論文ID: CJ-24-0124
Background: In contrast to the well-known prognostic values of the cardiorenal linkage, it remains unclear whether impaired cognitive function affects cardiac prognosis in relation to cardiac sympathetic innervation and renal function in patients with heart failure (HF).
Methods and Results: A total of 433 consecutive HF patients with left ventricular ejection fraction (LVEF) <50% underwent the Mini-Mental State Examination (MMSE) and a neuropsychological test for screening of cognition impairment or subclinical dementia. Following metaiodobenzylguanidine (MIBG) scintigraphy, patient outcomes with a primary endpoint of lethal cardiac events (CEs) were evaluated for a mean period of 14.8 months. CEs were documented in 84 HF patients during follow-up. MMSE score, estimated glomerular filtration rate (eGFR) and standardized heart-to-mediastinum ratio of MIBG activity (sHMR) were significantly reduced in patients with CEs compared with patients without CEs. Furthermore, overall multivariate analysis revealed that these parameters were significant independent determinants of CEs. The cutoff values of MMSE score (<26), sHMR (<1.80) and eGFR (<47.0 mL/min/1.73 m2) determined by receiver operating characteristic (ROC) analysis successfully differentiated HF patients at more increased risk for CEs from other HF patients.
Conclusions: Impairment of cognitive function is not only independently related to but also synergistically increases cardiac mortality risk in association with cardiac sympathetic function and renal function in patients with HF.
The world’s population is aging rapidly, and the number of elderly people with heart disease is increasing.1,2 In addition, clinicians are faced with mental health issues that are unique to the elderly. A recent large North American survey showed that the frequencies of cognitive impairment in elderly patients with chronic heart failure (HF) were 24% for mild cognitive impairment and 15% for dementia.3 The risk of dementia in elderly HF patients has been reported to be 2–4-fold higher than that in normal subjects,4 and both the frequency and severity of dementia have been reported as positively correlated with the severity of HF.5 Memory and attention deficits are the main symptoms of dementia, and they significantly reduce self-care quality of life.6 Dementia also causes non-adherence to medications, and affects the prognosis and condition of heart disease itself. Recent studies have suggested that decreased cerebral perfusion due to HF is a potential cause of Alzheimer’s disease.7,8 Other cardiovascular risk factors (e.g., hypertension, lipid abnormalities, and diabetes mellitus), apolipoprotein E, and atrial fibrillation have also been reported to increase the risk of dementia in patients with heart disease.9 However, it has been also reported that cerebral blood flow improved after heart transplantation,10 that cognitive function improved after cardiac resynchronization therapy (CRT),11 and that cognitive function improved after treatment of anemia and prescription of statins.12,13 Dementia in patients with chronic HF is mainly due to aging, complications of neurological diseases including Alzheimer’s disease, and cardiac output reduction, among others. It is thought that a treatable form of dementia, known as reversible pathology, may exist if these conditions can be differentiated and the cause of dementia is reduced cardiac output due to cardiac disease.
Chronic kidney disease (CKD) is also a risk factor for cognitive dysfunction and is a particularly higher risk for the development of cerebrovascular dementia characterized by impaired executive function.14 Renal dysfunction and albuminuria have both been shown to be independent risk factors for the development of dementia in patients with CKD.15 Many factors are thought to be involved in the mechanism of cognitive dysfunction in patients with CKD. Vascular risk factors include cerebrovascular disease, hypertension, diabetes mellitus, dyslipidemia, myocardial infarction, atrial fibrillation and smoking, and nonvascular risk factors include anemia, hyperparathyroidism, aluminum, drugs, sleep disorders and depression.16
Recently, the concept of a pathological condition known as the brain-heart-renal linkage has been proposed and the mechanism elucidated.17 Sympathetic hyperactivity, renin-angiotensin-aldosterone (RAS) activation, endothelin (ET) activation, insulin resistance, increased oxidative stress, and renal anemia are thought to overlap and contribute to the development of cerebrovascular disease; however, there have been few reports.
The aim of this study was to determine prognostic incremental values for risk stratification of HF patients with reduced left ventricular ejection fraction (LVEF) with a focus on the prognostic interactions of cognitive function assessed by the Mini-Mental State Examination (MMSE), renal function and cardiac sympathetic nerve system.
