Elevated Serum Non-HDL (High-Density Lipoprotein) Cholesterol and Triglyceride Levels as Residual Risks for Myocardial Infarction Recurrence Under Statin Treatment.

Objective- Secondary prevention for recurrent myocardial infarction (MI) is one of the most important therapeutic goals in patients with old MI (OMI). Although statins are widely used for this purpose, there remains considerable residual risk even after LDL (low-density lipoprotein cholesterol) is well controlled by statins. Approach and Results- We examined clinical impacts of nHDL (nonhigh-density lipoprotein cholesterol) and its major components triglyceride and LDL as residual risks for acute MI recurrence, using the database of our CHART (Chronic Heart Failure Analysis and Registry in the Tohoku District)-2 Study, the largest-scale cohort study of cardiovascular patients in Japan. We enrolled 1843 consecutive old MI patients treated with statins (mean age 67.3 years, male 19.2%) in the CHART-2 Study. The incidence of recurrent acute MI during the median 8.6-year follow-up was compared among the groups divided by the levels of nHDL (<100, 100-129, and ≥130 mg/dL), LDL (<70, 70-99, and ≥100 mg/dL), triglyceride (<84, 84-149, and ≥150 mg/dL), and combination of LDL and triglyceride. Kaplan-Meier curves and multiple Cox proportional hazards models showed that higher levels of nHDL, but not LDL or triglyceride alone, were associated with higher incidence of recurrent acute MI. Furthermore, higher triglyceride levels were associated with higher incidence of recurrent MI in patients with LDL <100 mg/dL but not in those with LDL ≥100 mg/dL. Conclusions- These results indicate that management of residual risks for acute MI recurrence should include nHDL management considering both LDL and triglyceride in old MI patients under statin treatment. Clinical Trial Registration- URL: http://www.clinicaltrials.gov . Unique identifier: NCT00418041.

Comparable prognostic impact of BNP levels among HFpEF, Borderline HFpEF and HFrEF: a report from the CHART-2 Study.

We aimed to compare the usefulness of plasma levels of B-type natriuretic peptide (BNP) for long-term risk stratification among patients with heart failure (HF) with preserved left ventricular ejection fraction (LVEF) (HFpEF), borderline HFpEF, and HF with reduced LVEF (HFrEF) in the same HF cohort. In the CHART-2 Study (N = 10,219), we categorized 4301 consecutive Stage C/D HF patients (mean age 68.7 years, female 32.4%) into 3 groups: HFpEF (LVEF ≥ 50%, N = 2893), borderline HFpEF (LVEF 40-50%, N = 666), and HFrEF (LVEF ≤ 40%, N = 742). During the median 6.3-year follow-up, all-cause deaths occurred in 887 HFpEF, 330 borderline HFpEF, and 330 HFrEF patients. Although median BNP levels increased from HFpEF, borderline HFpEF to HFrEF (85.3, 126 and 208 pg/ml, respectively, P < 0.001), the relationship between log2 BNP levels and the mortality risk was comparable among the 3 groups. As compared with patients with BNP < 30 pg/ml, those with 30-99, 100-299 and ≥ 300 pg/ml had comparably increasing mortality risk among the 3 groups (hazard ratio 2.5, 4.7 and 7.8 in HFpEF, 2.1, 4.2 and 7.0 in borderline HFpEF, and 3.0, 4.7 and 9.5 in HFrEF, respectively, all P < 0.001). BNP levels have comparable prognostic impact among HFpEF, borderline HFpEF, and HFrEF patients.

Prognostic impacts of serum uric acid levels in patients with chronic heart failure: insights from the CHART-2 study.

