- Email: [email protected]
Association of plasma xanthine oxidoreductase activity with severity and clinical outcome in patients with chronic heart failure
International Journal of Cardiology 228 (2017) 151–157
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Association of plasma xanthine oxidoreductase activity with severity and clinical outcome in patients with chronic heart failure Yoichiro Otaki a, Tetsu Watanabe a,⁎, Daisuke Kinoshita a, Miyuki Yokoyama a, Tetsuya Takahashi a, Taku Toshima a, Takayuki Sugai a, Takayo Murase b, Takashi Nakamura c, Satoshi Nishiyama a, Hiroki Takahashi a, Takanori Arimoto a, Tetsuro Shishido a, Takuya Miyamoto a, Isao Kubota a a b c
Department of Cardiology, Pulmonology, and Nephrology, Yamagata University School of Medicine, Yamagata, Japan Radioisotope and Chemical Analysis Center, Laboratory Management Department, Sanwa Kagaku Kenkyusho Co., Ltd., Mie, Japan Pharmacological Study Group, Pharmaceutical Research Laboratories, Sanwa Kagaku Kenkyusho Co., Ltd., Mie, Japan
a r t i c l e
i n f o
Article history: Received 22 July 2016 Accepted 5 November 2016 Available online 09 November 2016 Keywords: Xanthine oxidoreductase Uric acid Heart failure Cardiac prognosis
a b s t r a c t Background: Oxidative stress due to purine degradation is associated with the development of chronic heart failure (CHF). Xanthine oxidoreductase (XOR) is a rate-limiting enzyme of purine degradation that plays a key role in uric acid (UA) production with a resultant increase in reactive oxygen species. However, the relationship between plasma XOR activity and CHF severity and clinical outcome remains unclear. Methods and results: We measured XOR activity in 440 patients with CHF and 44 control subjects. Abnormally high and low XOR activities were identified based on the results for 95% of the control subjects (high and low XOR activities ≥120 and b33 pmol/100 μL/h, respectively). The prevalence rates of high and low XOR activities increased with advancing New York Heart Association functional class. There were 158 cardiac events during a median follow-up period of 1034 days. Multivariate Cox proportional hazard regression analysis showed that both high and low XOR activities were significantly associated with cardiac events in patients with CHF after adjustment for confounding risk factors including serum UA and loop diuretic use. Kaplan–Meier analysis revealed that the cardiac event rate was significantly higher in patients with either high or low XOR activity. The net reclassification index was significantly improved by adding XOR activity to the basic risk factors. Conclusions: We provide the first evidence of an association of plasma XOR activity with CHF severity and clinical outcome. Plasma XOR activity could be used to identify high-risk CHF patients and could be a therapeutic target for XOR inhibitors. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Chronic heart failure (CHF) is a major and increasing public health problem with a high mortality rate [1]. Oxidative stress is closely associated with CHF development and the major cause of excessive oxidative stress in heart failure is considered to be increased levels of reactive oxygen species (ROS) [2,3]. Xanthine oxidoreductase (XOR) is a rate-limiting enzyme of the last step of purine degradation in nucleic acid metabolism [4,5]. When XOR catalyzes the oxidation of hypoxanthine to xanthine and xanthine to uric acid (UA), ROS are generated in this process. Therefore, XOR is recognized as a significant source of ROS contributing the development of oxidative stress-related tissue injury [4]. Both XOR level and activity in myocardial tissue are reportedly increased in animal heart failure
⁎ Corresponding author at: Department of Cardiology, Pulmonology and Nephrology, Yamagata University School of Medicine, 2-2-2 Iida-Nishi, Yamagata 990-9585, Japan. E-mail address: [email protected] (T. Watanabe).
http://dx.doi.org/10.1016/j.ijcard.2016.11.077 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
models [6,7]. Circulating XOR can exacerbate oxidative stress in myocardial tissues, and levels are increased in patients with acute phase myocardial infarction [8–10]. Hyperuricemia increases the risk of heart failure through XORdependent ROS production and is considered a therapeutic target [11]. As XOR activity is very low in healthy human [12,13], no existing studies have examined plasma XOR activity in patients with CHF. The aims of the present study were to (1) investigate the association between XOR activity and UA level, and (2) assess the prognostic importance of XOR activity in patients with CHF. 2. Methods 2.1. Study subjects This was a prospective observational study to elucidate the clinical utility of plasma XOR activity in patients with CHF. We included 440 patients who were admitted to our hospital for the diagnosis or treatment of CHF, as well 44 age- and gender-matched control subjects without heart disease. The diagnosis of CHF was made by two cardiologists who used the generally accepted Framingham criteria, including a history of dyspnea,
152
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
symptomatic exercise intolerance with signs of pulmonary congestion or peripheral edema, and radiological or echocardiographic evidence of left ventricular enlargement or dysfunction [14]. Transthoracic echocardiography was performed by physicians who were blinded to the biochemical data. The diagnoses of hypertension, diabetes mellitus, and hyperlipidemia were established on the basis of medical records or history of medical therapy. Exclusion criteria included acute coronary syndrome within 3 months preceding admission, active hepatic disease, pulmonary disease, or malignant disease. Demographic and clinical data including age, gender, New York Heart Association (NYHA) functional class, and medications at discharge were collected from patients' medical records and interviews.
conditions in the remaining 227 (51%) patients. CKD and acidic urine were identified in 189 (43%) and 133 (30%), respectively. As XOR activity in CHF was not normally distributed, we defined the reference interval as 33–120 pmol/100 μL/h based on the results for 95% of control subjects. The CHF patients were divided into three groups according to XOR activity: low XOR (b33 pmol/100 μL/h, n = 57), normal XOR (33–120 pmol/100 μL/h, n = 268), and high XOR (≥120 pmol/100μL/h, n = 115). The prevalence rates of high and low XOR activities increased with advancing NYHA functional class (Fig. 1).
2.2. Biochemical markers
3.2. Comparisons of clinical characteristics among XOR groups
Venous blood and urine samples were obtained in the early morning within 24 h after admission. XOR activity assay was performed using stable isotope-labeled substrate and liquid chromatography high-resolution mass spectrometry (Sanwa Kagaku Kenkyusho Co., ltd, Japan) [15]. The XOR detection limit was ≥33 pmol/100 μL/h. Because XOR activity was not normally distributed, the reference interval was defined as 33–120 pmol/100 μL/h based on the results for 95% of the control subjects. Serum UA, creatinine, and urine pH were measured at the same time. Each subject's estimated glomerular filtration rate (eGFR) was calculated with the modification of diet in renal disease equation using the Japanese coefficient according to the Kidney Disease Outcomes Quality Initiative (K/DOQI) clinical guidelines [16,17]. Blood samples were obtained to measure brain natriuretic peptide (BNP) levels. Once collected, the samples were transferred to chilled tubes containing 4.5 mg ethylenediaminetetraacetic acid disodium salt and aprotinin (500 U/mL) and centrifuged at 1000 ×g for 15 min at 4 °C. The clarified plasma samples were frozen, stored at −70 °C, and thawed immediately before the assay. BNP concentrations were measured using a commercially available radioimmunoassay specific for human BNP (Shiono RIA BNP assay kit, Shionogi Co. Ltd., Tokyo, Japan) [18,19]. 2.3. Endpoints and follow-up Patients were prospectively followed for a median of 1034 days (interquartile range 613 to 1743). Patients were followed by telephone or medical record twice a year until 2555 days after discharge. Nine patients were lost to follow-up because they moved or we were unable to contact them. The primary end points were cardiac events including progressive heart failure requiring rehospitalization, acute coronary syndrome, and cardiac death defined as death due to progressive heart failure, acute coronary syndrome, or sudden cardiac death. Sudden cardiac death was defined as death without definite premonitory symptoms or signs, and was diagnosed by the attending physician after patients were transported to the emergency department. The secondary endpoint was cardiovascular death. There were 16 non-cardiac deaths during the follow-up period. The study was approved by the institutional ethics committee and all patients provided written informed consent to participate. 2.4. Statistical analysis Continuous variable normality was checked with Shapiro–Wilk tests. Due to the nonnormal distribution of BNP, we used log10-transformed BNP values in all analyses. XOR activity was also non-normally distributed, so for the analyses we divided the patients into three groups based on the XOR activity reference interval. All values are expressed as means ± SD or median (interquartile range). We performed t-tests and chi-square tests to compare continuous and categorical variables, respectively. The association between CHF severity and XOR activity was analyzed by chi-square test. The association between serum UA and plasma XOR activity was assessed with analysis of variance (ANOVA). The Cox proportional hazard analysis was carried out to identify independent predictors for cardiac events. Predictors that were significant in the univariate analysis were entered into the multivariate analysis. Cardiac event-free curves were constructed according to the Kaplan–Meier method and compared using log-rank tests. We calculated the C index, net reclassification index (NRI), and integrated discrimination index (IDI) to measure the quality of improvement for correct reclassification following the addition of XOR activity to the model. P b 0.05 was considered statistically significant. Statistical analyses were performed using standard software packages (JMP version 11; SAS Institute Inc., Cary, NC and R 3.0.2 software with additional packages including Rcmdr, Epi, pROC, and PredictABEL).
