sign in sign up
<<
Home , Hyperuricemia, Hypoalbuminemia

Circulation Journal ORIGINAL ARTICLE Official Journal of the Japanese Circulation Society http://www.j-circ.or.jp Renal Disease The Impact of on In-Hospital Mortality and Incidence of Acute Injury in Patients Undergoing Percutaneous Coronary Intervention Sang-Ho Park, MD; Won-Yong Shin, MD, PhD; Eun-Young Lee, MD, PhD; Hyo-Wook Gil, MD, PhD; Se-Whan Lee, MD; Seung-Jin Lee, MD, PhD; Dong-Kyu Jin, MD, PhD; Sae-Yong Hong, MD, PhD

Background: There is very little information about the relationship between hyperuricemia, acute kidney injury (AKI) and in-hospital mortality.

Methods and Results: With a retrospective analysis of the medical records, 1,247 patients who had percutane- ous coronary intervention (PCI) were investigated. AKI was defined as an increase in of ≥0.5 mg/dl or ≥50% over baseline within 7 days of PCI. The association of AKI with clinical, biochemical and procedural vari- ables were examined. In addition, the association of hyperuricemia with in-hospital mortality was also examined. Of the 1,247 patients in the study population, 51 (4.1%) experienced AKI after PCI, 15 of whom required . In-hospital mortality occurred in 1.6% (20 of 1,247) in 19.6% (10 of 51) of AKI individuals, and 0.8% (10 of 1,186) of the non-AKI participants (odd ratios, 28.927; 95% confidence intervals, 11.411–73.328; P<0.001). In our study, the most powerful predictors of these variables were acute myocardial infarction, baseline estimated glomerular filtra- tion rate (eGFR) <60 ml · min–1 · 1.73 m–2, diabetics mellitus, and hyperuricemia. Notably, the incidence of AKI after PCI markedly increased in diabetic or hyperuricemic patients with a baseline eGFR of <60 ml · min–1 · 1.73 m–2.

Conclusions: It is clear that AKI develops due to multiple risk factors. Our results indicate that hyperuricemia is independently associated with an increased risk of in-hospital mortality and AKI in patients treated with PCI. (Circ J 2011; 75: 692 – 697)

Key Words: Ischemic heart disease; Kidney; Percutaneous coronary intervention; Renal failure

cute kidney injury (AKI) is often encountered in role in the pathogenesis of AKI.19,21,22 Hyperuricemia might patients who have coronary artery disease. Previous be one of the factors deteriorating renal injury.17 However, in A studies have documented the adverse prognostic the literature, there is very little information about the rela- impact of chronic renal insufficiency after percutaneous coro- tionship between hyperuricemia and incidence of AKI/in-hos- nary intervention (PCI).1,2 AKI after PCI is associated with pital mortality. Therefore, we retrieved the clinical and labo- increased 1-year mortality in patients undergoing PCI.3–5 ratory data from our interventional cardiology database and , , diabetes mellitus (DM), determined the effect of on the incidence of AKI and advanced age and increased volume of contrast media are in-hospital mortality in patients treated with PCI. well-known risk factors for the development of AKI.6–9 Meta- bolic syndrome,10 ,10–12 ,7,13,14 anemia15,16 and hyperuricemia14 have recently been reported Methods as new and possible risk markers for AKI after PCI. Hyperuri- Study Population cemia has been considered as a risk factor for , With a retrospective analysis of the medical records, we iden- , DM, chronic kidney disease and isch- tified all patients who had coronary interventional interven- emic heart disease.17–21 Furthermore, hyperuricemia is accom- tion at a single center from August 2006 to December 2009. panied by enhanced synthesis of , The average age of the patients was 61.01±12.24 years in activated renin–angiotensin–aldosterone system, increased males (n=777) and 69.67±8.98 years in females (n=470). No endothelin-1 and inhibited system, which play a subgroups of patients (for example, those with acute myocar-

Received July 2, 2010; revised manuscript received October 1, 2010; accepted November 4, 2010; released online January 14, 2011 Time for primary review: 18 days Department of Internal Medicine, Soonchunhyang University Cheonan Hospital, Cheonan, Korea Mailing address: Won-Yong Shin, MD, PhD, Department of Internal Medicine, Soonchunhyang University Cheonan Hospital, 23-20 Bongmyung-Dong, Cheonan, Chungnam 330-721, Korea. E-mail: [email protected] ISSN-1346-9843 doi: 10.1253/circj.CJ-10-0631 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected]

Circulation Journal Vol.75, March 2011 Hyperuricemia and Acute Kidney Injury in PCI 693