AIMS: Prognostic impacts of serum uric acid (UA) levels in patients with chronic heart failure (CHF) remain inconclusive, especially for the whole range of serum UA levels. METHODS AND RESULTS: In the Chronic Heart Failure Registry and Analysis in the Tohoku District-2 (CHART-2) study, we enrolled 4652 consecutive patients with CHF and classified them into four groups based on baseline serum UA levels by the Classification and Regression Tree: G1 (<3.8 mg/dL, N = 313), G2 (3.8-7.1 mg/dL, N = 3070), G3 (7.2-9.2 mg/dL, N = 1018), and G4 (>9.2 mg/dL, N = 251). Mean age was 71 ± 12, 69 ± 12, 68 ± 13, and 69 ± 15 years in G1, G2, G3, and G4, respectively (P < 0.001). During the median follow-up of 6.3 years, in G1, G2, G3, and G4, 111 (35%), 905 (29%), 370 (36%), and 139 (55%) patients died and 79 (25%), 729 (24%), 300 (29%), and 115 (46%) experienced heart failure hospitalization, respectively (both P < 0.001). G1 was characterized by a significantly high prevalence of women as compared with G2, G3, and G4 (59%, 32%, 24%, and 23%, respectively). Serum creatinine levels (0.8 ± 0.4, 0.9 ± 0.4, 1.2 ± 0.6, and 1.4 ± 0.8 mg/dL, respectively), prevalence of atrial fibrillation (34%, 39%, 45%, and 50%, respectively), and diuretics use (36%, 45%, 67%, and 89%, respectively) increased from G1, G2, G3 to G4 (all P < 0.001), while left ventricular ejection fraction decreased from G1, G2, G3 to G4 (59 ± 15, 58 ± 15, 54 ± 15, and 52 ± 17%, respectively, P < 0.001). Multivariable Cox proportional hazards models showed that, as compared with G2, both G1 and G4 had increased incidence of all-cause death [adjusted hazard ratio (aHR) 1.34, 95% confidence interval (CI) 1.08-1.67, P = 0.009; aHR 1.28, 95% CI 1.02-1.61, P = 0.037, respectively] and heart failure admission (aHR 1.39, 95% CI 1.09-1.78, P = 0.008 and aHR 1.35, 95% CI, 1.06-1.71, P = 0.014, respectively). This U-shaped relationship was evident in the elderly patients. Furthermore, abnormal transitions to either higher or lower levels of serum UA from G2 were associated with increased mortality (aHR 1.29, 95% CI 1.06-1.57, P = 0.012; aHR 1.57, 95% CI 1.12-2.20, P = 0.009). CONCLUSIONS: These results demonstrate that serum UA levels have the U-shaped prognostic effects and abnormal transitions to either higher or lower levels are associated with poor prognosis in the elderly patients with CHF.

Temporal changes in left ventricular ejection fraction and their prognostic impacts in patients with Stage B heart failure.

BACKGROUND: We have recently demonstrated that left ventricular ejection fraction (LVEF) dynamically changes over time with prognostic impacts in Stage C/D patients, namely, those who have a current or past history of heart failure (HF). However, it is unknown whether this is also the case in asymptomatic Stage B patients, namely, those who have a risk of HF, but do not have a history of HF. METHODS: In our CHART-2 Study (N = 10,219), we enrolled 4005 Stage B patients and divided them into 3 groups by LVEF; preserved EF (pEF, LVEF ≥50%, N = 3526), mid-range EF (mrEF, LVEF 41-49%, N = 302), and reduced EF (rEF, LVEF ≤40%, N = 177). We examined the prognostic impacts of LVEF transitions among the 3 groups in comparison with 4477 patients with Stage C/D HF. RESULTS: Stage B were characterized by less severe clinical status and better prognosis compared with Stage C/D. Stage B in mrEF and rEF at baseline dynamically transitioned to other groups at 1-year, whereas those in pEF unchanged; at 1-year, mrEF transitioned to pEF/rEF by 50/16%, and rEF transitioned to pEF/mrEF by 25/31%, respectively, whereas pEF transitioned to mrEF/rEF by only 3.6/0.7%, respectively, which were consistent with findings in findings with Stage C/D. Although LVEF decrease was directly associated with all-cause mortality in both the Stage B and Stage C/D with pEF, factors related to LVEF changes were different between the 2 groups. CONCLUSIONS: In Stage B, LVEF dynamically changes with prognostic impacts as in Stage C/D, whereas different determination factors may be involved in the 2 stages. CLINICAL TRIAL REGISTRATION: Chronic Heart Failure Analysis and Registry in the Tohoku District (CHART)-2 (NCT00418041).

Prognostic impacts of dynamic cardiac structural changes in heart failure patients with preserved left ventricular ejection fraction.

