- Email: [email protected]
FMR1 CGG repeat length and ovarian function in a donor population
Risks of Adverse Outcomes Compared to Fertile Women (AORs 95% CIs)
Singletons
Low Birthweight (<2,500g) Small-for-Gestation Birthweight Premature (<37 weeks) Infant Death*
Twins
Subfertile
ART
Subfertile
ART
1.35 (1.23, 1.49) 0.94 (0.86, 1.08) 1.37 (1.26, 1.49) 1.58 (1.03, 2.43)
1.90 (1.76, 2.05) 1.19 (1.11, 1.28) 1.90 (1.78, 2.04) 1.77 (1.20, 2.60)
1.16 (1.03, 1.30) 1.01 (0.88, 1.17) 1.23 (1.10, 1.37) 1.81 (1.18, 2.77)
1.14 (1.07, 1.22) 0.93 (0.85, 1.01) 1.20 (1.12, 1.28) 0.95 (0.69, 1.31)
*Limited to births 7/04-12/09 with indicators of subfertility but no ART treatment (subfertile), and women with ART treatment. The risks of perinatal and infant outcomes were modeled by plurality using logistic regression, adjusted for parental ages, race/ethnicity, education, payor status, and maternal pre-existing conditions (diabetes and chronic hypertension), and infant gender(s). Adjusted odds ratios (AORs) and 95% confidence intervals are reported. Fertile women were the reference group. RESULTS: The study population included 451,975 pregnancies: 447,510 fertile, 8,778 subfertile, and 13,687 ART; and 459,623 singleton and 10,352 twin pregnancies. The risks of adverse perinatal and infant outcomes by fertility status and plurality are shown below. The risks were significantly increased for singleton and twin infants born to both subfertile and ARTtreated women for low birthweight and prematurity, for singletons born to ART women for small-for-gestational age and infant death, and for singletons and twins born to subfertile women for infant death. CONCLUSIONS: The risks for adverse perinatal and infant outcomes are significantly increased for both subfertile and ART-treated women, even after stratifying by plurality and adjusting for confounding factors. Supported by: NIH grant R01 HD067270. P-170 Tuesday, October 18, 2016 FMR1 CGG REPEAT LENGTH AND OVARIAN FUNCTION IN A DONOR POPULATION. N. Banks,a G. Patounakis,a C. M. Owen,a M. W. Healy,a B. W. Whitcomb,b M. E. Hartman,c K. Devine,c K. S. Richter,c A. DeCherney,a E. Levens,c M. J. Hill.a aNICHD, Bethesda, MD; bUniversity of Massachusetts Amherst, Amherst, MA; cShady Grove Fertility Reproductive Science Center, Rockville, MD. OBJECTIVE: Previous studies have suggested that low FMR1 CGG repeat length (<26) may be a predictor of decreased functional ovarian reserve, but this has not been confirmed. Our objective is to study FMR1 CGG repeat length in a fertile donor population. DESIGN: Retrospective cohort study. MATERIALS AND METHODS: From 2011-2015, a total of 326 egg donors had available FMR1 testing and underwent 514 stimulation cycles. The primary outcome was mature oocyte (MII) yield. Analysis was performed separately using patient maximum and minimum repeat length alleles. Generalized estimating equations (GEE) modeled the association of repeat length with MII yield while adjusting for female age and multiple cycles. A separate analysis with GEE modeling was performed grouping patients by FMR1 allele combinations using the definitions low repeats <26, normal 26-34 and high 34-55 and allele combinations designated by low-low, lownormal, et cetera. RESULTS: FMR1 CGG repeat length ranged from 19-44. Repeat length was not associated with MII yield for either the maximum or minimum allele. FMR1 groups included 31 low-low cycles, 115 low-normal, 11 low-high, 308 Repeat length by allele combinations
Allele Difference in number of Combination MIIs retrieved (unadjusted) Low-low Low-normal Low-high Normalnormal Normal-high
Difference in number of MIIs retrieved*
-0.05 (-2.13-2.03), p¼0.96 2.71 (-1.25-6.67), p¼0.18 -1.3 (-6.25-3.65), p¼0.6 Reference
0.22 (-2.07-2.52), p¼0.84 0.09 (-5.24-5.42), p¼0.97 -0.97 (-6.32-4.38), p¼0.72 Reference
-0.9 (-3.73-1.93), p¼0.53
-0.78 (-4.1-2.53), p¼0.64
*Adjusted for age and antral follicle count
FERTILITY & STERILITYÒ
