Abstract
Aims
Red blood cell distribution width‐to‐albumin ratio (RAR), an innovate biomarker of inflammation, can independently predict adverse cardiovascular outcomes. However, the association between RAR and prognosis in patients with non‐ischaemic heart failure (NIHF) remains unclear.
Methods and results
A total of 2077 NIHF patients admitted to the Heart Failure Care Unit, Fuwai Hospital, were consecutively enrolled from December 2006 to October 2017 in this retrospective study. The primary endpoint was a composite outcome of all‐cause mortality and heart transplantation. The correlation between RAR and the composite outcome was assessed by the Kaplan–Meier survival analysis and the Cox regression analysis. Incremental predictive values and the clinical performance of RAR for all‐cause mortality or heart transplantation were also assessed based on a 12‐variable traditional risk model. The median follow‐up time in this study was 1433 (1341, 1525) days. As the gender no longer satisfied the Cox proportional risk assumption after 1150 days, we set 1095 days as the follow‐up time for analysis. A total of 500 patients reached the composite outcome. Multivariable Cox regression showed that per log2 increase of RAR was significantly associated with a 132.9% [hazard ratio 2.329, 95% confidence interval (CI) 1.677–3.237, P < 0.001] increased risk of all‐cause mortality or heart transplantation. Better model discrimination [concordance index: 0.766 (95% CI 0.754–0.778) vs. 0.758 (95% CI 0.746–0.770), P < 0.001], calibration (Akaike information criterion: 1487.3 vs. 1495.74; Bayesian information criterion: 1566.25 vs. 1569.43; Brier score: 1569.43 vs. 1569.43; likelihood ratio test P < 0.001), and reclassification (integrated discrimination improvement: 1.35%, 95% CI 0.63–2.07%, P < 0.001; net reclassification improvement: 13.73%, 95% CI 2.05–27.18%, P = 0.034) were improved after adding RAR to the traditional model (P < 0.001 for all). A higher overall net benefit was also obtained in the threshold risk probability of 20–55%.
Conclusions
High level of RAR was an independent risk factor of poor outcome in NIHF.
Keywords: Non‐ischaemic heart failure, Red blood cell distribution width‐to‐albumin ratio, Inflammation, Prognosis
Introduction
Heart failure (HF) is a complex, multifactorial syndrome resulting from an impaired heart function with high morbidity and mortality. It is a rapidly growing global public health problem with prevalence up to 1–2% in the adult population. 1 Therefore, detection of novel biomarkers and mechanisms to identify high‐risk HF patients are critical for the effective management and treatment.
Systemic inflammation has been recognized as a common pathophysiology feature of both acute and chronic HF, which is the results of myocardial injury, oxidative stress, mitochondrial impairments, and neurohumoral activation. 2 HF patients tend to have higher levels of serum biomarkers associated with inflammation, including high‐sensitivity C‐reactive protein (hs‐CRP), interleukin (IL)‐6, IL‐8, adiponectin, and tumour necrosis factor‐α, each of which has been found to correlate with clinical events. 3 , 4
It is well known that a complete blood count (CBC) is part of a routine test for patients on admission to the hospital. Certain markers from CBC have been shown to be strongly associated with a worse prognosis in HF patients, such as neutrophil‐to‐lymphocyte ratio (NLR) 5 and platelet‐to‐lymphocyte ratio (PLR). 6 Red blood cell distribution width (RDW) has been newly proved to be an effective marker in assessing HF prognosis. It represents the heterogeneity of red cell volume and has traditionally been used in the differential diagnosis of anaemia and also as a marker of inflammation. 7 Serum albumin (ALB) level, which reflected nutritional status 8 and anti‐inflammatory effects, 9 has also been regarded as a simple prognostic factor for acute HF. 10 The combination of these two easy‐obtained markers, RDW‐to‐ALB ratio (RAR), has been emerging as a powerful prognostic marker in ischaemic heart disease (IHD), 11 HF, 12 sepsis, 13 and aortic aneurysms, 14 and more importantly, the predictive value is better than the solely RDW or ALB.
The prognostic value of RAR in non‐ischaemic HF (NIHF) might be different from that in IHD, which showed higher concentration of inflammation‐related biomarkers. 3 Thus, the prognostic role of inflammation‐related biomarkers in HF studies might be primarily determined by patients with ischaemic aetiology. To answer that question, we conducted this retrospective study to comprehensively assess the prognostic value of RAR in an NIHF cohort.
Materials and methods
Study population
This is a single‐centre retrospective analysis of 5124 consecutive HF patients admitted to the Heart Failure Center of Fuwai Hospital (Beijing, China) from December 2006 to October 2017. HF was defined and treated according to the definitions established in the European Society of Cardiology Guidelines for the diagnosis and treatment of acute and chronic HF. 15 The exclusion criteria were as follows: (i) age < 18 years (n = 179) or had a history of IHD (n = 2011); (ii) life expectancy < 1 year due to malignancy, cardiac amyloidosis, or other end‐stage disease (n = 173); (iii) severe respiratory disease, systemic infection, or autoimmune diseases (n = 144); and (iv) more than 50% data were missing at admission or lack of baseline RDW and ALB (n = 540). Ultimately, 2077 patients with NIHF were included in the study. This retrospective study was performed in line with the Declaration of Helsinki, with the approval from the ethics committee of Fuwai Hospital (2018‐1041).