Table 1. Comparison of Clinical, Laboratory and Angiographical Findings Between Non-AKI and AKI Characteristics Non-AKI (n=1,196) AKI (n=51) P value* Age (years) 64.04±11.92 69.33±9.65 <0.001 Diabetes mellitus – no. (%) 378 (31.6%) 34 (66.7%) <0.001 Hyperuricemia – no. (%) 160/1,074 (14.9%) 24/49 (49.0%) <0.001 Baseline eGFR (ml · min–1 · 1.73 m–2) 85.80±33.21 45.32±26.24 <0.001 Fasting blood sugar (mg/dl) 117.80±38.58 144.57±73.46 0.01 Albumin (g/dl) 4.25±0.46 3.76±0.56 <0.001 Baseline BUN (mg/dl) 14.72±5.92 24.55±12.19 <0.001 Baseline creatinine (mg/dl) 0.81±0.26 1.52±0.71 <0.001 Hemoglobin (g/dl) 13.53±1.89 11.86±2.09 <0.001 Hematocrit (%) 40.01±5.32 35.76±5.76 <0.001 Uric acid (mg/dl) 5.12±1.54 6.51±2.23 <0.001 Glycated hemoglobin (%) 6.81±1.52 7.54±1.41 <0.001 hsCRP (mg/L) 8.91±25.62 22.91±45.55 0.04 LVEF (%) 54.86±12.59 47.07±14.05 <0.001 Contrast volume (ml) 306.08±132.59 359.39±159.26 0.006 AMI – no. (%) 416/1,196 (34.8%) 30/51 (58.8%) <0.001 Multivessel disease – no. (%) 721/1,196 (60.3%) 41/51 (80.4%) 0.004 In-hospital mortality 10 (0.8%) 10 (19.6%) <0.001 Categorical variables are expressed as no. (%) and continuous variables as mean ± SD, unless otherwise stated. *Continuous variables were compared by means of the Student’s t-test; categorical variables were compared with the Pearson’s χ2 test. AKI, acute kidney injury; eGFR, estimated glomerular filtration rate (by Cockroft-Gault formula); BUN, urea nitro- gen; hsCRP, high sensitive C-reactive ; LVEF, left ventricular ejection fraction; AMI, acute myocardial infarction.

Table 2. Comparison of Clinical, Laboratory and Angiographical Findings Between Elective PCI and Emergent PCI Characteristics Elective PCI (n=937) Emergent PCI (n=310) P value* Age (years) 65.03±11.22 61.92±13.41 <0.001 Body mass index (kg/m2) 24.58±3.41 23.99±3.42 0.018 Diabetes mellitus – no. (%) 340 (36.3%) 72 (23.2%) <0.001 Hyperuricemia – no. (%) 134/816 (16.4%) 50/307 (16.3%) 0.993 Baseline eGFR (ml · min–1 · 1.73 m–2) 83.61±33.39 86.20±35.31 0.256 Fasting blood sugar (mg/dl) 117.26±40.73 123.83±41.06 0.014 Albumin (g/dl) 4.22±0.48 4.26±0.45 0.181 Baseline BUN (mg/dl) 15.03±6.58 15.29±6.61 0.499 Baseline creatinine (mg/dl) 0.84±0.31 0.83±0.34 0.369 Hemoglobin (g/dl) 13.25±1.89 14.10±1.89 <0.001 Hematocrit (%) 39.32±5.36 41.40±5.24 <0.001 Uric acid (mg/dl) 5.15±1.60 5.24±1.59 0.360 (%) 6.89±1.57 6.67±1.32 0.049 hsCRP (mg/L) 8.24±13.48 23.15±35.99 0.05 LVEF (%) 56.86±11.96 47.39±12.40 <0.001 Contrast volume (ml) 308.83±135.94 306.29±128.52 0.773 AMI – no. (%) 190/937 (20.3%) 256/310 (82.6%) <0.001 Multivessel disease – no. (%) 579/931 (61.8%) 183/310 (59.0%) 0.387 AKI – no. (%) 30/937 (3.2%) 21/310 (6.8%) 0.006 In-hospital mortality – no. (%) 6/937 (0.6%) 14/310 (4.5%) <0.001 Categorical variables are expressed as no. (%) and continuous variables as mean ± SD, unless otherwise stated. *Continuous variables were compared by means of the Student’s t-test; categorical variables were compared with the Pearson’s χ2 test. PCl, percutaneous coronary intervention. Other abbreviations see in Table 1. dial infarction (AMI) or ) were excluded. PCI was per- (2) estimated glomerular filtration rate (eGFR) based on the formed using the standard technique. The specific type of Cockcroft–Gault formula23 <10 ml · min–1 · 1.73 m–2, (3) intra- revascularization procedure was at the discretion of the inter- vascular administration of iodinated contrast media within ventional cardiologist. All other treatments were at the dis- 72 h, and (4) a serum creatinine concentration of <0.4 mg/dl. cretion of the physicians, whether before, during or after the After we examined the medical records of 1,373 patients, procedure. Exclusion criteria were as follows: (1) , 126 patients were excluded based on exclusion criteria.

