OBJECTIVE: To evaluate the variability of follicle stimulating hormone (FSH) and estradiol (E2) results from different laboratories. DESIGN: Prospective analyses of sera separated and frozen, obtained from 15 female patients in the follicular phase. All kits used for radioimmunoassay (RIA) for FSH were from the Second International Reference Preparation (IRP). Four different kits were used for FSH and three for E2. SETTING: Sera obtained from each patient were separated into five test tubes and frozen. Analysis of all samples was done on the same day in each one of five participating hospital-based RIA laboratories (North Shore, Cornell, Yale, Mount Sinai, and Norfolk). PARTICIPANTS: Fifteen consecutive patients from the assisted reproductive technology program at North Shore University Hospital participated in the study. MAIN OUTCOME MEASURES: After FSH and E2 levels were tabulated for each laboratory, mean levels were calculated. RESULTS: Using the Bonferroni adjusted pairwise multiple comparisons analysis, significant differences were found between three groups of laboratories within the FSH results and three different groups within the E2 results. CONCLUSIONS: [1] Different results may be obtained on the same sera for FSH and E2 at different laboratories; [2] for FSH, even though the same IRP is used, different results are reached; [3] for E2, even when the same kits are used, results may be significantly different. These results suggest that specific FSH and E2 levels used to predict chances for achieving a viable pregnancy through in vitro fertilization should be interpreted with caution across institutions.
OBJECTIVE: To evaluate the variability of follicle stimulating hormone (FSH) and estradiol (E2) results from different laboratories. DESIGN: Prospective analyses of sera separated and frozen, obtained from 15 female patients in the follicular phase. All kits used for radioimmunoassay (RIA) for FSH were from the Second International Reference Preparation (IRP). Four different kits were used for FSH and three for E2. SETTING: Sera obtained from each patient were separated into five test tubes and frozen. Analysis of all samples was done on the same day in each one of five participating hospital-based RIA laboratories (North Shore, Cornell, Yale, Mount Sinai, and Norfolk). PARTICIPANTS: Fifteen consecutive patients from the assisted reproductive technology program at North Shore University Hospital participated in the study. MAIN OUTCOME MEASURES: After FSH and E2 levels were tabulated for each laboratory, mean levels were calculated. RESULTS: Using the Bonferroni adjusted pairwise multiple comparisons analysis, significant differences were found between three groups of laboratories within the FSH results and three different groups within the E2 results. CONCLUSIONS: [1] Different results may be obtained on the same sera for FSH and E2 at different laboratories; [2] for FSH, even though the same IRP is used, different results are reached; [3] for E2, even when the same kits are used, results may be significantly different. These results suggest that specific FSH and E2 levels used to predict chances for achieving a viable pregnancy through in vitro fertilization should be interpreted with caution across institutions.
Authors: A H Watt; A T Legedza; E S Ginsburg; R L Barbieri; R N Clarke; M D Hornstein Journal: J Assist Reprod Genet Date: 2000-05 Impact factor: 3.412