Clinical data
Patient demographics, medical history, laboratory test results, echocardiographic data, and medications at admission were collected from the hospital information system of Fuwai Hospital and recorded by trained physicians. Fasting venous blood samples were taken from all patients within 24 h after admission and measured in laboratory department for immediate testing using standard techniques. RDW was measured by automatic haematology analyser (Sysmex, XN2000, Japan), and ALB was measured by biochemical analyser (Hitachi Labospect 008AS, Japan).
Definition
RAR was calculated based on the following formula: RAR = RDW (%)/ALB (g/dL). Body mass index (BMI) was defined as weight (kg)/height (m2). According to the WHO BMI cut‐off points for Asian populations, 16 patients were divided into underweight (BMI < 18.5 kg/m2), normal (18.5 ≤ BMI < 23.0 kg/m2), overweight (23.0 ≤ BMI < 27.5 kg/m2), and obesity (BMI ≥ 27.5 kg/m2).
Outcomes and follow‐up
The composite endpoint was defined as the combination of all‐cause mortality and heart transplantation. Patients' follow‐up was conducted either by their clinic visit or through telephone at the 3rd, 6th, and 12th month and every 3–6 months thereafter after discharge. The outcome events were collected by patients' follow‐up, information from their relatives, or through electronic medical records.
Statistical analysis
Continuous variables were presented as mean ± standard deviation, or median and interquartile range, depending on whether or not they were normally distributed. Continuous normally distributed variables were tested with Student's t‐test, skewed variables were tested with the Mann–Whitney U test, and categorical variables were tested with χ 2 tests or Fisher's exact test as appropriate.
We used Spearman's ρ correlation coefficients to explore the correlation between RAR and other clinical variables and visualized by a correlation heat map. Predictors associated with the composite endpoint were analysed using univariate and multivariate Cox proportional hazards models. Endpoints were assessed with the Kaplan–Meier method and compared with the log‐rank test. Multiple measurements and plots were performed to assess the potential incremental prognostic value of RAR based on a 12‐variable‐based traditional model, which includes age, sex, systolic blood pressure (SBP), haemoglobin, sodium, estimated glomerular filtration rate (eGFR), log2 N‐terminal pro‐brain natriuretic peptide (NT‐proBNP), New York Heart Association (NYHA) class, left ventricular ejection fraction (LVEF), diabetes mellitus, treatment with angiotensin‐converting enzyme inhibitors (ACE‐Is)/angiotensin receptor blockers (ARBs), and treatment with beta‐blockers. The concordance index (C‐index) was employed for assessing model discrimination. Model calibration was evaluated by the Hosmer and Lemeshow test, the Bayesian information criterion (BIC), the Akaike information criterion (AIC), and the Brier score. The net reclassification improvement (NRI) and the integrated discrimination improvement (IDI) were introduced to calculate model reclassification performance. The clinical usefulness of the RAR was assessed by decision curve analysis (DCA). Subgroup analysis was also performed to explore the consistency of the association between RAR and composite endpoints of HF patients.
All analyses were conducted by SPSS 25.0 (IBM, Chicago, IL, USA) and R Version 4.1.3. A two‐tailed P < 0.05 was considered statistically significant.
Results
Baseline characteristics of the study population
A total of 2077 NIHF patients were finally enrolled in this study. The baseline clinical characteristics are shown in Table 1 . The mean age of this cohort was 52.77 ± 15.75 years, and 66.4% were male. Compared with those who did not reach the primary outcomes, patients who died or had heart transplantation tended to have lower BMI, heart rate, and blood pressure; higher RAR, hs‐CRP, NT‐proBNP, and total bilirubin; and lower ALB, haemoglobin, eGFR, and lipid profile. Patients with events showed a significantly worse cardiac function, as shown by the NYHA class and LVEF. In addition, these patients were less likely to have hypertension, hyperlipidaemia, and anaemia or to receive standard HF therapy 15 (P < 0.05).
Table 1.
