Introduction

There have been several reports of a relationship between raised serum uric acid levels and cardiovascular disease, but they have usually been plagued by confounding factors (e.g. obesity, hypertension), whose relevance could not be excluded. The NHANES I study provides a database that has allowed scientists at the Albert Einstein College of Medicine, Bronx, New York, to make a fairly definitive assessment of the association.

Method

A 1995 report from the First National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Follow-up Study (NHEFS), 1971-1987, found that the baseline uric acid level was an independent predictor of mortality and ischemic heart disease in women, but not in men. An additional 5 years of follow-up was available that provided sufficient data for more detailed analysis.

The baseline survey, sponsored by the US National Center for Health Statistics, was conducted from 1971-1975 on a probability sample of over 20,000 persons in the US population. A sub-sample of nearly 7,000 subjects provided detailed information, which included serum uric acid levels in 6,651 persons. After exclusions for existing illnesses (including gout and cardiovascular diseases) and missing data, there were 5,926 subjects aged 25 to 74 years in the analysis.

Baseline questions provided information on conditions such as diabetes, heart disease, hypertension, and gout, as well as smoking status, and menstrual cycle status. Outcome measures were obtained from medical records and death certificates. Deaths were analyzed for all causes, total cardiovascular disease, and ischemic heart disease.

Subjects were stratified into quartiles of baseline uric acid level, by sex. Mortality rates were adjusted for age and race. Hazard regression analyses were adjusted for age, race, body mass index, cholesterol level, smoking status, alcohol consumption, and history of hypertension and diabetes.

Results

The mean serum uric acid level for the total collective was 32.7 mmol/L (5.5 mg/dL). Men had a significantly higher mean than women and black persons higher means than whites persons. Mean values increased with age in women, but not in men.

Higher quartiles of uric acid levels were associated with higher blood pressures, higher cholesterol levels, increased body mass indices, raised serum creatinine levels, diuretic use, increased alcohol consumption, and diabetes (women only). Smoking status and previous kidney disease were not associated with increased uric acid levels.

During the follow-up period (average 16.4 years) there were 1,593 deaths (26.9%). Of these, 731 (45.9%) were attributed to cardiovascular disease (ischemic heart disease (IHD), stroke, and other).

Death rates for ischemic heart disease were significantly higher in the 4th uric acid quartile, when compared with the 1st, for both sexes:

	Uric Acid in mmol/L (mg/dL)	IHD Death Rate/1000 patient-years	Risk Ratio (95% CI)
MEN
1st Quartile	<32.1 (<5.4)	4.59	1.0 (defined)
2nd Quartile	32.1 - 36.3 (5.4-6.0)	4.58	1.00 (0.13-3.34)
3rd Quartile	36.4 - 41.6 ( 6.1-7.0)	5.80	1.26 (0.45-3.76)
4th Quartile	>41.6 (>7.0 )	8.14	1.77 (1.08-3.98)
WOMEN
1st Quartile	<23.8 (<4.0 )	2.02	1.0 (defined)
2nd Quartile	23.8 - 28.0 (4.0-4.49)	2.31	1.15 (0.51-3.89)
3rd Quartile	28.1 - 33.3 (4.5-5.6)	3.02	1.50 (0.89-4.01)
4th Quartile	>33.3 (>5.6)	6.05	3.00 (1.45-6.28)

Separate multivariate Cox regression analyses revealed that, after controlling for age, race, body mass index, cholesterol level, smoking status, alcohol consumption, history of hypertension and diabetes, and recent diuretic use, hazard ratios in men for each 6 mmol/L (1 mg/dL) increase in uric acid were 1.09 and 1.17 for cardiovascular disease and IHD mortality, respectively. The same ratios for women were 1.26 and 1.30, respectively.

Comment

As the authors of the study say, "in this population . . .serum uric acid level bore a continuous, independent, specific, and significant positive relationship to cardiovascular mortality". The results differ from those of another analysis, based on Framingham Heart Study data.¹This may be due to the different populations examined - the Framingham population is nearly entirely white, whereas the NHANES is much more representative of the US population in general; moreover, the latter were slightly older. Cardiovascular mortality was twice as high in the NHANES population as in the Framingham study.

In addition to gout, hyperuricemia is seen with raised alcohol intake, obesity, diabetes, hyperlipidemia, hypertension, renal disease, IHD, and diuretic use. These conditions correspond largely to those seen in the insulin resistance syndrome, or syndrome X. It may be that an increase in serum uric acid is an expression of insulin resistance - studies have found a correlation between increased uric acid levels and decreased insulin-stimulated glucose uptake.

Do increased uric acid levels have a causal role in cardiovascular disease? This is not yet known, although uric acid has been shown to play a role in platelet adhesiveness and aggregation, as well as hypertensive renal disease.

What can one do with the information from this study? It represents a piece in the puzzle depicting how coronary artery disease develops. From a more practical viewpoint, it shows that serum uric acid determination can be an additional indicator of possible risk for cardiovascular disease. And any advanced warning is useful, if it leads to the appropriate corrective measures.