedures (N ¼ 173) No. (%) Lower limb amputation a 8 (4.6) EVAR surgery b 75 (43.4) Open aortic surgery c 25 (14.4) Carotid surgery d 20 (11.5) Femoral/infrainguinal surgery e 35 (20.1) Angioplasty alone 1 (0.6) Other surgery f 9 (5.2) EVAR, Endovascular aneurysm repair. a Above-knee amputation, foot amputation, below-knee amputation. b EVAR abdominal aortic aneurysm (AAA), thoracic endovascular aortic repair (TEVAR), TEVAR plus right femoral aneurysm, TEVAR plus left carotid subclavian bypass. c Aortobifemoral bypass, infarenal AAA, open AAA, open AAA plus aortobifemoral bypass. d Carotid repair, carotid aneurysm, carotid angioplasty, carotid bypass, carotid endarterectomy. e Bilateral or unilateral popliteal aneurysm, femoral aneurysm, femoral bypass, femoral-tibial synthetic graft, iliac thrombectomy with or without stenting with femoral exposure, common femoral artery repair. f Forearm amputation, arm bypass, rib resection, toe amputation. significantly between the delirium group (3.9 days) and the delirium with dementia (13.6 days) group (P ¼.024; Table VI). The sitter day range is displayed in Fig 2. Overall, only 15 patients were identified who required >200 sitter hours (8.3 days), accounting for 69.7% of the annual budget for sitter costs. Furthermore, eight of the 15 patients had pre-existing diagnosed dementia and represented 43.7% of the total sitter use. DISCUSSION This prospective study demonstrated the incidence of delirium to be 11.6%; it was associated with the type of vascular procedure, MoCA score, and pre-existing dementia. Previous studies have reported delirium incidence from 5% to 39%. 8-12,19 The variance in incidence reports may be attributed to the design of the studies that focus on a limited or select subset of vascular procedures 10,11 and use of exclusion criteria such as dementia or specific age cutoffs. 12,24 Our prospective study identifies preoperative risk factors for delirium in a vascular surgery population that is representative of actual surgical practice without excluding patients who have been diagnosed with major cognitive disorder (dementia). Our delirium rates may be lower because of improved medical care of vascular patients (amputations occurred in only 4.6% of our population) as well as advances in the field of vascular surgery, resulting in less invasive surgical techniques. Type of anesthesia or sedation was not investigated for the purpose of this study and its relation to POD. This study identified a number of independent preoperative risk factors for POD: open aortic surgery, previous delirium, lower limb amputation, dementia, and MoCA scores #15, which are indicative of moderate to severe major cognitive impairment. Our finding that open aortic surgery and amputation are risk factors for delirium is consistent with other study findings. 10,26 Predictive factors that have been identified in other studies are age, 14,30,31 cognitive impairment, 9 previous cerebrovascular accident, 12,25,32 previous delirium 25,33 and specific types of surgery. 9,12 Our study MoCA score of #15 was associated with an OR of 6.13 for POD, and patients diagnosed with pre-existing dementia had an OR of 5.63 of POD. Vascular cognitive impairment had been associated with symptomatic ischemic or hemorrhagic events; however, it is now recognized that progressive decline in cognitive functioning may result from cerebrovascular disease even in the absence of a clinically detectable precipitating event. 34 Delirium in general is known to occur more commonly superimposed on pre-existing cognitive impairment, 35 and delirium risk may be greater in a dementia of vascular origin compared with other dementias, possibly reflecting the relatively widespread neuronal disturbance associated with vascular disorders. 36 Thus, the use of a validated scale to assess cognition preoperatively is an important aspect of clinical care in a population that is known to have large percentages of patients with cognitive impairment. The majority of our study vascular surgery patients (68.8%) experienced some form of cognitive impairment (MoCA scores <24), which would imply at least some mild cognitive impairment. Cardiac and vascular patients are highly susceptible to cerebrovascular disease, which can lead to mild cognitive impairment that progresses into vascular dementia, giving rise to the term vascular cognitive impairment. 