Have Patients Urinate for Their Hearts’ Sake

A new multi-marker approach for the prediction of heart failure incidence…. 

The association between microalbuminuria and poor outcomes in patients with diabetes mellitus has long been known. In the last decade, similar associations between microalbuminuria and various outcomes have been identified in the general population and in those with HIV. Researchers’ understanding of this relationship was furthered by the observation that higher albumin excretion rates, even within the range previously categorized as “normal,” were associated with greater mortality risk. Finally, it has been demonstrated that in terms of mortality risk, urinary albumin excretion is a better risk stratifier than traditional risk factors, such as cholesterol.

The analysis used the Framingham Offspring Study to examine the comparative abilities of a panel of candidate biomarkers (C-reactive protein, plasminogen activator inhibitor-1 homocysteine, aldosterone-to-renin ratio, B-Type natriuretic peptide, and urine albumin-to-creatinine ratio) along with the urinary albumin excretion rate to predict new episodes of heart failure (HF). The Framingham Offspring Study enrolled 5,124 individuals who were children of the original Framingham cohort participants. These individuals have been examined approximately every 4 years, and this study analyzed the examination that took place between 1995 and 1998. HF events were adjudicated by 3 physicians using the Framingham HF criteria. As a secondary analysis, these associations were also assessed with respect to events that were not associated with acute ischemia or myocardial infarction (e.g., nonischemic).

At the sixth visit during which the biomarkers were measured, men were on average 59 years old and women were 58 years old. Hypertension was found in 30% of the men and 25% of the women, and the proportions of individuals with diabetes were, respectively, 12% and 9%. Mean values of biomarkers were all within normal limits, as expected for people who are not experiencing acute events. Urine albumin-to-creatinine ratios were, respectively, 4.88 (2.15, first quartile; 10.93, third quartile) and 8.55 (3.57, first quartile; 17.24, third quartile) for men and women.

Over a mean follow-up of 9.4 years, the biomarker panels were significantly associated with risk for HF (P = .00005). Removing one biomarker at a time from the multivariable model, B-Type natriuretic peptide and the urinary albumin-to-creatinine ratio emerged as the biomarkers most central to the prediction of HF risk. The hazard ratio per increment of 1 standard deviation of the log was 1.52 (95% CI, 1.24-1.87; P < .0001) for B-Type natriuretic peptide and 1.35 (95% confidence interval [CI], 1.11-1.66; P = .002) for the urinary albumin-to-creatinine ratio. When the analysis was restricted to events without an obvious ischemic etiology, the relationships remained the same. Per increment of 1 standard deviation of the log, B-Type natriuretic peptide was associated with a hazard ratio of 1.78 (95% CI, 1.37-2.31; P < .0001) and the urinary albumin-to-creatinine ratio was associated with a hazard ratio of 1.39 (95% CI, 1.06-1.82; P = .02).

The medical community clearly understands and uses exceptionally well a number of different parameters as cardiac risk stratifiers. Pre-test probability is a concept that is frequently discussed when considering an evaluation for coronary artery disease.

One of the novel aspects of this analysis is the association between B-Type natriuretic peptide and the increased risk for new HF. However, the most important and practical aspect of this analysis is the association between the albumin-to-creatinine ratio and HF. This study adds to a growing literature that underscores the strength of the associations between abnormalities in urinary albumin excretion and cardiovascular events or death.

As we continue to focus on maximizing quality of care in an environment of ever-increasing pressure to contain costs, the noninvasive nature of urine albumin measurement and its lesser cost, compared with markers — such as B-Type natriuretic peptide and C-reactive protein — must be considered. Given a lesser cost and better predictive value, does it make sense not to make every one of your patients pee in the cup?