AIMS: We aimed to examine temporal changes in left ventricular (LV) structures and their prognostic impacts in patients with heart failure (HF) and preserved ejection fraction (HFpEF). METHODS AND RESULTS: In the Chronic Heart Failure Analysis and Registry in the Tohoku District-2 (CHART-2) study (n = 10 219), we divided 2698 consecutive HFpEF patients (68.9 ± 12.2 years, 32.1% female) into three groups by LV hypertrophy (LVH) and enlargement (LVE) at baseline: (-)LVH/(-)LVE (n = 989), (+)LVH/(-)LVE (n = 1448), and (+)LVH/(+)LVE (n = 261). We examined temporal changes in LV structures and their prognostic impacts during a median 8.7-year follow-up. From (-)LVH/(-)LVE, (+)LVH/(-)LVE to (+)LVH/(+)LVE at baseline, the incidence of the primary outcome, a composite of cardiovascular death or HF admission, significantly increased. Among 1808 patients who underwent echocardiography at both baseline and 1 year, we noted substantial group transitions from baseline to 1 year; the transition rates from (-)LVH/(-)LVE to (+)LVH/(-)LVE, from (+)LVH/(-)LVE to (-)LVH/(-)LVE, from (+)LVH/(-)LVE to (+)LVH/(+)LVE, and from (+)LVH/(+)LVE to (+)LVH/(-)LVE were 27% (182/671), 22% (213/967), 6% (59/967), and 26% (44/170), respectively. In the univariable Cox proportional hazard model, patients who transitioned from (+)LVH/(-)LVE to (+)LVH/(+)LVE or remained in (+)LVH/(+)LVE had the worst subsequent prognosis [hazard ratio (HR) 4.65, 95% confidence interval (CI) 3.09-6.99, P < 0.001; HR 4.01, 95% CI 2.85-5.65, P < 0.001, respectively], as compared with those who remained in (-)LVH/(-)LVE. These results were unchanged after adjustment for the covariates including baseline LV ejection fraction (LVEF) and 1-year LVEF change. CONCLUSION: In HFpEF patients, LV structures dynamically change over time with significant prognostic impacts, where patients who develop LVE with LVH have the worst prognosis.

Clinical benefits and risks of antithrombotic therapy in patients with atrial fibrillation with comorbidities - A report from the CHART-2 Study.

BACKGROUND: The benefits of antithrombotic therapy (ATT) for atrial fibrillation (AF) are occasionally offset by major bleeding complications. However, the clinical benefits and risks of ATT in AF patients, with special references to comorbidities, such as heart failure (HF), coronary artery disease (CAD), and the patterns of AF, remain to be fully elucidated. METHODS: A total of 3221 consecutive AF patients from our Chronic Heart Failure Analysis and Registry in the Tohoku District-2 (CHART-2) Study (N = 10,219) were divided into 4 groups based on ATT at enrollment; no-ATT, anticoagulant alone, antiplatelet alone, and both of them (AC&AP). Then, efficacy and safety outcomes including thromboembolic events, major bleeding, and mortality were evaluated among the 4 groups. RESULTS: Anticoagulant monotherapy was associated with reduced risk of ischemic stroke in patients with but not in those without HF, CAD, or non-paroxysmal AF. Although there was no significant difference in major bleeding among the 4 groups, a composite of thromboembolism and major bleeding occurred more frequently in the AC&AP group, even in combination with anticoagulants and single antiplatelet therapy, indicating that the combination therapy is more harmful than anticoagulant monotherapy for AF patients, especially for those with HF or CAD. Lastly, no-ATT group was associated with worse prognosis compared with other 3 groups. CONCLUSIONS: These results indicate that ATT is beneficial for AF patients particularly for those with HF, CAD, or non-paroxysmal AF and that among ATT, anticoagulant monotherapy may be most profitable for both clinical benefits and risks for AF patients.

Risk of de-novo heart failure and competing risk in asymptomatic patients with structural heart diseases.

AIMS: Asymptomatic patients with structural heart diseases are classified as a population at high risk for heart failure (HF) in Stage B. However, limited data are available regarding incidence and related factors of de-novo HF (DNHF) considering competing risk in this population. METHODS AND RESULTS: In 3362 Stage B patients (mean age 68 yrs, male 76%) from the CHART-2 Study (N = 10,219), we examined incidence of death and DNHF, defined as the first episode of either HF hospitalization or HF death, and factors related to DNHF. RESULTS: During the median 6.0-year follow-up, 627 deaths (31/1000 person-years) and 293 DNHF (15/1000 person-years) occurred. Among the 627 deaths, 212 (34%) and 325 (52%) were specified as cardiovascular and non-cardiovascular deaths, respectively. During the follow-up of 271 DNHF hospitalizations, we observed 124 deaths, including 65 (52%) cardiovascular and 47 (40%) non-cardiovascular deaths. The competing risk model showed that age, diabetes mellitus, stroke, atrial fibrillation, diastolic blood pressure, hemoglobin levels, estimated glomerular filtration ratio and left ventricular ejection fraction was significantly associated with DNHF. Bayesian structural equation modeling showed that many of these cardiac and non-cardiac variables contribute to DNHF by affecting each other, while diabetes mellitus was independently associated with DNHF. CONCLUSIONS: Stage B patients had a high incidence of DNHF as well as that of death due to both cardiovascular and non-cardiovascular causes. Thus, management of Stage B patients should include multidisciplinary approaches considering both cardiac and non-cardiac factors, in order to prevent DNHF as well as non-HF death as a competing risk. TRIAL REGISTRATION: clinicaltrials.gov identifier: NCT00418041.

