2018 Volume 82 Issue 12 Pages 2998-3004
Background: The strategy for cardiovascular surgery in dementia patients is controversial, so we aimed to investigate whether preoperative dementia and its severity might affect the outcomes of cardiovascular surgery by evaluating with the Mini-Mental State Examination (MMSE).
Methods and Results: The study group comprised 490 patients undergoing cardiovascular surgery. Their preoperative cognitive status was evaluated using the MMSE, and analysis was performed to compare the patients with MMSE score <24 (dementia group, n=51) or MMSE score 24–30 (non-dementia group, n=439). Furthermore, the effect of the severity of dementia was analyzed with a cut-off MMSE score of 19/20. Risk factors for surgical outcomes were explored using multivariate logistic regression analysis. Hospital mortality was 11.8% in the dementia group and 2.1% in the non-dementia group (P=0.002). Regarding the postoperative morbidities, the incidence of cerebrovascular disorder (P=0.001), pneumonia (P=0.039), delirium (P=0.004), and infection (P=0.006) was more frequent in dementia group. Among the patients with MMSE <20, hospital mortality was as high as 25%, and the rate of delirium was 58%. Multivariate logistic regression analysis revealed that MMSE score <24 (P=0.003), lower serum albumin (P=0.023) and aortic surgery (P=0.036) were independent risk factors for hospital death.
Conclusions: Preoperative dementia affects the outcomes of cardiovascular surgery with regard to hospital death and delirium. The surgical indication for patients with MMSE <20 might be difficult, but surgery with an appropriate strategy should be considered for patients with MMSE <24.
Population aging is a common issue in developed countries. In particular, the shift towards an aging society in Japan is proceeding at unprecedented speed in this century. The government of Japan reported that the population aged >65 years was 35 million in 2017 and accounted for 27.6% of the total population. As more than 40% of the elderly population suffers from cardiac disease,1 the candidates for cardiovascular surgery in this group may continue to increase.2 Advances in cardiopulmonary bypass, operative techniques and perioperative care have enabled surgery to be performed in elderly patients with acceptable rates of perioperative mortality and morbidity.3–7
Editorial p 2939
According to this aging progression, the elderly population with cognitive disorders may also increase. In 2013 the Minister of Health, Labor and Welfare of Japan estimated that the population of dementia patients was 4.62 million, and one-quarter of the elderly have some form of cognitive disorder, including dementia. For both the elderly patient with dementia and the treating surgeon, clinical decision making is often difficult, and weighing the risk against the benefit of surgery may be a complex matter.
Clinicians can diagnose the syndromes of dementia based on the standard criteria of the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5).8 Briefly, the DSM-5 diagnosis of dementia, which corresponds to a “major neurocognitive disorder”, requires substantial impairment to be present in ≥1 cognitive domains and must be sufficient to interfere with independence in everyday activities. The Mini-Mental State Examination (MMSE) is one of the most widely used instruments used in dementia screening, and it is highly cited because of its brief method for detecting suspected dementia.9 The maximum MMSE score is 30 points, and consists of cognitive functions in orientation, registration, attention, calculation, recall and language. Bour et al reported that the optimal cut-off score for the diagnosis of dementia was 23/24, classifying dementia for MMSE score <24, and this value has been verified as a standard criterion for estimating the neurocognitive disorder.10,11 Furthermore, to evaluate dementia severity, the relationship between the MMSE score and the Clinical Dementia Rating (CDR) categories has been reported. The CDR categories consists of severe, moderate, and mild dementia,12 and it has been suggested that these correspond to MMSE score <10, <20, and ≥20, respectively.13
The aims of this study were to investigate whether preoperative dementia affects the outcomes of cardiovascular surgery after assessment of risk factors using MMSE scoring, and whether the surgical results may be altered by the severity of dementia.
A total of 1,396 patients who underwent cardiac surgery at Nagoya University Hospital from January 2012 to December 2015 were included in this study. Patients were evaluated by the MMSE; grip and leg strength; and 6-minute walk distance (6MWD) on admission. Inclusion criteria were as follows: (1) age ≥60 years, (2) elective cardiac surgery (coronary artery bypass grafting (CABG), valve surgery, thoracic aortic surgery), (3) absence of central nervous system disease, or severe hearing or vision disorders, and (4) agreement for evaluation of cognitive function. Finally, 490 patients were analyzed. There were no patients who had been determined as unsuitable for surgery according to the MMSE score after being referred to the Department of Cardiac Surgery during the study period. The surgeries included CABG (n=122), valve surgery (n=162), and aortic surgery (n=206). All patients underwent perioperative rehabilitation. The Ethics Committee of Nagoya University Hospital approved this retrospective study (No. 2016-197), and the approval included a waiver of individual consent.