A total of 433 consecutive patients with symptomatic HF, echocardiographic LVEF <50%, and cardiac sympathetic function assessed by 111 MBq 123I-labeled metaiodobenzylguanidine (MIBG) cardiac scintigraphy who were admitted between April 2017 and March 2022 were retrospectively enrolled from the Teine Keijinkai Hospital HF database. The study was approved by the Ethics Committee of Teine Keiinkai Hospital (approval no. 2-023090-00) in accordance with the tenets of the Declaration of Helsinki. Due to the retrospective and observational nature of this study, the need to obtain written informed consent from each patient was waived. The enrollment criteria included symptomatic HF requiring hospitalization and LVEF <50% determined by transthoracic echocardiography. Patients who refused to undergo MIBG myocardial scintigraphy of their own volition or who were unable to undergo MIBG myocardial scintigraphy due to a serious medical condition or inability to maintain bed rest, patients who refused resuscitative treatment and patients with malignant tumors with a clearly poor prognosis, patients with noncardiac disease (severe emphysema, cirrhosis, etc.) whose expected survival was <1 year, and patients <20 years of age were excluded. The selection process and the number of HF patients selected for this study are shown in Figure 1. A definite diagnosis of HF was made by several experienced cardiologists with reference to the Framingham criteria, including typical symptoms such as distended jugular veins, peripheral edema, pulmonary rales, S3 or S4 gallop, and tachycardia. Chest X-rays and 2-dimensional transthoracic echocardiography were performed to confirm the final diagnosis of HF and to exclude other conditions with similar symptoms and signs. In addition to a history of myocardial infarction and/or coronary revascularization, ischemic and nonischemic etiologies were differentiated by electrocardiography, echocardiography, cardiac scintigraphy, or a combination of these, as appropriate, and coronary artery information was obtained by computed tomography, magnetic resonance imaging, and invasive selective coronary angiography. Blood tests for hemoglobin (Hb), creatinine, serum sodium and B-type natriuretic peptide (BNP) were performed immediately before discharge from the hospital after compensated HF and optimal medical treatment. Kidney function was assessed by the use of a standard estimated glomerular filtration rate (eGFR) formula.
Study flowchart. MIBG, metaiodobenzylguanidine; MMSE, Mini-Mental State Examination.
Measurement of Cognitive Function
The policy for the current study was to use the MMSE for the assessment and evaluation of dementia because it has a short test time (10–15 min) and can be easily administered without requiring special equipment.18 The MMSE is a 30-point cognitive function test consisting of 11 items: time perception, place perception, immediate and delayed playback of 3 words, calculation, object naming, sentence recitation, 3 levels of verbal commands, written commands, written text, and graphic imitation. An MMSE score ≤23 suggests dementia and an MMSE score ≤27 suggests mild cognitive impairment (MCI). Recently, the frequency of juvenile dementia has been reported to be increasing,19 and the MMSE was administered to patients admitted with HF, regardless of whether they were young or not, because the coexistence of severe cognitive function may influence treatment decisions. The MMSE was administered by experienced physical therapists from the hospital’s rehabilitation department after HF had been compensated and the consciousness of the HF patients was clear; however, the physical therapists were unaware of the laboratory data and clinical information for the enrolled patients, including cardiac nuclear medicine tests and echocardiography.
Echocardiographic AssessmentUsing a commercially available ultrasound system equipped with a 2.5-MHz variable frequency transducer, the echocardiographer performed a 2-dimensional echocardiographic examination from the paracentral long-axis image and apical 4-, 3- and 2-chamber images in the left lateral decubitus position. The following echocardiographic parameters were measured under stable conditions before discharge: left atrial diameter (LAD; mm), LV end-diastolic diameter (LVDd; mm), LVEF (%) calculated by the biplane modified Simpson method, LV end-diastolic volume (EDV; mL), LV end-systolic volume (ESV; mL) and septal E/e′.20
Assessment of Cardiac Sympathetic InnervationCardiac sympathetic innervation was quantitatively assessed in patients, after compensation of HF and initiation of optimal and possible medical therapy before discharge, by cardiac imaging with MIBG using a γ camera with a general-purpose low-energy collimator for 15–30 min (early images) and 4 h (late images) after intravenous tracer injection as described previously.21 Cardiac 123I-MIBG activity was measured as the heart-to-mediastinum ratio (HMR) by an experienced nuclear medicine technologist unaware of the clinical data, using dedicated MIBG software (smart MIBG software, Tokyo, Japan), which automatically defines regions of interest in the superior mediastinum and the entire heart on a planar anterior image. The kinetics of 123I-MIBG washout from the heart were calculated as the washout rate (WR) from early and late cardiac 123I-MIBG activity with a decay correction. A mathematical method established in a cross-calibration phantom experiment has standardized cardiac MIBG HMR (sHMR) measured with late images for medium-energy collimator conditions.22 In addition to reducing MIBG HMR variability, this method allows comparison of sHMR data regardless of interfacility data acquisition differences, such as data obtained with low- or medium-energy collimator or from different study populations. The technicians were unaware of the clinical information of the enrolled patients during the nuclear cardiology and echocardiography examinations.