Baseline characteristics of patients with non‐ischaemic heart failure
| Variable | Total patients (N = 2077) | Without events (N = 1577) | With events (N = 500) | P value |
|---|---|---|---|---|
| Male, n (%) | 1380 (66.4) | 1039 (65.9) | 341 (68.2) | 0.368 |
| Age | 52.77 ± 15.75 | 52.50 ± 15.52 | 53.61 ± 16.44 | 0.172 |
| BMI (kg/m2) | 24.73 ± 4.77 | 25.14 ± 4.82 | 23.49 ± 4.36 | <0.001 |
| BSA (m2) | 1.78 ± 0.25 | 1.79 ± 0.25 | 1.73 ± 0.23 | <0.001 |
| BMI, n (%) | ||||
| Underweight | 138 (7.4) | 91 (6.5) | 47 (10.2) | <0.001 |
| Normal | 595 (31.8) | 407 (28.9) | 188 (40.6) | <0.001 |
| Overweight | 671 (35.8) | 520 (36.9) | 151 (32.6) | 1.000 |
| Obesity | 468 (25.0) | 391 (27.8) | 77 (16.6) | <0.001 |
| Heart rate (b.p.m.) | 82.71 ± 19.08 | 83.52 ± 19.48 | 80.14 ± 17.52 | <0.001 |
| SBP (mmHg) | 118.09 ± 20.97 | 120.74 ± 20.60 | 109.76 ± 19.93 | <0.001 |
| DBP (mmHg) | 72.47 ± 13.93 | 73.94 ± 14.16 | 67.86 ± 12.04 | <0.001 |
| NYHA functional class, n (%) | <0.001 | |||
| I | 38 (1.8) | 38 (2.4) | 0 (0.0) | |
| II | 396 (19.1) | 345 (21.9) | 51 (10.2) | |
| III | 1039 (50.1) | 819 (52.0) | 220 (44.0) | |
| IV | 601 (29.0) | 372 (23.6) | 229 (45.8) | |
| Biochemical indicators | ||||
| White blood cell (×109/L) | 7.15 ± 2.27 | 7.15 ± 2.19 | 7.14 ± 2.54 | 0.943 |
| Neutrophil count (×109/L) | 4.70 ± 1.95 | 4.65 ± 1.87 | 4.86 ± 2.17 | 0.03 |
| Lymphocyte count (×109/L) | 1.77 ± 0.74 | 1.83 ± 0.71 | 1.60 ± 0.81 | <0.001 |
| NLR | 2.54 (1.84, 3.77) | 2.44 (1.78, 3.47) | 3.01 (2.05, 4.68) | <0.001 |
| RAR | 3.41 (3.02, 3.97) | 3.33 (2.98, 3.83) | 3.71 (3.25, 4.47) | <0.001 |
| PLR | 129.63 ± 78.86 | 125.36 ± 65.78 | 143.11 ± 109.40 | <0.001 |
| Hb (g/L) | 140.13 ± 23.90 | 141.59 ± 23.33 | 135.52 ± 25.08 | <0.001 |
| RBC (×1012/L) | 4.67 ± 0.79 | 4.73 ± 0.78 | 4.48 ± 0.80 | <0.001 |
| RDW (%) | 13.80 (12.90, 15.05) | 13.50 (12.80, 14.70) | 14.50 (13.50, 15.90) | <0.001 |
| Haematocrit (%) | 42.05 ± 6.76 | 42.46 ± 6.64 | 40.77 ± 6.98 | <0.001 |
| Platelets (×109/L) | 198.26 ± 71.36 | 202.92 ± 71.33 | 183.58 ± 69.51 | <0.001 |
| Albumin (g/dL) | 4.02 ± 0.53 | 4.06 ± 0.52 | 3.88 ± 0.54 | <0.001 |
| eGFR (mL/min/1.73 m2) | 74.30 ± 25.66 | 75.06 ± 25.03 | 71.89 ± 27.44 | 0.016 |
| hs‐CRP (mg/L) | 3.27 (1.52, 8.79) | 2.88 (1.43, 7.84) | 4.28 (2.05, 11.26) | <0.001 |
| FPG (mmol/L) | 5.08 (4.58, 5.79) | 5.08 (4.60, 5.82) | 5.07 (4.51, 5.70) | 0.107 |
| HbA1c (%) | 6.10 (5.70, 6.70) | 6.10 (5.70, 6.70) | 6.20 (5.80, 6.70) | 0.027 |
| ALT (IU/L) | 25.00 (19.00, 34.00) | 24.00 (19.00, 34.00) | 27.00 (20.75, 36.00) | <0.001 |
| AST (IU/L) | 24.00 (15.00, 39.00) | 24.00 (16.00, 40.00) | 23.00 (14.00, 37.00) | 0.027 |
| TBiL (μmol/L) | 23.40 (16.10, 35.60) | 21.70 (15.40, 32.20) | 30.05 (19.78, 47.52) | <0.001 |
| TG (mmol/L) | 1.29 (0.96, 1.82) | 1.36 (1.01, 1.91) | 1.13 (0.85, 1.54) | <0.001 |
| TC (mmol/L) | 4.13 (3.42, 4.96) | 4.27 (3.55, 5.06) | 3.82 (3.14, 4.61) | <0.001 |
| HDL‐C (mmol/L) | 1.01 ± 0.32 | 1.03 ± 0.32 | 0.95 ± 0.31 | <0.001 |
| LDL‐C (mmol/L) | 2.53 (2.01, 3.19) | 2.59 (2.06, 3.25) | 2.39 (1.84, 2.96) | <0.001 |
| ESR (mm/h) | 7.00 (3.00, 16.00) | 7.00 (2.00, 14.00) | 8.00 (3.00, 19.00) | 0.003 |
| NT‐proBNP (pg/mL) | 1997.80 (898.10, 4463.80) | 1667.30 (761.20, 3761.25) | 3414.00 (1644.10, 6411.05) | <0.001 |
| Echocardiographic index | ||||
| LAD (mm) | 46.00 (41.00, 52.00) | 46.00 (41.00, 51.00) | 49.00 (44.00, 55.00) | <0.001 |
| LVEDD (mm) | 63.31 ± 13.39 | 62.53 ± 12.68 | 65.68 ± 15.11 | <0.001 |
| LVEF (%) | 38.77 ± 14.99 | 39.68 ± 14.63 | 35.99 ± 15.73 | <0.001 |
| Medication | ||||
| Digoxin | 1338 (64.4) | 1036 (65.7) | 302 (60.4) | 0.036 |
| ACE‐I/ARB | 1233 (59.4) | 1020 (64.7) | 213 (42.6) | <0.001 |
| Beta‐blocker | 1723 (83.0) | 1356 (86.0) | 367 (73.4) | <0.001 |
| MRA | 1598 (76.9) | 1235 (78.3) | 363 (72.6) | 0.01 |
| Diuretic | 1732 (83.4) | 1313 (83.3) | 419 (83.8) | 0.83 |
| Comorbidities | ||||
| Hypertension | 838 (40.3) | 684 (43.4) | 154 (30.8) | <0.001 |
| Diabetes | 395 (19.0) | 308 (19.5) | 87 (17.4) | 0.321 |
| Anaemia | 277 (13.3) | 178 (11.3) | 99 (19.8) | <0.001 |
| Hyperlipidaemia | 558 (26.9) | 456 (28.9) | 102 (20.4) | <0.001 |
