Introduction

In Japan, the aging population of individuals with cardiovascular diseases has increased [1], and the prevalence of cognitive impairment among these patients has become a significant concern [2]. Patients with cardiovascular disease who experience cognitive dysfunction have a poor life prognosis [3], increased risk of rehospitalization [4], and decreased activities of daily living (ADL) functionality [5]. Therefore, the assessment and intervention of cognitive function in patients with cardiovascular disease are recommended in rehabilitation guidelines for cardiovascular diseases [6].

Cardiovascular diseases and associated risk factors have been reported as contributors to age-related cognitive decline [7,8]. Factors such as bed rest during hospitalization and environmental changes pose risks for cognitive impairment [9,10]. The cognitive function of patients with cardiovascular disease may decline during hospitalization. However, treatments for cardiovascular disease often result in increased cardiac output and cerebral blood flow [11,12]. Previous studies have indicated increased cerebral blood flow is associated with improved cognitive function [13–15]. Therefore, treatment may improve cognitive function in patients with cardiovascular disease.

However, evidence regarding cognitive function changes in patients with cardiovascular disease during hospitalization remains insufficient. Research reporting changes in cognitive function during hospitalization for patients with cardiovascular disease is scarce, and the factors influencing these changes still need to be clarified. Consequently, this study aimed to elucidate the percentage of patients with cardiovascular disease experiencing changes in cognitive function during hospitalization and identify the characteristics associated with such changes.

Materials and methods

We accessed medical records to conduct this retrospective study on August 28, 2022. The study population consisted of patients admitted to the Cardiology Department of our hospital from September 2019 to March 2022 who underwent cardiac rehabilitation interventions and for whom cognitive function assessment at the time of admission was feasible. Individuals who declined cognitive function assessment, those with severe visual impairment or hearing difficulties hindering assessment, those discharged within one week of the initial cognitive function evaluation, and those with dementia were excluded.

On admission, cognitive function, physical function, and clinical background factors were assessed, with a follow-up cognitive function reassessment at discharge. Cognitive function was evaluated using the Mini-Mental State Examination (MMSE) [16] and the Japanese version of the Montreal Cognitive Assessment (MoCA-J) [17]. Initially, the MMSE was conducted upon admission, and individuals with scores ≤23 were classified as having cognitive impairment [16,18]. Individuals scoring ≥24 underwent subsequent evaluation with the MoCA-J. MoCA-J scores ≤25 indicated mild cognitive impairment (MCI), and scores ≥26 indicated non-cognitive impairment [17]. Individuals in the cognitive impairment group underwent MMSE reassessment at discharge, whereas those in the MCI and non-cognitive impairment groups underwent MoCA-J reassessment. Improvement was defined as a change of ≥2 points in MMSE or MoCA-J scores from those at admission; individuals showing an increase were categorized as the improvement group, those showing a decrease as the decline group, and the remainder as the maintenance group [19–22]. Occupational therapists performed the cognitive function tests according to the manual. Physical function was assessed using grip strength and the Short Physical Performance Battery (SPPB) [23–25].

Clinical background factors extracted from medical records included diagnosis on admission, age, sex, body mass index (BMI), left ventricle ejection fraction, serum albumin (Alb), hemoglobin, brain natriuretic peptide (BNP), C-reactive protein (CRP), estimated glomerular filtration rate (eGFR), Geriatric Nutritional Risk Index (GNRI), admission blood pressure and heart rate, pre-admission ADL (Katz index), presence of dementia, comorbidities (hypertension, diabetes, dyslipidemia, atrial fibrillation, chronic kidney disease, chronic obstructive pulmonary disease, cerebrovascular disease, angina pectoris, myocardial infarction, heart failure admission, lower limb peripheral arterial disease, and malignant tumors), care level, facility admission, employment status, history of falls, length of hospital stay, and outcomes. The GNRI was calculated as 14.89 × serum Alb value + 41.7 × (admission weight/ideal weight) [26].

Comparisons of patient background factors among the cognitive impairment, MCI, and non-cognitive impairment groups at admission were analyzed using the Kruskal–Wallis or χ² test. Three-group comparisons of cognitive function changes were conducted using the Kruskal–Wallis or χ² test for each group based on cognitive function at admission. A multinomial logistic regression analysis was performed using a change in cognitive function during hospitalization as the outcome. Four models were created with a maximum of five independent variables to account for overfitting. Model 1 used severity of cardiac disease, Model 2 used laboratory data, Model 3 used comorbidities, and Model 4 used pre-admission activity status and physical function at admission as independent variables. Comparisons of background factors between those with and without follow-up cognitive function during hospitalization were analyzed using the Mann–Whitney U or χ² test. Statistical analyses were performed using R and EZR software [27]. R is an open-source free software. The statistical significance level was set at 5%.