Follow-up ProtocolHF patients in our database have regular outpatient follow-up by cardiologists for ≥1 year if they survive. Fatal cardiac events (CEs), including sudden cardiac death, death due to progressive pump failure, fatal sustained arrhythmia (including cardiac arrest, pulseless electrical activity, sustained ventricular tachyarrhythmias, and ventricular fibrillation) and appropriate implantable cardioverter defibrillator (ICD) therapy for fatal ventricular arrhythmias, were analyzed as the primary endpoint in this study. However, by reviewing medical records and performing outcome analyses, patients’ clinical course and outcome were confirmed retrospectively. Sudden cardiac death was defined as a witnessed cardiac arrest and death within 1 h of the acute onset of symptoms or an unexpected death in a patient who had been well for the previous 24 h.
Statistical AnalysisData are presented as mean±SD. The mean values were compared between 2 groups by unpaired t-tests, and categorical variables were compared by chi-squared tests. Kaplan-Meier analysis was used to generate time-dependent cumulative event-free survival curves using the significant key parameters selected in the multivariate analysis. Log-rank tests were also used to compare curves when needed. The Cox proportional hazard model was used to estimate hazard ratios and 95% confidence intervals (CIs) for the significant variables, using the statistically appropriate number of significant variables identified in the univariable analysis and dependent on the number of CEs. To determine the optimal cutoff values for independent significant parameters, receiver operating characteristic (ROC) analysis was performed. Analyses were performed using SAS for Windows, version 9.4 (SAS Institute, Cary, NC, USA) and Mathematica software for Windows (version 14.0, Wolfram Research Inc., Champaign, IL, USA), with P values <0.05 considered significant.
Following measurement of cognitive impairment by the MMSE, and calculation of eGFR, cardiac sympathetic innervation was quantified as the sHMR. Two representative cases are shown in Figure 2. Despite almost identical levels of cardiac dysfunction, Case 1 with decreased values for MMSE score (21), eGFR (26.4 mL/min/1.73 m2) and late sHMR (1.30) died due to progressive pump failure, while Case 2 with normal values for MMSE score (30), eGFR (67.8 mL/min/1.73 m2) and late sHMR (2.73) did not have a CE during the observation period.
Case 1: 70-year-old woman undergoing cardiac resynchronization therapy had nonischemic cardiomyopathy with reduced LVEF (23.2%), and low MMSE score (21), decreased eGFR (26.4 mL/min/1.73 m2) and decreased standardized late HMR (1.30). She died during the follow-up period due to progressive pump failure. Case 2: 67-year-old male with ischemic cardiomyopathy had reduced LVEF (34.1%) but normal MMSE score (30) and nearly normal eGFR (67.8 mL/min/1.73 m2) and standardized late HMR (2.73). No cardiac events occurred during the follow-up period. eGFR, estimated glomerular filtration rate; HMR, heart-to-mediastinum ratio of MIBG activity; LVEF, left ventricular ejection fraction; MMSE, Mini-Mental State Examination.
The patient population included 334 males (77.1%), the mean age was 68.1±13.1 years and the mean LVEF was 26.6±9.6%. In 84 HF patients (19.3%), CEs recorded during a mean follow-up period of 14.8 months included 12 sudden cardiac deaths, 54 cardiac deaths due to refractory and progressive pump failure, 3 fatal ventricular tachyarrhythmias and 15 appropriate ICD treatments for fatal ventricular tachyarrhythmias.
HF patients with CEs were older than HF patients without CEs, and HF patients with CEs had higher NYHA functional class, lower eGFR (39.0±19.7 vs. 49.1±25.8 mL/min/1.73 m2, P=0.0010), lower Hb levels (11.9±2.1 vs. 12.8±2.3 mg/dL, P=0.0011) and lower MMSE scores (25.7±4.0 vs. 28.8±2.7, P<0.0001) than HF patients without CEs (Table 1). Late sHMR was significantly lower in HF patients with CEs than in HF patients without CEs (1.66±0.39 vs. 2.06±0.52, P<0.0001) (Table 1). Among the significant univariate variables (Table 2), MMSE score, eGFR and late sHMR together with NYHA functional class, history of ventricular tachycardia/ ventricular fibrillation and use of antivasopressin medication were significant independent prognostic factors in the multivariate Cox analysis. Hazard ratios and 95% CIs (P values) for MMSE score, eGFR and late sHMR were 0. 882 and 0.841–0.929 (P<0.0001), 0.980 and 0.969–0.990 (P<0.0001), and 0.493 and 0.305–0.793 (P=0.0036), respectively.