Data Collection and DefinitionsThe patients’ demographic, procedural and postoperative outcome data were collected. Follow-up data were obtained from hospital charts, referring physicians at the follow-up institutions, or by contacting the patients. The 30-day mortality included all-cause death occurring within 30 days of the procedure, even if the patient was discharged from the hospital within that time. Hospital death included any death during hospitalization after surgery.
The MMSE was used to assess the patients’ general cognitive function. As shown in the Figure, patients were distributed into 2 groups according to a conventional cut-off score of 23/24.9 Patients with MMSE score <24 (n=51) were classified as the dementia group, while those with score of 24–30 (n=439) were classified as the non-dementia group. Statistical analysis was performed to compare the 2 groups. Furthermore, patients in the dementia group were classified according to the severity of dementia, as described in CDR and MMSE studies.12,13 Patients with MMSE score <20 were classified as the moderate dementia group (n=12), while those with a score ≥20 were classified as the mild dementia group (n=39), and the results of the groups were compared. There were no patients evaluated as having severe dementia.
Study flowchart. CABG, coronary artery bypass grafting; MMSE, Mini-Mental State Examination.
Preoperative physical activity level was examined by the 6MWD, handgrip strength and knee extensor muscle strength. The 6MWD test was performed according to the standardized procedure described by the ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories.14 Briefly, patients were instructed to walk the length of a predetermined course at their own pace while attempting to cover as much distance as possible in the 6 min. At the end of the time period, patients were instructed to stop walking, and the distance covered was measured. Handgrip strength was measured by standard adjustable-handle JAMAR dynamometer, and knee extensor muscle strength was measured by a hand-held dynamometer. Measurements were performed twice for each hand and leg, and the average of the highest value of each measurement was calculated.
We analyzed 20 preoperative variables and surgical procedures to determine the risk factors for hospital death, non-discharge to home (including death or transfer to convalescent hospital), major complications, and delirium. Major complications consisted of cerebrovascular disorder, pneumonia, dysphagia, wound infection, and bleeding. Cerebrovascular disorder included cerebrovascular bleeding/infarction with de novo neurological symptoms and a lesion on radiological image. Pneumonia and dysphagia were diagnosed by respective specialists. Wound infection included cases of superficial and/or deep surgical site infection, and bleeding included cases of re-exploration performed for hemostasis. To assess the cutoff score for the continuous value of preoperative physical activity measurement for logistic regression analysis, receiver-operating characteristics (ROC) curves were obtained. The calculated cutoff values were 350 m for 6MWD, 22 kg for grip strength, and 16 kg for leg strength. The variables identified by univariate analysis were included in the multivariate logistic regression analysis.
Statistical AnalysisCategorical variables were compared using a univariate logistic regression model or Fisher’s exact test for the analysis. Continuous variables are expressed as the mean±standard deviation and were analyzed using Student’s t-test. Potential risk factors for the surgical outcomes, as determined by univariate analysis (with P value <0.2), were explored with multivariate logistic regression analysis. P<0.05 was considered to be significant. All statistical analyses were performed using the SPSS 24 software program (IBM SPSS Statistics, NY, USA).
The baseline characteristics of the patients are shown in Table 1. In the analysis of 51 patients in the dementia group and 439 in the non-dementia group, the mean age and EuroSCORE II of the dementia group were higher than those for the non-dementia group (P<0.005, P=0.027, respectively). In dementia group, body surface area and serum albumin level were less than in the non-dementia group (P=0.032, 0.016, respectively). Additionally, the preoperative physical activity level of the dementia group was significantly lower in comparison with the non-dementia group (6MWD, grip strength, and leg strength, each P<0.005). The numbers of surgical procedures were 12 coronary, 15 valve and 24 aorta in the dementia group, and 110, 147 and 182, respectively, in the non-dementia group (P=0.189). Coronary surgery included 6 off-pump (50.0%) in the dementia group and 41 off-pump (37.3%) in the non-dementia group (P=0.534). Transcatheter aortic valve replacement (TAVR) was not performed in this study period.