Clinical characteristics and prognostic factors in elderly patients with chronic heart failure -A report from the CHART-2 study.

BACKGROUND: Since most of the randomized clinical trials for heart failure (HF) were designed to exclude elderly patients, limited data are available on their clinical characteristics, prognosis, and prognostic factors. METHODS: We compared clinical characteristics, prognosis, and prognostic factors among Stage C/D HF patients in our CHART-2 Study (N = 4876, mean 69 years, women 32%, 6.3-year follow-up) by age (G1, ≤64 years, N = 1521; G2, 65-74 years, N = 1510; and G3, ≥75 years, N = 1845). RESULTS: From G1 to G3, the prevalence of women, left ventricular ejection fraction (LVEF) and plasma levels of B-type natriuretic peptide (BNP) increased (all P < 0.001). Similarly, 5-year mortality increased (9.9, 17.3 to 39.9%, P < 0.001) along with a decrease in proportion of cardiovascular death and an increase in non-cardiovascular death in both sexes. While all-cause and cardiovascular mortality was comparable between the sexes, women had significantly lower incidence of non-cardiovascular death than men in G2 and G3, which was attributable to the higher incidence of cancer death and pneumonia death in men than in women. Although NYHA functional class III-IV, chronic kidney disease, cancer, LVEF, and BNP had significant impacts on all-cause death in all groups, their impacts were less evident in G3 as compared with G1. CONCLUSIONS: The elderly HF patients, as compared with younger HF patients, were characterized by more severe clinical background, increased proportion of non-cardiovascular death and worse prognosis with different impacts of prognostic factors across the age groups.

Association between temporal changes in C-reactive protein levels and prognosis in patients with previous myocardial infarction - A report from the CHART-2 Study.

BACKGROUND: Several studies have reported that C-reactive protein (CRP), an inflammatory biomarker, predicts cardiovascular events independently of low-density lipoprotein cholesterol levels. However, no study examined whether temporal changes in CRP levels are associated with clinical events in patients with previous myocardial infarction (MI). METHODS AND RESULTS: We examined 2184 consecutive patients with previous MI and CRP data at baseline in the Chronic Heart Failure Registry and Analysis in the Tohoku district-2 (CHART-2) Study. During the median 6.4 years follow-up, 592 all-cause, 245 cardiovascular, and 273 non-cardiovascular deaths occurred. Patients with CRP ≥ 2.0 mg/L at baseline had significantly increased incidence of all-cause (hazard ratio (HR) 1.68, P < 0.001) and non-cardiovascular death (HR 1.86, P < 0.001), compared with those with CRP < 2.0 mg/L. Temporal changes in CRP levels were associated with prognosis; among patients with CRP ≥ 2.0 mg/L at baseline, those with CRP ≥ 2.0 mg/L at 1-year had significantly increased incidence of all-cause (HR 2.12, P < 0.001), cardiovascular (HR 2.31, P < 0.001), and non-cardiovascular death (HR 2.29, P < 0.001). Among patients with CRP < 2.0 mg/L at baseline, those with CRP ≥ 2.0 mg/L at 1-year had significantly increased incidence of all-cause (HR 1.76, P < 0.001) and cardiovascular death (HR 2.10, P = 0.001). These results remained significant after adjusted with the inverse probability of treatment weighted models using propensity sore. Furthermore, as compared with patients with CRP < 2.0 mg/L at both baseline and 1-year, those with CRP ≥ 2.0 mg/L at both baseline and 1-year had increased incidence of all-cause, cardiovascular, and non-cardiovascular death. CONCLUSIONS: These results provide the evidence that temporal increases in CRP levels are associated with increased clinical events in patients with previous MI.

Conversion formula from B-type natriuretic peptide to N-terminal proBNP values in patients with cardiovascular diseases.