ACE‐I, angiotensin‐converting enzyme inhibitor; ALT, alanine aminotransferase; ARB, angiotensin receptor blocker; AST, aspartate aminotransferase; BMI, body mass index; BSA, body surface area; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; ESR, erythrocyte sedimentation rate; FPG, fasting plasma glucose; Hb, haemoglobin; hs‐CRP, high‐sensitivity C‐reactive protein; LAD, left atrial diameter; LVEDD, left ventricular end‐diastolic diameter; LVEF, left ventricular ejection fraction; MRA, mineralocorticoid receptor antagonist; NLR, neutrophil‐to‐lymphocyte ratio; NT‐proBNP, N‐terminal pro‐brain natriuretic peptide; NYHA, New York Heart Association; PLR, platelet‐to‐lymphocyte ratio; RAR, RDW‐to‐albumin ratio; RBC, red blood cell; RDW, red blood cell distribution width; SBP, systolic blood pressure; TBiL, total bilirubin; TC, total cholesterol; TG, triglyceride.
Relationship between red blood cell distribution width‐to‐albumin ratio and clinical variables
Spearman's correlation showed that RAR had a positive relation with sex, age, NYHA class, hs‐CRP, NT‐proBNP, left atrial diameter, and anaemia and that BMI, SBP, haemoglobin, eGFR, triglyceride, total cholesterol, left ventricular end‐diastolic diameter, and treatment of ACE‐I/ARB, beta‐blocker, and mineralocorticoid receptor antagonist were negatively correlated with RAR (Supporting Information, Table S1 ). These correlations were visualized by a heat map (Figure 1 ).
Figure 1.
Correlation heat map between red blood cell distribution width (RDW)‐to‐albumin ratio (RAR) and clinical variables. (A) Correlation between RAR and inflammatory markers. (B) Correlation between RAR and clinical variables. ACE‐I, angiotensin‐converting enzyme inhibitor; ALT, alanine aminotransferase; ARB, angiotensin receptor blocker; AST, aspartate aminotransferase; BMI, body mass index; BSA, body surface area; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; ESR, erythrocyte sedimentation rate; FPG, fasting plasma glucose; Hb, haemoglobin; HR, heart rate; hs‐CRP, high‐sensitivity C‐reactive protein; LAD, left atrial diameter; LVEDD, left ventricular end‐diastolic diameter; LVEF, left ventricular ejection fraction; MRA, mineralocorticoid receptor antagonist; NLR, neutrophil‐to‐lymphocyte ratio; NT‐proBNP, N‐terminal pro‐brain natriuretic peptide; NYHA, New York Heart Association; PLR, platelet‐to‐lymphocyte ratio; RBC, red blood cell; SBP, systolic blood pressure; TBiL, total bilirubin; TyG, triglyceride–glucose index.
Predictive ability of red blood cell distribution width‐to‐albumin ratio for the composite endpoint
The median follow‐up time in this study was 1433 (1341, 1525) days. As the gender no longer satisfied the Cox proportional risk assumption after 1150 days, we set 1095 days as the follow‐up time for analysis. A total of 500 (24.1%) patients reached the composite endpoint: 410 patients had all‐cause death, and 90 patients underwent heart transplantation. As shown in Table 2 , RAR was independently associated with a more than four‐fold risk of all‐cause mortality or heart transplantation [hazard ratio (HR) 4.106, 95% confidence interval (CI) 3.268–5.159, per log2 increase, P < 0.001] in NIHF patients. The association remains, even after adjusting for age and sex (Model 2) and other traditional risk factors, such as SBP, haemoglobin, sodium, eGFR, log2 NT‐proBNP, NYHA class, LVEF, diabetes mellitus, treatment with ACE‐Is/ARBs, and treatment with beta‐blockers (Model 3).
Table 2.
The association between red blood cell distribution width‐to‐albumin ratio and composite endpoints of heart failure patients
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| RAR (per log2 increase) | 4.106 (3.268, 5.159) | <0.001 | 4.314 (3.414, 5.451) | <0.001 | 2.329 (1.677, 3.237) | <0.001 |
| RAR tertiles | ||||||
| Tertile 1 (≤3.15) | Reference | Reference | Reference | |||
| Tertile 2 (3.15–3.76) | 1.793 (1.393, 2.306) | 0.004 | 1.798 (1.397, 2.314) | <0.001 | 1.225 (0.923, 1.525) | 0.159 |
| Tertile 3 (≥3.76) | 3.425 (2.705, 4.336) | <0.001 | 3.463 (2.732, 4.391) | <0.001 | 1.703 (1.270, 2.285) | <0.001 |
| P for trend | <0.001 | <0.001 | <0.001 | |||
CI, confidence interval; HR, hazard ratio; RAR, red blood cell distribution width‐to‐albumin ratio.
