End points for comparative effectiveness research in heart failure

Novel measures to assess ventricular assist device patient-reported outcomes: Findings from the MCS A-QOL study

BACKGROUND

Generic and heart failure-specific measures do not capture unique aspects of living with a ventricular assist device (VAD). Using state-of-the-science psychometric measurement methods, we developed a measurement system to assess post-ventricular assist device adjustment and health-related quality of life (HRQOL).

METHODS

Patients were recruited from 10/26/16-2/29/20 from 12 U.S. VAD programs. We created a dataset of participants (n = 620) enrolled before left (L)VAD implantation, with data at 3- or 6- months post-implantation (group1 [n = 154]), and participants enrolled after LVAD implantation, with data at one timepoint (group 2 [n = 466]). We constructed 5 item banks: 3 modified from existing measures and 2 new measures. Analyses included item response theory (IRT) modeling, differential item functioning tests for systematic measurement bias, and indicators of reliability and validity.

RESULTS

Of 620 participants, 56% (n = 345) were implanted as destination therapy, 51% (n = 316) were <12 months post-implantation, mean age = 57.3 years, 78% (n = 485) male, 70% (n = 433) White, 58% (n = 353) married/partnered, and 58% (n = 357) with >high school education. We developed 5 new VAD item banks/measures: 6-item VAD Team Communication; 12-item Self-efficacy Regarding VAD Self-care; 11-item Being Bothered by VAD Self-care and Limitations; 7-item Satisfaction with Treatment; and 11-item Stigma. Cronbach's alpha reliability ranged from good (≥0.80) to excellent (≥0.90) for item banks/measures. All measures, except VAD Team Communication, demonstrated at least moderate correlations (≥0.30) with construct validity indicators.

CONCLUSIONS

These measures meet IRT modeling assumptions and requirements; scores demonstrate reliability and validity. Use of these measures may assist VAD clinicians to inform patients about VADs as a treatment option and guide post-VAD interventions.

A qualitative enquiry regarding the factors influencing return to work after a cardiac incident in the Western Cape, South Africa

BACKGROUND

Cardiovascular disease is amongst the top three leading causes of mortality in South Africa and the world. The effects of cardiovascular disease can be seen in limitations of function within all spheres of life, including work function. Cardiac rehabilitation programs have been documented to improve functional abilities, however further research is required with regard to activity limitation, participation restrictions, facilitators, and barriers of return to work, after cardiac rehabilitation.

OBJECTIVE

This qualitative study aimed to determine and explore the barriers and enablers of return to work for individuals with cardiovascular disease who completed a cardiac rehabilitation program in the Western Cape, South Africa. The qualitative exploration was part of a study that resulted in a return-to-work program for individuals with cardiac disease.

METHOD

Qualitative methodology was employed for this study. Semi-structured interviews were used with 10 participants that attended a cardiac rehabilitation program. Thematic analysis of the qualitative data was completed.

RESULTS

The participants described the barriers and facilitators of their return-to-work process and the factors that were of critical importance in the resumption of the worker role or the result of unemployment. The four major themes were 1) Cardiac conditions negatively affect function, 2) Being unemployed, 3) Participation in ADLs and lifestyle modification enhances one's activity levels, and 4) Return to work.

CONCLUSION

Lifestyle modification and reasonable accommodation enable a successful return to work after a cardiac incident.

Clinical effect of Danshen decoction in patients with heart failure: A systematic review and meta-analysis of randomized controlled trials

BACKGROUND

The incidence of heart failure (HF) is increasing year by year, posing a great threat to human health. Although pharmacotherapy has been able to significantly prolong patient survival, pharmacotherapy for HF still has limitations due to its complex pathogenesis and considerable individual variability, there is a great need to explore complementary and alternative therapies to slow down the progression of HF. Danshen decoction is used to treat several cardiovascular diseases including HF, but the efficacy of stabilization is uncertain. This meta-analysis evaluated the clinical efficacy of Danshen Decoction for the treatment of HF.

