Prognostic value of six-minute walk distance in patients with heart failure: A meta-analysis

Cardiopulmonary exercise testing for evaluation of peak oxygen consumption is a gold standard method to evaluate exercise capacity.1,2 However, this test is expensive and time-consuming, and patients with substantial heart failure cannot conduct maximal symptom-limited exercises.3 The six-minute walk distance (6MWD), a simple submaximal exercise test, has been recommended as an alternative to cardiopulmonary exercise testing.4 This test has been increasingly incorporated and accepted into contemporary heart failure practice.5 Several studies have investigated the prognostic value of 6MWD in predicting mortality or hospital readmission in heart failure patients.6 However, the prognostic value of 6MWD remains controversial.7,8 This meta-analysis aimed to evaluate the prognostic significance of baseline 6MWD for prediction of death or readmission in patients with heart failure. Two authors independently searched Pubmed, Embase and Web of Science to identify relevant studies up to November 2017 by using the following keywords: ‘6-min walk’ OR ‘six-min walk’ OR ‘walk test’ AND ‘heart failure’ OR ‘cardiac failure’ AND ‘death’ OR ‘mortality’ OR ‘hospital readmission’ or ‘rehospitalization’ AND ‘follow-up’. The inclusion criteria were as follows: (a) observational studies enrolling patients with heart failure; (b) baseline 6MWD as exposure; (c) reported multivariable-adjusted risk estimates of all-cause mortality, readmission or combined endpoint. Studies not providing the quantitative risk estimate for the endpoint or enrolling acute decompensated patients with heart failure were excluded. The Newcastle–Ottawa scale (NOS) was used to assess the methodological quality of the included studies. Data analyses were performed using RevMan (version 5.2, Oxford, UK). For categorical analysis, we pooled risk estimates for the shortest versus longest 6MWD category. For analysing prognostic values as continuous values, we recalculated the risk ratio (RR) to each 50 m 6MWD reduction by the following formula: RR50 = exp (ln (RR1) × 50). Among the 486 relevant records identified from the electronic medical database search, 13 studies comprising 6076 patients with heart failure were finally included in the meta-analysis (Table 1). For methodological quality assessment, the NOS scores ranged from 5 to 8. The meta-analysis showed that the shortest 6MWD was associated with an increased risk of all-cause mortality (RR 1.96; 95% confidence interval (CI) 1.10–3.50; I2 = 77.0%; p = 0.005; Supplementary Material Figure S1-A online) in a random effect model. In addition, each 50 m decrease in 6MWD significantly increased by 18%, indicating a high risk of all-cause mortality (RR 1.18; 95% CI 1.07–1.29; I2 = 51.0%; p = 0.11; Supplementary Figure S1-B) in a random effect model. Each 50 m decrease in 6MWD was associated with 43% risk of readmission (RR 1.43; 95% CI 1.10–1.86; I2 = 88.0%; p < 0.001; Supplementary Figure S2) in a random effect model meta-analysis. For the combined endpoint, the shortest 6MWD was associated with high risk of the combined endpoint of death or readmission (RR 1.86; 95% CI 1.40–2.46; I2 = 0.0%; p = 0.76; Supplementary Figure S3-A) in a fixed-effect model meta-analysis. A random effect model meta-analysis suggested that each 50 m decrease in 6MWD was associated with a 20% risk of the combined endpoint of death or readmission (RR 1.20; 95% CI 1.09–1.32; I2 = 52.0%; p = 0.15; Supplementary Figure S3-B). We opted not to perform the publication bias test as the number of studies analysed was less than the recommended arbitrary minimum number of studies. Main characteristics of the included studies. Data calculated from original study. LVEF: left ventricular ejection fraction; HR: hazard ratio; OR: odds ratio; CI: confidence interval; NOS: Newcastle–Ottawa Scale; P: prospective; R: retrospective; U: unclear; HF: heart failure; NYHA: New York Heart Association; NP: not provided; DM: diabetes mellitus; CAD: coronary artery disease; CRP: C-reactive protein; NT-proBNP: N-terminal pro-brain natriuretic peptide; hs-CRP: high-sensitivity CRP; SBP: systolic blood pressure; eGFR: estimated glomerular filtration rate; CCI: Charlson Comorbidity Index; DBP: diastolic blood pressure; BNP: brain natriuretic peptide. Main characteristics of the included studies. Data calculated from original study. LVEF: left ventricular ejection fraction; HR: hazard ratio; OR: odds ratio; CI: confidence interval; NOS: Newcastle–Ottawa Scale; P: prospective; R: retrospective; U: unclear; HF: heart failure; NYHA: New York Heart Association; NP: not provided; DM: diabetes mellitus; CAD: coronary artery disease; CRP: C-reactive protein; NT-proBNP: N-terminal pro-brain natriuretic peptide; hs-CRP: high-sensitivity CRP; SBP: systolic blood pressure; eGFR: estimated glomerular filtration rate; CCI: Charlson Comorbidity Index; DBP: diastolic blood pressure; BNP: brain natriuretic peptide. This meta-analysis demonstrates that short 6MWD is independently associated with high risk of all-cause mortality, readmission and the combined endpoint of death or readmission in patients with heart failure. Patients with heart failure with the shortest 6MWD presented an approximately 96% and 86% higher risk of all-cause mortality and the combined endpoint of death or readmission than those with the longest 6MWD. In addition, each 50 m 6MWD reduction exhibited significantly increased all-cause mortality (18%), readmission (43%) and combined endpoint of death or readmission risk (20%). Furthermore, tele-6MWD showed a high predictive role with respect to readmission or all-cause mortality in patients with heart failure, and the values were comparable to the prognostic impact of the conventional 6MWD.9 Overall, these findings suggesting that 6MWD is a strong indicator of poor prognosis in patients with heart failure. Several potential limitations should be noted. First, substantial heterogeneity was found across studies, and differences in age, gender, standardization of the test protocol, aetiology or severity of heart failure and follow-up time can partly explain the heterogeneity. Second, variables adjusted for risk estimates varied in individual studies, and many well-established variables were not fully adjusted in several studies. Lack of adjustment of important confounders can lead to overestimation of summary effect sizes. Third, various cutoff values of 6MWD were reported, and we failed to establish an optimal threshold of the 6MWD in this meta-analysis. Fourth, the limited numbers of studies on individual outcomes prevented us performing subgroup analysis. Fifth, the analysis patients predominantly included elderly men. Thus, the generalizability of the current findings to women and young men should be presented with caution. Finally, the use of heart failure medications that may have changed the prognosis was poorly reported. This meta-analysis suggests that short 6MWD is associated with high risk of all-cause mortality, readmission and combined endpoint of death or readmission in patients with heart failure. The 6MWD test can be applied for stratification of adverse outcomes among patients with heart failure. The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article. The author(s) received no financial support for the research, authorship and/or publication of this article.