Chronic heart failure (CHF) is one of the major causes of mortality and affects 1%-2% adults worldwide (1, 2). Approximately 74% of CHF patients have at least one non-cardiac comorbidity (3), which increase the medical complexity and mortality rates. Chronic kidney disease (CKD) is a progressive condition characterized by decreased estimated glomerular filtration rate (eGFR) and/or presence of proteinuria. More than 50% of deaths in patients with CKD are resulted from cardiovascular disease (4). CKD is highly prevalent in CHF with the comorbidity rate ranging from 33% to 48% (3, 5, 6). A meta-analysis including 57 studies concluded that patients with the cooccurrence of CKD and CHF experienced nearly 2-fold higher risk of mortality (7). Hyperuricemia is one of the major metabolic diseases that caused by both an increased production and a decreased excretion of uric acid (UA). Various studies have shown that hyperuricemia was associated with higher risk of mortality in general population and patients with cardiovascular disease (8–10). Hyperuricemia affects about half of patients with CHF (11), and accounts for increased risk of hospitalization, cardiovascular mortality, and all-cause mortality regardless of the ejection fraction (EF) phenotype of CHF (12, 13).

Hyperuricemia is a common finding in CKD due to reduced excretion of UA (14), and also an independent risk factor for the development of CKD (15). The comorbidity of hyperuricemia and CKD poses worse outcomes in general population (16–18). Although, similar findings were not generated in several studies based on patients with CHF (19, 20). In the Beta-Blocker Evaluation of Survival Trial, hyperuricemia is associated with hospitalization and mortality in CHF patients without CKD, but not those with CKD (19). In addition, CHART-2 study and EVEREST study demonstrated that elevated serum uric acid is not associated with mortality in CHF patients with eGFR <60ml/min/1.73m² and CHF patients with eGFR <30ml/min/1.73m² (20, 21). However, these previous studies did not account for proteinuria in the definition of CKD, which have limited information on the association between hyperuricemia, CKD, and adverse outcomes. Moreover, their findings should be validated in other populations because the characteristics of heart failure vary greatly across different regions and ethnicities (22, 23). Therefore, the aim of our study was to comprehensively examine the joint association of hyperuricemia and CKD with in-hospital mortality and long-term mortality in a population of hospitalized CHF patients in China.

In the current study, we identified 9,429 CHF patients who were hospitalized between 1 January 2011 and 30 June 2019. 18 patients were excluded due to the age of <18 years, 1 patient was excluded due to pregnancy, and 43 patients were excluded due to the diagnosis of acute kidney injury. A total of 9,367 CHF patients were included in the in-hospital mortality analysis ( Figure 1 ). Of these, the mean age was 62.64 ± 14.61 years, 6,244 (66.66%) were men, 275 (2.94%) had in-hospital mortality (182 were CV mortality), and 1,323 (14.12%) lost to follow-up. Finally, a total of 7,769 CHF patients were included in the long-term mortality analysis ( Figure 1 ). The patients lost to follow-up were younger, and had a lower proportion of hyperuricemia but similar proportion of CKD ( Supplementary Table 1 ). Patients in the hyperuricemia+/CKD- group were younger and more likely to be men. They also had higher BMI, heart rate, and higher prevalence of HFrEF, atrial fibrillation, and valvular heart disease, while patients in the hyperuricemia-/CKD+ group had higher blood pressure, and higher prevalence of diabetes, stroke, anemia, chronic pulmonary disease, and connective tissue disease. Those in the hyperuricemia+/CKD+ group had higher serum concentration of NT-proBNP and higher prevalence of NYHA class IV heart failure ( Table 1 ). Patients included in the long-term mortality analysis exhibited similar characteristics ( Supplementary Table 2 ).

Hyperuricemia+/CKD+ group had the highest incidence rate of in-hospital mortality (5.80%), followed by hyperuricemia-/CKD+ group (4.28%), hyperuricemia+/CKD- group (2.10%), and hyperuricemia-/CKD- group (1.41%) ( Table 2 ). After adjusting for potential confounders, the risk of in-hospital mortality was twice as high in the hyperuricemia+/CKD+ group compared with the hyperuricemia-/CKD- group (OR [95% CI], 2.12 [1.46-3.08]). Those with hyperuricemia alone and CKD alone also exhibited a significantly higher risk of in-hospital mortality compared with the hyperuricemia-/CKD- group (OR [95% CI], 1.58 [1.01-2.46] and 1.67 [1.10-2.55], respectively) ( Table 2 ). Analyses of asymptomatic hyperuricemia and gout, defining CKD as only proteinuria, using serum UA >420 μmol/L in men and >360 μmol/L in women as the cutoff point to define hyperuricemia all generated similar results ( Table 2 ). Similar findings were also observed in the CV mortality analyses ( Supplementary Table 3 ).