Comparison of Clinical Data, Medications, 2-Dimensional Echocardiographic Parameters and MIBG Scintigraphy Parameters Between Groups With and Without Cardiac Events
| Cardiac events group (n=84) |
Non-cardiac events group (n=349) |
P value | |
|---|---|---|---|
| Age (years) | 71.8±10.1 | 67.3±13.6 | 0.0038 |
| Sex (male/female) | 65/19 | 266/80 | 0.9525 |
| NYHA (I/II/III/IV) | 62/11/5/6 | 336/12/1/0 | <0.0001 |
| Mini-Mental State Examination Score | 25.7±4.0 | 28.8±2.7 | <0.0001 |
| Past history | |||
| Hypertension | 28 (33.3%) | 179 (51.2%) | 0.0028 |
| Diabetes mellitus | 35 (41.5%) | 124 (35.7%) | 0.3254 |
| Dyslipidemia | 40 (47.6%) | 171 (48.9%) | 0.8205 |
| Atrial fibrillation | 29 (34.5%) | 128 (36.6%) | 0.7118 |
| VT/Vf | 32 (38.1%) | 66 (18.9%) | 0.0003 |
| Etiology | |||
| Ischemic | 42 (50%) | 160 (45.8%) | 0.4935 |
| Device implantation | |||
| ICD | 19 (22.6%) | 59 (16.9%) | <0.0001 |
| CRT | 22 (26.1%) | 48 (13.7%) | 0.0081 |
| Laboratory data | |||
| Hemoglobin (g/dL) | 11.9±2.1 | 12.8±2.3 | 0.0011 |
| eGFR (mL/min/1.73 m2) | 39.0±19.7 | 49.1±25.8 | 0.0010 |
| Sodium (mmol/L) | 137.8±3.7 | 138.9±3.3 | 0.0190 |
| BNP (pg/mL) | 874.2±844.2 | 694.4±900.0 | 0.0970 |
| Medications | |||
| ACE-I/ARB | 54 (65.1%) | 222 (63.7%) | 0.2312 |
| β-blocker | 76 (90.4%) | 334 (95.7%) | 0.0844 |
| Loop diuretic | 68 (80.9%) | 260 (74.5%) | 0.2133 |
| Anti-aldosterone agent | 49 (58.3%) | 213 (61.0%) | 0.6998 |
| Antivasopressin agent | 55 (65.4%) | 156 (44.6%) | 0.0005 |
| Amiodarone | 42 (50.0%) | 100 (28.6%) | 0.0003 |
| Statin | 45 (53.5%) | 204 (58.5%) | 0.4177 |
| Echocardiographic parameters | |||
| M-mode | |||
| LVDd (mm) | 57.8±10.5 | 56.5±8.5 | 0.2408 |
| LVDs (mm) | 50.7±12.4 | 48.1±10.1 | 0.0452 |
| IVSTd (mm) | 10.2±2.4 | 10.3±2.3 | 0.7306 |
| PTWd (mm) | 9.3±2.2 | 9.5±2.1 | 0.4795 |
| LAD (mm) | 47.6±8.9 | 46.4±8.6 | 0.2363 |
| Modified Simpson method | |||
| LVEF (%) | 24.9±9.7 | 27.0±9.7 | 0.0770 |
| EDV (mL) | 164.8±85.8 | 145.5±53.0 | 0.0093 |
| ESV (mL) | 127.7±80.4 | 108.3±47.9 | 0.0054 |
| Tissue Doppler method | |||
| Septal E/e′ | 19.2±9.9 | 16.4±7.5 | 0.0139 |
| Findings of MIBG imaging | |||
| Washout ratio (%) | 51.6±17.1 | 40.3±16.8 | <0.0001 |
| Early standardized HMR | 2.05±0.46 | 2.32±0.47 | <0.0001 |
| Late standardized HMR | 1.66±0.39 | 2.06±0.52 | <0.0001 |
| GWTG-Heart Failure Risk Score | 43.0±5.7 | 37.0±5.5 | <0.0001 |
Values are shown as mean±1 standard deviation. ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; BNP, B-type natriuretic peptide; CRT, cardiac resynchronization therapy; EDV, left ventricular end-diastolic volume; ESV, left ventricular end-systolic volume; eGFR, estimated glomerular filtration rate; GWTG-Heart Failure Risk Score, Get With The Guideline-Heart Failure Risk Score; HMR, heart-to-mediastinum ratio; ICD, implantable cardioverter-defibrillator; IVSTd, end-diastolic interventricular septal wall thickness; LAD, left atrial diameter; LV, left ventricular; LVDd, end-systolic left ventricular diameter; LVEF, left ventricular ejection fraction; NS, no significance; NYHA, New York Heart Association Classification; PWTd, end-diastolic posterior wall thickness; VT/Vf, ventricular tachycardia/ventricular fibrillation.