METHODS

The registration number assigned to this meta-analysis on the PROSPERO platform is CRD42022351918. Four databases were searched, and randomized controlled trials (RCTs) of Danshen decoction combined with conventional treatment (CT) of HF were screened, CT included medical treatments other than Danshen Decoction to which the patient was treated (including but not limited to angiotensin converting enzyme inhibitor, angiotensin receptor blocker, angiotensin receptor neprilysin inhibitors, β-blockers, diuretics, mineralcorticoid recept antagonist etc.). The clinical efficacy rate (CER), left ventricular ejection fraction (LVEF), left ventricular end-diastolic dimension (LVEDD), left ventricular end-systolic diameter (LVESD), brain natriuretic peptide (BNP), N-terminal pro-B type natriuretic peptide (NT-proBNP) and hypersensitive C-reactive protein (hs-CRP) were included as outcome indicators. The GRADE grading scale was used to grade the above indicators. The Cochrane risk-of-bias tool and the Jadad quality scale were used to assess the methodological quality of RCTs. Finally, RevMan V.4.5 software was used for data synthesis, 95% confidence intervals (CI) for dichotomous data, risk ratios (RR), and mean differences (MD) for continuous variables were calculated, Chi-square and I2 were used for heterogeneity assessment.

RESULTS

Nine RCTs with a total of 855 patients were included in this study, and all included RCTs had a low overall quality risk of bias and high quality of reported information. The results of the meta-analysis showed that compared with the use of CT, CER (%) was significantly improved due to Danshen decoction combined with CT (MD = 3.95, 95% CI [2.58, 6.04], P < 0.00001), LVEF (%) was significantly improved (MD = 5.46, 95% CI [5.32, 5.60], P < 0.00001), LVEDD (mm) was significantly reduced (MD = -5.27, 95% CI [-6.21, -4.32], P < 0.00001), LVESD (mm) was significantly reduced (MD = -4.60, 95% CI [-5.87, -3.32], P < 0.00001), BNP (pg/mL) was significantly reduced (MD = -88.61, 95% CI [-121.98, -55.24], P < 0.00001), NT-proBNP (pg/mL) was significantly decreased (SMD = -3.33, 95% CI [-5.92, -0.73], P = 0.01), hs-CRP (mg/L) was significantly decreased (MD = -2.73, 95% CI [-4.11, -1.34], P = 0.0001). The quality of the GRADE evidence for all outcomes was moderate to low and no RCTs reported adverse events.

CONCLUSION

Our research demonstrates that Danshen decoction is an effective and safe treatment option for HF. Nevertheless, considering the limitations of methodological and the quality of RCTs, more rigorous, large-scale, multicenter randomized clinical trials are needed to further evaluate the efficacy and safety of Danshen decoction in the treatment of HF patients.

Non-cardiology vs. cardiology care of patients with heart failure and reduced ejection fraction is associated with lower use of guideline-based care and higher mortality: Observations from The Swedish Heart Failure Registry

BACKGROUND

Patients with heart failure (HF) are often cared for by non-cardiologists. The implications are unknown.

METHODS

In a nationwide HF cohort with reduced ejection fraction (HFrEF), we compared demographics, clinical characteristics, guideline-based therapy use and outcomes in non-cardiology vs. cardiology in-patient and out-patient care.