This study was approved by the Tokyo Police Hospital Ethics Committee (approval number: 19-A20). Medical records were used, and the Ethics Committee waived the requirement for informed consent. All participants were informed of their inclusion in the study and that they were free to withdraw at any time.

Results

A total of 828 individuals were admitted during the study period, and 611 underwent MMSE assessments. Among these, 215 individuals scored ≤23 on the MMSE, whereas 396 individuals scored ≥24. Among the 396 individuals with MMSE scores ≥24, 327 underwent MoCA-J assessments. Of the total study population, 215 individuals (41.3%) were classified as having cognitive impairment, 224 (40.2%) as having MCI, and 103 (18.5%) as having non-cognitive impairment. Background factors among patients in the three groups are presented in Table 1. Age, sex, BMI, Alb, BNP, CRP, eGFR, GNRI, heart rate, pre-hospitalization Katz index, diabetes, dyslipidemia, cerebrovascular disease, care levels of Japanese public long-term care insurance, nursing home resident, employment status, length of hospital stay, outcomes, grip strength, and SPPB were significantly different among the three groups.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Patient characteristics stratified by cognitive function at admission.

https://doi.org/10.1371/journal.pone.0309306.t001

The results of the cognitive function assessment at the time of discharge classified 90 patients (35.9%) into the maintenance group, 117 (46.6%) into the improvement group, and 44 (17.5%) into the decline group (Table 2). There was a statistically significant difference among the three groups only in cognitive function at admission (P = 0.026). Other patient characteristics were not significantly different (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Patient characteristics stratified by change in cognitive function.

https://doi.org/10.1371/journal.pone.0309306.t002

The results of the multivariate analysis of factors associated with changes in cognitive function during hospitalization are shown in Table 3. Multivariate analysis identified MCI at admission, cognitive decline at admission, and age as factors associated with cognitive improvement when the reference was cognitive maintenance (MCI, odds ratio (OR) = 5.62, 95% confidence interval (CI): 1.69–18.7; Cognitive impairment, OR = 12.3, 95% CI: 3.47–43.5; Age, OR = 0.96, 95% CI = 0.92–0.99). Sex, log-BNP, and length of hospital stay were not significantly associated with improved cognitive function when the reference was cognitive maintenance. With reference to maintenance of cognitive function, age, sex, log-BNP, days in hospital, and cognitive function at admission were not significantly associated with cognitive decline. Similarly, models were created to investigate laboratory data, comorbidities, pre-admission activity status, and physical function. However, only age and cognitive function at admission were associated with improved cognitive function during hospitalization, and no factors related to cognitive decline during hospitalization were extracted.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Examination of factors associated with changes in cognitive function.

https://doi.org/10.1371/journal.pone.0309306.t003

The background factors of individuals who dropped out of cognitive function assessment follow-up during hospitalization, compared to those of individuals who continued, are presented in Table 4. The dropout group had a higher proportion of younger individuals, males, those not requiring care, and those with higher BMI, higher Alb, higher Hb, lower BNP, lower CRP, higher eGFR, higher GNRI, higher pre-hospitalization Katz index, lower rate of lower limb peripheral arterial disease, higher employment rate, shorter length of hospital stay, higher in-hospital death rate, higher grip strength, higher SPPB score, and better cognitive function at admission.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Characteristics of patients who dropped out and those who continued follow-up.

https://doi.org/10.1371/journal.pone.0309306.t004

Discussion

Cognitive impairment in patients with cardiovascular diseases is associated with poor prognosis, increased risk of rehospitalization, and functional outcomes. Prevention of cognitive impairment is crucial for disease management. However, the actual changes in cognitive function during hospitalization in patients with cardiac disease and the factors associated with these changes are unknown. To the best of our knowledge, tThis study is the first to elucidate the changes in cognitive function during hospitalization for patients with cardiovascular disease and to investigate their characteristics. The novelty of this research lies in the two-point assessment during hospitalization, which revealed that 46.6% of patients with cardiovascular disease experienced cognitive improvement, whereas 17.5% had cognitive decline. Those with MCI or cognitive impairment at admission, as well as younger patients, were associated with improved cognitive function during hospitalization.