Model 1: unadjusted. Model 2: age and sex. Model 3: age, sex, systolic blood pressure, haemoglobin, sodium, estimated glomerular filtration rate, log2 N‐terminal pro‐brain natriuretic peptide, New York Heart Association class, left ventricular ejection fraction, diabetes mellitus, treatment with angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers, and treatment with beta‐blockers.
Moreover, when treating RAR as a categorical variable, this association still existed in the NIHF cohort. RAR was divided into three tertiles according to its distribution: RAR ≤ 3.15 as Tertile 1, 3.15 < RAR < 3.76 as Tertile 2, and RAR ≥ 3.76 as Tertile 3. Tertile 1 was set as the reference group. As shown in Figure 2 , the highest tertile had the worst event‐free survival, and the comparison between different tertiles was significant (log‐rank test P < 0.001 for all). Similar to before, higher RAR tertile was also associated with a higher risk of the primary outcome (Table 2 ); however, when adjusted by all 12 traditional risk factors, this association was only pronounced between Tertile 3 and Tertile 1 (HR 1.703, 95% CI 1.270–2.285, P < 0.001; P for trend < 0.001).
Figure 2.
Kaplan–Meier survival curves based on red blood cell distribution width‐to‐albumin ratio (RAR) tertile on event‐free survival.
Predicting value of red blood cell distribution width‐to‐albumin ratio in non‐ischaemic heart failure
RAR could significantly improve model discrimination as the C‐index was higher after RAR was added to the traditional model [C‐index 0.766 (95% CI 0.754–0.778) vs. 0.758 (95% CI 0.746–0.770), P < 0.001]. Better model calibration was also observed as a lower AIC, BIC, and Brier score was obtained (AIC: 1487.3 vs. 1495.74; BIC: 1566.25 vs. 1569.43; Brier score: 1569.43 vs. 1569.43; likelihood ratio test P < 0.001). Calibration plot further strengthened this result (Figure 3 ). Moreover, the performance of model reclassification was also improved after RAR was incorporated, with an IDI of 1.35% (95% CI 0.63–2.07, P < 0.001) and an NRI of 13.73% (95% CI 2.05–27.18, P = 0.034) (Table 3 ). The DCA analysis was conducted to further assess its clinical usefulness of RAR and demonstrated that RAR‐incorporated model (Model 2) had a higher overall net benefit in predicting composite endpoints within a risk threshold probability of 20–55% than that of the traditional model (Figure 4 ).
Figure 3.
Calibration plot for the red blood cell distribution width‐to‐albumin ratio‐incorporated model.
Table 3.
Performance of the models to predict the risk of all‐cause mortality or heart transplantation
| Variable | Model 1 | Model 2 |
|---|---|---|
| Discrimination | ||
| C‐index | 0.758 (0.746, 0.770) | 0.766 (0.754, 0.778) |
| χ 2 test | P < 0.001 | |
| Calibration | ||
| H–L test | χ 2 = 9.180 (P = 0.327) | χ 2 = 9.894 (P = 0.356) |
| Brier score | 0.161 | 0.158 |
| AIC | 1495.74 | 1487.3 |
| BIC | 1569.43 | 1566.25 |
| Likelihood ratio test | P < 0.001 | |
| Reclassification | ||
| IDI (%) | Reference | 1.35 (0.63, 2.07), P < 0.001 |
| NRI (%) | Reference | 13.73 (2.05, 27.18), P = 0.034 |
AIC, Akaike information criterion; BIC, Bayesian information criterion; C‐index, Harrell's concordance index; H–L, Hosmer and Lemeshow test; IDI, integrated discrimination improvement; NRI, net reclassification index.
Model 1: age, sex, systolic blood pressure, haemoglobin, sodium, estimated glomerular filtration rate, log2 N‐terminal pro‐brain natriuretic peptide, New York Heart Association class, left ventricular ejection fraction, diabetes mellitus, treatment with angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers, and treatment with beta‐blockers. Model 2: Model 1 + red blood cell distribution width‐to‐albumin ratio.
Figure 4.
Decision curve analysis for the red blood cell distribution width‐to‐albumin ratio (RAR)‐incorporated model. Model 1: age, sex, systolic blood pressure, haemoglobin, sodium, estimated glomerular filtration rate, log2 N‐terminal pro‐brain natriuretic peptide, New York Heart Association class, left ventricular ejection fraction, diabetes mellitus, treatment with angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers, and treatment with beta‐blockers. Model 2: Model 1 + RAR.
Subgroup analysis
Subgroup analysis (Figure 5 ) revealed that no obvious interaction was observed in the association between the continuous variable of RAR and prognostic impact among different subgroups of sex, NYHA class, eGFR, type of HF, BMI, and complicated with diabetes mellitus or hypertension. However, this association seemed more pronounced in patients with age ≥ 65 years (P = 0.012) and NT‐proBNP ≤ 2763 pg/mL (P = 0.007).
Figure 5.