37 Only 5.7% of patients had been previously recognized as having dementia in our prospective study population, which highlights the fact that cognitive impairment among patients presenting for vascular surgery is more prevalent than previously appreciated and is infrequently diagnosed or investigated before admission. The type of vascular surgery procedure performed has been found to be associated with significantly different rates of delirium. 8-10,12,26 In the literature, EVAR has been shown to be associated with lower rates of delirium. 8 Major amputation is associated with the highest rates of delirium compared with other vascular surgery procedures 8,12,38 and in retrospective database analysis of vascular surgery patients is an independent relative risk factor. 10,12 Patients who require aggressive treatment such as amputation have been recognized as having more medical comorbidities, such as diabetes mellitus, 39 cerebral vascular disease, 9,12 end-stage renal failure, 20 and cognitive impairment, 23 which may partially account for the high rates of delirium. Our patients scheduled for lower limb amputation showed significantly lower MoCA scores compared with the other surgical groups. The scoring range for patients

6 6 Styra et al Journal of Vascular Surgery Table III. Montreal Cognitive Assessment (MoCA) scores by vascular surgery procedure with further analysis of the visuospatial/executive and memory domains in the lower limb amputation group Surgery type MoCA visuospatial/executive (range, 0-5) MoCA memory (range, 0-5) MoCA total scores (range, 0-30) P P All surgery types a d Lower limb amputation a d Open aortic surgery b Carotid surgery c EVAR b Femoral/infrainguinal b Other surgery c EVAR, Endovascular aneurysm repair. Lower limb amputation excludes toe amputation. Values are reported as mean 6 standard deviation. a Indicates significance level of P <.01 for multivariate analysis of variance with all surgery types. b Indicates significance level of P <.01 for Tukey test with lower limb amputation. c Indicates significance level of P <.05 for Tukey test with lower limb amputation. Table IV. Multivariate logistic regression analysis on predictive variables for postoperative delirium (POD) patients Variable OR 95% CI P Amputation b Open aortic surgery b Dementia (pre-existing) b MoCA score (#15) a Previous delirium a CI, Confidence interval; MoCA, Montreal Cognitive Assessment; OR, odds ratio. MoCA score #15 indicates moderate to severe cognitive impairment. Lower limb amputation excludes toe amputation. Age, heavy alcohol use, substance use, psychiatric history (depression and schizophrenia), renal disease, smoking, previous cerebrovascular accident or transient ischemic attack, opioids, illicit drugs, psychotropic medication, and cognitive enhancers were not predictive of POD. a Indicates significance level of P <.05. b Indicates significance level of P <.01. Fig 1. Receiver operating characteristic (ROC) curve for the final prediction model. print & web 4C=FPO who required amputation was consistent with a moderate to severe major neurocognitive disorder. The MoCA has been found to be a valid preoperative screening tool for vascular cognitive impairment among patients undergoing vascular surgery, with the majority of patients (68.8%, which is analogous to our study) identified as having an undiagnosed cognitive impairment. 23 The MoCA has also been used to follow cognition in cardiovascular patients and to measure the effect of interventions, such as carotid artery stenting, on subsequent cognitive function. 40,41 A review of cognitive decline after open aortic surgery and EVAR was unable to identify a postoperative difference in cognitive decline between the two operations in the longer term, although the incidence of POD was found to be higher in open aortic surgery compared with EVAR. 42 Thus, the MoCA is a valid and sensitive tool that can be incorporated in preoperative screening to assess cognitive function and risk of delirium in the vascular surgery population. In our study, only nine patients had a diagnosis of preexisting dementia compared with 11 patients identified by the MoCA. Others may have been picked up by other risk factors, such as previous delirium; however, previous delirium is often not documented in clinical practice and often depends on reports of the patient or family, who may not be aware of past events. The MoCA is a tool that can be used in combination with other variables to assist in identifying cognitive impairment. A limitation of the study is that we did not measure anxiety or depression in our patients at the time of testing, especially in those who were scheduled for amputation. Patients who are awaiting surgery have been shown to have increased anxiety and depressive symptoms. It has been established that patients with major depressive disorder commonly experience deficits across multiple domains of cognitive function, which can be further affected by comorbid anxiety. This may affect the MoCA and have interpretive implications; however,

