Introduction

In humans, uric acid (UA) is the end product of purine catabolism [1]. Its serum levels, governed by the production (liver) and elimination (mainly the kidney) rates, are influenced by several variables, such as genetically determined factors (i.e., activity of synthesizing enzymes or renal transport systems), racial and demographic characteristics (i.e., gender), and morbidity (i.e., renal failure, malignancies) [2].

The role of UA in cardiovascular and renal disease has been intensively investigated, although not without controversy [3, 4]. Over recent years, there has been renewed debate concerning the nature of the association between raised serum UA concentrations and cardiovascular disease (CVD), and controversy still exists on whether hyperuricemia is simply a risk marker (due to its strong association with cardiovascular risk factors), or an independent risk factor for atherosclerosis [5]. In general population samples at relatively low risk for CVD, UA is a very weak predictor of cardiovascular morbidity and mortality, once the effect of known con-founders is accounted for [6]. On the contrary, UA seems to be a significant independent predictor of CVD in certain categories of patients at high cardiovascular risk, such as diabetics [7], patients with stroke [8], heart failure [9], and angiographically proven coronary artery disease [10].

Less is known about UA as a potential prognosticrisk factor for outcomes in patients affected specifically by acute myocardial infarction [11–15], and several studies suggest that a higher UA is independently associated with poorer survival in these patients. However, studies on this topic differ in number size, time of UA measurement (early phase vs within the first 48 h), and type of reperfusion [thrombolysis vs. percutaneous coronary intervention (PCI)].

We prospectively assessed in 856 STEMI patients consecutively admitted to our Intensive Cardiac Care Unit (ICCU) after primary PCI, whether UA levels are associated with in-ICCU mortality and complications.

Methods

Study population

From 1st January 2005 to 31st December 2009, 856 consecutive patients with STEMI (within 12 h from symptoms’ onset) were admitted to our Intensive Cardiac Care Unit (ICCU), which is located at a tertiary center.

In our hospital, in Florence, the reperfusion strategy of STEMI patients is represented by primary PCI [16–19]. Patients are first evaluated by the Medical Emergency System staff in the pre-hospital setting and then directly admitted to the catheterization laboratory or transferred to it after a rapid stabilization in the Emergency Department (ED). After primary PCI, they are admitted to our ICCU.

A successful procedure was defined as an infarct artery stenosis <20% associated with TIMI (Thrombolysis in Myocardial Infarction) grade 3 flow. Failure PCI was defined as resulting in TIMI grade 0–2 flow, regardless of the degree of residual stenosis [16].

The diagnosis of STEMI was based on the criteria of the American College of Cardiology/American Heart Association [17].

On ICCU admission, after PCI, in a fasting blood sample the following parameters were measured: glucose (g/l), troponin I (ng/ml), uric acid (mg/dl) [14], NT-pro Brain Natriuretic Peptide (NT-BNP) (pg/ml) [14], leukocyte count (×103/μl), fibrinogen (mg/dl), erythrocyte sedimentation rate (ESR), glycated hemoglobin (%), cholesterol (mg/dl) and triglycerides (mg/dl). Creatinine (mg/dl) was also measured in order to calculate glomerular filtration rate (ml/min/1.73 m2). Glucose values and Tn I were measured three times a day, and peak glucose and peak Tn I were considered [20], respectively.

Transthoracic two-dimensional echocardiography was performed on ICCU admission in order to measure left ventricular ejection fraction (LVEF).

In-ICCU mortality and in-ICCU complications were recorded. [21].

The study was approved by an appropriate ethics committee, and all patients gave informed consent to participate.

Statistical analysis

Data have been processed by means of SPSS 13.0 statistical package (SPSS Inc, Chicago, IL, USA). A p value <0.05 was considered statistically significant. Data are reported as frequencies (percentages) and medians [95% Confidence interval (CI)] and analyzed by means of χ2 (or Fisher’s exact text, when appropriate) and Mann–Whitney U test, respectively. Moreover, study population has been divided by tertiles of uric acid levels in order to investigate which variables differed between the three subgroups. Logistic regression analysis was carried out considering as outcomes intra-ICCU mortality and complications. In these two multivariable analyses, candidate variables were chosen as those that demonstrated significantly differences at univariable analysis or were clinically relevant. Backward procedure (probability for entry: 0.05; probability for removal: 0.10) was repeated until all variables in the model reached statistical significance.

Results

Table 1 depicts the clinical characteristics of the 856 consecutive STEMI patients included in the study. In more than half of the cases (54.2%) the acute myocardial infarction was anterior. The incidence of PCI failure was 5.4%. In-ICCU mortality rate was 3.3% (28/856) while in-ICCU complications were detected in the 28.0% (240/856).

Table 1 Clinical characteristics of patients with ST elevation myocardial infarction included in the study
Full size table

Tertiles of uric acid (UA tertile) are shown in Table 2. STEMI patients in the third UA tertile were the oldest (p < 0.001) and showed the highest BMI (p < 0.001) and triglyceride values (p < 0.001) and the lowest eGFR (p < 0.001) and EF (p < 0.001). Killip classes III–IV were more frequent in the third UA tertile that was associated with the highest values of peak Tn I (p = 0.005), NT-proBNP (p < 0.001), and fibrinogen (p = 0.036). Admission glucose and peak glycemia showed a progressive significant increase among UA tertiles (p = 0.006 and p < 0.001, respectively). The incidence of in-ICCU complications was significantly higher in the third UA tertile while in-ICCU mortality rate did not show statistical differences among UA tertiles.