Circulation Journal Vol.75, March 2011 694 PARK SH et al.

Table 3. Univariate Variables Associated With AKI of more than 70% stenosis in ≥2 coronary arteries. We exam- ined clinical (age, sex, body mass index, smoking, hyperten- P value* OR 95%CI sion, DM, eGFR, AMI), laboratory (hemoglobin, hematocrit, Age (years) 0.002 1.043 1.016–1.072 albumin, , low-density lipoprotein , Diabetes mellitus <0.001 4.328 2.388–7.845 high-density lipoprotein cholesterol, fasting blood glucose, AMI 0.001 2.679 1.514–4.737 glycated hemoglobin (HbA1c), serum creatinine, serum blood Multivessel disease 0.005 2.701 1.340–5.444 urea nitrogen (BUN), high sensitive C-reactive protein eGFR (per ml · min–1 · 1.73 m–2) <0.001 0.940 0.925–0.955 (hsCRP), N-terminal pro B-type natriuretic peptide, serum eGFR <60 ml · min–1 · 1.73 m–2 <0.001 12.746 6.255–25.975 sodium, left ventricular ejection fraction (LVEF) and serum Hyperuricemia <0.001 5.484 3.056–9.841 uric acid), procedual (the amount/type of contrast media and Uric acid (per mg/dl) <0.001 1.546 1.329–1.798 number of involved coronary vessels) characteristics. Serum uric acid was measured in 1,123 patients because of 124 miss- Contrast amount (per Ml) 0.007 1.002 0.1001–1.004 ing values. Serum uric acid was measured regardless of fast- Hypoalbuminemia <0.001 6.478 3.338–12.574 ing status in cases undergoing emergent PCI, while it was Fasting blood sugar (mg/dl) <0.001 1.009 1.005–1.013 measured in fasting status in cases undergoing elective PCI. Hemoglobin (g/dl) <0.001 0.636 0.546–0.739 Iopromide (Ultravist, Bayer Health, 370 mg I/ml, osmolality Hematocrit (%) <0.001 0.891 0.852–0.931 770 mOsm/kg water) and ioversol (Optiray, Mallinckrodt, Glycated hemoglobin (%) 0.002 1.269 1.091–1.476 Amersham Health, 320 mg I/ml, osmolality 702 mOsm/kg hsCRP (mg/L) 0.002 1.010 1.004–1.016 water) were used as non-ionic, low-osmolar contrast agents, LVEF (%) <0.001 0.959 0.939–0.979 and iodixanol (Vispaque, Amersham Health, 320 mg I/ml, Baseline BUN (mg/dl) <0.001 1.130 1.096–1.164 osmolality 290 mOsm/kg water) was used as a non-ionic, iso- Baseline creatinine (mg/dl) <0.001 30.312 14.533–63.225 osmolar contrast agent. *Univariate binary logistic regression. OR, odds ratio; CI, confidence interval. Other abbreviations see in Statistical Analysis Table 1. Data are presented as mean ± SD or a percentage of the total. Continuous variables were compared by means of the Stu- dent’s t-test; categorical variables were compared with the Pearson’s χ2 test. Logistic regression (univariate analysis) Table 4. Independent Predictors Associated With AKI by was used to identify that AKI and hyperuricemia or uric acid Multivariate Analysis were associated with in-hospital mortality after PCI. Univar- P value* OR 95%CI iate and multivariate logistic regression were used to identify Contrast amount (per 1 ml) 0.045 1.003 1.000–1.005 the risk factors affecting AKI after PCI. The strength of the Diabetes mellitus 0.021 3.665 1.217–11.035 association between risk factors and AKI after PCI was Hemoglobin (g/dl) 0.017 0.372 0.166–0.835 expressed as odd ratios (OR) with 95% confidence intervals AMI 0.024 3.062 1.155–8.112 (CI). Models were developed with stepwise techniques and eGFR <60 ml · min–1 · 1.73 m–2 <0.001 5.743 1.749–18.857 by considering variables that were clinically relevant. Hyperuricemia <0.001 4.739 1.961–11.449 *Multivariate binary logistic regression for the variables that had a Results statistical significance in univariate analysis. Models were devel- oped with stepwise techniques and by considering variables that Comparison of Clinical, Laboratory and Procedural were clinically relevant. Characteristics Between AKI and Non-AKI (Table 1) Abbreviations see in Tables 1,3. The study was carried out with 1,247 patients, and there were 777 men and 470 women. Comparison of clinical, biochemi- cal and procedural characteristics between AKI and non-AKI Definition and Materials are shown in Table 1. Of the 1,247 patients in the study pop- Hyperuricemia was defined as a serum uric acid concentra- ulation, 51 (4.1%) experienced AKI after PCI, 15 of whom tion of greater than 7.0 mg/dl in men and 6.0 mg/dl in pre- required hemodialysis. As seen in Table 1, these patients menopausal women.24 The relationship between uric acid and were older; more often had DM, hyperuricemia, AMI and was found in terms of frank hyperuri- multivessel disease; showed lower values in baseline eGFR, cemia (defined as more than 6mg/dl in women and more than albumin, hemoglobin, hematocrit and LVEF; and revealed 7 mg/dl in men).17 Based on the above 2 references, we higher values in baseline creatinine, baseline BUN, hsCRP, defined hyperuricmeia as a serum uric acid concentration of amount of contrast media, fasting blood glucose, HbA1c and ≥7 mg/dl for males and of ≥6.5 mg/dl for females because uric acid. In-hospital mortality was 19.6% (10 of 51) in the postmenopausal female patients were enrolled in our study. AKI inividuals and 0.8% (10 of 1,186) in the non-AKI. AKI was defined as an increase in serum creatinine of ≥0.5 mg/dl or ≥50% over baseline within 7 days of PCI. AMI Comparison of Clinical, Laboratory and Procedural was considered to have occurred if 2 of the following 3 crite- Characteristics Between Elective PCI and Emergent ria were met: typical chest pain for ≥30 min, T eleva- PCI (Table 2) tion, development of ST-T wave changes and new Q waves We analyzed the data, divided by emergent and elective PCI or newly developed bundle branch block on electrocardiogra- group. As seen in Table 2, there was no statistical difference phy. DM was defined as a fasting blood glucose concentration in serum uric acid concentration, hyperuricemia, BUN, cre- of ≥125 mg/dl, or a clinical diagnosis of diabetes with dietary, atinine, eGFR, contrast volume, albumin, and mulitvessel oral or insulin treatment. Hypoalbuminemia was defined as a disease between emergent and elective PCI. However, statis- serum albumin concentration of ≤3.5 g/dl. Multivessel coro- tical significance was shown in age, BMI, DM, fasting blood nary disease was defined as a coronary angiographic finding sugar, hemoglobin, hematocrit, glycated hemoglobin, hsCRP,