BACKGROUND: Although B-type natriuretic peptide (BNP) and N-terminal prohormone B-type natriuretic peptide (NT-proBNP) are released in equimolar proportions, their values differ depending on clinical conditions. A useful conversion formula between BNP and NT-proBNP remains to be developed for the clinical use. AIM: To develop a conversion formula from BNP to NT-proBNP. METHODS: In the derivation cohort, 923 patients with chronic heart failure, in whom both BNP and NT-proBNP values were available, were enrolled from our SUPPORT (Supplemental Benefit of ARB in Hypertensive Patients with Stable Heart Failure using Olmesartan) trial. The validation cohort included 1154 consecutive patients with or at risk of cardiovascular diseases, in whom both BNP and NT-proBNP values were measured simultaneously at Tohoku University Hospital. We regressed log10 NT-proBNP onto log10 BNP and factors influencing BNP and NT-proBNP values. RESULTS: We adopted the model with the smallest Akaike information criterion consisting of log10 BNP, age, sex, BMI, creatinine clearance (CCr), hemoglobin, and atrial fibrillation (AF). As compared with the previously reported conversion formulas, the present conversion formula utilized non-linear transformation by spline function, and exhibited the strongest correlation between actual and calculated values of NT-proBNP (r = 0.928). The root mean squared error (RMSE) of the present conversion formula was smallest compared with the previously reported conversion formulas, indicating that this formula most effectively converts BNP values to NT-proBNP values. CONCLUSIONS: We have developed a useful conversion formula from BNP to NT-proBNP values, using age, sex, BMI, CCr, hemoglobin, and AF, which could be widely used in daily clinical practice.

Gender differences in prognostic relevance of self-care behaviors on mortality and hospitalization in patients with heart failure - A report from the CHART-2 Study.

BACKGROUND: Self-care behaviors (ScB) are associated with symptoms and outcomes in patients with heart failure (HF). However, little is known about gender differences in the prognostic relevance of ScB in HF patients. METHODS: We examined gender differences in ScB of HF patients regarding its prognostic associations with mortality and HF hospitalization with a reference to ScB dimensions. The European Heart Failure Self-Care Behavior Scale (EHFScBS) was used to evaluate ScB in 2233 patients with Stage C/D HF in the CHART-2 Study. RESULTS: Male patients (n=1583) were younger (71 vs. 73 yrs) and had lower ScB (median 33 vs. 31) (all p<0.001) than females (n=650). During the median follow-up of 2.57 years, patients with high ScB (score 12-32, n=1090), as compared with low ScB patients (score 33-60, n=1143), had significantly increased all-cause mortality in males [adjusted hazard ratio (aHR) 1.44, p=0.02] but not in females (aHR 0.80, p=0.40) (p for interaction 0.02), while ScB was not significantly associated with incidence of HF hospitalization in both genders. Among the 3 dimensions in EHFScBS, complying with regimen was associated with decreased mortality in females, but not in males (p for interaction 0.003), while asking for help was related with increased incidence of HF hospitalization in males (aHR 1.34, p=0.072) but not in females (aHR 0.98, p=0.931) (p for interaction 0.048). CONCLUSIONS: There were gender differences in the prognostic relevance of self-care with mortality and incidence of HF hospitalization, suggesting that self-care should be implemented considering gender differences to improve prognosis.

Prognostic impact of residual stenosis after percutaneous coronary intervention in patients with ischemic heart failure - A report from the CHART-2 study.

BACKGROUND: Complete revascularization with PCI is not always achieved in patients with ischemic HF. Therefore, this study aimed to elucidate the prognostic impact of residual coronary stenosis (RS) after percutaneous coronary intervention (PCI) in patients with ischemic heart failure (HF). METHODS: We analyzed a total of 1307 patients with symptomatic HF and a history of PCI registered in our Chronic Heart Failure Analysis and Registry in the Tohoku District-2 (CHART-2) Study. RS that was defined as the presence of ≥70% luminal stenosis in major coronary arteries at the last coronary angiography. RESULTS: Among the study population, 851 patients (65.1%) had RS. During a median follow-up period of 3.2 years, patients with RS had higher all-cause mortality than those without it even after propensity score matching (21.9 vs. 11.6%, log-rank P = 0.027). Multivariable Cox hazard analysis also showed the negative impact of RS on all-cause death in ischemic HF patients [hazard ratio (HR):1.62, 95% confidence interval (CI): 1.07-2.46, P = 0.024]. Importantly, when divided all subjects into three subgroups by left ventricular ejection fraction (LVEF) [LVEF < 40% (HFrEF), LVEF 40-49% (HFmrEF), and LVEF ≥ 50% (HFpEF)], inverse probability of treatment weighted method provided a similar result that RS after PCI was an independent risk factor for death in the HFpEF [HR(95%CI); 1.94(1.22-3.09), P < 0.01] and HFmrEF [4.47(1.13-14.98), P < 0.01] groups, but not in the HFrEF group [1.20(0.59-2.43), P = 0.62]. CONCLUSIONS: These results indicate that RS after PCI could aggravate long-term prognosis of ischemic HF patients with moderate- to well-preserved EF, but not those with reduced EF.