| Dementia group (n=51) |
Non-dementia group (n=439) |
P value | |
|---|---|---|---|
| MMSE score | 11–23 | 24–30 | |
| Age (years) | 75.4±5.9 | 71.5±7.2 | <0.005 |
| Sex (M/F) | 32/19 | 298/141 | 0.528 |
| Body surface area (m2) | 1.5±0.2 | 1.6±0.4 | 0.032 |
| Hypertension | 33 (65%) | 248 (56%) | 0.297 |
| Hyperlipidemia | 12 (24%) | 82 (19%) | 0.341 |
| Diabetes mellitus | 15 (29%) | 158 (36%) | 0.439 |
| Smoking | 27 (53%) | 189 (43%) | 0.093 |
| Hemodialysis | 1 (2%) | 23 (5%) | 0.496 |
| Atrial fibrillation | 5 (10%) | 85 (19%) | 0.125 |
| COPD | 3 (6%) | 8 (2%) | 0.084 |
| History of cerebrovascular disease | 11 (22%) | 51 (12%) | 0.071 |
| History of cardiovascular surgery | 12 (24%) | 71 (16%) | 0.160 |
| History of heart failure | 15 (29%) | 132 (30%) | 1.000 |
| Ejection fraction (%) | 61.5±11.4 | 60.7±12.2 | 0.695 |
| Serum albumin level (g/dL) | 3.7±0.6 | 3.9±0.5 | 0.016 |
| EuroSCORE II (%) | 7.3±6.8 | 5.1±5.6 | 0.027 |
| Grip strength (kg) | 21.8±7.6 | 26.0±9.2 | <0.005 |
| Leg strength (kg) | 17.3±8.0 | 23.8±10.0 | <0.005 |
| 6-minute walk distance (m) | 361.5±80.0 | 422.3±120.6 | <0.005 |
| Procedures | 0.189 | ||
| Coronary artery bypass grafting | 12 (23.5%) | 110 (25.0%) | |
| Valve surgery | 15 (29.4%) | 147 (33.5%) | |
| Aortic surgery | 24 (47.1%) | 182 (41.5%) |
Data are n (%) or mean±standard deviation. COPD, chronic obstructive pulmonary disease; MMSE, Mini-Mental State Examination; EuroSCORE II, European System for Cardiac Operative Risk Evaluation II.
With regard to the 30-day mortality rate, there were 3 deaths (5.9%) in the dementia group, and 5 (1.1%) in the non-dementia group (P=0.041). The hospital mortality rate in the dementia group was higher than in the non-dementia group (11.8% vs. 2.1%, P=0.002; Table 2). There were no significant differences in operation time, cardiopulmonary bypass time, ventilation time, or ICU stay, except for postoperative hospital stay, which was longer in the dementia group than in the non-dementia group (36.1 vs. 26.8 days, P=0.014). The number of patients who transferred to a convalescent hospital was 16 (31.4%) in the dementia group and 64 (14.6%) in the non-dementia group (P=0.005). Mid-term follow-up for these transferred patients indicated that 13 patients (81.3%) in the dementia group were discharged to home on average 59.9±66.0 months later. Meanwhile, 51 patients (79.7%) in the non-dementia group were discharged home 61.4±41.3 months later, which presented no difference between groups (P=0.940).
| Dementia group (n=51) |
Non-dementia group (n=439) |
P value | |
|---|---|---|---|
| 30-day mortality | 3 (5.9%) | 5 (1.1%) | 0.041 |
| Hospital mortality | 6 (11.8%) | 9 (2.1%) | 0.002 |
| Operation time (min) | 414±152 | 390±136 | 0.247 |
| CPB time (min) | 191±79 | 191±78 | 0.975 |
| Ventilation time (h) | 5,009±8,896 | 3,887±9,442 | 0.412 |
| ICU stay (days) | 7.4±11.1 | 4.7±7.1 | 0.102 |
| Hospital stay (days) | 36.1±27.2 | 26.8±25.1 | 0.014 |
| Transfer to convalescent hospital | 16 (31.4%) | 64 (14.6%) | 0.005 |
| Morbidities | |||
| Arrhythmia | 27 (53%) | 178 (41%) | 0.098 |
| Cerebrovascular disorder | 9 (18%) | 19 (4%) | 0.001 |
| Pneumonia | 9 (18%) | 36 (8%) | 0.039 |
| Dysphagia | 9 (18%) | 33 (7%) | 0.259 |
| Wound infection | 7 (14%) | 9 (2%) | 0.006 |
| Bleeding | 2 (4%) | 11 (2%) | 1.000 |
| Delirium | 18 (35%) | 84 (19%) | 0.004 |
Data are n (%) or mean±standard deviation. CPB, cardiopulmonary bypass; ICU, intensive care unit.