3. Results 3.1. Baseline characteristics of control subjects and CHF patients The patients' baseline characteristics are presented in Table 1. There were 266 and 174 patients in NYHA functional classes II and III/IV, respectively. Hypertension, diabetes mellitus, and hyperlipidemia were identified in 355 (81%), 116 (26%), and 115 (26%) patients, respectively. The etiology of heart failure was ischemic heart disease in 104 (24%) patients, dilated cardiomyopathy in 109 (25%), and other
As shown in Table 2, patients in the high XOR group were in a more severe NYHA functional class and had higher levels of BNP and lower left ventricular ejection fraction (LVEF) compared to the normal XOR group. The prevalence rate of dilated cardiomyopathy was greater in the high XOR group compared to normal XOR group. On the other hand, the low XOR group was older and in a more severe NYHA functional class than the normal XOR group. In addition, the low XOR group had higher levels of BNP and creatinine and lower levels of body mass index and eGFR compared to the normal XOR group. There was no significant difference with regard to gender, prevalence rates of hypertension, diabetes mellitus, hyperlipidemia, or acidic urine, UA level, or medications excluding loop diuretics. 3.3. UA level and XOR activity in patients with CHF Serum UA levels were not significantly different among the three XOR groups (Table 2). However, serum UA was higher in males and patients with CKD, acidic urine, and those prescribed loop diuretics (Fig. 2A–D). Since serum UA is affected by UA excretion, we examined whether XOR activity is related to serum UA in patients without CKD or acidic urine. The high XOR group had a higher mean serum UA Table 1 Clinical characteristics of 44 control subjects and 440 chronic heart failure patients. Variables
Control n = 44
All patients n = 440
P value
Age (years old) Male/female Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) NYHA II/III Etiology IHD/DCM/others Biochemical data Log BNP (pg/mL) Creatinine (mg/dL) eGFR (mL/1.73/m2) Urine pH UA (mg/dL) XOR (pmol/100 μL/h) Acidic urine, n (%) CKD, n (%) High/normal/low XOR ratio, n Echocardiographic data LVEDD (mm) LVEF (%) Medication ACEIs and/or ARBs, n (%) β-Blockers, n (%) Loop diuretics, n (%) Aldosterone blockers, n (%) XOR inhibitors, n (%) Statins, n (%)
67 ± 8 23/21
70 ± 12 269/171 355 (81%) 116 (26%) 115 (26%) 266/174
0.1005 0.1503
104/109/227
5.3 ± 1.4 79.7 (50.7–95.9)
1/42/1
2.48 ± 0.54 1.01 ± 0.94 65 ± 25 6.2 ± 0.8 6.4 ± 2.3 73.2 (43.6–121.5) 133 (30%) 189 (43%) 115/268/57
0.0013
b0.0001
56 ± 10 48 ± 18 313 (71%) 286 (65%) 272 (62%) 124 (28%) 54 (12%) 102 (23%)
Data are expressed as mean ± SD, number (percentage). ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; CKD, chronic kidney disease; DCM, dilated cardiomyopathy; eGFR, estimated glomerular filtration rate; IHD, ischemic heart disease; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; UA, uric acid; XOR, xanthine oxidoreductase.
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
compared with patients in the low and normal XOR groups without CKD or acidic urine (Fig. 2E–F). 3.4. Clinical outcomes and XOR activity There were 158 cardiac events including 60 cardiac deaths during the study period. Univariate and multivariate Cox proportional hazard regression analyses were performed to determine the impact of XOR activity on cardiac prognosis. Both high and low XOR activities were significantly related to cardiac events in patients with CHF. Also, age, male gender, NYHA functional class, log10 BNP, creatinine, eGFR, UA, acidic urine, CKD, left ventricular end diastolic diameter (LVEDD), LVEF, and loop diuretics use were related to future cardiac events (Table 3). Multivariate Cox proportional hazard regression analysis showed that both high and low XOR activities were significantly associated with cardiac events after adjusting for age, male gender, NYHA functional class, log10 BNP, eGFR, UA, acidic urine, LVEF, and loop diuretic use (high XOR activity, hazard ratio, 2.77; 95% confidence interval [CI], 1.88–4.08, P b 0.0001; low XOR activity, hazard ratio, 1.93; 95% CI, 1.17– 3.20, P = 0.0107; Table 3). Kaplan–Meier analysis demonstrated that the cardiac event rate and cardiac mortality were greater in the high XOR group compared to the normal XOR group (Fig. 3A–B). Kaplan– Meier analysis also showed that the cardiac event rate was higher in the low XOR group compared to the normal XOR group, but cardiac mortality was not different between the low and normal XOR groups (Fig. 3C–D). 3.5. Improving reclassification by adding XOR activity to predict cardiac events To examine whether model fit and discrimination were improved by adding XOR to the basic predictors of future cardiac events, we evaluated C index differences and improvements in NRI and IDI. According to multivariate Cox proportional hazard regression analysis, age, male gender, NYHA functional class, log10 BNP, eGFR, UA, acidic urine, and LVEF were entered into the baseline prediction model. The C index in the prediction model with XOR activity was significantly greater than that in the baseline model (0.746 vs. 0.807, P = 0.0006; Fig. 4 and Table 4). In addition, the prediction model with XOR activity significantly improved NRI and IDI compared to the baseline model (NRI, 0.2589, 95% CI, 0.1147–0.4031, P = 0.0004; IDI, 0.1080, 95% CI, 0.0767–0.1394, P b 0.0001; Table 4). To more precisely examine the relationship between XOR activity and cardiac event, all patients were divided into six groups according to XOR sextile: first sextile (≤ 36.2 pmol/100 μL/h, n = 74), second sextile (36.2–50.7 pmol/100 μL/h, n = 73), third sextile (50.7– 72.8 pmol/100 μL/h, n = 73), fourth sextile (72.8–101.2 pmol/100 μL/h, n = 73), fifth sextile (101.2–169.1 pmol/100 μL/h, n = 73), sixth sextile
153
Table 2 Comparisons of clinical characteristics among high, normal, and low XOR groups.
Variables Age (years old) Male/female BMI (kg/m2) Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) NYHA II/III–IV Etiology IHD/DCM/others Biochemical data Log BNP (pg/mL) Creatinine (mg/dL) eGFR (mL/1.73/m2) Urine pH UA (mg/dL) Acidic urine, n (%) CKD, n (%) Echocardiographic data LVEDD (mm) LVEF (%) Medication ACEIs and/or ARBs, n (%) β-Blockers, n (%) Loop diuretics, n (%) Aldosterone blockers, n (%) XOR inhibitors, n (%) Statins, n (%)
Low XOR n = 57
Normal XOR n = 268
High XOR n = 115
75 ± 11⁎ 38/19 20.8 ± 3.8⁎ 48 (84%) 13 (23%) 20 (35%) 28/29†
71 ± 11 158/110 22.4 ± 3.4 216 (81%) 67 (25%) 62 (23%) 179/89
68 ± 14 73/42 22.7 ± 4.0 91 (79%) 36 (31%) 33 (29%) 59/56†
17/10/30
54/62/152
33/37/45†
2.60 ± 0.54⁎ 1.48 ± 1.81⁎ 53 ± 25⁎ 6.4 ± 0.9 6.3 ± 2.4 16 (28%) 34 (60%)†
2.42 ± 0.55 0.91 ± 0.56 68 ± 26 6.2 ± 0.8 6.2 ± 2.3 73 (27%) 108 (40%)
2.55 ± 0.58⁎ 1.01 ± 0.98 64 ± 22 6.1 ± 0.8 6.8 ± 2.2 44 (38%) 47 (41%)
57 ± 9 52 ± 18
55 ± 10 48 ± 17
57 ± 9 44 ± 17⁎
40 (70%) 39 (68%) 43 (75%)† 16 (28%) 10 (17%) 20 (35%)
191 (71%) 172 (64%) 152 (57%) 72 (27%) 31 (12%) 55 (20%)
82 (71%) 75 (65%) 77 (67%) 36 (31%) 13 (11%) 27 (23%)
Data are expressed as mean ± SD, number (percentage). ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; BMI, body mass index; CKD, chronic kidney disease; DCM, dilated cardiomyopathy; eGFR, estimated glomerular filtration rate; IHD, ischemic heart disease; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; VHD, valvular heart disease; UA, uric acid; XOR, xanthine oxidoreductase. ⁎ P b 0.05 vs. normal XOR group by ANOVA with Scheffe post hoc test. † P b 0.05 vs. normal XOR group by chi-square test.