Increased risk of cancer death in patients with chronic heart failure with a special reference to inflammation-A report from the CHART-2 Study.

BACKGROUND: Although several factors, including heart failure (HF) and inflammation, are known to increase the incidence of cancer, it remains unknown whether HF may increase cancer mortality, especially with a reference to inflammation. METHODS AND RESULTS: We examined 8843 consecutive cardiovascular patients without a prior history of cancer in our CHART-2 Study (mean 68 yrs., female 30.9%). As compared with patients without HF (Stage A/B, N = 4622), those with HF (Stage C/D, N = 4221) were characterized by higher prevalence of diabetes, previous myocardial infarction, atrial fibrillation, and stroke. During the median 6.5-year follow-up (52,675 person-years), 282 cancer deaths occurred. HF patients had significantly higher cancer mortality than those without HF in both the overall (3.7 vs, 2.8%, hazard ratio (HR) 1.42, 95% confidence interval (CI) 1.12-1.79, P = 0.004) and the propensity score-matched cohorts (HR 1.46, 95%CI 1.10-1.93, P = 0.008), which was confirmed in the competing risk models. The multivariable Cox proportional hazard model in the matched cohort showed that HF was associated with increased cancer mortality in patients with C-reactive protein (CRP) ≥ 2.0 mg/L (HR 1.87, 95%CI 1.18-2.96, P = 0.008) at baseline, but not in those with CRP < 2.0 mg/L (HR 0.89, 95%CI 0.54-1.45, P = 0.64) (P for interaction = 0.03). Furthermore, temporal changes in CRP levels were associated with cancer death in the overall cohort; HF patients with CRP ≥ 2.0 mg/L at both baseline and 1-year had significantly increased cancer death, while those with CRP ≥ 2.0 mg/L at baseline and < 2.0 mg/L at 1-year not. CONCLUSIONS: These results provide the first evidence that HF is associated with increased cancer death, especially when associated with prolonged inflammation.

The 3A3B score: The simple risk score for heart failure with preserved ejection fraction - A report from the CHART-2 Study.

BACKGROUND: Few simple risk models, without echocardiography have been developed for patients with heart failure (HF) and preserved left ventricular ejection fraction (LVEF) (HFpEF). METHODS: To develop a risk score to predict all-cause death for HFpEF patients, we examined 1277 HF patients with LVEF ≥50% and BNP ≥100 pg/ml in the CHART-2 Study, a large-scale prospective cohort study for HF in Japan. We selected the optimal subset of covariates for the score with Cox proportional hazard models and random survival forests (RSF). RESULTS: During the median 5.7-year follow-up, 576 deaths occurred. Cox models and RSF analyses consistently indicated age ≥75 years, albumin <3.7 g/dl, anemia, BMI <22 kg/m2, BNP ≥300 pg/ml (or NT-proBNP ≥1400 pg/ml), and BUN ≥25 mg/dl, as the important 6 prognostic variables. Incorporating these 6 variables, we developed a scoring system (3A3B score, with 2 points given to age ≥75 years and 1 point to the others based on the hazard ratios. The discrimination ability of the risk score was excellent (c-index 0.708). Regarding model goodness-of-fit, the overall gradient in 5-year risk was well captured by the score. The predictive accuracy of the 3A3B score was confirmed in the external validation cohorts from the TOPCAT trial (N = 835, c-index 0.652) and the ASIAN-HF registry (N = 170, c-index 0.741). CONCLUSIONS: We developed a simple risk score to predict long-term prognosis of HFpEF patients. The 3A3B score, comprising 6 commonly available parameters in daily practice, has potential utility in the risk stratification and management of HFpEF patients.