Uric Acid as a Prognostic Marker in Acute Heart Failure — New Expectations from an Old Molecule

Heart failure is one of the most prominent clinical diagnoses in modern medicine, be it from the clinical, prognostic, epidemiological or socio-economic point of view. In times of flow-chart guided diagnostic and therapeutic decision trees there is a growing need for diagnostic tools to objectively channel our options as early as possible. Accordingly, a growing number of markers and indicator parameters have been evaluated to assess disease status and putative further disease development. Nearly every month another parameter or assessment method is reported to contribute to the estimation of heart failure status and progression. These markers are becoming increasingly costly, complex to obtain or to interpret and most of them may never make it into the everyday clinical setting. Recently, an increasing interest in uric acid (UA) has emerged, as a number of studies have shown that hyperuricaemia is a constant feature of metabolic imbalance within heart failure pathophysiology. The independent prognostic significance of hyperuricaemia in chronic heart failure (CHF) has been observed to be comparable and even stronger than many other established prognosticators in CHF 1. If confirmed in prospective and appropriately powered studies, UA could become a widely used marker to evaluate CHF status. In contrast to many novel markers, UA is an easily measured and widely available parameter. This concept has now been extended to acute heart failure: in this issue of the European Journal of Heart Failure, Pascual-Figal et al. report on the prognostic value of hyperuricaemia in patients hospitalized due to acute decompensated heart failure 2. High uric acid levels in these patients indicated a higher risk of death and/or re-hospitalisation. Although this study is well in line with the observations in chronic heart failure, the exact role of UA itself and the underlying mechanism of hyperuricaemia in this context is still a matter of debate. Often in the discussion of hyperuricaemia in heart failure patients the dependency of UA levels on renal function is stressed and increasing UA levels are thought to be merely a reflection of impaired renal function. Undoubtedly, renal impairment - often in a state of chronic compensated renal insufficiency - is frequent in CHF. UA excretion depends on renal function and might be further compromised by diuretic treatment in these patients. Therefore, an association between UA levels and impaired renal function is a common finding, also confirmed in the study by Pascual-Figal et al. Several lines of evidence suggest however, that overproduction of UA is the dominating factor in CHF rather than undersecretion to account for elevated UA levels (Fig. 1). This may be of importance in the distinction from gout, where undersecretion indeed is the dominant cause accounting for about 80% of clinical events. Metabolic imbalance with increased catabolic turnover and metabolite diversion such as from increased PRPP-synthetase activity may contribute to accelerated substrate supply for purine degradation via the xanthine oxidase (XO) pathway. Inflammatory immune activation and increased oxygen radical accumulation may further contribute in the setting of CHF to up-regulation of XO activity. In accordance, direct measurement of XO in CHF has shown up-regulated enzyme activity 345. Further, in numerous studies the association of UA with mortality, clinical and pathophysiologic surrogate markers remained significant after adjustment for renal function and diuretic treatment. Moreover, the concept of up-regulated XO activity as the primary cause of hyperuricaemia in CHF is supported by the therapeutic effect of XO inhibition in a number of studies For overview see 6). All this suggests that overproduction of UA might indeed be the predominant underlying mechanism of hyperuricaemia in CHF. The increased accumulation of reactive oxygen species (ROS) as an inevitable by-product of XO activity is understood as the principal underlying pathologic characteristic of hyperuricaemia in heart failure 7. The distinction from gout in view of pathophysiologic mechanisms opens the field for the discussion of another point: What cut off level for UA is best to characterise the increased risk of heart failure patients? The range of normal values of UA refers to a Gaussian distribution of physiologic UA levels in male and female patients. The cut-off for hyperuricaemia as currently applied in many studies indicate the (gender independent) limits of serum solubility of UA where precipitation of UA (i.e. acute gout) becomes imminent. This does, however, by no means reflect the associations with morbidity and mortality that have been observed for hyperuricaemia in heart failure patients. What UA levels should be used here to identify patients with increased metabolic imbalance? Pascual-Figal et al. found significantly increased mortality and increased hospitalisation rates for heart failure using the conventional cut-offs for UA normal levels derived from gout related data (i.e. 7 mg/dL in men and 6 mg/dL in women). When tested in quartiles of this study population, a significant UA cut-off for increased mortality was observed by the authors at 7.2 and 7.0 mg/dL in male and female patients, respectively. We do not know whether this is the best cut off in this dataset. Calculation of receiver operating characteristic (ROC) curves would be required to identify UA cut-off levels with the optimum specificity/sensitivity balance. The authors did not perform ROC analyses for UA in their study. In a previous study in chronic heart failure, using ROC analysis we identified and validated a UA level of 565 μmol/L (i.e. 9.5 mg/dL) as the optimum cut-off to predict impaired prognosis in CHF patients. Whether this margin could be applied as well in acute heart failure needs to be tested in future studies. The most prominent prognostic markers in acute heart failure are natriuretic peptides and it would be interesting to see a comparison between UA and such established prognosticators. This is, unfortunately, not reported in the current study and such comparison would be required to fully evaluate the potential of UA as prognostic marker in acute heart failure. Notably, the importance of UA as a prognostic marker in heart failure is not dependent on the question whether UA itself is actively involved in disease processes or whether it merely reflects the level of XO activity (and hence ROS accumulation). Some recent evidence suggests UA to be a principal endogenous danger signal mediating immune response upon cell injury 8. Accordingly, iv infusion of UA in a mouse model caused increased endotoxin stimulated tumour necrosis factor-α production and hence pro-inflammatory immune activation 9. In contrast, the antioxidant capacity of UA has repeatedly been reported 10. The debate about whether UA contributes to or prevents detrimental effects in heart failure patients is ongoing. In conclusion, hyperuricaemia has emerged as a significant predictor of disease state and progression of chronic heart failure. The work by Pascual-Figal et al has extended this concept to patients with acute heart failure. If confirmed, UA may add valuable information for clinical evaluation of heart failure status. The assessment of UA is widely available at low cost, which may be an advantage for widespread use of this marker in clinical and ambulatory settings. Whether UA itself actively contributes to disease processes or is a mere marker of disease status needs to be clarified. On the basis of up-regulated XO activity accounting for ROS accumulation, observed hyperuricaemia could become a therapeutic target in HF patients. Promising results from a number of small scale and pilot studies suggest a role for XO inhibitors in tailored patient populations.