The distribution of serum uric acid values in a population unselected as to gout or hyperuricemia: Tecumseh, Michigan 1959–1960

Abstract

Serum uric acid determinations were made for 6,000 study subjects from the Tecumseh Community Health Study, Tecumseh, Michigan, 1959–1960. These 6,000 subjects represent a natural population without prior selection for either hyperuricemia or gout. Male subjects, of whom there were 2,987, had serum uric acid values ranging from 1.0 to 13.6 mg. per 100 ml. with an arithmetic mean of 4.9 mg. per 100 ml. and a standard deviation of 1.40 mg. per 100 ml. Female subjects, 3,013 in all, had serum uric acid values ranging from 1.0 to 11.6 mg. per 100 ml. with an arithmetic mean of 4.2 mg. per 100 ml. and a standard deviation of 1.16 mg. per 100 ml. The sex specific distribution curve for male subjects is broad and slightly skewed to the high value end of the scale. The curve for female subjects, by contrast, is narrow, peaking sharply at a value well below that of the curve in male subjects and is somewhat more skewed toward the upper end of the scale. The age-sex specific mean serum uric acid values for both sexes are lowest in the four year olds with rising trend of values in the five to nine and ten to fourteen year age groups. At about puberty the curves begin to separate. The curve for male subjects continues to rise to a peak at ages twenty to twenty-four years; it then falls slightly and plateaus at a level of about 5.2 mg. per 100 ml. For female subjects, there is a slight rise in serum uric acid values beyond puberty but the curve shortly falls again and plateaus at a level of about 4.0 mg. per 100 ml. until the age of menopause, when it rises gradually to approach closely that of male subjects in the early fifties. The data with reference to relative distribution above arbitrarily defined cutting points suggest that these points, commonly used in clinical medicine to define “hyperuricemia,” are unrealistically low and, in addition, fail to take into account important differences associated with age. The observed serum uric acid level for each individual subject has been adjusted or standardized to that of the appropriate age-sex group. The distribution curves of the present data show no suggestion of bimodality and suggest genetic polymorphism.