Prior longitudinal studies assessing MMSE in outpatient populations indicated a decline of -0.9 points over one year [28] and -0.2 points over 1.5 years in community-dwelling older adults [29]. Cooley et al. reported a -0.9-point decline in MoCA scores over one year in a longitudinal study [30]. Studies conducted in Japan targeting older adults reported that 30% experienced a decrease of ≥2 points in MoCA-J scores, whereas 30% showed an improvement over one year [22]. In contrast, in the present study, those with cognitive decline at admission were likelier to have improved cognitive function at discharge.

Cerebral blood flow improves with treatment in patients with cardiovascular disease [11,12], and studies on healthy individuals have indicated improved cognitive function with increased cerebral blood flow [13–15]. Based on prior studies on community-dwelling individuals, it can be inferred that patients with cardiovascular disease exhibit improved cognitive function more efficiently. This may be due to correcting a temporary decrease in cerebral blood flow caused by cardiac disease through treatment, resulting in improved cognitive function. However, this study did not evaluate cerebral blood flow and other factors, which presents a future research opportunity.

In our study, 17.5% of patients experienced a decline in cognitive function during hospitalization. Older adults in the hospital have a high prevalence of delirium and depressive symptoms [9], with a 2.4 times greater likelihood of cognitive decline [31]. Older patients with cardiovascular disease often have lower ADL capabilities at admission and lower activity levels in the ward [32], making them more susceptible to the effects of bed rest during hospitalization. Furthermore, patients with cardiovascular disease may experience reduced physical activity due to intravenous treatment [33]. The association between physical activity and cognitive function has been revealed in numerous studies [34,35], and it is plausible that the decline in cognitive function observed in our study was influenced by bed rest during hospitalization. This study has a small sample size and limited variables due to its retrospective nature. Therefore, future prospective studies are needed to examine in detail the factors related to changes in cognitive function.

Considering the results of this study, a certain number of patients with cardiovascular disease will experience cognitive decline during hospitalization, making screening tests for cognitive function important. Interventions that maintain or improve cognitive function should then be considered. Targeted interventions for these individuals could contribute to improved cognitive function, prevention of rehospitalization, and better life prognosis. However, it is essential to note that older patients with cardiovascular disease require individualized focus, and guidelines emphasize the importance of providing comprehensive cardiac rehabilitation interventions tailored to each patient [6]. Individualized assessments, especially risk assessments for cognitive function changes during hospitalization, and interventions by specialized teams are crucial.

The strength of this study is that it reveals previously unknown changes in cognitive function during hospitalization, which is essential for long-term prognosis and disease management in patients with cardiac disease. However, this study has several limitations, including its small sample size and single-center setting. Future large-scale multicenter cohort studies are necessary to elucidate the factors related to cognitive function changes during hospitalization in patients with cardiovascular diseases. We performed a multivariate analysis, but overfitting problems limited the number of independent variables. Future studies should use large data sets and conduct detailed analyses using multivariate and subgroup analyses. In such studies, investigating additional factors related to cognitive function, such as physical activity, quality of life, education level, delirium, and depressive symptoms, is essential. Furthermore, the high dropout rate in our study suggests the possibility of sampling bias. The patients analyzed in this study may represent a more severe subset of patients with cardiovascular disease, as those who dropped out had fewer complications and were discharged earlier. Efforts to evaluate the factors contributing to dropout and establish measures to minimize dropout are needed. Therefore, high-quality prospective studies are required in the future. Additionally, the data in our study are limited to the hospitalization period, and the post-discharge course is unknown. Because changes in cognitive function during hospitalization could be temporary, further detailed investigations are required to determine whether cognitive decline during hospitalization is associated with cognitive decline, the onset of dementia or MCI after discharge, as well as with life prognosis, rehospitalization, and functional prognosis. Finally, this study did not consider the effects of treatment during hospitalization. Future studies should include a more detailed examination of treatment and the resulting changes in cardiac function.

Conclusions

On admission, 41.3% of patients with cardiovascular disease exhibited cognitive impairment, and 40.2% had MCI. Cognitive decline occurred in 17.5% of the patients during hospitalization, whereas cognitive improvement occurred in 46.6%. These findings demonstrate the importance of assessing cognitive changes in hospitalized patients with cardiovascular disease. Future studies are needed to identify factors associated with changes in cognitive function.

Acknowledgments

We express our gratitude to Dr. Masafumi Kasao, Dr. Tadayuki Kadohira, Dr. Mitsunobu Kaneko, Dr. Yuugo Nara, Nurse Airi Shimoda, Nurse Yuka Yamazaki, Nurse Rikako Endo, Nurse Saori Yoshida, and the Rehabilitation Department staff for their valuable cooperation and support in conducting this research.