De novo mutations producing unstable hemoglobins or hemoglobins M : I. Establishment of a Depository and use of data to test for an association of de novo mutation with advanced parental age.

Hemoglobins M and unstable hemoglobins cause clinical syndromes that are transmitted in autosomal dominant fashion. Pedigrees of 50 probands with de novo mutations producing unstable Hb disease or Hb M disease were compiled. Cases were ascertained (1) by screening the relevant literature published from 1950 through 1980 and (2) through personal communication. Additional pedigree data on several published cases were collected, and a depository containing all available information rekated to de novo Hb mutants was established. The 50 probands were born in 14 countries between 1922 and 1976. Paternity was tested in 36% of the cases, and no instance of false paternity was noted.The data were used to test for an association of advanced parental age with the appearance of de novo mutants. Paternal ages at the probands' births ranged from 20 to 50 years, with a mean of 32.7 years. Maternal ages ranged from 18 to 43 years, with a mean of 28.5 years. For each year and country (or, where necessary, for the nearest possible year and/or a demographically similar country), the cumulative frequency distributions of the ages of parents who had a child in that country and year were computed; the ages of each proband's father and mother were then expressed as percentiles on these distributions. The distribution of paternal age percentiles was shifted toward the upper end of the range, with 11 of the 50 paternal ages falling between the 90th and 100th percentiles. The distribution of maternal age percentiles was more complex, with one peak (10 of 50 ages) falling between the 30th and 40th percentiles and a second peak (10 of 50 ages), between the 90th and 100th percentiles. These distributions, though suggestive of an association of advanced parental age and the appearance of de novo mutations that cause unstable Hb disease or methemoglobinemic cyanosis, were not significantly different from those uniform distributions expected in the absence of a parental age effect.