Subgroup analysis of the association between red blood cell distribution width‐to‐albumin ratio and composite endpoints. Adjusted by age, sex, systolic blood pressure, haemoglobin, sodium, estimated glomerular filtration rate (eGFR), log2 N‐terminal pro‐brain natriuretic peptide (NT‐proBNP), New York Heart Association (NYHA) class, left ventricular ejection fraction, diabetes mellitus, treatment with angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers, and treatment with beta‐blockers. BMI, body mass index; CI, confidence interval; HF, heart failure; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, HF with preserved ejection fraction; HFrEF, HF with reduced ejection fraction; HR, hazard ratio.
Discussion
In this study, we comprehensively assessed the predictive value of RAR in an NIHF cohort. We found that RAR was significantly associated with a higher risk of all‐cause mortality or heart transplantation. Better model discrimination, calibration, and reclassification were also obtained after RAR was added to the traditional model. The RAR‐incorporated model also showed good clinical usefulness in this cohort.
As a simple and inexpensive inflammatory indicator, a high level of RAR has been reported to be associated with poor prognosis in various inflammation‐related diseases. 11 , 12 , 13 , 14 , 17 A cohort study based on data from the MIMIC‐III database evaluated RAR's prognostic effect in short‐ and long‐term mortality of patients with HF. 12 The study showed that the risk of 90 day mortality was up to 1.65 and 2.70 times, respectively, in the second tertile (RAR 4.33–5.44) and the third tertile (RAR > 5.44) compared with the first tertile (RAR < 4.33). Moreover, the risk of 1 year mortality has increased to 1.61 and 2.40 times, respectively, in higher RAR groups. Given the complexity of HF, our study explored the predictive value of RAR in a large Chinese cohort of NIHF, and it was the first one to evaluate the incremental prognostic value of RAR and its clinical usefulness. It was found that adding RAR into the traditional well‐established model significantly improved reclassification performance. DCA showed good clinical benefit in the combined model within the 20–55% threshold range. However, the pathophysiological mechanisms are still not clear.
RDW is a classical indicator used to evaluate the size and variability of circulating erythrocytes. Recently, increased RDW has been associated with poor outcomes in patients with several cardiovascular diseases including HF. 18 Several pathophysiological mechanisms have been proposed, including inflammation, oxidative stress, neurohormonal and adrenergic system activation, undernutrition, ineffective erythropoiesis, or reduced iron mobilization. 7 Inflammation, which has been recognized as a major pathophysiological contributor to HF, may impair bone marrow function with the release of premature erythrocytes into the circulation and increased RDW. 19 RDW was significantly associated with established inflammatory markers such as IL‐6 and CRP. 20
As for RDW and anaemia, nutritional deficiencies and heterozygous haemoglobinopathies are present in many forms of anaemia and are characterized by different degrees of anisocytosis. 21 However, several studies about HF patients have demonstrated the prognostic value of RDW independent of anaemia. 22 , 23 , 24 Similar result was seen in our study that RAR was still an independent risk factor even after adjusting haemoglobin in NIHF patients. ALB may be a reflection of the nutritional status and systemic information. 25 A reduction in serum ALB level may be related to blood volume overload, chronic inflammation, liver congestion, malnutrition, and cachexia, which is associated with poor prognosis in patients with HF. 26 , 27 Therefore, as a combination of RDW and ALB, RAR is a potential novel biomarker that can be easily and quickly obtained and may be a superior tool to other single markers for risk stratification of HF patients. 18
HF is associated with measures of systemic inflammation, and relevant studies have shown the value of inflammatory mediators in evaluating the prognosis of HF patients. 28 In our research, RAR positively correlated with other inflammatory markers such as NLR, PLR, and hs‐CRP (Figure 1 ). NLR and PLR were also treated as novel inflammatory markers. There are also accumulating data indicating their unique role in predicting mortality in patients with HF. 29 , 30 The systemic inflammation status of HF may be the cause of the increase of NLR and PLR. 2 During an overwhelming inflammatory response, lymphocytopaenia and lymphocyte hypoactivity occur due to B‐cell and T‐cell apoptosis, contributing to greater mortality. 31
Moreover, in the subgroup analysis of this study, we found that the association between RAR and outcome was pronounced in NIHF patients with age ≥ 65 years or NT‐proBNP ≤ 2763 pg/mL. That may imply that RAR may be an ideal biomarker for predicting the prognosis of elderly patients, and it should be used with caution when we evaluate patients with higher levels of NT‐proBNP as it might be covered by it. Furthermore, the current study found no significant interaction in the predictive value of RAR across LVEF categories. This result was in line with previous reports demonstrating the correlation between inflammation and poor outcomes in HF independent of LVEF. 32 We are cautious about the findings of subgroup analysis as these results may be influenced by heterogeneity among different populations. These results require further research to be confirmed.
This study had several limitations. First, it was a retrospective single‐centre cohort study; even though we adjusted as many clinically relevant variables as possible and performed subgroup analysis, potential confounders likely remained. Second, only the baseline RAR result was assessed, and we could not further explore the association between the RAR change during hospitalization and prognosis. Finally, our study was mainly focused on Chinese ethnicity, and the generality and applicability of our results may be compromised.
Conclusions
RAR provides significant prognostic value and could further remarkably improve stratification capabilities in NIHF patients.
