Introduction
Chronic kidney disease (CKD) is a public health burden worldwide due to its rapidly expanding patient populations, high risk of progression into end-stage kidney disease (ESKD), and poor prognosis of morbidity and mortality [
1,
2]. The leading mortality of these patients is cardiovascular (CV) related deaths. With CKD progression, the CV outcomes become worse, including CV death, re-infarction, congestive heart failure, stroke, and resuscitation [
3]. The most common cause of CKD is diabetes mellitus (DM) [
4]. The 2002 National Cholesterol Education Program report designated DM a coronary heart disease risk equivalent, and DM is placed in the highest risk category [
5]. Furthermore, DM and CKD are both potent independent risk factors for CV events and progression to ESKD [
6,
7]. DM has huge burden of atherosclerosis related intimal thickening and CKD also causes medial calcification [
8]. Therefore, patients with both conditions at the same time are therefore at exceedingly high risk of adverse events and would end with poor patient outcome.
The serum level of uric acid (UA) is also a risk factor for kidney disease [
9], cardiovascular disease (CVD) [
10‐
12], and atherosclerosis [
13]. Serum UA is an independent risk factor for CKD, even in those without diabetes [
14,
15]. Two large epidemiologic studies showed that UA is a major predictor for the incidence of renal disease [
15,
16]. Moreover, hyperuricemia is often prevalent in CKD patients, and that is associated with a higher incidence of ESKD [
16]. A number of studies showed that UA independently predicts the development of type 2 DM [
17‐
19] and the progression of CKD [
20]. For about 20 years, UA is known to be a potential risk factor for CKD and CVD with pathological implications [
21,
22,
23]. Given the complex interplay among hyperuricemia, DM and the progression of CKD, we are interested to explore the complicated interactions regarding renal and patients outcomes. Here, we aimed to investigate the effects of DM and hyperuricemia on patient mortality and the development of ESKD in a large cohort of CKD patients.
Discussion
The principal finding of this study is that the conditions hyperuricemia and DM when occurred together further increased risk of ESKD and all-cause mortality compared with the conditions existed alone. Results were independent of traditional risk factors such as age, gender, BP, and smoking in patients with CKD. Results are consistent with previous studies [
15,
31,
32] showing that hyperuricemia is an independent risk factor for ESKD in the general population and in patients with CKD [
33‐
36]. The higher risk on all-cause mortality and ESKD remained significant after adjustment for multiple confounding factors. However, other epidemiologic studies revealed uncertain conclusion because of differences in methodologies and impact on serum UA concentrations by even subtle changes in kidney function in the general population [
37]. The causal role of serum UA in kidney disease, hypertension, or DM remains debatable regarding the general population [
37]. For patients of CKD stage 3 to 5, we reported earlier in a retrospective study that hyperuricemia is associated with higher risk of incident renal replacement therapy and all-cause mortality [
33]. The potential mechanisms that hyperuricemia contributes to CKD progression include a poorer renal perfusion via stimulation of afferent arteriolar vascular smooth muscle cell proliferation [
16,
38‐
40]. Hyperuricemia may lead to acute UA nephropathy [
41], chronic urate nephropathy [
42], gout related renal injury and anesthesia related nephropathy. Many conditions associated with hyperuricemia in CKD patients could also contribute the progression of CKD.
To our knowledge, this is the first study to show that hyperuricemia is a risk equivalent to DM for all-cause mortality (HR = 1.48 vs. HR = 1.52) and ESKD (HR = 1.34 vs. HR = 1.59) in patients with CKD stage 3 to 5. Gout is a risk factor for CVD [
43], CV mortality [
44] and all-cause mortality [
44,
45]. The possible mechanism is related to hyperuricemia [
46]. Hyperuricemia in the absence of gout has a risk of stroke 1.47 times higher [
47] and a risk of coronary heart disease 1.34 times higher [
48]. From a retrospective study on claims database study, gout has a risk equivalent to DM for the incidence of stroke [
46]. Hyperuricemia is linked to impaired production of nitric oxide [
49,
50] the activation of renin-angiotensin system [
51]. Both of the above factors cause endothelial dysfunction [
52,
53], and further contribute to hypertension [
54,
55] and hyperuricemia [
56,
57]. Studies showed that UA stimulates the proliferation of vascular smooth muscle cells [
58‐
60]. Hyperuricemia-related monosodium urate crystals [
61,
62] may cause atherosclerosis with more coagulation [
63]. Both hyperuricemia and DM are linked to CVD and all-cause mortality. In addition, hyperuricemia-related gout usually requires treatment with nonsteroidal anti-inflammatory drug (NSAID). NSAID is also associated with higher CV mortality and all-cause mortality [
64‐
66]. Finally, the hyperuricemia may be related to the use of diuretics, a medication which is typically used in patients with heart failure with pulmonary edema and unstable heart function. Both low cardiac output and diuretic therapy reduce UA excretion. Hence hyperuricemia is likely a good maker for poor heart function and higher risk for mortality [
67]. In summary, regarding all-cause mortality, hyperuricemia has a risk equivalent to DM.