7 Journal of Vascular Surgery Styra et al 7 Volume -, Number - Table V. Incidence of postoperative delirium (POD) and dementia in retrospective vascular surgery cohort Patient group Total sample (N ¼ 434; 100%) POD (n ¼ 90; 20.7%), No. (%) Non-POD (n ¼ 344; 79.2%), No. (%) P Emergency (45.5) 24 (54.4).005 a Elective (17.8) 320 (82.1).000 a Dementia (72.0) 7 (38.9).001 a No dementia (17.6) 337 (82.4).000 a a Indicates significance level of P <.01. Table VI. Sitter days per cohort Days per patient Patient group Total (N ¼ 434; 100%) P Total unit sitter days, % Mean Range Delirium alone 72 (16.6).024 a <1-30 Delirium with dementia 18 (4.1).024 a Dementia alone 7 (1.6) d < < a Indicates significance level of P <.05. anxiety and depression (measured by the Hamilton Anxiety and Depression Scale) were not significant in a study of preoperative vascular patients in differentiating those who had MoCA scores above and below The authors concluded that it was feasible to identify cognitive impairment using the MoCA in this high-risk surgical group preoperatively. Further research in preoperative cognitive assessment as well as other contributing variables (anxiety, depression) should be undertaken. Pre-existing cognitive impairment is considered among the strongest predictors of delirium for individuals older than 50 years 34 who are undergoing any type of surgery. Age was not independently identified by our study as a significant risk factor for POD. Emergency vascular surgery patients were not included in our prospective study because of the critical nature of these patients on arrival. Our retrospective study captured both emergency and elective patients and had a higher incidence of delirium (20.7%). This higher incidence may be accounted for by the emergency patients, who had a 45.5% incidence of delirium. Other studies have found that inclusion of emergency patients increases delirium incidence. 10 Potential consideration for the higher rates may be related to increased hemodynamic instability and increased disease severity by their emergency presentation. The retrospective chart review was consistent with our model of patients diagnosed with dementia who had a high (72.0%) incidence of delirium. Identification of the other comorbidities that would be associated with the etiology of delirium in these patients would have been helpful to identify other predictive variables beyond the dementia-delirium dichotomy. A drawback of the retrospective chart review is that we are able to identify cognitive status in only two categoriesddementia (major Fig 2. Range of sitter hours for postoperative delirium (POD) and POD with dementia patients. neurocognitive impairment) or nondementiadbecause we are dependent on the clinical information provided in the chart. A clearer picture based on levels of severity of the dementia may further allow us to identify specific cutoff scores that could improve our assessment of health care resource use. Patients identified as having severe cognitive impairment could be selected to receive more intensive intervention programs to prevent delirium and functional decline. The total hours of sitter care and the normal distribution spread of delirious patients with and without dementia appear to represent two different groups of patients in regard to the need for hospital support services. This has major implications for both patients morbidity and health care costs. The extended need for sitters by patients with baseline dementia who develop POD is associated with increased health care expenditures and negative impact on outcome of the patient. This retrospective study further highlights the need for identification of the patient with cognitive impairment to decrease morbidity of patients by considering procedure modification and informed care. Patients and families need to be informed print & web 4C=FPO

