Declining sperm counts in the United States? A critical review.

PURPOSE Recent reports suggest declining sperm counts in the United States. These reports did not include all available data and did not account for geographic variations noted in prior studies. We examined all available data on U.S. sperm counts and evaluated whether geographic variations account for the decline suggested. MATERIALS AND METHODS We reviewed all 29 U.S. studies from 1938 to 1996 reporting manually counted semen analyses of 9,612 fertile or presumably fertile men. We determined mean sperm concentrations by geographic location with weighted analysis of variance, and assessed any changes with time by linear regression analysis. RESULTS Mean sperm concentrations from New York were significantly higher than from all other U.S. cities (98.6 versus 71.6 x 10(6) sperm per cc, respectively, p = 0.006). There has been no statistically significant change with time for mean sperm concentrations reported from New York (p = 0.49) or from U.S. cities other than New York (p = 0.62). Analysis without separating by location revealed a decline (p = 0.047). CONCLUSIONS Sperm concentrations are highest in New York compared to other U.S. cities. When accounting for this geographic difference and examining all available data, there appears to be no significant change in sperm counts in the U.S. during the last 60 years. Further studies addressing the causes of geographic variations are needed.