Results of Univariate and Multivariate Analyses
| Univariate analysis | Multivariate Cox proportional-hazards model analysis |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| χ2 | HR | 95% CI | P value | χ2 | HR | 95% CI | P value | |||
| Lower | Upper | Lower | Upper | |||||||
| Age | 30.1 | 1.048 | 1.029 | 1.069 | <0.0001 | 3.88 | 1.022 | 1.000 | 1.048 | 0.0487 |
| NYHA functional class | 26.6 | 2.042 | 1.614 | 2.507 | <0.0001 | 6.92 | 1.441 | 1.105 | 1.817 | 0.0085 |
| MMSE | 45.5 | 0.841 | 0.804 | 0.881 | <0.0001 | 20.9 | 0.882 | 0.841 | 0.929 | <0.0001 |
| VT/Vf | 17.1 | 2.692 | 1.713 | 4.165 | <0.0001 | 15.1 | 2.765 | 1.676 | 4.496 | 0.0001 |
| Hemoglobin | 21.6 | 0.792 | 0.713 | 0.876 | <0.0001 | 3.21 | 0.877 | 0.756 | 1.012 | 0.0733 |
| eGFR | 23.6 | 0.978 | 0.969 | 0.987 | <0.0001 | 16.1 | 0.980 | 0.969 | 0.990 | <0.0001 |
| Antivasopressin agents | 22.7 | 2.939 | 1.875 | 4.708 | <0.0001 | 4.22 | 1.724 | 1.025 | 2.963 | 0.0398 |
| ESV | 4.89 | 1.003 | 1.002 | 1.007 | 0.0269 | 0.44 | 0.998 | 0.944 | 1.002 | 0.5036 |
| Late standardized HMR | 26.4 | 0.318 | 0.205 | 0.492 | <0.0001 | 8.49 | 0.493 | 0.305 | 0.793 | 0.0036 |
CI, confidence interval; HR, hazard ratio; MMSE, Mini-Mental State Examination. Other abbreviations as in Table 1.
The cutoff values for MMSE score, eGFR and late sHMR, which defined the prognosis of patients with HF, were calculated by ROC analysis and were 26, 47.0 mL/min/1.73 m2 and 1.80, respectively (Figure 3A). HF patients with an MMSE score <26, eGFR <47.0 mL/min/1.73 m2 or late sHMR <1.80 had significantly lower CE-free rates than the other HF patients (Figure 3B) (hazard ratio, 95% CIs and P values were 5.801, 3.767–9.053 and P<0.0001, 2.751, 1.723–4.537 and P<0.0001, and 3.132, 2.007–5.002 and P<0.0001, respectively.) Combining 2 of the 3 parameters (MMSE score, eGFR and late sHMR) more clearly discriminated HF patients, allowing 3 risk categories to be identified: low, intermediate and high risk (Figure 4A). A cumulative score of the 3 parameters (1 point given for MMSE score <26, 0 points for MMSE score >26, 1 point for eGFR <47.0 mL/min/1.73 m2, 0 points for eGFR >47.0 mL/min/1.73 m2, 1 point for late sHMR <1.80, and 0 points for late sHMR >1.80) was used to identify the population as Scores 0, 1, 2, and 3 (Table 3)) was shown to increase the risk of CE in HF patients (Figure 4B). (The respective hazard ratios, 12 95% CIs and P values are shown in Figure 4A,4B.)
(A) Using ROC analysis, cutoff values of 26, 47.0 mL/min/1.73 m2 and 1.80 were calculated for MMSE score, eGFR and standardized late HMR, respectively. (B) Kaplan-Meier event-free curves using the cutoff values calculated from ROC analysis clearly distinguished high-risk and low-risk patients. eGFR, estimated glomerular filtration rate; HMR, heart-to-mediastinum ratio of MIBG activity; MMSE, Mini-Mental State Examination; ROC, receiver-operating characteristic.
