Health Status And Its Association With Clinical Outcomes In Patients With Acute Heart Failure

Background Heart failure(HF)negatively impacts patients’ health status,including physical limitation,functional status,and quality of life.Health status in patients with HF is as important as clinical outcome.Kansas City Cardiomyopathy Questionnaire(KCCQ),one of the most widely used HF-specific tool,plays an important role in trials,clinical care and health monitoring.Current data mainly focus on the KCCQ in outpatients with HF,but data on KCCQ score and the influence factors in HF inpatients are scarce.Furthermore,the temporal changes of short-and long-term health status in heart failure were debated.On the other hand,most studies reported the prognostic value of KCCQ score in chronic HF,but little is known on its predictive value in acute HF.Additionally,data on the predictive effects of KCCQ in chronic HF were mainly from western countries,data from China was sparse.The changes of serial KCCQ assessment were also an important prognosticator in HF,but the linear association with clinical outcomes remains uncertain.Objective(1)To evaluate the health status of patients with HF measured by shortened KCCQ and the influence factors in patients hospitalized for HF,and further evaluate the temporal changes of health status in HF after 1 year of discharge.(2)Evaluate the prognostic value of KCCQ-12 in acute HF and in chronic HF separately,as well as the prognostic value of changes in KCCQ-12 in HF.Methods The China Patient-centered Evaluative Assessment of Cardiac Events prospective heart failure study(China PEACE 5p-HF study)enrolled 4907 patients hospitalized for HF from 52 hospitals located in 20 provinces.Patients’ baseline characteristics were collected through medical records,patient interviews,local examinations and lab tests.Patients completed KCCQ-12 and EQ-5D within 48 hours of hospital admission.At 1,6,and 12 months after discharge,local investigators interviewed patients and sought information on health status(KCCQ-12,EQ-5D)and pre-defined clinical outcomes(all-cause death,hospitalizations,and the composite of cardiovascular death or HF hospitalization).KCCQ-12 contained four domains,physical limitation,symptomatic frequency,quality of life and social limitation;the overall score of KCCQ-12 ranged from 0 to 100,and the higher score represented better health.(1)We use multivariable logistic and linear regression to analyze the influence factors of KCCQ-12 in acute HE.Repeated measures ANO VA was performed to analyze the temporal changes of KCCQ-12 and EQ-5D,and further evaluate the changes of 4 domains.The mixed effects model was performed to analyze the risk factors of longitudinal KCCQ changes.(2)Restricted cubic splines were used to analyze the linear association between KCCQ score and clinical outcomes.Multivariable Cox model was performed to evaluate the associations between KCCQ-12 score(during hospitalization,1 month post discharge,and the changes of two)and clinical outcome.Results(1)Totally 4869 patients completed baseline questionnaires.The median age(interquartile range,IQR)was 67(57-75),and 37.6%were women.The median(IQR)KCCQ-12 was 43.1(26.7-60.6).Women,the elderly(≥75 years),patient with lower education level,patients divorced or living alone were more likely to have worse score.Patients with comorbidities of atrial fibrillation(AF),renal dysfunction,COPD,anemia,diabetes,depression were more likely to have lower score.Patient with higher NT-proBNP level,NYHA class Ⅲ-Ⅳ,multiple comorbidities,hyponatremia and lower albumin level more frequently had lower scores.There was no significant difference of KCCQ-12 score among patients with preserved,middle ranged and reduced ejection fraction.Patients with worse KCCQ-12 score more frequently reported lower EQ-5D index score and VAS.In multivariable analysis,women,aged 75 years or older,AF,COPD,diabetes,BMI>24kg/m2,depression,higher NT-proBNP,NYHA class Ⅲ-Ⅳ,albumin<35g/L,use of aldosterone were the independent influence factors of worse KCCQ score.New-onset HF and the use of β-blockers were independent influence factors of better KCCQ score.During 1-year follow-up,the number of answered questionnaires was 4869 at baseline,4267 at 1 month,3679 at 6 months and 2529 at 1 year.The median score(IQR)of 4 timepoint assessments were 43.1(26.7,60.6),71.3(56.3,81.9),78.8(62.5,87.5),79.0(62.5,88.5).The median scores in physical domain were 50.0(33.3,75.0),83.3(58.3,100),83.3(66.7,100),83.3(66.7,100),in symptomatic frequency were 50.0(25.0,70.0),85.0(70.0,100),90.0(70.0,100),90.0(70.0,100),in quality of life were 25.0(12.5,37.5),37.5(25.0,62.5),50.0(37.5,75.0),62.5(37.5,75.0),in social limitation were 50.0(25.0,75.0),91.7(66.7,100),91.7(66.7,100),87.5(66.7,100).EQ-5D index at 4 timepoints were 0.71(0.44,0.82),0.83(0.71,1.00),0.84(0.78,1.00),0.84(0.77,1.00).The pattern in subgroups of clinical interest was similar to that observed in the whole population.Women,aged 75 or older,stroke,COPD,renal dysfunction,depression,HF with preserved ejection fraction,NYHA class Ⅲ-Ⅳ and baseline higher NT-proBNP levels persistently remained worse scores over time.(2)Totally 4898 patients had baseline KCCQ-12 score.The proportions of patients scored 0-24,25-49,50-74,and 75-100 were 22.5%,38.0%,29.2%and 10.3%.Within 1 year of discharge,the crude mortality of four groups was 29.6%,17.6%,12.3%,9.7%,the hospitalization rate was 52.7%,47.3%,44.8%,40.0%,and the composite event rate was 50.6%,38.2%,33.2%,25.5%(all P<0.05),respectively.The linear associations of KCCQ-12 score with all-cause death,hospitalization and composite outcome were satisfied.After adjustment,each 10 points decrease in KCCQ-12 score was associated with a 11%increase in all-cause death,a 4%increase in hospitalizations,and a 7%increase in composite events.Among 4267 patients who completed KCCQ-12 at 1 month of discharge,the proportions of patients scored 0-24,25-49,50-74,and 75-100 were 5.5%,13.2%,37.1%,44.2%.The linear associations between KCCQ score and above clinical outcomes were satisfied.After adjustment,each 10 points decrease in KCCQ-12 score was associated with a 16%increase in all-cause death,a 14%increase in hospitalizations,and a 19%increase in composite events.There were 75.6%patients experiencing improvement,11.4%with stable status,and 13.0%experiencing deterioration.After adjusting baseline characteristics and KCCQ-12 score,the changes of KCCQ score were linearly associated with 1-year clinical outcomes.Each 5 points decrease in KCCQ-12 score was associated with a 7%increase in all-cause death,an 8%increase in hospitalizations,and a 7%increase in composite events.Conclusions(1)Health status in patients hospitalized for HF were greatly impaired.Women,older age,AF,COPD,diabetes,BMI>24kg/m2,depression,higher NT-proBNP levels,NYHA class III-IV,albumin<35g/L were the independent influence factors of worse KCCQ score,while new-onset HF and use of β-blockers were protective factors of better KCCQ score.Health status showed a marked improvement at first month after discharge,but subsequently had a small improvement at 6 months and remained stable at 1 year follow-up.The pattern over time remained consistent in four domains of KCCQ-12 and subgroups of interest.(2)KCCQ-12 score was negatively associated with 1-year clinical outcomes in both acute HF and chronic HF.The changes of serial KCCQ assessment were also linearly associated with increased risk of all-cause death,hospitalization,and composite events during 1 year follow up.