Regarding the postoperative morbidities, univariate analysis revealed that the incidence of cerebrovascular disorder (P=0.001), pneumonia (P=0.039), delirium (P=0.004), wound infection (including both superficial and deep infection) (P=0.006) was more frequent in the dementia group than in the non-dementia group.
Analysis According to the Severity of DementiaAccording to our definitions, the 51 patients in the dementia group were divided into 2 groups according to their MMSE scores: those with a score of 20–23 were classified as the mild dementia group (n=39), and those with a score of 11–19 were classified as the moderate dementia group (n=12) (Table 3). With regard to the hospital mortality, there were 3 deaths (25%) in the moderate dementia group and 3 (8%) in the mild dementia group, with no significant difference in mortality (P=0.134). Regarding morbidity, pneumonia (P=0.013) and delirium (P=0.020) occurred more frequently in the moderate dementia group.
| Moderate dementia group (n=12) |
Mild dementia group (n=39) |
P value | |
|---|---|---|---|
| MMSE score | 11–19 | 20–23 | |
| Age (years) | 73.6±5.0 | 75.9±6.0 | 0.231 |
| Hospital mortality | 3 (25%) | 3 (8%) | 0.134 |
| Hospital stay (days) | 44.9±38.4 | 33.4±23.0 | 0.341 |
| Morbidities | |||
| Arrhythmia | 27 (53%) | 21 (54%) | 0.098 |
| Cerebrovascular disorder | 3 (25%) | 6 (15%) | 0.424 |
| Pneumonia | 5 (42%) | 4 (10%) | 0.013 |
| Dysphagia | 2 (17%) | 7 (18%) | 1.000 |
| Wound infection | 2 (17%) | 5 (13%) | 0.661 |
| Bleeding | 0 (0%) | 2 (5%) | 1.000 |
| Delirium | 7 (58%) | 11 (28%) | 0.020 |
Data are n (%) or mean±standard deviation. MMSE, Mini-Mental State Examination.
The details of the univariate analysis of preoperative and surgical factors for various outcomes are summarized in Table 4. Potential risk factors for the surgical outcomes were determined with P<0.2 and the results for the various outcomes are summarized in Table 5. With regard to hospital mortality, MMSE score <24, serum albumin level, and aortic surgery were identified as independent risk factors (P=0.003, 0.023, 0.036, respectively). Age >75 years, hemodialysis, 6MWD <350 m, aortic surgery, and EuroSCORE II were independent risk factors for non-discharge to home; otherwise, a low MMSE score had no effect. There were no obvious risk factors for major complications except for aortic surgery (P=0.042). Furthermore, only MMSE score <24 (P=0.017) and EuroSCORE II (P<0.001) were identified as independent risk factors for delirium.
| Hospital mortality |
Non-discharge to home |
Major complication |
Delirium | |
|---|---|---|---|---|
| MMSE <24 | 0.002 | <0.001 | <0.001 | 0.002 |
| Age >75 years | 0.153 | <0.001 | 0.735 | 0.261 |
| Age >80 years | 0.047 | 0.015 | 0.491 | 0.311 |
| Sex | 1.000 | 0.099 | 0.025 | 0.443 |
| Body surface area | 0.116 | 0.982 | 0.309 | 0.285 |
| Hypertension | 0.853 | 0.217 | 0.664 | 0.634 |
| Hyperlipidemia | 0.947 | 0.915 | 0.402 | 0.494 |
| Diabetes mellitus | 0.853 | 0.361 | 0.773 | 0.240 |
| Smoking | 0.989 | 0.774 | 0.169 | 0.570 |
| Hemodialysis | 0.749 | 0.014 | 0.003 | 0.129 |
| Atrial fibrillation | 1.000 | 0.922 | 0.072 | 0.621 |
| COPD | 0.378 | 0.629 | 0.988 | 0.748 |
| History of cerebrovascular disease | 0.938 | 0.777 | 0.600 | 0.748 |
| History of cardiovascular surgery | 0.330 | 0.238 | 0.178 | 0.263 |
| History of heart failure | 0.779 | 0.972 | 0.008 | 0.753 |
| Ejection fraction | 0.916 | 0.224 | 0.713 | 0.204 |
| Serum albumin level | 0.013 | 0.002 | 0.144 | 0.128 |
| EuroSCORE II | 0.033 | <0.001 | 0.003 | <0.001 |
| Grip strength <22 kg | 0.169 | <0.001 | 0.707 | 0.109 |
| Leg strength <16 kg | 0.014 | <0.001 | 0.265 | 0.252 |
| 6MWD <350 m | 0.562 | <0.001 | 0.196 | 0.205 |
| Surgical procedure | 0.040 | 0.003 | <0.001 | 0.159 |
P values are given. 6MWD, 6-minute walk distance. Other abbreviations as in Table 1.