Funding
This research was supported by the Key Projects in the National Science and Technology 6 Pillar Program of the 13th Five‐Year Plan Period (grant numbers 2017YFC1308300, 2017YFC1308305), Beijing, People's Republic of China, and the CAMS Innovation Fund for Medical Science (grant number 2020‐I2M‐1‐002).
Conflict of interest
The authors report no conflicts of interest in this work.
Supporting information
Table S1. Association between RAR and clinical variables in heart failure patients.
Acknowledgements
We thank all the patients and practitioners who took part in the research.
Zhou, P. , Tian, P.‐C. , Zhai, M. , Huang, Y. , Zhou, Q. , Zhuang, X.‐F. , Liu, H.‐H. , Wang, J.‐X. , Zhang, Y.‐H. , and Zhang, J. (2024) Association between red blood cell distribution width‐to‐albumin ratio and prognosis in non‐ischaemic heart failure. ESC Heart Failure, 11: 1110–1120. 10.1002/ehf2.14628.
Ping Zhou and Peng‐Chao Tian contributed equally to this manuscript.
Contributor Information
Yu‐Hui Zhang, Email: yuhuizhangjoy@163.com.
Jian Zhang, Email: fwzhangjian62@126.com.
References
- 1. Ziaeian B, Fonarow GC. Epidemiology and aetiology of heart failure. Nat Rev Cardiol 2016;13:368‐378. doi: 10.1038/nrcardio.2016.25 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Gaudino M, Di Franco A, Rong LQ, et al. Inflammation in heart failure: JACC state‐of‐the‐art review. J Am Coll Cardiol 2020;75:1324‐1340. doi: 10.1016/j.jacc.2022.04.029 [DOI] [PubMed] [Google Scholar]
- 3. Kažukauskienė I, Baltrūnienė V, Rinkūnaitė I, Žurauskas E, Vitkus D, Maneikienė VV, et al. Inflammation‐related biomarkers are associated with heart failure severity and poor clinical outcomes in patients with non‐ischemic dilated cardiomyopathy. Life (Basel) 2021;11:1006. doi: 10.3390/life11101006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Nymo SH, Hulthe J, Ueland T, McMurray J, Wikstrand J, Askevold ET, et al. Inflammatory cytokines in chronic heart failure: Interleukin‐8 is associated with adverse outcome. Results from CORONA. Eur J Heart Fail 2014;16:68‐75. doi: 10.1093/eurjhf/hft125 [DOI] [PubMed] [Google Scholar]
- 5. Curran FM, Bhalraam U, Mohan M, Singh JS, Anker SD, Dickstein K, et al. Neutrophil‐to‐lymphocyte ratio and outcomes in patients with new‐onset or worsening heart failure with reduced and preserved ejection fraction. ESC Heart Fail 2021;8:3168‐3179. doi: 10.1002/ehf2.13424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ye GL, Chen Q, Chen X, Liu YY, Yin TT, Meng QH, et al. The prognostic role of platelet‐to‐lymphocyte ratio in patients with acute heart failure: A cohort study. Sci Rep 2019;9:10639. doi: 10.1038/s41598-019-47143-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Targoński R, Sadowski J, Starek‐Stelmaszczyk M, Targoński R, Rynkiewicz A. Prognostic significance of red cell distribution width and its relation to increased pulmonary pressure and inflammation in acute heart failure. Cardiol J 2020;27:394‐403. doi: 10.5603/CJ.a2018.0103 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Che R, Huang X, Zhao W, Jiang F, Wu L, Zhang Z, et al. Low serum albumin level as a predictor of hemorrhage transformation after intravenous thrombolysis in ischemic stroke patients. Sci Rep 2017;7:7776. doi: 10.1038/s41598-017-06802-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Grimm G, Haslacher H, Kampitsch T, Endler G, Marsik C, Schickbauer T, et al. Sex differences in the association between albumin and all‐cause and vascular mortality. Eur J Clin Invest 2009;39:860‐865. doi: 10.1111/j.1365-2362.2009.02189.x [DOI] [PubMed] [Google Scholar]
- 10. Gopal DM, Kalogeropoulos AP, Georgiopoulou VV, Tang WW, Methvin A, Smith AL, et al. Serum albumin concentration and heart failure risk: The Health, Aging, and Body Composition Study. Am Heart J 2010;160:279‐285. doi: 10.1016/j.ahj.2010.05.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Huang M, Liu F, Li Z, Liu Y, Su J, Ma M, et al. Relationship between red cell distribution width/albumin ratio and carotid plaque in different glucose metabolic states in patients with coronary heart disease: A RCSCD‐TCM study in China. Cardiovasc Diabetol 2023;22:39. doi: 10.1186/s12933-023-01768-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ni Q, Wang X, Wang J, Chen P. The red blood cell distribution width‐albumin ratio: A promising predictor of mortality in heart failure patients—A cohort study. Clin Chim Acta 2022;527:38‐46. doi: 10.1016/j.cca.2021.12.027 [DOI] [PubMed] [Google Scholar]