RESULTS

Between 2000 and 2016, 36,076 patients with HFrEF were enrolled in the Swedish HF registry (19,337 [54%] in-patients overall), with 44% of in-patients and 45% of out-patients managed in non-cardiology settings. Predictors of treatment in non-cardiology were age > 75 years (adjusted odds ratio for non-cardiology 1.20; 95% confidence interval 1.14-1.27), lower education level (0.71; 0.66-0.76 for university vs. compulsory), valve disease (1.24; 1.18-1.31) and systolic blood pressure (SBP) >120 mmHg (1.05; 1.00-1.10). Non-cardiology care was significantly associated with lower use of beta-blockers (0.80; 0.74-0.86) and devices (intracardiac defibrillator [ICD] and/or cardiac resynchronization therapy [CRT]: 0.63; 0.56-0.71), and less frequent specialist follow-up (0.61; 0.57-0.65). Over 1-year follow-up the risk of all-cause mortality (adjusted hazard ratio 1.09; 1.03-1.15) was higher but the risk of first HF (re-) hospitalization was lower (0.93; 0.89-0.97) in non-cardiology vs. cardiology care.

CONCLUSIONS

In HFrEF, non-cardiology care was independently associated with older ageand lower education. After covariate adjustment, non-cardiology care was associated with lower use of beta-blockers and devices, higher mortality, and lower risk of HF hospitalization. Access to cardiology care may not be equitable and this may have implications for use of guideline-based care and outcomes.

Baroreflex activation therapy reduces frequency and duration of hypertension-related hospitalizations in patients with resistant hypertension

Purpose

Baroreflex activation therapy (BAT) has been shown to lower blood pressure in patients with resistant hypertension. The purpose of this study was to analyze whether this translates into a reduction of more relevant clinical endpoints.

Methods

Patients with resistant hypertension were treated with the second-generation BAT system. Records on hospitalization (dates of admission and discharge, main diagnosis) were obtained from medical insurance companies.

Results

Records on hospitalization were available for a period of 1 year before BAT in two patients and 2 years in 22 patients. The total number of hospitalizations per patient was 3.3 ± 3.5/year before BAT and 2.2 ± 2.7/year after BAT (p = 0.03). Hospitalizations related to hypertension were significantly decreased from 1.5 ± 1.6/year before BAT to 0.5 ± 0.9/year after BAT (p < 0.01). The cumulative duration of hypertension-related hospital stays was significantly reduced from 8.0 ± 8.7 days/year before BAT to 1.8 ± 4.8 days/year after BAT (p < 0.01). Office cuff blood pressure was 183 ± 27 mmHg over 102 ± 17 mmHg under 6.6 ± 2.0 antihypertensive drugs before BAT and 157 ± 32 mmHg over 91 ± 20 mmHg (both p < 0.01) under 5.9 ± 1.9 antihypertensive drugs (p = 0.09 for number of drugs) at latest follow-up. Daytime ambulatory blood pressure was 164 ± 21 mmHg over 91 ± 14 mmHg before BAT and 153 ± 21 mmHg (p = 0.03) over 89 ± 15 mmHg (p = 0.56) at latest follow-up. Heart rate was 75 ± 16 bpm before BAT and 72 ± 12 bpm at latest follow-up (p = 0.35).

Conclusions

Rate and duration of hypertension-related hospitalizations in patients with severe resistant hypertension were lowered after BAT. Whether the response is mediated through improvements in blood pressure control requires further studies.

Electronic supplementary material

The online version of this article (10.1007/s10286-020-00670-9) contains supplementary material, which is available to authorized users.

Aortic Stenosis and Heart Failure: Disease Ascertainment and Statistical Considerations for Clinical Trials

Aortic stenosis is a progressive disease that develops over decades, and once symptomatic and untreated, is associated with poor survival. Transcatheter aortic valve replacement has evolved significantly in the past decade and has expanded its indication from surgically inoperable and high-risk patients to patients with intermediate risk. Assessment of heart failure-related outcomes include the use of functional assessments, disease-specific quality of life surveys and standardised ascertainment of events, such as hospitalisations. Multiple statistical approaches are currently being tested to account for recurrent events such as hospitalisations for heart failure or to combine binary and continuous outcomes, both intended to assess the holistic burden of the disease, as opposed to the traditional analysis of time to first event.