(A) Kaplan-Meier event-free survival curves clearly classify patients into low, intermediate and high risk groups using 2 of the 3 prognostic variables: MMSE score, eGFR and standardized late HMR. (B) In the Kaplan-Meier event-free survival curves, accumulation of 3 abnormal parameters (i.e., MMSE score <26, eGFR <47.0 mL/min/1.73 m2, and standardized late HMR <1.80) was associated with a synergistic reduction in event-free survival. The number of abnormal parameters is indicated by the score category (0–3). eGFR, estimated glomerular filtration rate; HMR, heart-to-mediastinum ratio of MIBG activity; MMSE, Mini-Mental State Examination.
Comparison of Clinical Data, Medications, 2-Dimensional Echocardiographic Parameters and MIBG Scintigraphy Parameters in Score Groups 0–3
| Group 0 (n=115) |
Group 1 (n=168) |
Group 2 (n=119) |
Group 3 (n=31) |
P value | |
|---|---|---|---|---|---|
| Age (years) | 62.1±14.3 | 68.3±12.3 | 71.7±12.0 | 75.8±5.2 | <0.05*,**,***,†,‡ |
| Sex (male/female) | 95/20 | 127/41 | 90/29 | 22/9 | NS |
| NYHA (I/II/III/IV) | 115/0/0/0 | 159/7/1/1 | 102/11/4/2 | 22/5/1/3 | <0.05*,**,***,†,‡,# |
| Mini-Mental State Examination score | 29.9±0.3 | 29.3±1.8 | 26.5±4.0 | 23.0±3.9 | <0.05*,**,***,†,‡,# |
| Past history | |||||
| Hypertension | 52 (45.2%) | 77 (45.8%) | 66 (55.4%) | 12 (38.7%) | NS |
| Diabetes mellitus | 39 (33.9%) | 67 (39.8%) | 55 (46.2%) | 14 (45.2%) | NS |
| Dyslipidemia | 57 (49.5%) | 77 (45.8%) | 62 (52.1%) | 15 (48.4%) | NS |
| Atrial fibrillation | 41 (35.6%) | 56 (33.3%) | 47 (39.4%) | 13 (41.9%) | NS |
| VT/Vf | 17 (14.7%) | 41 (24.4%) | 26 (21.6%) | 14 (45.1%) | <0.05*,**,***,‡,# |
| Etiology | |||||
| Ischemic | 53 (46.1%) | 71 (42.3%) | 62 (52.1%) | 16 (51.6%) | NS |
| Device implantation | |||||
| ICD | 15 (13.0%) | 40 (23.8%) | 29 (24.3%) | 12 (38.7%) | <0.05*,**,***, |
| CRT | 11 (9.5%) | 32 (19.0%) | 18 (15.1%) | 9 (29.1%) | <0.05*,*** |
| Laboratory data | |||||
| Hemoglobin (g/dL) | 13.9±2.1 | 12.4±2.4 | 12.0±2.1 | 11.3±2.0 | <0.05*,**,***,‡ |
| eGFR (mL/min/1.73 m2) | 70.0±19.7 | 45.5±23.1 | 32.6±16.7 | 26.3±11.1 | <0.05*,**,***,†,‡,# |
| Sodium (mmol/L) | 139.4±3.4 | 138.8±3.4 | 138.1±3.4 | 138.2±3.9 | <0.05** |
| BNP (pg/mL) | 497.2±610.2 | 614.3±642.2 | 865.6±960.1 | 1,690.5±1,672.4 | <0.05**,***,†,‡,# |
| Medications | |||||
| ACE-I/ARB | 101 (87.8%) | 89 (52.9%) | 68 (57.4%) | 14 (45.0%) | <0.05*,**,*** |
| β-blocker | 113 (98.3%) | 158 (94.0%) | 111 (93.2%) | 27 (87.0%) | <0.05 |
| Loop diuretic | 71 (61.7%) | 133 (79.2%) | 96 (80.6%) | 27 (87.0%) | <0.05*,**,***, |
| Anti-aldosterone agent | 80 (69.5%) | 102 (60.7%) | 61 (51.2%) | 18 (58.1%) | <0.05** |
| Antivasopressdin agent | 32 (27.8%) | 87 (51.7%) | 71 (59.6%) | 20 (64.5%) | <0.05*,**,*** |
| Amiodarone | 23 (20.0%) | 53 (31.5%) | 45 (37.8%) | 21 (67.7%) | <0.05*,**,***,‡,# |
| Statin | 70 (60.8%) | 94 (55.9%) | 69 (57.9%) | 16 (51.6%) | NS |
| Echocardiographic parameters | |||||
| M-mode | |||||
| LVDd (mm) | 54.9±8.4 | 57.5±8.4 | 56.8±9.5 | 60.0±8.4 | <0.05*,***,# |
| LVDs (mm) | 46.1±9.8 | 48.8±9.9 | 49.0±11.1 | 54.7±12.4 | <0.05*,**,***,‡,# |