(N169.1 pmol/100 μL/h, n = 74). Our analysis revealed a U-shaped relationship between XOR activity and the cardiac event rate in patients with CHF (Fig. 5). 4. Discussion 4.1. Main findings In the present study, (1) the prevalence rates of high and low XOR activities increased with advancing NYHA functional class, (2) serum UA levels were not different among XOR groups, but higher UA was noted in the high XOR group in patients without CKD or acidic urine, (3) multivariate logistic analysis revealed that abnormal XOR activity level was closely associated with poor clinical outcomes in patients with CHF, (4) the prediction model including XOR activity had an improved C index, NRI, and IDI, and (5) there was a U-shape relationship between XOR activity and the cardiac event rate. 4.2. UA level and XOR activity in CHF
Fig. 1. The association between XOR activity and NYHA functional class. NYHA, New York Heart Association; XOR, xanthine oxidoreductase.
Serum UA level is reportedly affected by several factors including gender, CKD, acidic urine, and loop diuretic use [20]. Although XOR plays a pivotal role in UA production, a significant association between serum UA level and XOR activity was not observed in CHF patients, in particular those with CKD or acidic urine. Since UA excretion is dependent on acidic urine and kidney dysfunction [21,22], it is plausible that serum UA level is affected by UA excretion rather than XORdependent UA production in these conditions. XOR activity, rather than serum UA level, is reported to be actively involved in hemodynamic impairment in patients with CHF [11].
154
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
Fig. 2. The association between serum UA and XOR activity. Serum UA level in the setting of male gender (A), CKD (B), acidic urine (C), and loop diuretic use (D). Association between serum UA and XOR activity in patients with and without CKD (E) or those with and without acidic urine (F). CKD, chronic kidney disease; UA, uric acid; XOR, xanthine oxidoreductase.
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
155
Table 3 Univariate and multivariate Cox proportional hazard analyses of predicting cardiac events in patients with chronic heart failure. Univariate analysis
Multivariate analysis
Variables
HR
95% CI
P value
HR
95% CI
P value
Age (years old) Male vs. female Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) NYHA III/IV vs. II Log BNPa Creatininea eGFRa UAa Acidic urine CKD High XOR vs. normal XOR Low XOR vs. normal XOR LVEDDa LVEFa Loop diuretics use
1.02 1.58 1.01 1.15 1.02 2.97 1.73 1.09 0.67 1.35 1.47 1.63 3.56 3.02 1.22 0.81 3.02
1.01–1.04 1.13–2.23 0.67–1.50 0.81–1.63 0.71–1.45 2.16–4.08 1.46–2.05 0.99–1.20 0.57–0.80 1.16–1.58 1.06–2.03 1.19–2.23 2.50–5.06 1.95–4.67 1.03–1.45 0.68–0.96 2.01–4.46
0.0014 0.0082 0.9494 0.4297 0.9300 b0.0001 b0.0001 0.0933 b0.0001 0.0002 0.0191 0.0023 b0.0001 b0.0001 0.0212 0.0165 b0.0001
1.01 1.94
1.00–1.03 1.31–2.89
0.0834 0.0011
1.75 1.43
1.21–2.54 1.15–1.78
0.0031 0.0012
0.95 1.03 1.06
0.78–1.13 0.84–1.26 0.73–1.54
0.5420 0.7749 0.7568
2.77 1.93
1.88–4.08 1.17–3.20
b0.0001 0.0107
1.09 1.89
0.91–1.33 1.23–2.89
0.3129 0.0037
CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HR, hazard ratio; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; UA, uric acid; XOR, xanthine oxidoreductase. a Per 1-SD increase.
Filippatos et al. concluded that serum UA level is not associated with cardiac events in patients with comorbid CHF and CKD because serum UA level does not represent XOR activity in patients with CKD [22]. Furthermore, plasma XOR activity is an independent predictor of cardiac events in patients with CKD [23]. These findings support our contention that plasma XOR activity provides useful clinical information in addition to serum UA level in patients with CHF.
4.3. XOR activity and clinical outcome in CHF Accumulating evidence indicates that XOR activity in myocardial tissue is upregulated and closely associated with the development of CHF [5]. XOR-induced ROS generation induces several pathophysiological processes such as myocardial apoptosis and impaired sarcomere contraction through abnormal ryanodine receptor oxidation, finally leading
Fig. 3. Impact of plasma XOR activity on cardiac events in patients with CHF. Kaplan–Meier analyses of high XOR activity to predict all cardiac events (A) and cardiac mortality (B) in patients with CHF. Kaplan–Meier analyses of low XOR activity to predict all cardiac events (C) and cardiac mortality (D) in patients with CHF. CHF, chronic heart failure; XOR, xanthine oxidoreductase.
156
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
Fig. 5. The relationship between plasma XOR activity sextile and cardiac event rate. Fig. 4. Receiver operating characteristic curve of the baseline model and those including XOR activity in patients with CHF. CHF, chronic heart failure; XOR, xanthine oxidoreductase.
to cardiac dysfunction [24,25]. Similarly, the high XOR group had more severe CHF and cardiac systolic dysfunction compared to subjects with levels in the reference range. Our results represent the first evidence that high XOR activity is closely associated with poor clinical outcomes in patients with CHF after adjusting for confounding risk factors. Furthermore, the prediction model significantly improved by adding XOR activity, suggesting that it could be a useful parameter to predict future cardiac events in patients with CHF. Surprisingly, low XOR activity is also related to CHF severity and clinical outcome. The patients with low XOR had lower body mass index compared to other groups, suggesting the relationship between low XOR activity and cardiac cachexia. It was reported that cancer induced cachexia decreases protein synthesis in the liver [26]. Thus, low XOR activity may result from a decrease in XOR protein synthesis secondary to CHF induced cachexia. In addition, it should be noted that patients with low XOR had several other risk factors for poor clinical outcomes including advanced age, severe CHF, high BNP, and CKD.
5. Study limitations As this analysis was performed at a single center and XOR activity is highly population dependent, a validation study is needed to better delineate the prognostic value of XOR activity. Second, further investigations are required to determine the pathological mechanism by which abnormal XOR activity deteriorates cardiac prognosis. Finally, although the mean LVEF in the present study was equivalent to those reported in the Japanese heart failure study and heart failure registry [27,28], it was relatively high compared to that seen in Western countries [29]. 6. Conclusions We provide the first evidence of an association between plasma XOR activity with CHF severity and clinical outcome. Measuring XOR activity could help identify high-risk patients and XOR may be a useful therapeutic target in CHF. Conflict of interest None.
4.4. Clinical perspectives Acknowledgment of grant support Considering the role of XOR, XOR inhibitor therapy could be beneficial in the setting of CHF, but it is still under discussion. Some reports raise the possibility that the cardio-protective effects of XOR inhibitors depend on XOR activity, rather than hyperuricemia, in patients with CHF [11,22]. Similarly, we found that high XOR activity is superior prognostic parameter to serum UA level in patients with CHF. XOR inhibitors may not necessarily be beneficial for CHF patients with hyperuricemia if it is not due to heightened XOR activity. Therefore, measuring plasma XOR activity is important and may identify those patients with CHF likely to respond to XOR inhibitors.