Hyperuricemia was also a risk equivalent to DM for ESKD in this CKD cohort in Taiwan, which has the highest incidence of ESKD worldwide [
68]. In addition to potential mechanisms that hyperuricemia contributes to CKD progression by reducing renal perfusion via stimulated proliferation of afferent arteriolar vascular smooth muscle cells [
16,
38‐
40], and the over-use of NSAID for gout attack also threatened CKD progression. Once gout attack, patients got used to taking NSAIDs for pain relief even if definite evidence of renal toxicity of NSAIDs. From a Nationwide study in Taiwan (109,400 incident chronic ESKD patients from 1998 to 2009) [
69], adjusted odds ratio (OR) was 2.73 (95% CI: 2.62–2.84) for nonselective NSAIDs and 2.17 (95% CI: 1.83–2.57) for celecoxib. Compared with the non-users, users of oral NSAID were 3.74 times more likely to develop dialysis-required ESKD. This severe renal risk could be even greater for people who had recently used the parenteral form of NSAIDs (adjusted OR: 8.66) [
69]. About 30% dialytic patients still took NSAID 1 year before the initiation of dialysis (2018 Annual Report on Kidney Disease, Taiwan) [
70]. Moreover, the number of patients taking NSAID was likely under-estimated because its over-the-counter availability in Taiwan.
Our patients with both DM and hyperuricemia had more increased risk of ESKD (HR = 2.46) than either DM or hyperuricemia alone. The joint effect was greater than the additive HR (1.59*1.34 = 2.13) of DM and hyperuricemia. The discrepancy may be due to synergistic or potentiating effects. First, DM and hyperuricemia share some similar mechanism for renal injury in patients with CKD, but other UA associated mechanisms for renal injury may be independent from DM (like acute UA nephropathy [
41], chronic urate nephropathy [
42], gout related renal injury and NSAID related nephropathy [
69]). Such additional mechanisms of hyperuricemia and gout-related renal injury could lead to higher risk of ESKD in DM-related CKD. Therefore, even hyperuricemia and DM shared similar mechanisms for CKD progression, the combined risks of DM and hyperuricemia for ESKD appeared higher than DM or hyperuricemia alone. Second, there are several kinds of synergistic effects between DM and hyperuricemia on CKD progression. Initial hyperuricemia is an independent risk factor for the progression of diabetic kidney disease (DKD) [
71]. High serum UA levels potentiate CKD progression in patients with type 2 DM [
72]. Initially, activation of the renin-angiotensin system causes glomerular hyperfiltration [
51,
73], a finding characterizes diabetic kidney disease (DKD) and CKD. Thus, hyperfiltration is potentiated under hyperuricemia in patients with both CKD and type 2 DM. In addition, UA stimulates proliferation of vascular smooth muscle cells and their oxidative stress [
59], leading to progression of CKD. The oxidative stress and inflammation are typical findings of DKD [
74,
75]. Furthermore, UA-related alleles of SLC2A9 rs11722228, SLC2A9 rs3775948, ABCG2 rs2231142 affect DKD susceptibility in the Chinese patients with type 2 DM [
76]. A clinical study on 15-year follow-up supported the contribution of hyperuricemia on CKD progression [
71]. In another clinical study, febuxostat preserves eGFR in patients of DKD, at levels beyond glycemic control [
77]. Low-doses allopurinol reduce the severity of proteinuria in type 2 DM, probably through decreased serum UA [
78]. Therefore, UA can be considered as a mediator of DKD [
79] and the joint effect of hyperuricemia and DM on ESKD could be synergistic.
There are some limitations of our present study. First, causal effect of hyperuricemia on ESKD and all-cause mortality cannot be established. Second, we did not record the gout condition and the usage of NSAID. Third, our results cannot be generalized to patients covering all stages of CKD. Finally, for patients with CKD under critical status, the result cannot be generalized to this population, neither. Despite these limitations, DM and hyperuricemia having joint effect on ESKD and all-cause mortality remains a robust finding.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.