| IVSTd (mm) | 10.4±2.3 | 10.1±2.1 | 10.7±2.5 | 9.6±1.8 | <0.05# |
| PTWd (mm) | 9.6±2.1 | 9.3±1.9 | 9.6±2.3 | 8.6±2.5 | <0.05# |
| LAD (mm) | 43.4±8.7 | 47.2±7.8 | 48.3±8.7 | 48.9±9.6 | <0.05*,**,*** |
| Modified Simpson method | |||||
| LVEF (%) | 28.3±10.1 | 26.5±9.5 | 26.1±9.5 | 23.2±10.2 | <0.05*** |
| EDV (mL) | 138.7±50.3 | 152.5±54.9 | 147.8±70.4 | 169.9±81.6 | <0.05*,*** |
| ESV (mL) | 102.2±45.6 | 114.3±49.4 | 112.4±64.3 | 134.1±77.8 | <0.05*,*** |
| Tissue Doppler method | |||||
| Septal E/e′ | 13.5±5.4 | 17.0±7.8 | 18.6±8.3 | 23.4±11.0 | <0.05*,**,***,‡,# |
| Findings of MIBG imaging | |||||
| Washout ratio (%) | 32.8±12.9 | 41.6±16.5 | 49.9±18.1 | 55.1±13.4 | <0.05*,**,***,†,‡,# |
| Early standardized HMR | 2.52±0.36 | 2.32±0.46 | 2.04±0.49 | 1.93±0.32 | <0.05*,**,***,†,‡,# |
| Late standardized HMR | 2.36±0.43 | 2.02±0.47 | 1.69±0.45 | 1.50±0.18 | <0.05*,**,***,†,‡,# |
| GWTG-Heart Failure Risk Score | 34.3±4.9 | 38.3±5.2 | 41.1±6.1 | 44.2±4.2 | <0.05*,**,***,†,‡,# |
Values are shown as mean±1 standard deviation. *Group 0 vs. Group 1, P<0.05. **Group 0 vs. Group 2, P<0.05. ***Group 0 vs. Group 3, P<0.05. †Group 1 vs. Group 2, P<0.05. ‡Group1 vs. Group 3, P<0.05. #Group 2 vs. Group 3, P<0.05. Abbreviations as in Table 1.
The Supplementary Figure shows the incremental prognostic value to the Get With The Guidelines-Heart Failure (GWTG Heart Failure) Risk Score, which has been reported as a useful risk score for chronic HF in Japan as well as in Europe and the USA.23,24
Our study results showed that the MMSE score was an independent predictor of prognosis, as were the late sHMR and eGFR values, which have been conventionally considered prognostic parameters for HF patients. Furthermore, the results suggested that cognitive assessment may significantly and additively stratify risk in HF patients when combined with assessment of cardiac sympathetic nerve function and renal dysfunction.
Link Between HF and Cognitive FunctionHF is often associated with complications such as ischemic heart disease, hypertension and diabetes mellitus, which can also be risks for cognitive impairment.1
An important pathway from HF to cognitive impairment is hemodynamic stress due to HF. It has been reported that reduced blood flow to the ventral hippocampus in the brain, which is responsible for memory, is associated with the severity of cognitive impairment.3 Changes in neuroendocrine hormones, such as elevated levels of cortisol and catecholamines in the blood and activation of the renin-angiotensin (RAS) system associated with HF, also promote cognitive dysfunction.25
HF patients with persistent cardiac dysfunction for >1 year show progressive cognitive impairment, and it has been reported that cognitive dysfunction also progresses in HF patients who have a recovery of cardiac function.25 Therefore, early and appropriate treatment of HF is also important to prevent the progression of cognitive impairment. In addition, halting cognitive impairment is also thought to reduce the frailty cycle in HF patients.