This work was in part supported by the consigned research fund from Sanwa Kagaku Kenkyusho Co., Ltd. References [1] D.M. Lloyd-Jones, M.G. Larson, E.P. Leip, A. Beiser, R.B. D'Agostino, W.B. Kannel, et al., Lifetime risk for developing congestive heart failure: the Framingham Heart Study, Circulation 106 (2002) 3068–3072. [2] J.M. Hare, Oxidative stress and apoptosis in heart failure progression, Circ. Res. 89 (2001) 198–200.
Table 4 Statistics for model fit and improvement with the addition of XOR activity on the prediction of cardiac events. Baseline model
C index (P value) 0.746
NRI (95% CI, P value) Reference
IDI (95% CI, P value) Reference
+XOR activity
0.807 (P = 0.0006)
0.2589 (0.1147–0.4031, P = 0.0004)
0.108 (0.0767–0.1394, P b 0.0001)
Baseline model includes age, gender, NYHA functional class, log BNP, eGFR, UA, acidic urine, LVEF. BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; IDI, integrated discrimination index; LVEF, left ventricular ejection fraction; NYHA functional class, New York Heart Association functional class; NRI, net reclassification index; 95% CI, 95% confidence interval; UA, uric acid; XOR, xanthine oxidoreductase.
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157 [3] Y. Otaki, H. Takahashi, T. Watanabe, A. Funayama, S. Netsu, Y. Honda, et al., HECTtype ubiquitin E3 ligase ITCH interacts with thioredoxin-interacting protein and ameliorates reactive oxygen species-induced cardiotoxicity, 2016. J. Am. Heart Assoc. 5. http://dx.doi.org/10.1161/jaha.115.002485. [4] M.G. Battelli, A. Bolognesi, L. Polito, Pathophysiology of circulating xanthine oxidoreductase: new emerging roles for a multi-tasking enzyme, Biochim. Biophys. Acta 2014 (1842) 1502–1517. [5] P. Pacher, A. Nivorozhkin, C. Szabo, Therapeutic effects of xanthine oxidase inhibitors: renaissance half a century after the discovery of allopurinol, Pharmacol. Rev. 58 (2006) 87–114. [6] J.W. de Jong, R.G. Schoemaker, R. de Jonge, P. Bernocchi, E. Keijzer, R. Harrison, et al., Enhanced expression and activity of xanthine oxidoreductase in the failing heart, J. Mol. Cell. Cardiol. 32 (2000) 2083–2089. [7] L.C. Amado, A.P. Saliaris, S.V. Raju, S. Lehrke, M. St John, J. Xie, et al., Xanthine oxidase inhibition ameliorates cardiovascular dysfunction in dogs with pacing-induced heart failure, J. Mol. Cell. Cardiol. 39 (2005) 531–536. [8] S. Tan, S. Gelman, J.K. Wheat, D.A. Parks, Circulating xanthine oxidase in human ischemia reperfusion, South. Med. J. 88 (1995) 479–482. [9] Y. Yokoyama, J.S. Beckman, T.K. Beckman, J.K. Wheat, T.G. Cash, B.A. Freeman, et al., Circulating xanthine oxidase: potential mediator of ischemic injury, Am. J. Phys. 258 (1990) G564–G570. [10] O.S. Ali, H.M. Abdelgawad, M.S. Mohammed, R.R. El-Awady, Ischemic heart diseases in Egypt: role of xanthine oxidase system and ischemia-modified albumin, Heart Vessel. 29 (2014) 629–637. [11] J.M. Hare, B. Mangal, J. Brown, C. Fisher Jr., R. Freudenberger, W.S. Colucci, et al., Impact of oxypurinol in patients with symptomatic heart failure. Results of the OPTCHF study, J. Am. Coll. Cardiol. 51 (2008) 2301–2309. [12] M.A. Newaz, N.N. Adeeb, Detection of xanthine oxidase in human plasma, Med. J. Malaysia 53 (1998) 70–75. [13] T. Murase, M. Nampei, M. Oka, N. Ashizawa, K. Matsumoto, A. Miyachi, et al., Xanthine oxidoreductase activity assay in tissues using stable isotope-labeled substrate and liquid chromatography high-resolution mass spectrometry, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 1008 (2016) 189–197. [14] P.A. McKee, W.P. Castelli, P.M. McNamara, W.B. Kannel, The natural history of congestive heart failure: the Framingham study, N. Engl. J. Med. 285 (1971) 1441–1446. [15] T. Murase, M. Oka, M. Nampei, A. Miyachi, T. Nakamura, A highly sensitive assay for xanthine oxidoreductase activity using a combination of [C, N]xanthine and liquid chromatography/triple quadrupole mass spectrometry, J. Label. Compd. Radiopharm. 1008 (2016) 189–197. [16] K.U. Eckardt, J.S. Berns, M.V. Rocco, B.L. Kasiske, Definition and classification of CKD: the debate should be about patient prognosis—a position statement from KDOQI and KDIGO, Am. J. Kidney Dis. 53 (2009) 915–920.
157
[17] K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratificationAm. J. Kidney Dis. 39 (2002) S1-266. [18] Y. Otaki, T. Watanabe, H. Takahashi, A. Hirayama, T. Narumi, S. Kadowaki, et al., Association of heart-type fatty acid-binding protein with cardiovascular risk factors and all-cause mortality in the general population: the Takahata study, PLoS One 9 (2014) e94834. [19] Y. Otaki, H. Takahashi, T. Watanabe, S. Kadowaki, T. Narumi, Y. Honda, et al., Electrocardiographic left ventricular hypertrophy Cornell product is a feasible predictor of cardiac prognosis in patients with chronic heart failure, Clin. Res. Cardiol. 103 (2014) 275–284. [20] Y. Otaki, T. Watanabe, H. Takahashi, H. Hasegawa, S. Honda, A. Funayama, et al., Acidic urine is associated with poor prognosis in patients with chronic heart failure, Heart Vessel. 28 (2013) 735–741. [21] A. Kanbara, M. Hakoda, I. Seyama, Urine alkalization facilitates uric acid excretion, Nutr. J. 9 (2010) 45. [22] G.S. Filippatos, M.I. Ahmed, J.D. Gladden, M. Mujib, I.B. Aban, T.E. Love, et al., Hyperuricaemia, chronic kidney disease, and outcomes in heart failure: potential mechanistic insights from epidemiological data, Eur. Heart J. 32 (2011) 712–720. [23] B. Gondouin, N. Jourde-Chiche, M. Sallee, L. Dou, C. Cerini, A. Loundou, et al., Plasma xanthine oxidase activity is predictive of cardiovascular disease in patients with chronic kidney disease, independently of uric acid levels, Nephron 131 (2015) 167–174. [24] D.B. Sawyer, D.A. Siwik, L. Xiao, D.R. Pimentel, K. Singh, W.S. Colucci, Role of oxidative stress in myocardial hypertrophy and failure, J. Mol. Cell. Cardiol. 34 (2002) 379–388. [25] D.R. Gonzalez, A.V. Treuer, J. Castellanos, R.A. Dulce, J.M. Hare, Impaired Snitrosylation of the ryanodine receptor caused by xanthine oxidase activity contributes to calcium leak in heart failure, J. Biol. Chem. 285 (2010) 28938–28945. [26] A.A. Narsale, R.T. Enos, M.J. Puppa, S. Chatterjee, E.A. Murphy, R. Fayad, et al., Liver inflammation and metabolic signaling in ApcMin/+ mice: the role of cachexia progression, PLoS One 10 (2015) e0119888. [27] N. Shiba, K. Nochioka, M. Miura, H. Kohno, H. Shimokawa, Trend of westernization of etiology and clinical characteristics of heart failure patients in Japan—first report from the CHART-2 study, Circ. J. 75 (2011) 823–833. [28] Y. Otaki, T. Watanabe, H. Takahashi, A. Funayama, D. Kinoshita, M. Yokoyama, et al., Comorbid renal tubular damage and hypoalbuminemia exacerbate cardiac prognosis in patients with chronic heart failure, Clin. Res. Cardiol. 105 (2016) 162–171. [29] W. von Scheidt, C. Zugck, M. Pauschinger, R. Hambrecht, O. Bruder, A. Hartmann, et al., Characteristics, management modalities and outcome in chronic systolic heart failure patients treated in tertiary care centers: results from the EVIdence based TreAtment in Heart Failure (EVITA-HF) registry, Clin. Res. Cardiol. 103 (2014) 1006–1014.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Association of plasma xanthine oxidoreductase activity with severity and clinical outcome in patients with chronic heart failure Yoichiro Otaki a, Tetsu Watanabe a,⁎, Daisuke Kinoshita a, Miyuki Yokoyama a, Tetsuya Takahashi a, Taku Toshima a, Takayuki Sugai a, Takayo Murase b, Takashi Nakamura c, Satoshi Nishiyama a, Hiroki Takahashi a, Takanori Arimoto a, Tetsuro Shishido a, Takuya Miyamoto a, Isao Kubota a a b c
Department of Cardiology, Pulmonology, and Nephrology, Yamagata University School of Medicine, Yamagata, Japan Radioisotope and Chemical Analysis Center, Laboratory Management Department, Sanwa Kagaku Kenkyusho Co., Ltd., Mie, Japan Pharmacological Study Group, Pharmaceutical Research Laboratories, Sanwa Kagaku Kenkyusho Co., Ltd., Mie, Japan
a r t i c l e
i n f o
Article history: Received 22 July 2016 Accepted 5 November 2016 Available online 09 November 2016 Keywords: Xanthine oxidoreductase Uric acid Heart failure Cardiac prognosis