Circulation Journal Vol.75, March 2011 Hyperuricemia and Acute Kidney Injury in PCI 695

Table 5. Incidence of AKI According to Hyperuricemia and eGFR Incidence (%) OR 95%CI P value* eGFR >– 60, hyperuricemia (–) 1.1 Reference eGFR >– 60, hyperuricemia (+) 1.1 0.939 0.325–5.295 0.953 eGFR <60, hyperuricemia (–) 8.3 7.437 3.105–17.815 <0.001 eGFR <60, hyperuricemia (+) 35.6 31.767 31.767–74.808 <0.001 Hyperuricemia (–), non-hypereruricemic patient group; hyperuricemia (+), hyperuricemic patient group. *Binary logistic regression. Abbreviations see in Tables 1,3.

Figure. Comparison of serum uric acid concentration between non- death and death group. Death gro­ up showed a higher serum uric acid concentration than non-death group (6.042±2.67 vs. 5.169±1.57 mg/dl, P= 0.018)

LVEF, AMI, AKI, and in-hospital mortality. D had a significantly higher risk of AKI than group A (OR, 40.722; 95%CI, 13.872–119.541; P<0.001). Group C had a Predictors of AKI (Tables 3,4) significantly higher risk of AKI than group A (OR, 11.422; We used univariate analysis to study different possible risk 95%CI, 3.636–35.871; P<0.001). Although there was no sig- factors for the development of AKI. Univariate variables nificant difference in the incidence of AKI between groups A associated with AKI were age, baseline eGFR, baseline cre- and B, there was a slight tendency for the incidence of AKI atinine and BUN, DM, AMI, dose of contrast media, hemo- to increase in group B as compared to group A (OR, 3.460; globin, hematocrit, hyperuricemia, uric acid, fasting blood 95%CI, 0.969–12.358; P=0.056). Group D had a higher inci- glucose, HbA1c, hsCRP, hypoalbuminemia, LVEF and mul- dence of AKI than group C (OR, 3.565; 95%CI, 1.714–7.417; tivessel involvement (Table 3). We performed multivariate P<0.001). analysis for the variables that had a statistical significance in univariate analysis. Multivariate analysis showed that the Incidence of AKI in the Study Groups Based on Baseline significant predictors of AKI after PCI were a baseline eGFR eGFR and Hyperuricemic Status (Table 5) of <60 ml · min–1 · 1.73 m–2, AMI, DM, hyperuricemia, amount The incidence of AKI was 1.1% (8/714) in non-hyperurice- of contrast media and hemoglobin (Table 4). mic patients with a baseline eGFR of ≥60 ml · min–1 · 1.73 m–2 (group 1), 1.1% (1/95) in hyperuricemic patients with a base- Incidence of AKI in the Study Groups Based on Baseline line eGFR of ≥60 ml · min–1 · 1.73 m–2 (group 2), 8.3% (15/180) eGFR and Diabetic Status in non-hyperuricemic patients with a baseline eGFR of <60 ml · The observed incidence of AKI after PCI was related to base- min–1 · 1.73 m–2 (group 3), and 35.6% (21/59) in hyperurice- line eGFR and the presence of diabetes and hyperuricemia. mic patients with a baseline eGRF of <60 ml · min–1 · 1.73 m–2 Incidence of AKI was 0.6% (4/636) in non-diabetic patients (group 4). Group 4 had a significantly higher risk of AKI than with a baseline eGFR of ≥60 ml · min–1 · 1.73 m–2 (group A), group 1 (OR, 31.767; 95%CI, 31.767–74.808; P<0.001). 2.1% (6/280) in diabetic patients with a baseline eGFR of Group 3 had a significantly higher risk of AKI than group 1 ≥60 ml · min–1 · 1.73 m–2 (group B), 6.7% (12/178) in non-dia- (OR, 7.437; 95%CI, 3.105–17.815; P<0.001). There was no betic patients with a baseline eGFR of <60 ml · min–1 · 1.73 m–2 significant difference in the incidence of AKI between groups (group C) and 20.5% (25/122) in diabetic patients with a 1 and 2 (OR, 0.939; 95%CI, 0.325–5.295; P=0.953) (Table 4). baseline eGFR of <60 ml · min–1 · 1.73 m–2 (group D). Group Group 4 had a higher incidence of AKI than group 3 (OR,

Circulation Journal Vol.75, March 2011 696 PARK SH et al.