Association between quality of life and prognosis of candidate patients for heart transplantation: a cross-sectional study

OBJECTIVE

to verify the association between the prognostic scores and the quality of life of candidates for heart transplantation.

METHOD

a descriptive cross-sectional study with a convenience sample of 32 outpatients applying to heart transplantation. The prognosis was rated by the Heart Failure Survival Score (HFSS) and the Seattle Heart Failure Model (SHFM); and the quality of life by the Minnesota Living With Heart Failure Questionnaire (MLHFQ) and the Kansas City Cardiomyopathy Questionnaire (KCCQ). The Pearson correlation test was applied.

RESULTS

the correlations found between general quality of life scores and prognostic scores were (HFSS/MLHFQ r = 0.21), (SHFM/MLHFQ r = 0.09), (HFSS/KCCQ r = -0.02), (SHFM/KCCQ r = -0.20).

CONCLUSION

the weak correlation between the prognostic and quality of life scores suggests a lack of association between the measures, i.e., worse prognosis does not mean worse quality of life and the same statement is true in the opposite direction.

Agreement between heart failure patients and providers in assessing New York Heart Association functional class

BACKGROUND

Uncertainty persists regarding whether patient assessment of New York Heart Association (NYHA) functional classification should be preferred over provider assessment among patients with heart failure (HF).

OBJECTIVES

To compare patient against provider NYHA assessments, and both to distance walked on a 6-minute walk test (6MWT) among patients with HF.

METHODS

In this prospective study, we enrolled 101 HF patients who self-assessed NYHA classification. Health care providers who were blinded to patient ratings of NYHA also rated NYHA. Patients completed a 6MWT according to a standardized protocol. We used Spearman coefficients (rs) to evaluate the correlations between variables.

RESULTS

Patient- and provider-determined NYHA class were poorly correlated, but the relationship was statistically significant (rs = 0.40, p < 0.001). Patients consistently reported better NYHA class (class I: 72% vs 15%) than providers. Provider-determined NYHA had a stronger correlation with 6MWT distance (rs = -0.36, p < 0.001 vs. rs = -0.22, p = 0.03). Providers assigned a worse class to older patients who had comorbidity; patients with dyspnea and longer HF duration assigned themselves a worse class.

CONCLUSION

Patients and providers exhibited poor agreement in NYHA assignment.

The clinical epidemiology of fatigue in newly diagnosed heart failure

BACKGROUND

Fatigue is a common and distressing but poorly understood symptom among patients with heart failure (HF). This study sought to evaluate the prevalence, predictors, and prognostic value of clinically documented fatigue in newly diagnosed HF patients from the community.

METHODS

This retrospective cohort study consisted of 12,285 newly diagnosed HF patients receiving health care services through the Geisinger Health System, with passive data collection through electronic medical records (EMR). Incident HF, fatigue, and other study variables were derived from coded data within EMRs. A collection of 87 candidate predictors were evaluated to ascertain the strongest independent predictors of fatigue using logistic regression. Patients were followed for all-cause mortality for an average of 4.8 years. The associations between fatigue and 6-month, 12-month, and overall mortality were evaluated via Cox proportional hazards regression models.

RESULTS

Clinically documented fatigue was found in 4827 (39%) newly diagnosed HF patients. Depression demonstrated the strongest association with fatigue. Fatigue was often part of a symptom cluster, as other HF symptoms including dyspnea, chest pain, edema, syncope, and palpitations were significant predictors of fatigue. Volume depletion, lower body mass index, and abnormal weight loss were also strong predictors of fatigue. Fatigue was not significantly associated with either 6-month (HR = 1.12, p = 0.16) or overall mortality (HR = 1.00, p = 0.89) in adjusted models.

CONCLUSIONS

Fatigue is a commonly documented symptom among newly diagnosed HF patients, and its origins may lie in both psychologic and physiologic factors. Though fatigue did provide a prognostic signal in the short-term, this was largely explained by physiologic confounders. Proper therapeutic remediation of fatigue in HF relies on identifying underlying factors.