During the mean follow-up of 4.81 years, a total of 2,883 deaths were identified. Among them, 1,433 were CV death, 1,059 were non-CV death, and 391 deaths could not be classified due to lack of information. The hyperuricemia+/CKD+ group had the highest incidence rate of long-term mortality among all 4 groups ( Table 2 ). Compared with the hyperuricemia-/CKD- group, the adjusted HRs and 95% CI for long-term mortality were 1.25 (1.11-1.41) for the hyperuricemia+/CKD- group, 1.37 (1.22-1.53) for the hyperuricemia-/CKD+ group, and 1.59 (1.43-1.76) for the hyperuricemia+/CKD+ group ( Table 2 ). The results did not materially change in analyses of asymptomatic hyperuricemia and gout, analyses defining CKD as only reduced eGFR and only proteinuria, sensitivity analyses, and CV mortality analyses ( Table 2 and Supplementary Table 3 ).

The long-term mortality risk was proportional to serum UA level, regardless of the CKD status or CKD subtypes ( Table 3 ). Although, further analyses showed that the association of serum UA with in-hospital mortality was more evident in patients with eGFR ≥60 ml/min/1.73m² (P-interaction =0.02), it was not significant in patients with eGFR <60ml/min/1.73m² ( Table 3 ). The association with long-term mortality was more evident in patients without proteinuria compared with those with proteinuria (P-interaction =0.02) ( Table 3 ). Furthermore, the in-hospital and long-term mortality risk increased with the upgrade of risk classification of CKD progression, and did not vary between patients with and without hyperuricemia (P-interaction =0.29 for in-hospital mortality and 0.55 for long-term mortality) ( Table 4 ).

In the subgroup analyses, we did not find significant interactions between hyperuricemia/CKD groups and age, sex, hypertension, EF phenotype, and NYHA classification in relation to both in-hospital mortality and long-term mortality (P-interaction >0.10 for all). However, the association between hyperuricemia/CKD groups and in-hospital mortality appeared to be more pronounced in CHF patients without myocardial infarction (P-interaction =0.05) ( Figures 2 , 3 ).

In this retrospective cohort study, we found that hyperuricemia and CKD were individually and cumulatively associated with increased risk of in-hospital mortality and long-term mortality in patients with CHF. These findings persisted after adjusting for a wide spectrum of confounding factors and in sensitivity analyses. Moreover, the risks of both in-hospital mortality and long-term mortality were proportional to the serum UA level and risk classification of CKD progression. However, serum UA had a weaker association with in-hospital mortality in patients with eGFR <60 ml/min/1.73m², and a weaker association with long-term mortality in patients with proteinuria.

In recent years, the prognostic role of hyperuricemia in patients with heart failure has garnered particular interest. Findings from Metabolic Exercise Cardiac Kidney Index score database showed that each 60 μmol/L increase in UA was associated with 12% increased risk of mortality in patients with HFrEF (36), whereas Mantovani et al. found that hyperuricemia increased the mortality risk by 37% in general patients with CHF (12). CKD has a bidirectional relationship with hyperuricemia (37, 38), and may mediate the association between hyperuricemia and health outcomes (16). However, to the best of our knowledge, only a few studies examined the combined effect of hyperuricemia and CKD on mortality risk in patients with CHF, but generated inconsistent results with our study (19–21). For example, a study of 1,260 systolic heart failure patients from the Beta-Blocker Evaluation of Survival Trial found that hyperuricemia was associated with a 1.4-fold higher risk of mortality in patients without CKD, but not in those with CKD (19). However, this previous study only included patients with NYHA class III and IV systolic heart failure who were screened for participation in a clinical trial, while our study was performed in a real-world setting and included patients with all phenotypes of CHF. Thus, the study patients in our study might be more representative. In another cohort study including 4,652 patients with CHF, the association between serum UA and all-cause mortality was significantly modified by eGFR, and hyperuricemia was not associated with outcome in patients with CKD (20). Vaduganathan et al. also found that serum UA (per 300 μmol/L increase) was not associated with increased hazard of all-cause mortality in CHF patients with eGFR <30 ml/min/1.73m² (21). Notably, the definition of CKD in these studies did not account for proteinuria, which could not provide complete information on renal dysfunction. In the current study, we used a more complete definition of CKD, and observed that elevated serum UA was associated with increased risk of mortality in CHF patients with and without CKD, low eGFR, and proteinuria. Taken together, our study provided a more comprehensive understanding regarding the association between hyperuricemia, CKD, and mortality in patients with CHF by using well-characterized information and larger sample size.