a b s t r a c t Background: Oxidative stress due to purine degradation is associated with the development of chronic heart failure (CHF). Xanthine oxidoreductase (XOR) is a rate-limiting enzyme of purine degradation that plays a key role in uric acid (UA) production with a resultant increase in reactive oxygen species. However, the relationship between plasma XOR activity and CHF severity and clinical outcome remains unclear. Methods and results: We measured XOR activity in 440 patients with CHF and 44 control subjects. Abnormally high and low XOR activities were identified based on the results for 95% of the control subjects (high and low XOR activities ≥120 and b33 pmol/100 μL/h, respectively). The prevalence rates of high and low XOR activities increased with advancing New York Heart Association functional class. There were 158 cardiac events during a median follow-up period of 1034 days. Multivariate Cox proportional hazard regression analysis showed that both high and low XOR activities were significantly associated with cardiac events in patients with CHF after adjustment for confounding risk factors including serum UA and loop diuretic use. Kaplan–Meier analysis revealed that the cardiac event rate was significantly higher in patients with either high or low XOR activity. The net reclassification index was significantly improved by adding XOR activity to the basic risk factors. Conclusions: We provide the first evidence of an association of plasma XOR activity with CHF severity and clinical outcome. Plasma XOR activity could be used to identify high-risk CHF patients and could be a therapeutic target for XOR inhibitors. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Chronic heart failure (CHF) is a major and increasing public health problem with a high mortality rate [1]. Oxidative stress is closely associated with CHF development and the major cause of excessive oxidative stress in heart failure is considered to be increased levels of reactive oxygen species (ROS) [2,3]. Xanthine oxidoreductase (XOR) is a rate-limiting enzyme of the last step of purine degradation in nucleic acid metabolism [4,5]. When XOR catalyzes the oxidation of hypoxanthine to xanthine and xanthine to uric acid (UA), ROS are generated in this process. Therefore, XOR is recognized as a significant source of ROS contributing the development of oxidative stress-related tissue injury [4]. Both XOR level and activity in myocardial tissue are reportedly increased in animal heart failure
⁎ Corresponding author at: Department of Cardiology, Pulmonology and Nephrology, Yamagata University School of Medicine, 2-2-2 Iida-Nishi, Yamagata 990-9585, Japan. E-mail address: [email protected] (T. Watanabe).
http://dx.doi.org/10.1016/j.ijcard.2016.11.077 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
models [6,7]. Circulating XOR can exacerbate oxidative stress in myocardial tissues, and levels are increased in patients with acute phase myocardial infarction [8–10]. Hyperuricemia increases the risk of heart failure through XORdependent ROS production and is considered a therapeutic target [11]. As XOR activity is very low in healthy human [12,13], no existing studies have examined plasma XOR activity in patients with CHF. The aims of the present study were to (1) investigate the association between XOR activity and UA level, and (2) assess the prognostic importance of XOR activity in patients with CHF. 2. Methods 2.1. Study subjects This was a prospective observational study to elucidate the clinical utility of plasma XOR activity in patients with CHF. We included 440 patients who were admitted to our hospital for the diagnosis or treatment of CHF, as well 44 age- and gender-matched control subjects without heart disease. The diagnosis of CHF was made by two cardiologists who used the generally accepted Framingham criteria, including a history of dyspnea,
152
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
symptomatic exercise intolerance with signs of pulmonary congestion or peripheral edema, and radiological or echocardiographic evidence of left ventricular enlargement or dysfunction [14]. Transthoracic echocardiography was performed by physicians who were blinded to the biochemical data. The diagnoses of hypertension, diabetes mellitus, and hyperlipidemia were established on the basis of medical records or history of medical therapy. Exclusion criteria included acute coronary syndrome within 3 months preceding admission, active hepatic disease, pulmonary disease, or malignant disease. Demographic and clinical data including age, gender, New York Heart Association (NYHA) functional class, and medications at discharge were collected from patients' medical records and interviews.
conditions in the remaining 227 (51%) patients. CKD and acidic urine were identified in 189 (43%) and 133 (30%), respectively. As XOR activity in CHF was not normally distributed, we defined the reference interval as 33–120 pmol/100 μL/h based on the results for 95% of control subjects. The CHF patients were divided into three groups according to XOR activity: low XOR (b33 pmol/100 μL/h, n = 57), normal XOR (33–120 pmol/100 μL/h, n = 268), and high XOR (≥120 pmol/100μL/h, n = 115). The prevalence rates of high and low XOR activities increased with advancing NYHA functional class (Fig. 1).
2.2. Biochemical markers
3.2. Comparisons of clinical characteristics among XOR groups
Venous blood and urine samples were obtained in the early morning within 24 h after admission. XOR activity assay was performed using stable isotope-labeled substrate and liquid chromatography high-resolution mass spectrometry (Sanwa Kagaku Kenkyusho Co., ltd, Japan) [15]. The XOR detection limit was ≥33 pmol/100 μL/h. Because XOR activity was not normally distributed, the reference interval was defined as 33–120 pmol/100 μL/h based on the results for 95% of the control subjects. Serum UA, creatinine, and urine pH were measured at the same time. Each subject's estimated glomerular filtration rate (eGFR) was calculated with the modification of diet in renal disease equation using the Japanese coefficient according to the Kidney Disease Outcomes Quality Initiative (K/DOQI) clinical guidelines [16,17]. Blood samples were obtained to measure brain natriuretic peptide (BNP) levels. Once collected, the samples were transferred to chilled tubes containing 4.5 mg ethylenediaminetetraacetic acid disodium salt and aprotinin (500 U/mL) and centrifuged at 1000 ×g for 15 min at 4 °C. The clarified plasma samples were frozen, stored at −70 °C, and thawed immediately before the assay. BNP concentrations were measured using a commercially available radioimmunoassay specific for human BNP (Shiono RIA BNP assay kit, Shionogi Co. Ltd., Tokyo, Japan) [18,19]. 2.3. Endpoints and follow-up Patients were prospectively followed for a median of 1034 days (interquartile range 613 to 1743). Patients were followed by telephone or medical record twice a year until 2555 days after discharge. Nine patients were lost to follow-up because they moved or we were unable to contact them. The primary end points were cardiac events including progressive heart failure requiring rehospitalization, acute coronary syndrome, and cardiac death defined as death due to progressive heart failure, acute coronary syndrome, or sudden cardiac death. Sudden cardiac death was defined as death without definite premonitory symptoms or signs, and was diagnosed by the attending physician after patients were transported to the emergency department. The secondary endpoint was cardiovascular death. There were 16 non-cardiac deaths during the follow-up period. The study was approved by the institutional ethics committee and all patients provided written informed consent to participate. 2.4. Statistical analysis Continuous variable normality was checked with Shapiro–Wilk tests. Due to the nonnormal distribution of BNP, we used log10-transformed BNP values in all analyses. XOR activity was also non-normally distributed, so for the analyses we divided the patients into three groups based on the XOR activity reference interval. All values are expressed as means ± SD or median (interquartile range). We performed t-tests and chi-square tests to compare continuous and categorical variables, respectively. The association between CHF severity and XOR activity was analyzed by chi-square test. The association between serum UA and plasma XOR activity was assessed with analysis of variance (ANOVA). The Cox proportional hazard analysis was carried out to identify independent predictors for cardiac events. Predictors that were significant in the univariate analysis were entered into the multivariate analysis. Cardiac event-free curves were constructed according to the Kaplan–Meier method and compared using log-rank tests. We calculated the C index, net reclassification index (NRI), and integrated discrimination index (IDI) to measure the quality of improvement for correct reclassification following the addition of XOR activity to the model. P b 0.05 was considered statistically significant. Statistical analyses were performed using standard software packages (JMP version 11; SAS Institute Inc., Cary, NC and R 3.0.2 software with additional packages including Rcmdr, Epi, pROC, and PredictABEL).