Table 6. Association Between Serum Uric Acid, Also, there are several mechanisms by which uric acid con- 34 Hyperuricemia, AKI and In-Hospital tributes to AKI. However, to date, very little information is Mortality available in the literature on the role of hyperuricemia as a P value* OR 95%CI predictor of AKI after PCI. Hyperuricemia inhibits the nitric Serum uric acid (per mg/dl) 0.019 1.329 1.049–1.684 oxide system in the kidneys and increases endothelin-1 con- 19,22,34 Hyperuricemia 0.021 3.005 1.186–7.867 centrations. An increase in endothelin-1 and decrease in nitric oxide contribute to vasoconstriction, which reduces AKI <0.001 28.927 11.411–73.328 renal medullary blood flow.35,36 In our study, the incidence of *Univariate binary logistic regression. AKI after PCI was increased markedly in diabetic or hyper- Abbreviations see in Tables 1,3. uricemic patients with a baseline eGFR of <60 ml · min–1 · 1.73 m–2. Patients who are at higher risk of developing AKI, such as those with DM or renal failure, have an impaired 4.271; 95%CI, 2.069–8.818; P<0.001). endothelial function. At the same time, impaired endothelial function is an important finding of hyperuricemia.17,19–21 The Association Between Serum Uric Acid, Hyperuricemia, AKI current literature supports the role of hyperuricemia as an and In-Hospital Mortality (Figure, Table 6) independent, modifiable marker of cardiovascular damage In-hospital death group had a higher serum uric acid concen- and an independent predictor of cardiovascular event.18,20,37 tration than in-hospital survivor group (6.042±2.67 mg/dl vs. Mehran et al38 evaluated a total of 5,571 patients who under- 5.169±1.57, P=0.018) (Figure). As shown in Table 6, in- went PCI for the detection of risk factors for AKI, and have hospital mortality increased in the group developing AKI, demonstrated that multivessel coronary involvement is only a hyperuricemic group and patients with a high serum uric acid univariate predictor of AKI (P=0.003), which is consistent concentration. with our results. In patients with multivessel coronary involve­ ment, other vessels in the body, such as the renal artery, can be involved. If the renal artery is involved, renal blood supply Discussion might be decreased and the kidneys could be more susceptible This study indicated that hyperuricemia or high baseline to AKI. This might help to explain why hyperuricemic patients serum uric acid increased the incidence of AKI and in-hospi- were more susceptible to developing AKI in our study. tal mortality after PCI. Although the overall incidence of However, previous studies have reported that endothelial AKI after PCI was low (4.1%), diabetic or hyperuricemic function improves in patients with hyperuricemia and heart patients with chronic renal insufficiency (a baseline eGFR of failure or diabetes who received but not in those <60 ml · min–1 · 1.73 m–2) was at high risk of AKI, and AKI who received other uric acid-lowering .39,40 Allopurinol portended a very high risk of death in the hospital. In previ- is not a safe because its hypersensitivity reactions can be ous studies, the predictors of AKI were decreased eGFR, dia- fatal and there is no sufficient evidence to support treatment betes with chronic kidney disease, increased dose of contrast of hyperuricemia with this drug, Further stud- media, older age, anemia, drug toxicity, multivessel coronary ies are needed to determine whether individuals with asymp- involvement, conditions resulting in a low effective circulat- tomatic hyperuricemia should be treated with allopurinol. ing volume such as cardiogenic shock, use of an intra-aortic The present study has several limitations. First, this is a balloon pump, hypotension, congestive and retrospective study at a single center, and researchers were AMI.4,6,7,25–32 Our results were consistent with previous out- not blinded to the clinical and laboratory information includ- come studies documenting a higher incidence of AKI in ing variables of uric acid and renal function. Second, even decreased eGFR or pre-existing renal failure, diabetes with after multivariate analysis, hyperuricemia remained an inde- chronic kidney disease, AMI, low level of hemoglobin, older pendent predictor of AKI; however, multivariate models can- age, increased dose of contrast media, depressed LVEF and not entirely account for the baseline differences, and adjusted multivessel involvement. Therefore, although our study had effects still might be influenced by residual confounding fac- a retrospective design, it is inferred that our study model and tors. Finally, because the association of uric acid/hyperurice- results might be relevant and significant. Notably, in multi- mia and in-hospital mortality was evaluated only by a univari- variate analysis, the most powerful predictors of AKI were ate model, statistical power might be low and confoundable. pre-existing renal failure, DM, AMI, baseline hemoglobin, However, because our results are consistent with those of pre- dose of contrast media and hyperuricemia. vious studies in many variables and are adequately adjusted The mechanism of a higher incidence of AKI in AMI seems by multivariate logistic regression, it is believed that hyper- to be associated with hemodynamic instability. Since AMI is uricemia is a powerful predictor of AKI. At the same time, an emergent condition, adequate hydration before PCI is sel- our study had important implications that hyperuricemia domly performed. The incidences of AKI reported in previ- might impact in-hospital mortality after PCI. ous studies were generally lower than that in our study.6,7,25–31 Most of these studies were performed among stable patients, whereas those who were critically ill or in shock, such as Conclusions AMI, was included in our study. Therefore, to favorably influ- It is clear that AKI develops due to multiple risk factors. AKI ence prognosis therapeutic maneuvers that are directed both patients had a high incidence of vascular complications and at achieving hemodynamic stability and at preventing multi- increased in-hospital mortality. We found that serum uric failure are probably needed. acid is independently associated with an increased risk of In this study, the results of multivariate analysis showed AKI in patients with coronary artery disease treated with PCI that hyperuricemia was a powerful predictor of AKI. Several and that AKI and hyperuricemia might impact in-hospital epidemiological studies have shown that hyperuricemia is mortality after PCI. We believe that our results have clinical associated with cardiovascular events, metabolic syndrome, implications because hyperuricemia is identified as a modifi- stroke, hypertension, and dyslipidemia.17,21,33 able risk factor. Therefore, the serum uric acid concentration