It is well-known that CKD contributes to worse outcomes in patients with CHF by uremic toxins, fluid overload, enhancing systemic inflammation and oxidative stress, and activating the sympathetic nervous system and renin-angiotensin-aldosterone system (39–41). Several mechanisms may explain why the presence of both CKD and hyperuricemia cause a cumulative increase in risk of mortality. First, hyperuricemia may cause renal tubular injury, and accelerate the progression of CKD (42, 43), hence increases mortality risk. Second, increased xanthine oxidase activity, a major cause of hyperuricemia, can promote oxidative stress and upregulation of inflammatory cytokines, which further result in cardiac fibrosis and left ventricular dysfunction (44). It is still conflicting that whether UA is a direct risk factor of cardiovascular outcomes or just a passive marker of xanthine oxidase activity. However, in the current study, the risk of long-term mortality related to hyperuricemia was similar between patients with eGFR ≥60 ml/min/1.73m² (hyperuricemia mainly caused by reduced UA excretion) and patients with eGFR <60 ml/min/1.73m² (hyperuricemia mainly caused by increased xanthine oxidase activity). Recent studies of sodium-glucose cotransporter 2 inhibitors (SGLT2i) have suggested that the beneficial effect of SGLT2i on heart failure outcomes is mediated by their enhancement of renal UA excretion (45). Thus, elevated serum UA may directly contribute to the adverse outcomes in patients with CHF. Potential mechanisms include UA-induced vascular fibrosis, vascular inflammation, endothelial dysfunction, and nitric oxide reduction (46).

The findings of our study provide important clinical implications. Hyperuricemia is commonly recognized as a bystander in patients with CHF and CKD. However, our study showed that hyperuricemia and CKD cumulatively increased the risk of mortality in patients with CHF. This suggests that measurement of serum UA concentration might provide useful information on the assessment of outcome risk, regardless of kidney function. Moreover, these findings support the importance of including UA management in the treatment strategies of CHF regardless of the cause of hyperuricemia (increased production or decreased excretion). There is also a need to strengthen the patient education regarding limiting their intake of purine-rich food. Several novel medications for heart failure (e.g., angiotensin receptor neprilysin inhibitors, SGLT2i) have been demonstrated to lower serum UA concentration, which may be associated with improved outcomes (45, 47). Further clinical trials are warranted to assess the effect of UA-lowering treatment on adverse health outcomes in patients with CHF, CKD, and hyperuricemia.

The strengths of our study included the cohort design, large sample size, and well-characterized data on a broad spectrum of biological and demographic fields. However, several limitations should be noted. First, our study was conducted in a single center, although, over 70% of the study patients were from other 30 provinces of China. Therefore, the patients included in our study were still nationally representative, but may not be completely generalizable to other countries. Second, 14.12% of the patients were lost to follow-up, which may have introduced bias into the long-term mortality analysis. However, the proportion of CKD and many other basic characteristics were comparable between patients who lost to follow-up and those included in the long-term mortality analysis ( Supplementary Table 1 ). Therefore, the bias could be small. Third, due to the limitation of telephonic interview, we did not collect information on specific causes of death in all study patients. Fourth, data on serum UA, eGFR, and proteinuria were only obtained at baseline and were not assessed over time. Since these are all modifiable factors, the longitudinal trajectories of these parameters and their association with mortality in patients with CHF warrant investigation in the future.

In conclusion, hyperuricemia and CKD individually and cumulatively increase the risk of in-hospital mortality and long-term mortality in patients with CHF. These findings provided further evidence demonstrating the combined effect of hyperuricemia and CKD on heart failure outcomes, and highlighted the importance of the management of serum UA in CHF patients, with or without CKD.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the ethics committee of the Chinese People’s Liberation Army General Hospital. The ethics committee waived the requirement of written informed consent for participation.

Conceptualization: CW, WD, YB, and KH; Data curation: CW, HC, YZ, RW, DZ, LC, CR, QZ, ZL, YD, and JX; Formal analysis: CW and HC; Funding acquisition: KH; Investigation: CW, HC, and YZ; Methodology: CW and HC; Project administration: KH; Writing-original draft: CW and HC; Writing-review and editing: All authors. All authors contributed to the article and approved the submitted version.