3. Results 3.1. Baseline characteristics of control subjects and CHF patients The patients' baseline characteristics are presented in Table 1. There were 266 and 174 patients in NYHA functional classes II and III/IV, respectively. Hypertension, diabetes mellitus, and hyperlipidemia were identified in 355 (81%), 116 (26%), and 115 (26%) patients, respectively. The etiology of heart failure was ischemic heart disease in 104 (24%) patients, dilated cardiomyopathy in 109 (25%), and other
As shown in Table 2, patients in the high XOR group were in a more severe NYHA functional class and had higher levels of BNP and lower left ventricular ejection fraction (LVEF) compared to the normal XOR group. The prevalence rate of dilated cardiomyopathy was greater in the high XOR group compared to normal XOR group. On the other hand, the low XOR group was older and in a more severe NYHA functional class than the normal XOR group. In addition, the low XOR group had higher levels of BNP and creatinine and lower levels of body mass index and eGFR compared to the normal XOR group. There was no significant difference with regard to gender, prevalence rates of hypertension, diabetes mellitus, hyperlipidemia, or acidic urine, UA level, or medications excluding loop diuretics. 3.3. UA level and XOR activity in patients with CHF Serum UA levels were not significantly different among the three XOR groups (Table 2). However, serum UA was higher in males and patients with CKD, acidic urine, and those prescribed loop diuretics (Fig. 2A–D). Since serum UA is affected by UA excretion, we examined whether XOR activity is related to serum UA in patients without CKD or acidic urine. The high XOR group had a higher mean serum UA Table 1 Clinical characteristics of 44 control subjects and 440 chronic heart failure patients. Variables
Control n = 44
All patients n = 440
P value
Age (years old) Male/female Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) NYHA II/III Etiology IHD/DCM/others Biochemical data Log BNP (pg/mL) Creatinine (mg/dL) eGFR (mL/1.73/m2) Urine pH UA (mg/dL) XOR (pmol/100 μL/h) Acidic urine, n (%) CKD, n (%) High/normal/low XOR ratio, n Echocardiographic data LVEDD (mm) LVEF (%) Medication ACEIs and/or ARBs, n (%) β-Blockers, n (%) Loop diuretics, n (%) Aldosterone blockers, n (%) XOR inhibitors, n (%) Statins, n (%)
67 ± 8 23/21
70 ± 12 269/171 355 (81%) 116 (26%) 115 (26%) 266/174
0.1005 0.1503
104/109/227
5.3 ± 1.4 79.7 (50.7–95.9)
1/42/1
2.48 ± 0.54 1.01 ± 0.94 65 ± 25 6.2 ± 0.8 6.4 ± 2.3 73.2 (43.6–121.5) 133 (30%) 189 (43%) 115/268/57
0.0013
b0.0001
56 ± 10 48 ± 18 313 (71%) 286 (65%) 272 (62%) 124 (28%) 54 (12%) 102 (23%)
Data are expressed as mean ± SD, number (percentage). ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; CKD, chronic kidney disease; DCM, dilated cardiomyopathy; eGFR, estimated glomerular filtration rate; IHD, ischemic heart disease; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; UA, uric acid; XOR, xanthine oxidoreductase.
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
compared with patients in the low and normal XOR groups without CKD or acidic urine (Fig. 2E–F). 3.4. Clinical outcomes and XOR activity There were 158 cardiac events including 60 cardiac deaths during the study period. Univariate and multivariate Cox proportional hazard regression analyses were performed to determine the impact of XOR activity on cardiac prognosis. Both high and low XOR activities were significantly related to cardiac events in patients with CHF. Also, age, male gender, NYHA functional class, log10 BNP, creatinine, eGFR, UA, acidic urine, CKD, left ventricular end diastolic diameter (LVEDD), LVEF, and loop diuretics use were related to future cardiac events (Table 3). Multivariate Cox proportional hazard regression analysis showed that both high and low XOR activities were significantly associated with cardiac events after adjusting for age, male gender, NYHA functional class, log10 BNP, eGFR, UA, acidic urine, LVEF, and loop diuretic use (high XOR activity, hazard ratio, 2.77; 95% confidence interval [CI], 1.88–4.08, P b 0.0001; low XOR activity, hazard ratio, 1.93; 95% CI, 1.17– 3.20, P = 0.0107; Table 3). Kaplan–Meier analysis demonstrated that the cardiac event rate and cardiac mortality were greater in the high XOR group compared to the normal XOR group (Fig. 3A–B). Kaplan– Meier analysis also showed that the cardiac event rate was higher in the low XOR group compared to the normal XOR group, but cardiac mortality was not different between the low and normal XOR groups (Fig. 3C–D). 3.5. Improving reclassification by adding XOR activity to predict cardiac events To examine whether model fit and discrimination were improved by adding XOR to the basic predictors of future cardiac events, we evaluated C index differences and improvements in NRI and IDI. According to multivariate Cox proportional hazard regression analysis, age, male gender, NYHA functional class, log10 BNP, eGFR, UA, acidic urine, and LVEF were entered into the baseline prediction model. The C index in the prediction model with XOR activity was significantly greater than that in the baseline model (0.746 vs. 0.807, P = 0.0006; Fig. 4 and Table 4). In addition, the prediction model with XOR activity significantly improved NRI and IDI compared to the baseline model (NRI, 0.2589, 95% CI, 0.1147–0.4031, P = 0.0004; IDI, 0.1080, 95% CI, 0.0767–0.1394, P b 0.0001; Table 4). To more precisely examine the relationship between XOR activity and cardiac event, all patients were divided into six groups according to XOR sextile: first sextile (≤ 36.2 pmol/100 μL/h, n = 74), second sextile (36.2–50.7 pmol/100 μL/h, n = 73), third sextile (50.7– 72.8 pmol/100 μL/h, n = 73), fourth sextile (72.8–101.2 pmol/100 μL/h, n = 73), fifth sextile (101.2–169.1 pmol/100 μL/h, n = 73), sixth sextile
153
Table 2 Comparisons of clinical characteristics among high, normal, and low XOR groups.
Variables Age (years old) Male/female BMI (kg/m2) Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) NYHA II/III–IV Etiology IHD/DCM/others Biochemical data Log BNP (pg/mL) Creatinine (mg/dL) eGFR (mL/1.73/m2) Urine pH UA (mg/dL) Acidic urine, n (%) CKD, n (%) Echocardiographic data LVEDD (mm) LVEF (%) Medication ACEIs and/or ARBs, n (%) β-Blockers, n (%) Loop diuretics, n (%) Aldosterone blockers, n (%) XOR inhibitors, n (%) Statins, n (%)
Low XOR n = 57
Normal XOR n = 268
High XOR n = 115
75 ± 11⁎ 38/19 20.8 ± 3.8⁎ 48 (84%) 13 (23%) 20 (35%) 28/29†
71 ± 11 158/110 22.4 ± 3.4 216 (81%) 67 (25%) 62 (23%) 179/89
68 ± 14 73/42 22.7 ± 4.0 91 (79%) 36 (31%) 33 (29%) 59/56†
17/10/30
54/62/152
33/37/45†
2.60 ± 0.54⁎ 1.48 ± 1.81⁎ 53 ± 25⁎ 6.4 ± 0.9 6.3 ± 2.4 16 (28%) 34 (60%)†
2.42 ± 0.55 0.91 ± 0.56 68 ± 26 6.2 ± 0.8 6.2 ± 2.3 73 (27%) 108 (40%)
2.55 ± 0.58⁎ 1.01 ± 0.98 64 ± 22 6.1 ± 0.8 6.8 ± 2.2 44 (38%) 47 (41%)
57 ± 9 52 ± 18
55 ± 10 48 ± 17
57 ± 9 44 ± 17⁎
40 (70%) 39 (68%) 43 (75%)† 16 (28%) 10 (17%) 20 (35%)
191 (71%) 172 (64%) 152 (57%) 72 (27%) 31 (12%) 55 (20%)
82 (71%) 75 (65%) 77 (67%) 36 (31%) 13 (11%) 27 (23%)
Data are expressed as mean ± SD, number (percentage). ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; BMI, body mass index; CKD, chronic kidney disease; DCM, dilated cardiomyopathy; eGFR, estimated glomerular filtration rate; IHD, ischemic heart disease; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; VHD, valvular heart disease; UA, uric acid; XOR, xanthine oxidoreductase. ⁎ P b 0.05 vs. normal XOR group by ANOVA with Scheffe post hoc test. † P b 0.05 vs. normal XOR group by chi-square test.