| OR | 95% CI | P value | |
|---|---|---|---|
| Risk for hospital death | |||
| MMSE <24 | 5.715 | 1.798–18.165 | 0.003 |
| Serum albumin level | 0.329 | 0.126–0.858 | 0.023 |
| Aortic surgery | 2.709 | 1.067–6.880 | 0.036 |
| Risk for non-discharge to home | |||
| Age >75 years | 2.811 | 1.418–5.574 | 0.003 |
| Hemodialysis | 4.960 | 1.339–18.376 | 0.017 |
| 6MWD <350 m | 5.514 | 2.687–11.316 | 0.000 |
| Aortic surgery | 2.301 | 1.371–3.862 | 0.002 |
| EuroSCORE II | 1.077 | 1.012–1.146 | 0.020 |
| Risk for major complications | |||
| Aortic surgery | 1.689 | 1.019–2.801 | 0.042 |
| Risk for delirium | |||
| MMSE <24 | 2.257 | 1.156–4.407 | 0.017 |
| EuroSCORE II | 1.085 | 1.044–1.128 | <0.001 |
CI, confidence interval; OR, odds ratio. Other abbreviations as in Tables 1,4.
The results of the multivariate logistic regression analysis of 51 dementia patients are summarized in Table 6. There were no independent risk factors for hospital death. Aortic surgery was a risk factor for non-discharge to home and major complications. Regarding the risk for delirium, MMSE score <20 was identified as an independent risk factor (P=0.028).
| OR | 95% CI | P value | |
|---|---|---|---|
| Risk for hospital death: None | |||
| Risk for non-discharge to home: Aortic surgery | 2.387 | 1.065–5.349 | 0.035 |
| Risk for major complications: Aortic surgery | 3.017 | 1.296–7.025 | 0.010 |
| Risk for delirium: MMSE <20 | 5.515 | 1.200–25.348 | 0.028 |
Abbreviations as in Tables 1,5.
The present study analyzed the preoperative background and surgical outcomes among patients who underwent MMSE assessment. Regarding background, the dementia group with MMSE <24 was older and had higher EuroSCORE II, but lower serum albumin and exercise tolerance. Surgical outcomes showed higher rates of hospital mortality and morbidity, especially for cerebrovascular disorder, pneumonia, infection and delirium. Multivariate analysis revealed that preoperative MMSE score <24 was a risk factor for hospital death and delirium.
The outcomes of cardiovascular surgery in patients with preoperative dementia have seldom been assessed in studies, probably because patients with dementia are more likely to be excluded from the indications for surgery. Thus, the effect of the severity of dementia in cardiovascular surgery patients is unknown. A few studies of the impact of frailty in cardiac surgery, which included assessment of cognitive decline as one of the frailties, have been reviewed, with the conclusion of a higher likelihood of death, morbidity, functional decline and major adverse cardiac and cerebrovascular events.15 In the era of aging society and the spread of less invasive surgeries,16 we have to discuss appropriate strategies for mild to moderate dementia patients with subjective symptoms. Fundamentally, all surgical patients had to comprehend their treatment, and the surgical risks that are augmented by dementia, including mortality and morbidity, have to be described for patients and their families. As a result of this explanation, the strategy would be chosen by the patient and followed by the family’s consent.