Table 2 Tertiles of uric acid in the study population
Full size table

When evaluating UA values on the basis of gender-specific tertiles (Table 3), it was observed that in the first tertile men showed higher values of uric acid, while in the second and third tertiles no gender-related differences in uric acid values were observed.

Table 3 UA analyis on the basis of sex specific tertiles
Full size table

Logistic regression analysis

Uric acid (considered as continuous variable) was associated with mortality (1 mg/dl step crude OR: 1.24; 95% CI 1.03–1.51; p = 0.025). When adjusted for Tn I and renal failure (as inferred by eGFR < 60 ml/min/1.73 m2), uric acid lost its statistical significance (uric acid 1 mg/dl step: OR: 1.02; 95% CI 0.83–1.26; p = 0.858), while Tn I (100 pg/ml step OR: 1.002; 95% CI 1.000–1.003; p = 0.007) and renal failure (OR 9.16; 95% CI 3.60–23.32; p < 0.001) were independent predictors for in-ICCU mortality: Hosmer–Lemershow goodness-of-fit χ2 = 5.889; p = 0.660 (Table 4).

Table 4 Multivariate and univariate analysis
Full size table

Uric acid was associated with in-hospital complications (1 mg/dl step crude OR: 1.16; 95% CI 1.07–1.26; p < 0.001). At multivariable logistic regression analysis uric acid remained as an independent predictor for in-ICCU complications (1 mg/dl step OR: 1.11; 95% CI 1.01–1.21; p = 0.030) together with admission glycemia (1 g/dl step OR: 1.50; 95% CI 1.19–1.91; p < 0.001) and renal failure (OR: 1.46; 95% CI 0.99–2.16; p < 0.001): Hosmer–Lemershow goodness-of-fit χ2 = 3.554; p = 0.895.

Discussion

The present investigation describes an independent association between higher UA levels and in-ICCU complications in a large series of consecutive STEMI patients submitted to PCI.

Several factors may account for this finding. First, we confirm the strong association between UA levels and known cardiovascular risk factors, since UA values paralleled those of two of the components of the metabolic syndrome (that is BMI and triglycerides) [1, 5, 6]. Second, UA values appeared to be related to infarct size (as indicated by peak Tn I), hemodynamic derangement (as inferred by ejection fraction, Killip class and NT-proBNP) as well as to the metabolic and inflammatory acute responses to stress (as indicated by hyperglycemia and fibrinogen values). Overall it can be speculated that, in the early phase of STEMI patients, hyperuricemia stems from several complex mechanisms, ranging from pre-existing risk factors to the degree of myocardial ischemia, and to the acute metabolic response. The underlying mechanisms linking hyperuricemia to in-ICCU complications in the early phase of STEMI may be related to the pro-oxydant [22] and pro-inflammatory actions [23] attributed to uric acid and previously described in patients with overt ischemic conditions [24, 25]. In experimental models it is observed that, during tissue ischemia, the enzymatic effect of xantine oxidase is the production of reactive species of oxygen (ROS) and uric acid [25, 26]; hyperuricemia per se has been described to impair endothelium-dependent vasodilatation by reduction NO-synthase in animal experiments [27].

Previous studies investigated the relation between UA levels and mortality in patients with acute myocardial infarction, but the results are far from unanimous. Whereas Homayounfar et al. [12] report that hyperuricemia is not an independent prognostic risk factor for in-hospital death after AMI, Kojima et al. [13] find that the total mortality rate of patients whose serum UA concentrations are in the highest quartile is about 3.7 times higher than in those whose UA concentrations are in the lowest quartile. In their retrospective study (the Japanese Acute Coronary Syndrome Study), the Authors conclude that serum UA is a suitable marker for predicting AMI-related future. In their investigation, patients (admitted from January to December 2002) were enrolled within 48 h after onset of symptoms, and reperfusion was performed in 84% (mainly by means of mechanical reperfusion). Our group [14] recently documents in a homogeneous population of 466 STEMI patients all submitted to primary PCI within 12 h from symptoms’ onset, that uric acid levels, measured after mechanical revascularization on ICCU admission, is an independent risk factor for in-hospital mortality. In the present investigation, performed in a larger cohort of STEMI patients, all submitted to mechanical revascularization, we failed to confirm the independent association between uric acid and in-ICCU mortality, since when uric acid was adjusted for eGFR and Tn I (that is renal function and the extension of myocardial damage, respectively) it was no longer associated with early mortality. Discrepancies between the two studies can be related principally to the different number of patients enrolled, since the studies were performed in comparable STEMI populations.

In conclusion, in STEMI patients submitted to PCI, increased uric acid levels identify a subgroup more prone to in-ICCU complications, probably because hyperuricemia stems from several complex mechanisms ranging from pre-existing risk factors to the degree of myocardial ischemia and to the acute metabolic response. Hyperuricemia is not independently associated with early mortality when adjusted for renal function and the degree of myocardial damage.