Recurrent Hospitalization Among Patients With Atrial Fibrillation Undergoing Intracoronary Stenting Treated With 2 Treatment Strategies of Rivaroxaban or a Dose-Adjusted Oral Vitamin K Antagonist Treatment Strategy

Supplemental Digital Content is available in the text.

Background:

Patients with atrial fibrillation who undergo intracoronary stenting traditionally are treated with a vitamin K antagonist (VKA) plus dual antiplatelet therapy (DAPT), yet this treatment leads to high risks of bleeding. We hypothesized that a regimen of rivaroxaban plus a P2Y₁₂ inhibitor monotherapy or rivaroxaban plus DAPT could reduce bleeding and thereby have a favorable impact on all-cause mortality and the need for rehospitalization.

Methods:

Stented subjects with nonvalvular atrial fibrillation (n=2124) were randomized 1:1:1 to administration of reduced-dose rivaroxaban 15 mg daily plus a P2Y₁₂ inhibitor for 12 months (group 1); rivaroxaban 2.5 mg twice daily with stratification to a prespecified duration of DAPT of 1, 6, or 12 months (group 2); or the reference arm of dose-adjusted VKA daily with a similar DAPT stratification (group 3). The present post hoc analysis assessed the end point of all-cause mortality or recurrent hospitalization for an adverse event, which was further classified as the result of bleeding, a cardiovascular cause, or another cause blinded to treatment assignment.

Results:

The risk of all-cause mortality or recurrent hospitalization was 34.9% in group 1 (hazard ratio=0.79; 95% confidence interval, 0.66–0.94; P=0.008 versus group 3; number needed to treat=15), 31.9% in group 2 (hazard ratio=0.75; 95% confidence interval, 0.62–0.90; P=0.002 versus group 3; number needed to treat=10), and 41.9% in group 3 (VKA+DAPT). Both all-cause death plus hospitalization potentially resulting from bleeding (group 1=8.6% [P=0.032 versus group 3], group 2=8.0% [P=0.012 versus group 3], and group 3=12.4%) and all-cause death plus rehospitalization potentially resulting from a cardiovascular cause (group 1=21.4% [P=0.001 versus group 3], group 2=21.7% [P=0.011 versus group 3], and group 3=29.3%) were reduced in the rivaroxaban arms compared with the VKA arm, but other forms of rehospitalization were not.

Conclusions:

Among patients with atrial fibrillation undergoing intracoronary stenting, administration of either rivaroxaban 15 mg daily plus P2Y₁₂ inhibitor monotherapy or 2.5 mg rivaroxaban twice daily plus DAPT was associated with a reduced risk of all-cause mortality or recurrent hospitalization for adverse events compared with standard-of-care VKA plus DAPT.

Clinical Trial Registration:

URL: http://www.clinicaltrials.gov. Unique identifier: NCT01830543.

Use of outcome measures in pulmonary hypertension clinical trials

OBJECTIVES

To evaluate the use of surrogate measures in pulmonary hypertension (PH) clinical trials and how it relates to clinical practice.

BACKGROUND

Studies of pulmonary arterial hypertension (PAH) employ a variety of surrogate measures in addition to clinical events because of a small patient population, participant burden, and costs. The use of these measures in PH drug trials is poorly defined.

METHODS

We searched PubMed/MEDLINE/Embase for randomized or prospective cohort PAH clinical treatment trials from 1985 to 2013. Extracted data included intervention, trial duration, study design, patient characteristics, and primary and secondary outcome measures. To compare with clinical practice, we assessed the use of surrogate measures in a clinical sample of patients on PH medications at Duke University Medical Center between 2003 and 2014.