According to the severity of dementia in this study, among moderate dementia patients with MMSE <20, hospital mortality was up to 25%, and furthermore, the rates of occurrence of delirium and pneumonia were as high as 58%, and 42%, respectively. Although MMSE <20 was not a significant risk factor for hospital death, it was a significant risk factor for delirium. Preoperative dementia has been previously reported as a risk factor for delirium,17 and the MMSE can also predict delirium after cardiovascular surgery.18 Postoperative delirium is associated with negative outcomes that include prolonged hospitalization, persistent functional and cognitive deficits, and increased 30-day mortality.19 Meanwhile, with regard to postoperative pneumonia, cardiovascular surgery has been reported to be strongly associated with aspiration pneumonia in elderly patients, in whom the incidence rate of postoperative aspiration pneumonia was 9.8% in 1 study.20 Several risk factors for postoperative aspiration pneumonia have been reported, including advanced age, prolonged intubation and cerebrovascular disorder.21,22 Furthermore, the coexistence of cognitive impairment and oral weakness was reported to enhance the risk of aspiration pneumonia.23 In the present study, although there was no obvious relationship between pneumonia and physical frailty, additional univariate analysis indicated that MMSE <24 and MMSE <20 were significant risk factors for postoperative pneumonia (P=0.028, 0.003, respectively). Thus, in decision-making regarding cardiovascular surgery for patients with dementia, the surgical indication must be carefully discussed and, furthermore, maximum attention should be paid to preventing delirium and pneumonia.
Another possibility to consider is that dementia itself might have shortened the survival of the patients. Staekenborg et al have reported that a lower MMSE score (20±5) and the complication of cardiovascular disease were risk factors for death within 2 years after a diagnosis of dementia.24 Thus, the surgical indications for patients with moderate dementia (MMSE score <20) might still be controversial with high surgical risk and poor prognosis. And if a surgical strategy is selected, minimally invasive strategy represented by off-pump CABG, TAVR, and endovascular aneurysm repair, should be considered. Meanwhile, surgery should be considered for patients with mild dementia (MMSE <24) as it may be more effective than medical treatment in such cases. However, less invasive surgical methods should be implemented.
Regarding the perioperative management, although the postoperative rehabilitation program for cardiovascular surgery has been established, additional individual management might be necessary for each patient with dementia. In our institution, the postoperative mobilization program is based on the guidelines of the Japanese Circulation Society. Symptomatic dementia patients are referred to a geriatrician before surgery to get instructions regarding perioperative care. Generally, although cognitive impairment is the hallmark of dementia, the perioperative problems mainly consist of “behavioral and psychological symptoms of dementia (BPSD)” which is identified as aggression, agitation, depression, anxiety, delusions, hallucinations, delirium, apathy, and disinhibition.25 Non-pharmacological management of BPSD is recommended as the first line, and we have been promoting early mobilization, additional rehabilitation through occupational therapy, and family support. Secondly, a pharmacological approach is recommended with antipsychotic medication for BPSD, which includes risperidone, olanzapine and haloperidol.26,27 Additionally, a traditional Chinese medicine, yokukan-san, has been recommended for improving BPSD,28 and we have been trialling its administration. Thus, a comprehensive strategy including both non-pharmacological and pharmacological therapy with a multidisciplinary team should be instituted to determine the desirable perioperative control of cardiovascular surgery even in dementia patients.
Study LimitationsFirst, the present study only investigated the early postoperative period. The effect of the preoperative MMSE score over a longer follow-up period should be investigated. Second, the present study defined dementia using only the MMSE score. There are many tests with different approaches to the estimation of cognitive function, and because cognitive disorder is so complex, it would be better defined using several cognitive tests and neurological diagnosis. Third, we could not get information about the number of dementia patients who were determined as inadequate for surgery by physicians before referral during the study period. Since this study, less invasive strategies such as TAVR have become available, and we have had several cases of discussing the strategies for low MMSE score patients. Further study including surgically reassigned patients because of low MMSE score should be performed. Finally, the study only included a small number of participants from a single medical center. A larger number of patients with multiple center enrolment is needed for further discussion of appropriate strategies.
This study demonstrated that preoperative dementia affects the outcomes of cardiovascular surgery with regard to hospital death and the incidence of postoperative delirium. MMSE score might be a useful preoperative cognitive assessment for determining the strategy for elderly patients. Surgical intervention on patients diagnosed as having moderate dementia with MMSE score <20 remains controversial because of the high rates of mortality and morbidity, though surgery with a less invasive strategy should be considered for patients with MMSE score <24.
We thank Kiyonori Kobayashi, PT, MSc and Yoshihiro Nishida, MD, PhD (Department of Rehabilitation, Nagoya University Hospital, Nagoya, Japan) for their clinical assistance.
The authors declare no conflict of interest.