Link Between CKD and Cognitive FunctionThe incidence of CKD also increases with aging, and CKD itself is a risk factor for dementia. CKD is characterized by an increased risk of developing cerebrovascular dementia, characterized in particular by impaired executive function. In several studies, there was a close relationship between cerebral grey matter atrophy and reduced executive function in CKD patients, suggesting that prevention of brain atrophy is important for the prevention of cognitive decline.
The causes of cognitive dysfunction in patients with CKD include cerebrovascular disease, hypertension, diabetes mellitus, dyslipidemia, myocardial infarction, atrial fibrillation and smoking, while anemia, hyperparathyroidism, drugs, sleep disorders and depression are considered to be involved as nonvascular risk factors.14
Oxidative stress, which induces brain neuronal damage and impaired learning, and the RAS system, which induces oxidative stress in the brain, are considered as the mechanisms of cognitive dysfunction in CKD patients.15
Link Between Cardiac Sympathetic Function and Cognitive FunctionStudies of brain function have shown that brain activity in the anterior cingulate cortex and anterior insula is associated with sympathetic activity, with the former in particular also being involved in the control of cognitive task performance.26
Cognitive decline in dementia patients and the elderly is thought to involve a reduction or dysfunction of basal forebrain cholinergic neurons. Disease-specific mechanisms of cognitive decline in HF patients are thought to involve hypoperfusion and hypoxic cerebral blood flow.26
Systemic inflammation associated with HF induces elevated levels of cortisol and catecholamines in the blood and activation of the RAS system, which promote cognitive dysfunction.25
The results of the present study suggest that assessment of cardiac sympathetic nerve function using MIBG cardiac scintigraphy and measurement of cognitive function using a MMSE score are independent prognostic factors in HF patients and could additionally stratify risk.
Study LimitationsThis study was an observational and non-interventional cohort study of patients with irreversible systolic HF enrolled at a single institution.
A large multicenter intervention study based on the results of this study is expected to contribute to the identification of appropriate indications and timing of therapeutic interventions for improving cardiac output in order to increase cerebral blood flow (e.g., introduction of optimal oral medication, improvement of anemia, implantation of a biventricular pacemaker, heart transplantation) in HF patients with coexisting cognitive impairment who are at increased risk of death. In addition, such a study is expected to help with the indications and timing for introducing cardiac rehabilitation, which is believed to prevent cognitive decline.
Other tests to determine cognitive function such as Hasegawa’s Dementia Scale-Revised,27 Mini-Cog,28 Montreal Cognitive Assessment,29 Dementia Assessment Sheet for Community-based Integrated Care System-21 items,30 and the ABC dementia scale31 were not performed in this study.
Although a differential diagnosis of dementia with Lewy bodies was not performed, the absence of clinical findings such as marked visual hallucinations, rapid eye movement sleep behavior disorder, and parkinsonian symptoms, as well as the absence of decreased early sHMR on MIBG scintigraphy, indicated a low possibility of its involvement and development in the enrolled patient population.32,33 Furthermore, none of the patients enrolled in the study exhibited behavioral or psychological symptoms, such as verbal abuse or assault, or were bedridden, which makes it unlikely that Alzheimer’s disease was involved or had developed.34 However, the maximum observation period in the present study was 5 years. Longer observation periods may be needed to diagnose dementia and to assess dementia prognosis.35–37 A longer follow-up could have helped distinguish between cognitive dysfunction due to reduced cardiac output or organ dysfunction resulting from HF and cognitive dysfunction resulting from cranial nerve damage.
The addition of cognitive assessment to the evaluation of cardiac sympathetic function and renal function in patients with HF could be synergically useful for risk stratification and predicting prognosis.
The authors sincerely thank the staff of the Nuclear Medicine Laboratory, Teine-Keijinkai Hospital (Sapporo), Hokkaido, Japan for their clinical services and technical assistance.
No specific funding for this study.
The authors declare no conflict of interest.
The study was approved by the Ethics Committee of Teine Keiinkai Hospital (approval no. 2-023090-00).
M.N. and T.D. managed the project. R.H., T.T., K.K., H.I. and D.N. recruited the patients and collected the data. T.D., S.Y. and A.H. analyzed the data. T.N. directed the research. M.N., T.D. and T.N. wrote the manuscript. All authors read and discussed the manuscript.
The deidentified participant data will be shared on a request basis. Please directly contact the corresponding author to request data sharing. All data sets used will be available, including the study protocol. Data will be shared as soon as the IRB at Teine Keijinkai Hospital approves it, and will be available until the end of March, 2028. The data will be shared with anyone who wants to access it. Any analyses of the data will be approved and data will be shared as anonymized Excel file via email.
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-24-0124
