Heart failure (HF) is one of the major global public health issues and a common cause of hospitalization in middle-aged and older adults (1). Traditional and innovative therapies have been used to treat and manage HF, leading to a significant reduction in HF mortality (2, 3). However, the high rate of readmission of HF patients within 3 months of discharge is an emerging problem that can cause a clinical and economic burden. Readmission of HF patients within 3 months after discharge may be seen as a warning sign of a worse subsequent prognosis for the patient. A study from Japan concluded that patients with HF who were readmitted within 3 months had a higher risk of death at long-term follow-up (4); therefore, early identification of this group of patients with HF could optimize treatment more quickly, improve symptoms, increase patient survival, reduce readmission rates, and reduce the economic burden on families and society.

Several variables have been associated with readmission in patients with HF. Apostolos et al. conducted a propensity score-matched observational study that included HF patients with ejection fraction use from the U.S. Medicare-related OPTIMIZE-HF registry from March 2003 to December 2004 and found that systolic blood pressure (SBP) <120 mmHg at discharge was associated with an increased risk of readmission in these patients (5). However, this study included only patients with HF with an ejection fraction of ≥50%; therefore, it is not possible to infer whether an SBP of <120 mmHg at discharge is associated with an increased risk of readmission in all HF patients. Ioanna et al. also studied 671 HF patients from the OPTIMIZE-HF registry (6). They found that psychosocial factors were associated with 1-year readmission risk in HF patients. Still, their study sample size was small, and the questionnaire for assessing psychosocial factors was too cumbersome and complex for clinical replication. Therefore, we need to find an earlier and easier indicator to detect the risk of readmission and better manage this group of inpatients with HF.

Red cell distribution width (RDW), an indicator of changes in red blood cell (RBC) size and shape, is easily obtained from a complete blood count and is a simple, inexpensive measure (7). The RDW typically ranges from 11% to 15%, and a higher RDW value indicates greater variability in size (8). Traditionally, RDW has been used to differentiate the cause of anemia (9). Recent clinical evidence suggests that changes in RBC size are associated with the development and adverse outcomes of non-hematologic diseases, such as stroke, osteoporosis, dementia, and cardiovascular disease (10–13). Especially in patients with HF, RDW has been shown to be associated with first hospitalization and poor prognosis in HF (14, 15). However, most of the available evidence is focused on the United States and Europe, and there are few studies related to Asia, especially China.

Thus, this study aimed to investigate the relationship between RDW and the risk of 3-month readmission in hospitalized Chinese HF patients.

This retrospective cohort study focused on determining the association between RDW and the risk of readmission within 3 months in patients with HF. It suggested that RDW was linearly and positively associated with the risk of readmission within 3 months in middle-aged and elderly patients with HF in China after adjusting for several major confounding factors. In the multivariable-adjusted model, we found that every 1% increase in RDW was correlated with a 9% increased risk of readmission within 3 months (HR = 1.09, 95% CI: 1.03–1.15, P < 0.005) and the highest RDW group (tertile3) had a higher risk of 3-month readmission compared with the lowest group (tertile1). These results were consistent for each subgroup.

RDW is strongly associated with the prognosis of patients with HF. Consistent with our results, previous studies have shown that the RDW at admission is associated with a multifactorial adjusted risk of rehospitalization. Remo et al. showed that patients with high RDW had a risk ratio of 1.55 (95% CI: 1.08–2.22) for death or reoccurrence of HF from any cause compared with patients with a low RDW (20). Kyoung et al. found that readmission rates for older patients were higher in the highest group of the RDW category (7.4% in the lowest group and 15.8% in the highest group, P-trend < 0.001) (21). However, some studies have reported inconsistent results. Tan et al. included patients with HF in China by using multivariate logistic regression models to predict the risk of 3-month readmission and concluded that simple observation of RDW from clinical blood test reports could not be used to visually determine the risk of readmission (22); the area under the red cell volume distribution width curve also confirmed its weaker discriminatory ability (C statistic = 0.61); however, their study included was only 350 patients, and the study population was limited to the HF population in central China. Our study had a larger sample size, and the patients were not limited to central China to further investigate the relationship between RDW and the risk of readmission within 3 months.