- 13. Xu W, Huo J, Chen G, Yang K, Huang Z, Peng L, et al. Association between red blood cell distribution width to albumin ratio and prognosis of patients with sepsis: A retrospective cohort study. Front Nutr 2022;9:1019502. doi: 10.3389/fnut.2022.1019502 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Long J, Xie X, Xu D, Huang C, Liu Y, Meng X, et al. Association between red blood cell distribution width‐to‐albumin ratio and prognosis of patients with aortic aneurysms. Int J Gen Med 2021;14:6287‐6294. doi: 10.2147/IJGM.S328035 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JGF, Coats AJS, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Rev Esp Cardiol (Engl Ed) 2016;69:1167. doi: 10.1016/j.rec.2016.11.005 [DOI] [PubMed] [Google Scholar]
- 16. WHO Expert Consultation . Appropriate body‐mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363:157‐163. doi: 10.1016/S0140-6736(03)15268-3 [DOI] [PubMed] [Google Scholar]
- 17. Lu C, Long J, Liu H, Xie X, Xu D, Fang X, et al. Red blood cell distribution width‐to‐albumin ratio is associated with all‐cause mortality in cancer patients. J Clin Lab Anal 2022;36:e24423. doi: 10.1002/jcla.24423 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Liang L, Huang L, Zhao X, Zhao L, Tian P, Huang B, et al. Prognostic value of RDW alone and in combination with NT‐proBNP in patients with heart failure. Clin Cardiol 2022;45:802‐813. doi: 10.1002/clc.23850 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Xanthopoulos A, Giamouzis G, Dimos A, Skoularigki E, Starling R, Skoularigis J, et al. Red blood cell distribution width in heart failure: Pathophysiology, prognostic role, controversies and dilemmas. J Clin Med 2022;11:1951. doi: 10.3390/jcm11071951 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hullin R, Barras N, Abdurashidova T, Monney P, Regamey J. Red cell distribution width and prognosis in acute heart failure: Ready for prime time! Intern Emerg Med 2019;14:195‐197. doi: 10.1007/s11739-018-1995-7 [DOI] [PubMed] [Google Scholar]
- 21. Lippi G, Turcato G, Cervellin G, Sanchis‐Gomar F. Red blood cell distribution width in heart failure: A narrative review. World J Cardiol 2018;10:6‐14. doi: 10.4330/wjc.v10.i2.6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Pascual‐Figal DA, Bonaque JC, Redondo B, Caro C, Manzano‐Fernandez S, Sánchez‐Mas J, et al. Red blood cell distribution width predicts long‐term outcome regardless of anaemia status in acute heart failure patients. Eur J Heart Fail 2009;11:840‐846. doi: 10.1093/eurjhf/hfp109 [DOI] [PubMed] [Google Scholar]
- 23. Tseliou E, Terrovitis JV, Kaldara EE, Ntalianis AS, Repasos E, Katsaros L, et al. Red blood cell distribution width is a significant prognostic marker in advanced heart failure, independent of hemoglobin levels. Hellenic J Cardiol 2014;55:457‐461. [PubMed] [Google Scholar]
- 24. Bonaque JC, Pascual‐Figal DA, Manzano‐Fernandez S, et al. Red blood cell distribution width adds prognostic value for outpatients with chronic heart failure. Rev Esp Cardiol 2012;65:606‐612. doi: 10.1016/j.recesp.2011.12.006 [DOI] [PubMed] [Google Scholar]
- 25. Soeters PB, Wolfe RR, Shenkin A. Hypoalbuminemia: Pathogenesis and clinical significance. JPEN J Parenter Enteral Nutr 2019;43:181‐193. doi: 10.1002/jpen.1451 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Uthamalingam S, Kandala J, Daley M, Patvardhan E, Capodilupo R, Moore SA, et al. Serum albumin and mortality in acutely decompensated heart failure. Am Heart J 2010;160:1149‐1155. doi: 10.1016/j.ahj.2010.09.004 [DOI] [PubMed] [Google Scholar]
- 27. Ajoolabady A, Nattel S, Lip GYH, Ren J. Inflammasome signaling in atrial fibrillation: JACC state‐of‐the‐art review. J Am Coll Cardiol 2022;79:2349‐2366. doi: 10.1016/j.jacc.2022.03.379 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Doehner W. A direct link between haemodynamic failure and inflammatory activation in heart failure: The simplified approach to heart failure and to creation of life. Eur Heart J 2014;35:413‐415. [DOI] [PubMed] [Google Scholar]
- 29. Arruda‐Olson AM, Reeder GS, Bell MR, Weston SA, Roger VL. Neutrophilia predicts death and heart failure after myocardial infarction: A community‐based study. Circ Cardiovasc Qual Outcomes 2009;2:656‐662. doi: 10.1161/CIRCOUTCOMES.108.831024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Seropian IM, Romeo FJ, Pizarro R, Vulcano NO, Posatini RA, Marenchino RG, et al. Neutrophil‐to‐lymphocyte ratio and platelet‐to‐lymphocyte ratio as predictors of survival after heart transplantation. ESC Heart Fail 2018;5:149‐156. doi: 10.1002/ehf2.12199 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Drewry AM, Samra N, Skrupky LP, Fuller BM, Compton SM, Hotchkiss RS. Persistent lymphopenia after diagnosis of sepsis predicts mortality. Shock 2014;42:383‐391. doi: 10.1097/SHK.0000000000000234 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Shirazi LF, Bissett J, Romeo F, Mehta JL. Role of inflammation in heart failure. Curr Atheroscler Rep 2017;19:27. doi: 10.1007/s11883-017-0660-3 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Association between RAR and clinical variables in heart failure patients.