Objective: This study was undertaken to evaluate a decade of data on multifetal pregnancy reductions at centers with extensive experiences.
Study design: A total of 3513 completed cases from 11 centers in 5 countries were analyzed according to year (before 1990, 1991-1994, and 1995-1998), starting and finishing numbers of embryos or fetuses, and outcomes.
Results: With increasing experience there has been a considerable improvement in outcomes, with decreases in rates of both pregnancy loss and prematurity. Overall loss rates in the last few years were correlated strongly with starting and finishing numbers (starting number > or =6, 15.4%; starting number 5, 11.4%; starting number 4, 7.3%; starting number 3, 4.5%; starting number 2, 6.2%: finishing number 3, 18.4%; finishing number 2, 6.0%; finishing number 1, 6.7%). Birth weight discordance between surviving twins was increased with greater starting number. The proportion of cases with starting number > or =5 diminished from 23.4% to 15.9% to 12.2%. The proportion of patients >40 years old increased in the last 6 years to 9.3%. Gestational age at delivery did not vary with increasing maternal age but was inversely correlated with starting number.
Conclusion: Multifetal pregnancy reduction outcomes at our centers for both losses and early prematurity have improved considerably with experience. Reductions from triplets to twins and now from quadruplets to twins carry outcomes as good as those of unreduced twin gestations. Patient demographic characteristics continues to change as more older women use assisted reproductive technologies. In terms of losses, prematurity, and growth, higher starting numbers carry worse outcomes.