Circulation Journal Vol.75, March 2011 Hyperuricemia and Acute Kidney Injury in PCI 697 before PCI seems to be useful to assess the risk for the devel- cardiovascular events in high-risk patients. Am J Nephrol 2005; 25: opment of AKI and in-hospital mortality. 36 – 44. 19. Kang DH, Nakagawa T. Uric acid and chronic renal disease: Pos- sible implication of hyperuricemia on progression of renal disease. Semin Nephrol 2005; 25: 43 – 49. References 20. Madsen TE, Muhlestein JB, Carlquist JF, Horne BD, Bair TL, 1. Le Feuvre C, Dambrin G, Helft G, Tabet S, Beygui F, Legendre C, Jackson JD, et al. Serum uric acid independently predicts mortality et al. Comparison of clinical outcome following coronary stenting in patients with significant, angiographically defined coronary dis- or balloon angioplasty in dialysis versus non-dialysis patients. Am ease. Am J Nephrol 2005; 25: 45 – 49. J Cardiol 2000; 85: 1365 – 1368. 21. Dincer HE, Dincer AP, Levinson DJ. Asymptomatic hyperuricemia: 2 . Ting HH, Tahirkheli NK, Berger PB, McCarthy JT, Timimi FK, to treat or not to treat. Cleve Clin J Med 2002; 69: 594 – 608. Mathew V, et al. Evaluation of long-term survival after successful 22. Perlstein TS, Gumieniak O, Hopkins PN, Murphey LJ, Brown NJ, percutaneous coronary intervention among patients with chronic Williams GH, et al. Uric acid and the state of the intrarenal renin- renal failure. Am J Cardiol 2001; 87: 630 – 633. angiotensin system in humans. Kidney Int 2004; 66: 1465 – 1470. 3. Mccullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. 23. Cockcroft DW, Gault MH. Prediction of creatinine clearance from Acute renal failure after coronary intervention: Incidence, risk fac- serum creatinine. Nephron 1976; 16: 31 – 41. tors, and relationship to mortality. Am J Med 1997; 103: 368 – 375. 24. Terkeltaub R. Crystal deposition disease. In: Goldman L, Ausiello 4. Rihal CS, Textor SC, Grill DE, Berger PB, Ting HH, Best PJ, et al. D, editors. Cecil’ Textbook of medicine, 22nd edn. Philadelphia: Incidence and prognostic importance of acute renal failure after Saunders, 2004; 1702 – 1710. percutaneous coronary intervention. Circulation 2002; 105: 2259 – 25. Schwab SJ, Hlatky MA, Pieper KS, Davidson CJ, Morris KG, 2264. Skelton TN, et al. Contrast nephrotoxicity: A randomized con- 5. Gruberg L,Weissman NJ, Pichard AD, Waksman R, Kent KM, trolled trial of a nonionic and an ionic radiographic contrast agent. Satler LF, et al. Impact of renal function on morbidity and mortality N Engl J Med 1989; 320: 149 – 153. after percutaneous aortocoronary saphenous vein graft intervention. 26. Rudnick MR, Goldfarb S, Wexler L, Ludbrook PA, Murphy MJ, Am Heart J 2003; 145: 529 – 534. Halpern EF, et al. Nephrotoxicity of ionic and nonionic contrast 6. Parfrey PS, Griffiths SM, Barrett BJ, Paul MD, Genge M, Withers J, media in 1196 patients: A randomized trial: The Iohexol Coopera- et al. Contrast material induced renal failure in patients with diabe- tive Study. Kidney Int 1995; 47: 254 – 261. tes mellitus, renal insufficiency, or both: A prospective controlled 27. Weisberg LS, Kurnik PB, Kurnik BR. Risk of radiocontrast study. N Engl J Med 1989; 320: 143 – 149. nephropathy in patients with and without diabetes mellitus. Kidney 7. Rich MW, Crecelius CA. Incidence, risk factors, and clinical course Int 1994; 45: 259 – 265. of acute renal insufficiency after cardiac catheterization in patients 28. Baliga R, Ueda N, Walker PD, Shah SV. Oxidant mechanisms in 70 years of age or older: A prospective study. Arch Intern Med toxic acute renal failure. Am J Kidney Dis 1997; 29: 465 – 477. 1990; 150: 1237 – 1242. 