(N169.1 pmol/100 μL/h, n = 74). Our analysis revealed a U-shaped relationship between XOR activity and the cardiac event rate in patients with CHF (Fig. 5). 4. Discussion 4.1. Main findings In the present study, (1) the prevalence rates of high and low XOR activities increased with advancing NYHA functional class, (2) serum UA levels were not different among XOR groups, but higher UA was noted in the high XOR group in patients without CKD or acidic urine, (3) multivariate logistic analysis revealed that abnormal XOR activity level was closely associated with poor clinical outcomes in patients with CHF, (4) the prediction model including XOR activity had an improved C index, NRI, and IDI, and (5) there was a U-shape relationship between XOR activity and the cardiac event rate. 4.2. UA level and XOR activity in CHF
Fig. 1. The association between XOR activity and NYHA functional class. NYHA, New York Heart Association; XOR, xanthine oxidoreductase.
Serum UA level is reportedly affected by several factors including gender, CKD, acidic urine, and loop diuretic use [20]. Although XOR plays a pivotal role in UA production, a significant association between serum UA level and XOR activity was not observed in CHF patients, in particular those with CKD or acidic urine. Since UA excretion is dependent on acidic urine and kidney dysfunction [21,22], it is plausible that serum UA level is affected by UA excretion rather than XORdependent UA production in these conditions. XOR activity, rather than serum UA level, is reported to be actively involved in hemodynamic impairment in patients with CHF [11].
154
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
Fig. 2. The association between serum UA and XOR activity. Serum UA level in the setting of male gender (A), CKD (B), acidic urine (C), and loop diuretic use (D). Association between serum UA and XOR activity in patients with and without CKD (E) or those with and without acidic urine (F). CKD, chronic kidney disease; UA, uric acid; XOR, xanthine oxidoreductase.
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
155
Table 3 Univariate and multivariate Cox proportional hazard analyses of predicting cardiac events in patients with chronic heart failure. Univariate analysis
Multivariate analysis
Variables
HR
95% CI
P value
HR
95% CI
P value
Age (years old) Male vs. female Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) NYHA III/IV vs. II Log BNPa Creatininea eGFRa UAa Acidic urine CKD High XOR vs. normal XOR Low XOR vs. normal XOR LVEDDa LVEFa Loop diuretics use
1.02 1.58 1.01 1.15 1.02 2.97 1.73 1.09 0.67 1.35 1.47 1.63 3.56 3.02 1.22 0.81 3.02
1.01–1.04 1.13–2.23 0.67–1.50 0.81–1.63 0.71–1.45 2.16–4.08 1.46–2.05 0.99–1.20 0.57–0.80 1.16–1.58 1.06–2.03 1.19–2.23 2.50–5.06 1.95–4.67 1.03–1.45 0.68–0.96 2.01–4.46
0.0014 0.0082 0.9494 0.4297 0.9300 b0.0001 b0.0001 0.0933 b0.0001 0.0002 0.0191 0.0023 b0.0001 b0.0001 0.0212 0.0165 b0.0001
1.01 1.94
1.00–1.03 1.31–2.89
0.0834 0.0011
1.75 1.43
1.21–2.54 1.15–1.78
0.0031 0.0012
0.95 1.03 1.06
0.78–1.13 0.84–1.26 0.73–1.54
0.5420 0.7749 0.7568
2.77 1.93
1.88–4.08 1.17–3.20
b0.0001 0.0107
1.09 1.89
0.91–1.33 1.23–2.89
0.3129 0.0037
CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HR, hazard ratio; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; UA, uric acid; XOR, xanthine oxidoreductase. a Per 1-SD increase.
Filippatos et al. concluded that serum UA level is not associated with cardiac events in patients with comorbid CHF and CKD because serum UA level does not represent XOR activity in patients with CKD [22]. Furthermore, plasma XOR activity is an independent predictor of cardiac events in patients with CKD [23]. These findings support our contention that plasma XOR activity provides useful clinical information in addition to serum UA level in patients with CHF.
4.3. XOR activity and clinical outcome in CHF Accumulating evidence indicates that XOR activity in myocardial tissue is upregulated and closely associated with the development of CHF [5]. XOR-induced ROS generation induces several pathophysiological processes such as myocardial apoptosis and impaired sarcomere contraction through abnormal ryanodine receptor oxidation, finally leading
Fig. 3. Impact of plasma XOR activity on cardiac events in patients with CHF. Kaplan–Meier analyses of high XOR activity to predict all cardiac events (A) and cardiac mortality (B) in patients with CHF. Kaplan–Meier analyses of low XOR activity to predict all cardiac events (C) and cardiac mortality (D) in patients with CHF. CHF, chronic heart failure; XOR, xanthine oxidoreductase.
156
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157
Fig. 5. The relationship between plasma XOR activity sextile and cardiac event rate. Fig. 4. Receiver operating characteristic curve of the baseline model and those including XOR activity in patients with CHF. CHF, chronic heart failure; XOR, xanthine oxidoreductase.
to cardiac dysfunction [24,25]. Similarly, the high XOR group had more severe CHF and cardiac systolic dysfunction compared to subjects with levels in the reference range. Our results represent the first evidence that high XOR activity is closely associated with poor clinical outcomes in patients with CHF after adjusting for confounding risk factors. Furthermore, the prediction model significantly improved by adding XOR activity, suggesting that it could be a useful parameter to predict future cardiac events in patients with CHF. Surprisingly, low XOR activity is also related to CHF severity and clinical outcome. The patients with low XOR had lower body mass index compared to other groups, suggesting the relationship between low XOR activity and cardiac cachexia. It was reported that cancer induced cachexia decreases protein synthesis in the liver [26]. Thus, low XOR activity may result from a decrease in XOR protein synthesis secondary to CHF induced cachexia. In addition, it should be noted that patients with low XOR had several other risk factors for poor clinical outcomes including advanced age, severe CHF, high BNP, and CKD.
5. Study limitations As this analysis was performed at a single center and XOR activity is highly population dependent, a validation study is needed to better delineate the prognostic value of XOR activity. Second, further investigations are required to determine the pathological mechanism by which abnormal XOR activity deteriorates cardiac prognosis. Finally, although the mean LVEF in the present study was equivalent to those reported in the Japanese heart failure study and heart failure registry [27,28], it was relatively high compared to that seen in Western countries [29]. 6. Conclusions We provide the first evidence of an association between plasma XOR activity with CHF severity and clinical outcome. Measuring XOR activity could help identify high-risk patients and XOR may be a useful therapeutic target in CHF. Conflict of interest None.
4.4. Clinical perspectives Acknowledgment of grant support Considering the role of XOR, XOR inhibitor therapy could be beneficial in the setting of CHF, but it is still under discussion. Some reports raise the possibility that the cardio-protective effects of XOR inhibitors depend on XOR activity, rather than hyperuricemia, in patients with CHF [11,22]. Similarly, we found that high XOR activity is superior prognostic parameter to serum UA level in patients with CHF. XOR inhibitors may not necessarily be beneficial for CHF patients with hyperuricemia if it is not due to heightened XOR activity. Therefore, measuring plasma XOR activity is important and may identify those patients with CHF likely to respond to XOR inhibitors.