8 8 Styra et al Journal of Vascular Surgery preoperatively of the level of risk for delirium and the possible implications. A shift in health care to focus on preoperative factors is needed to advance positive overall outcomes. CONCLUSIONS MoCA scores can be used in the preoperative period to identify patients at high risk of POD, which can provide important predictive information to the clinician but also to the patient and family. Patients and family are often not aware of the risks of delirium associated with surgery, nor are they aware of the longer term morbidity that may occur especially when patients are presenting with unrecognized cognitive impairment. Predictors of POD include lower limb amputation and open aortic vascular surgery, impaired cognition, and previous delirium. Patients with pre-existing cognitive impairment undergoing major vascular surgery are at higher risk for delirium and the subsequent cognitive impairment that can follow a delirium, which further complicates the consequences of vascular disease therapy. 43 Patients undergoing open aortic surgery or amputation at a minimum should undergo a structured cognitive assessment, using a validated scale such as the MoCA, and counseling about their delirium risk. Provision of information about cognitive issues may allow procedure modification whenever possible. AUTHOR CONTRIBUTIONS Conception and design: RS, DB, JEW, LF, TL Analysis and interpretation: RS, EL, DB, TL Data collection: EL, DB, JEW Writing the article: RS, EL, MD, TL Critical revision of the article: RS, EL, MD, LF, TL Final approval of the article: RS, EL, MD, DB, JEW, LF, TL Statistical analysis: MD Obtained funding: RS, JEW, LF, TL Overall responsibility: RS REFERENCES 1. Diagnostic and statistical manual of mental disorders. 5th ed. Washington, DC: American Psychiatric Association; Neufeld KJ, Leoutsakos JM, Sieber FE, Wanamaker BL, Gibson Chambers JJ, Rao V, et al. Outcomes of early delirium diagnosis after general anesthesia in the elderly. Anesth Analg 2013;117: Inouye SK, Rushing JT, Foreman MD, Palmer RM, Pompei P. Does delirium contribute to poor hospital outcomes? A three-site epidemiologic study. J Gen Intern Med 1998;13: Gleason LJ, Schmitt EM, Kosar CM, Tabloski P, Saczynski JS, Robinson T, et al. Effect of delirium and other major complications on outcomes after elective surgery in older adults. JAMA Surg 2015;12: Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrel FE Jr, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA 2004;291: Franco K, Litaker D, Locala J, Bronson D. The cost of delirium in the surgical patient. Psychosomatics 2001;42: Ely EW, Siegel MD, Inouye SK. Delirium in the intensive care unit: an under-recognized syndrome of organ dysfunction. Crit Care Med 2001;22: Visser L, Prent A, van der Laan MJ, van Leeuwen BL, Izaks GJ, Zeebregts CJ, et al. Predicting postoperative delirium after vascular surgical procedures. J Vasc Surg 2015;62: Böhner H, Hummel T, Habel U, Miller C, Reinbott S, Yang Q, et al. Predicting delirium after vascular surgery: a model based on pre- and intraoperative data. Ann Surg 2003;238: Koebruggae B, van Wensen RJ, Bosscha K, Dautzenberg PL, Koning OH. Delirium after emergency/elective open and endovascular aortoiliac surgery at a surgical ward with a highstandard delirium care protocol. Vascular 2010;18: Benoit AG, Campbell BI, Tanner JR, Stalery JD, Wallbridge HR, Biehl DR, et al. Risk factors and prevalence of perioperative cognitive dysfunction in abdominal aneurysm patients. J Vasc Surg 2005;42: Katznelson R, Djaiani G, Mitsakakis N, Lindsay TF, Tait G, Friedman Z, et al. Delirium following vascular surgery: increased incidence with preoperative b-blocker administration. Can J Anaesth 2009;56: Dasgupta M, Dumbrell AC. Preoperative risk assessment for delirium after noncardiac surgery: a systematic review. J Am Geriatr Soc 2006;54: Lin Y, Chen J, Wang Z. Meta-analysis of factors which influence delirium following cardiac surgery. J Card Surg 2012;27: Ansaloni L, Catena F, Chattat R, Fortuna D, Franceschi C, Mascitti P, et al. Risk factors and incidence of postoperative delirium in elderly patients after elective and emergency surgery. Br J Surg 2010;97: Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 2005;53: Webb AJ, Pendlebury ST, Li L, Simoni M, Lovett N, Mehta Z, et al. Validation of the Montreal Cognitive Assessment versus Mini-Mental State Examination against hypertension and hypertensive arteriopathy after transient ischemic attack or minor stroke. Stroke 2014;45: Freitas S, Simões MR, Alves L, Vincente M, Santana I. Montreal Cognitive Assessment (MoCA): validation study for vascular dementia. J Int Neuropsychol Soc 2012;18: Böhner H, Schneider F, Stierstorfer A, Weiss U, Gabriel A, Friedrichs R, et al. Postoperative delirium following vascular surgery. Comparative results in a prospective study. Anaesthesist 2000;48: Sasajima Y, Sasajima T, Azuma N, Akazawa K, Saito Y, Inaba M, et al. Factors related to postoperative delirium in patients with lower limb ischaemia: a prospective cohort study. Eur J Vasc Endovasc Surg 2012;44: Sugimoto M, Kodama A, Narita H, Banno H, Tamamoto K, Komori K. Pre- and intraoperative predictors of delirium after open abdominal aortic aneurysm repair. Ann Vasc Dis 2015;8: Fick DM, Steis MR, Waller JL, Inouye SK. Delirium superimposed on dementia is associated with prolonged length of stay and poor outcomes in hospitalized older adults. J Hosp Med 2013;8: Partridge JS, Dhesi JK, Cross JD, Lo JW, Taylor PR, Bell R, et al. The prevalence and impact of undiagnosed cognitive impairment in older vascular surgical patients. J Vasc Surg 2014;60: Sasajima Y, Sasajima T, Uchida H, Kawai S, Haga M, Akasaka N, et al. Postoperative delirium in patients with

9 Journal of Vascular Surgery Styra et al 9 Volume -, Number - chronic lower limb ischaemia: what are the specific markers? Eur J Vasc Endovasc Surg 2000;20: Raats JW, Van Hoof-De Lepper CC, Feitsma MT, Meij JJ, Ho GH, Mulder PG, et al. Current factors of fragility and delirium in vascular surgery. Ann Vasc Surg 2015;29: Salata K, Katznelson R, Beattie WS, Carrol J, Lindsay TF, Djaiani G. Endovascular versus open approach to aortic aneruysm repair surgery: rates of postoperative delirium. Can J Anaesth 2012;59: McLennan SN, Mathias JL, Brennan LC, Stewart S. Validity of the Montreal Cognitive Assessment (MoCA) as a screening test for mild cognitive impairment (MCI) in a cardiovascular population. J Geriatr Psychiatry Neurol 2011;24: Inouye SK, van Dyck C, Allesi C, Balkin S, Siegal A, Horwitz R. Clarifying confusion: the confusion assessment method. Ann Intern Med 1990;113: Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, et al. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med 2001;29: Hollinger A, Siegemund M, Goettel N, Steiner LA. Postoperative delirium in cardiac surgery: an unavoidable menace? J Cardiothorac Vasc Anesth 2015;29: Balasundaram B, Holmes J. Delirium in vascular surgery. Eur J Vasc Endovasc Surg 2007;34: Bucerius J, Gummert JF, Borger MA, Walther T, Doll N, Falk V, et al. Predictors of delirium after cardiac surgery delirium: effect of beating-heart (off-pump) surgery. J Thorac Cardiovasc Surg 2004;127: Tse L, Schwarz SK, Bowering JB, Moore RL, Barr AM. Incidence of and risk factors for delirium after cardiac surgery at a quaternary care center: a retrospective cohort study. J Cardiothorac Vasc Anesth 2015;29: Richard E, Reitz C, Honig LH, Schupf N, Tang MX, Manly JJ, et al. Late-life depression, mild cognitive impairment, and dementia. JAMA Neurol 2013;70: Wahlund L, Bjorin GA. Delirium in clinical practice: experiences from a specialized delirium ward. Dement Geriatr Cogn Disord 1999;10: Robertsson B, Blennow K, Gottfries CG, Walin A. Delirium in dementia. Int J Geriatr Psychiatry 1998;13: Merino JG, Hachinski V. Implications and conclusionsd vascular cognitive impairment: evolution of the concept. J Int Neuropsychol Soc 2009;15: Pol RA, Van Leeuwen BL, Visser L, Izaks CJ, Van Den Dungen JJ, Tielliu FI, et al. Standardized frailty indicator as predictor for postoperative delirium after vascular surgery: a prospective cohort study. Eur J Vasc Endovasc Surg 2011;42: van Eijsden WA, Raats JW, Mulder PC, van der Laan L. New aspects of delirium in elderly patients with critical limb ischemia. Clin Interv Aging 2015;10: Shi GM, Jiang T, Zhang H, Li MH, Wang M, Liu YK, et al. Carotid endarterectomy and carotid artery stenting lead to improved cognitive performance in patients with severe carotid artery stenosis. Curr Neurovasc Res 2016;13: Yan Y, Yuan Y, Liang L, Chen T, Shen Y, Zhong C. Influence of carotid artery stenting on cognition of elderly patietns with severe stenosis of the internal carotid artery. Med Sci Monit 2014;20: Benson RA, Ozdemir BA, Matthews D, Loftus IM. A systematic review of postoperative cognitive decline following open and endovascular aortic aneurysm surgery. Ann R Coll Surg Engl 2017;99: MacLullich AM, Beaglehole A, Hall RJ, Meagher DJ. Delirium and long-term cognitive impairment. Int Rev Psychiatry 2009;21: Submitted Mar 19, 2018; accepted May 1, 2018.