RDW is a common clinical measure of heterogeneity in RBC size (23). A higher RDW indicates a defect in the maturation or degradation of blood cells (24). An elevated RDW has been shown to be associated with anemia, inflammatory diseases (e.g., infections and tumors), chronic kidney disease, diabetes mellitus, and cardiovascular diseases (e.g., HF, hypertension, stroke, and atrial fibrillation) (25–30). However, the specific mechanisms underlying the association between RDW and poor prognosis in chronic diseases, including HF, are not fully understood. Multiple interrelated pathological mechanisms, including nutritional deficiency, ineffective erythropoiesis, reduced iron mobilization, oxidative stress, chronic inflammation, adrenergic stimulation, and enhanced immune system activation, are thought to be associated with an increase in RDW and poor prognosis in HF patients (31, 32). The main role of erythrocytes is to transport oxygen and carbon dioxide from the lungs to the tissues and to maintain the acid-base balance of the body (33). Recent studies suggest that erythrocytes are involved in nitric oxide metabolism and have secretory functions (release of nitric oxide, nitric oxide metabolites, and adenosine triphosphate) (33). The mediators released by erythrocytes play a key role in cardiovascular regulation and have a direct impact on the heart. Nutritional deficiencies lead to inadequate erythropoiesis and increased RDW (34). Nutritional factors are critical for improving the medium-to-long-term prognosis of patients with HF. Chronic inflammation and oxidative stress can lead to increased RDW by suppressing effective bone marrow erythropoiesis and increasing RBC variability (35). On the one hand, inflammatory markers (interleukin-6 and tumor necrosis factor-α) disrupt erythropoiesis by directly inhibiting erythroid precursors, affect iron metabolism, and reduce erythropoiesis; it also decreases erythropoietin secretion from the kidney (36, 37). On the other hand, oxidative stress decreases erythrocyte survival and leads to increased circulating premature erythrocytes, resulting in anisocytosis and increased erythrocyte lysis, which leads to increased free radicals and adverse effects on the heart (38, 39). Thus, higher oxidative stress is another potential mechanism linking RDW to a poor prognosis in patients with HF.

Based on the subgroup analysis, we found that the association between RDW and the risk of readmission within 3 months was stronger in men than in women (HR: 1.18 vs. 1.06). Sex differences exist in almost all aspects of HF. First, there are differences in cardiac anatomy between men and women, with women without heart disease having a higher left ventricular ejection fraction than men (40). Women with HF with preserved ejection fraction have smaller, stiffer hearts and more frequent concentric remodeling than men with HF with preserved ejection fraction (41). Second, from a pathophysiological point of view, women with HF with reduced left ventricular ejection fraction have better adaptation of the myocardium to stress and a lower risk of ventricular tachycardia, atrial fibrillation, and sudden cardiac death than men (42, 43). Third, several neurohormonal modulators have been shown to have better therapeutic effects in women with HF, but even before the recent introduction of neurohormonal modulators into HF treatment, clinical outcomes for women with HF with preserved or reduced ejection fraction were consistently better than for men (44). Mechanistic studies to explain these sex-related differences in HF treatment are lacking, and further research is required to verify these phenomena. An analysis of PARADIGM-HF suggested that mortality and hospitalization rates are lower in women for reasons that are not yet clear (45). Moreover, future consideration of gender-based differences in treatment is needed to reduce the risk of readmission.

In the subgroup analysis, we found that the association between RDW and the risk of readmission within 3 months was stronger among patients with more co-morbidities (CCI score ≥3) and poor cardiac function (NYHA cardiac function class IV). Previous studies have demonstrated that co-morbidities are common in hospitalized elderly patients and are associated with long-term prognosis (46). Moreover, other studies have documented co-morbidities and cardiac function class as independent adverse prognostic factors in elderly patients with acute HF (47), consistent with our results.

Our study has the following strengths. First, this study had a larger sample size than previous studies. Second, the effect modification factor analysis allowed better use of the data to draw stable conclusions across different subgroups. However, this study has some limitations. First, because this was an observational study, we could not confirm a causal relationship between RDW and the risk of readmission within 3 months. Second, this study only assessed RDW values at the first admission. Therefore, it is impossible to evaluate the influence of dynamic changes in RDW on the risk of readmission within 3 months. Third, the subjects of this study were all patients with HF in China; therefore, there are some deficiencies in the extrapolation and generalizability of the study.

In conclusion, we found that RDW was positively and linearly related to the risk of readmission within 3 months in patients with HF. Patients with HF with higher RDW values also had a higher risk of readmission within 3 months. Further studies on the ability of RDW to predict the risk of readmission are needed.