This work was in part supported by the consigned research fund from Sanwa Kagaku Kenkyusho Co., Ltd. References [1] D.M. Lloyd-Jones, M.G. Larson, E.P. Leip, A. Beiser, R.B. D'Agostino, W.B. Kannel, et al., Lifetime risk for developing congestive heart failure: the Framingham Heart Study, Circulation 106 (2002) 3068–3072. [2] J.M. Hare, Oxidative stress and apoptosis in heart failure progression, Circ. Res. 89 (2001) 198–200.
Table 4 Statistics for model fit and improvement with the addition of XOR activity on the prediction of cardiac events. Baseline model
C index (P value) 0.746
NRI (95% CI, P value) Reference
IDI (95% CI, P value) Reference
+XOR activity
0.807 (P = 0.0006)
0.2589 (0.1147–0.4031, P = 0.0004)
0.108 (0.0767–0.1394, P b 0.0001)
Baseline model includes age, gender, NYHA functional class, log BNP, eGFR, UA, acidic urine, LVEF. BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; IDI, integrated discrimination index; LVEF, left ventricular ejection fraction; NYHA functional class, New York Heart Association functional class; NRI, net reclassification index; 95% CI, 95% confidence interval; UA, uric acid; XOR, xanthine oxidoreductase.
Y. Otaki et al. / International Journal of Cardiology 228 (2017) 151–157 [3] Y. Otaki, H. Takahashi, T. Watanabe, A. Funayama, S. Netsu, Y. Honda, et al., HECTtype ubiquitin E3 ligase ITCH interacts with thioredoxin-interacting protein and ameliorates reactive oxygen species-induced cardiotoxicity, 2016. J. Am. Heart Assoc. 5. http://dx.doi.org/10.1161/jaha.115.002485. [4] M.G. Battelli, A. Bolognesi, L. Polito, Pathophysiology of circulating xanthine oxidoreductase: new emerging roles for a multi-tasking enzyme, Biochim. Biophys. Acta 2014 (1842) 1502–1517. [5] P. Pacher, A. Nivorozhkin, C. Szabo, Therapeutic effects of xanthine oxidase inhibitors: renaissance half a century after the discovery of allopurinol, Pharmacol. Rev. 58 (2006) 87–114. [6] J.W. de Jong, R.G. Schoemaker, R. de Jonge, P. Bernocchi, E. Keijzer, R. Harrison, et al., Enhanced expression and activity of xanthine oxidoreductase in the failing heart, J. Mol. Cell. Cardiol. 32 (2000) 2083–2089. [7] L.C. Amado, A.P. Saliaris, S.V. Raju, S. Lehrke, M. St John, J. Xie, et al., Xanthine oxidase inhibition ameliorates cardiovascular dysfunction in dogs with pacing-induced heart failure, J. Mol. Cell. Cardiol. 39 (2005) 531–536. [8] S. Tan, S. Gelman, J.K. Wheat, D.A. Parks, Circulating xanthine oxidase in human ischemia reperfusion, South. Med. J. 88 (1995) 479–482. [9] Y. Yokoyama, J.S. Beckman, T.K. Beckman, J.K. Wheat, T.G. Cash, B.A. Freeman, et al., Circulating xanthine oxidase: potential mediator of ischemic injury, Am. J. Phys. 258 (1990) G564–G570. [10] O.S. Ali, H.M. Abdelgawad, M.S. Mohammed, R.R. El-Awady, Ischemic heart diseases in Egypt: role of xanthine oxidase system and ischemia-modified albumin, Heart Vessel. 29 (2014) 629–637. [11] J.M. Hare, B. Mangal, J. Brown, C. Fisher Jr., R. Freudenberger, W.S. Colucci, et al., Impact of oxypurinol in patients with symptomatic heart failure. Results of the OPTCHF study, J. Am. Coll. Cardiol. 51 (2008) 2301–2309. [12] M.A. Newaz, N.N. Adeeb, Detection of xanthine oxidase in human plasma, Med. J. Malaysia 53 (1998) 70–75. [13] T. Murase, M. Nampei, M. Oka, N. Ashizawa, K. Matsumoto, A. Miyachi, et al., Xanthine oxidoreductase activity assay in tissues using stable isotope-labeled substrate and liquid chromatography high-resolution mass spectrometry, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 1008 (2016) 189–197. [14] P.A. McKee, W.P. Castelli, P.M. McNamara, W.B. Kannel, The natural history of congestive heart failure: the Framingham study, N. Engl. J. Med. 285 (1971) 1441–1446. [15] T. Murase, M. Oka, M. Nampei, A. Miyachi, T. Nakamura, A highly sensitive assay for xanthine oxidoreductase activity using a combination of [C, N]xanthine and liquid chromatography/triple quadrupole mass spectrometry, J. Label. Compd. Radiopharm. 1008 (2016) 189–197. [16] K.U. Eckardt, J.S. Berns, M.V. Rocco, B.L. Kasiske, Definition and classification of CKD: the debate should be about patient prognosis—a position statement from KDOQI and KDIGO, Am. J. Kidney Dis. 53 (2009) 915–920.
157
[17] K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratificationAm. J. Kidney Dis. 39 (2002) S1-266. [18] Y. Otaki, T. Watanabe, H. Takahashi, A. Hirayama, T. Narumi, S. Kadowaki, et al., Association of heart-type fatty acid-binding protein with cardiovascular risk factors and all-cause mortality in the general population: the Takahata study, PLoS One 9 (2014) e94834. [19] Y. Otaki, H. Takahashi, T. Watanabe, S. Kadowaki, T. Narumi, Y. Honda, et al., Electrocardiographic left ventricular hypertrophy Cornell product is a feasible predictor of cardiac prognosis in patients with chronic heart failure, Clin. Res. Cardiol. 103 (2014) 275–284. [20] Y. Otaki, T. Watanabe, H. Takahashi, H. Hasegawa, S. Honda, A. Funayama, et al., Acidic urine is associated with poor prognosis in patients with chronic heart failure, Heart Vessel. 28 (2013) 735–741. [21] A. Kanbara, M. Hakoda, I. Seyama, Urine alkalization facilitates uric acid excretion, Nutr. J. 9 (2010) 45. [22] G.S. Filippatos, M.I. Ahmed, J.D. Gladden, M. Mujib, I.B. Aban, T.E. Love, et al., Hyperuricaemia, chronic kidney disease, and outcomes in heart failure: potential mechanistic insights from epidemiological data, Eur. Heart J. 32 (2011) 712–720. [23] B. Gondouin, N. Jourde-Chiche, M. Sallee, L. Dou, C. Cerini, A. Loundou, et al., Plasma xanthine oxidase activity is predictive of cardiovascular disease in patients with chronic kidney disease, independently of uric acid levels, Nephron 131 (2015) 167–174. [24] D.B. Sawyer, D.A. Siwik, L. Xiao, D.R. Pimentel, K. Singh, W.S. Colucci, Role of oxidative stress in myocardial hypertrophy and failure, J. Mol. Cell. Cardiol. 34 (2002) 379–388. [25] D.R. Gonzalez, A.V. Treuer, J. Castellanos, R.A. Dulce, J.M. Hare, Impaired Snitrosylation of the ryanodine receptor caused by xanthine oxidase activity contributes to calcium leak in heart failure, J. Biol. Chem. 285 (2010) 28938–28945. [26] A.A. Narsale, R.T. Enos, M.J. Puppa, S. Chatterjee, E.A. Murphy, R. Fayad, et al., Liver inflammation and metabolic signaling in ApcMin/+ mice: the role of cachexia progression, PLoS One 10 (2015) e0119888. [27] N. Shiba, K. Nochioka, M. Miura, H. Kohno, H. Shimokawa, Trend of westernization of etiology and clinical characteristics of heart failure patients in Japan—first report from the CHART-2 study, Circ. J. 75 (2011) 823–833. [28] Y. Otaki, T. Watanabe, H. Takahashi, A. Funayama, D. Kinoshita, M. Yokoyama, et al., Comorbid renal tubular damage and hypoalbuminemia exacerbate cardiac prognosis in patients with chronic heart failure, Clin. Res. Cardiol. 105 (2016) 162–171. [29] W. von Scheidt, C. Zugck, M. Pauschinger, R. Hambrecht, O. Bruder, A. Hartmann, et al., Characteristics, management modalities and outcome in chronic systolic heart failure patients treated in tertiary care centers: results from the EVIdence based TreAtment in Heart Failure (EVITA-HF) registry, Clin. Res. Cardiol. 103 (2014) 1006–1014.