29. Bakris GL, Lass N, Gaber AO, Jones JD, Burnett JC Jr. Radiocon- 8. Albert SG, Shapiro MJ, Brown WW, Goodgold H, Zuckerman D, trast medium-induced declines in renal function: A role for oxygen Durham R, et al. Analysis of radiocontrast-induced nephropathy by free radicals. Am J Physiol 1990; 258: F115 – F120. dual-labeled radionuclide clearance. Invest Radiol 1994; 29: 618 – 30. Yoshioka T, Fogo A, Beckman JK. Reduced activity of antioxidant 623. enzymes underlies contrast media-induced renal injury in volume 9. Best PJ, Lennon R, Ting HH, Bell MR, Rihal CS, Holmes DR, et depletion. Kidney Int 1992; 41: 1008 – 1015. al. The impact of renal insufficiency on clinical outcomes in patients 31. Brady HR, Singer GG. Acute renal failure. Lancet 1995; 346: undergoing percutaneous coronary interventions. J Am Coll Car- 1533 – 1540. diol 2002; 39: 1113 – 1119. 32. Pannu N, Wiebe N, Tonelli M. Prophylaxis strategies for contrast- 10. Toprak O, Cirit M, Yesil M, Byrne DW, Postaci N, Bayata S, et al. induced nephropathy. JAMA 2006; 295: 2765 – 2779. Metabolic syndrome as a risk factor for contrast-induced nephrop- 33. Hongo M, Hidaka H, Sakaguchi S, Nakanishi K, Ichikawa M, Hirota athy in non-diabetic elderly patients with renal impairment. Kidney N, et al. Association between serum uric acid levels and cardiometa- Blood Press Res 2006; 29: 2 – 9. bolic risk factors among Japanese junior high school students. Circ J 11. Iakovou I, Dangas G, Mehran R, Lansky AJ, Ashby DT, Fahy M, 2010; 74: 1570 – 1577. et al. Impact of gender on the incidence and outcome of contrast- 34. Ejaz AA, Mu W, Kang DH, Roncal C, Sautin YY, Henderson G, et induced nephropathy after percutaneous coronary intervention. J al. Could uric acid have a role in acute renal failure? Clin J Am Soc Invasive Cardiol 2003; 15: 18 – 22. Nephrol 2007; 2: 16 – 21. 12. Khanal S, Attallah N, Smith DE, Kline-Rogers E, Share D, O’Donnell 35. Persson PB, Hansell P, Liss P. Pathophysiology of contrast medium- MJ, et al. Statin therapy reduces contrast-induced nephropathy: An induced nephropathy. Kidney Int 2005; 68: 14 – 22. analysis of contemporary percutaneous interventions. Am J Med 36. Detrenis S, Meschi M, Musini S, Savazzi G. Lights and shadows on 2005; 118: 843 – 849. the pathogenesis of contrast-induced nephropathy: State of the art. 13. Cirit M, Toprak O, Yesil M, Kline-Rogers E, Share D, O’Donnell Nephrol Dial Transplant 2005; 20: 1542 – 1550. MJ, et al. Angiotensin-converting enzyme inhibitors as a risk factor 37. Okura T, Higaki J, Kurata M, Irita J, Miyoshi K, Yamazaki T, et al. for contrast-induced nephropathy. Nephron Clin Pract 2006; 104: Elevated serum uric acid is an independent predictor for cardiovas- c20 – c27. cular events in patients with severe coronary artery stenosis: Sub- 14. Toprak O, Cirit M, Esi E, Postaci N, Yesil M, Bayata S, et al. Hyper- analysis of the Japanese Coronary Artery Disease (JCAD) Study. uricemia as a risk factor for contrast-induced nephropathy in patients Circ J 2009; 73: 885 – 891. with chronic kidney disease. Catheter Cardiovasc Interv 2006; 67: 38. Mehran R, Aymong ED, Nikolsky E, Lasic Z, Iakovou I, Fahy M, 227 – 235. et al. A simple risk score for prediction of contrast-induced nephrop- 15. Dangas G, Iakovou I, Nikolsky E, Aymong ED, Mintz GS, Kipshidze athy after percutaneous coronary intervention: Development and NN, et al. Contrast-induced nephropathy after percutaneous coro- initial validation. J Am Coll Cardiol 2004; 44: 1393 – 1399. nary interventions in relation to chronic kidney disease and hemo- 39. George J, Carr E, Davies J, Belch JJ, Struthers A. High-dose allo- dynamic variables. Am J Cardiol 2005; 95: 13 – 19. purinol improves endothelial function by profoundly reducing vas- 16. Nikolsky E, Mehran R, Lasic Z, Mintz GS, Lansky AJ, Na Y, et al. cular oxidative stress and not by lowering uric acid. Circulation Low hematocrit predicts contrast-induced nephropathy after percu- 2006; 114: 2508 – 2516. taneous coronary interventions. Kidney Int 2005; 67: 706 – 713. 40. Waring WS, McKnight JA, Webb DJ, Maxwell SR. Lowering 17. Feig DI, Kang DH, Johnson RJ. Uric acid and cardiovascular risk. serum urate does not improve endothelial function in patients with N Engl J Med 2008; 359: 1811 – 1821. type 2 diabetes. Diabetologia 2007; 50: 2572 – 2579. 18. Short RA, Johnson RJ, Tuttle KR. Uric acid, microalbuminuria and

Circulation Journal Vol.75, March 2011