RESULTS

Between 1985 and 2013, 126 PAH trials were identified and analyzed. Surrogate measures served as primary endpoints in 119 trials (94.0%). Inclusion of invasive hemodynamics decreased over time (78.6%, 75.0%, 52.2%; P for trend = .02), while functional testing (7.1%, 60.0%, 81.5%; P for trend < .0001) and functional status or quality of life (0%, 47.6%, 62.8%; P for trend < .0001) increased in PAH trials over the same time periods. Echocardiography data were reported as a primary or secondary outcome in 32 trials (25.4%) with increased use from 1985-1994 to 1995-2004 (7.1% vs 35.0%, P = .04), but the trend did not continue to 2005-2013 (25.0%). In comparison, among 450 patients on PAH therapies at our institution between 2003 and 2013, clinical assessments regularly incorporated serial echocardiography and 6-minute walk distance tests (92% and 95% of patients, respectively) and repeat measurement of invasive hemodynamics (46% of patients).

CONCLUSIONS

The majority of PAH trials have utilized surrogate measures as primary endpoints. The use of these surrogate endpoints has evolved significantly over time with increasing use of patient-centered endpoints and decreasing or stable use of imaging and invasive measures. In contrast, imaging and invasive measures are commonly used in contemporary clinical practice. Further research is needed to validate and standardize currently used measures.

Composite outcome measures in a pragmatic clinical trial of chronic heart failure management: A comparative assessment

BACKGROUND

A number of composite outcomes have been developed to capture the perspective of the patient, clinician and objective measures of health in assessing heart failure outcomes. To date there has been a limited examination in the composition of these outcomes.

METHODS AND RESULTS

Three commonly used scoring systems in heart failure trials: Packer's composite, Patient Journey and the African American Heart Failure Trial (A-HeFT) scores were compared in assessing outcomes from the Which heart failure intervention is most cost-effective & consumer friendly in reducing hospital care (WHICH(?)) Trial. Comparability and interpretability of these outcomes and the influence of each component to the final outcome were examined. Despite all three composite outcomes incorporating mortality, hospitalisation and quality of life (QoL), the contribution of each individual component to the final outcomes differed. The component with the most influence in deteriorating condition for the Packer's composite was hospitalisation (67.7%), while in Patient Journey it was QoL (61.5%) and for A-HeFT composite score it was mortality (45.4%).

CONCLUSIONS

The contribution made by each component varied in subtle, but important ways. This study emphasises the importance of understanding the value system of the composite outcomes to enable meaningful interpretation of results.

Power and sample size calculations for the Wilcoxon-Mann-Whitney test in the presence of death-censored observations

We consider a clinical trial of a potentially lethal disease in which patients are randomly assigned to two treatment groups and are followed for a fixed period of time; a continuous endpoint is measured at the end of follow-up. For some patients; however, death (or severe disease progression) may preclude measurement of the endpoint. A statistical analysis that includes only patients with endpoint measurements may be biased. An alternative analysis includes all randomized patients, with rank scores assigned to the patients who are available for the endpoint measurement on the basis of the magnitude of their responses and with 'worst-rank' scores assigned to those patients whose death precluded the measurement of the continuous endpoint. The worst-rank scores are worse than all observed rank scores. The treatment effect is then evaluated using the Wilcoxon-Mann-Whitney test. In this paper, we derive closed-form formulae for the power and sample size of the Wilcoxon-Mann-Whitney test when missing measurements of the continuous endpoints because of death are replaced by worst-rank scores. We distinguish two approaches for assigning the worst-rank scores. In the tied worst-rank approach, all deaths are weighted equally, and the worst-rank scores are set to a single value that is worse than all measured responses. In the untied worst-rank approach, the worst-rank scores further rank patients according to their time of death, so that an earlier death is considered worse than a later death, which in turn is worse than all measured responses. In addition, we propose four methods for the implementation of the sample size formulae for a trial with expected early death. We conduct Monte Carlo simulation studies to evaluate the accuracy of our power and sample size formulae and to compare the four sample size estimation methods.