normal-normal, 54 normal-high and 0 high-high. Comparison between the groups showed no difference in MII yield (table 1). Low-high repeat group was associated with 882 additional IUs total gonadotropin use (p¼0.038) and 116 additional IUs gonadotropin use per MII retrieved (p¼0.04) compared to reference. CONCLUSIONS: As more couples have access to FMR1 CGG repeat length through pre-conception genetic screening, data to guide patient counseling are increasingly important. We do not find a significant correlation between FMR1 repeat length and MII yield in a fertile donor population. Our data are reassuring that ovarian response is comparable in young fertile women with low repeat length alleles, but more studies are needed to follow these women as they age. Supported by: Work supported in part by the NICHD intramural research program. P-171 Tuesday, October 18, 2016 LACK OF PROGESTERONE RISE POST-TRIGGER CORRELATES WITH SUB-OPTIMAL CLINICAL OUTCOMES DESPITE NORMAL L. Sekhon,a,b FERTILIZATION RATE. J. Rodriguez-Purata,a J. A. Lee,a M. Luna,a T. Mukherjee,a,b A. B. Copperman,a,b B. Sandler.a,b aReproductive Medicine Associates of New York, New York, NY; bObstetrics, Gynecology and Reproductive Science, Icahn School of Medicine at Mount Sinai, New York, NY. OBJECTIVE: A rise in LH and P4 after an r-hCG trigger, which mimics an endogenous LH flare, indicates oocyte maturation initiation. P4 level posttrigger has been characterized as an independent predictor of number of total and mature oocytes retrieved. This study sought to analyze clinical outcomes in patients with differing levels luteinization. DESIGN: Retrospective. MATERIALS AND METHODS: Patients underwent an IVF cycle with a hCG trigger between 2002-2015. Patients post-trigger b-hCG<40 mIU/mL were excluded. P4 (ng/mL) rise post-trigger was considered absent (<1), low (1.0-1.5), intermediate (1.6-2.3) or normal rise (>2.3). Cohorts were segregated by fertilization rate (<25%, 25-50%, 51-75%, >75%) and P4 level was compared. Pearson correlations were used to analyze correlation between age, BMI, FSH, AMH and total GND dosage with post-trigger P4 levels. A binary logistic regression analysis was used to model the influence of P4 level on pregnancy rate (PR), clinical PR, loss rate and live birth rate (LBR). RESULTS: A total of 16087 cycles were studied. Patient’s demographic, stimulation, laboratory parameters and clinical outcomes are shown in Table 1. Normal (n¼10797; P4: 4.42.0), intermediate rise (n¼3535; P4: 2.00.3), low rise (n¼1321; P4: 1.30.1) and absent luteinization (n¼434; P4: 0.80.2) cycles were identified. Age (r¼ -0.17), BMI (r¼ -0.09), day 3 FSH (r¼ -0.19) and total GND dose (r¼ -0.23) had a negative correlation and AMH (r¼0.18) had a positive correlation with the level of P4. Fertilization rate was observed similar between <2.3 or >2.3 ng/mL: 51.7% vs. 53.6% vs. 54.2% vs. 55.8%, respectively (p<0.05). P4 level was similar when compared according to fertilization rate (<25, 25-50, 51-75 and >75): 2.9 vs. 3.4 vs. 3.6 vs. 3.2 ng/mL, respectively. Raw data analysis showed PR (34.4% vs. 41.6% vs. 49.6% vs. 56.5%, respectively; p<0.05), clinical PR (26.4% vs. 32.5% vs. 39.7% vs. 46.5%, respectively; p<0.05), loss rate (16.2% vs. 17.2% vs. 18.1% vs.18.2%, respectively; NS) and LBR (17.4% vs. 23.8% vs. 30.9% vs. 38.0%, respectively; p<0.05) increased with increasing P4 level. After adjusting for age, FSH, AMH, BMI and total GND dose in a logistic regression model, elevated P4 was still associated with increased odds of pregnancy (OR 1.074 (95% CI 1.056 - 1.092), p<0.05), clinical pregnancy (OR 1.068 (95% CI 1.051 - 1.086), p<0.05); and live birth (OR 1.074 (95% CI 1.057 - 1.093), p<0.05), with no influence on loss rate (OR 1.009 (95% CI 0.988 - 1.030), NS).
e169
Singletons
Low Birthweight (<2,500g) Small-for-Gestation Birthweight Premature (<37 weeks) Infant Death*
Twins
Subfertile
ART
Subfertile
ART
1.35 (1.23, 1.49) 0.94 (0.86, 1.08) 1.37 (1.26, 1.49) 1.58 (1.03, 2.43)
1.90 (1.76, 2.05) 1.19 (1.11, 1.28) 1.90 (1.78, 2.04) 1.77 (1.20, 2.60)
1.16 (1.03, 1.30) 1.01 (0.88, 1.17) 1.23 (1.10, 1.37) 1.81 (1.18, 2.77)
1.14 (1.07, 1.22) 0.93 (0.85, 1.01) 1.20 (1.12, 1.28) 0.95 (0.69, 1.31)
*Limited to births 7/04-12/09 with indicators of subfertility but no ART treatment (subfertile), and women with ART treatment. The risks of perinatal and infant outcomes were modeled by plurality using logistic regression, adjusted for parental ages, race/ethnicity, education, payor status, and maternal pre-existing conditions (diabetes and chronic hypertension), and infant gender(s). Adjusted odds ratios (AORs) and 95% confidence intervals are reported. Fertile women were the reference group. RESULTS: The study population included 451,975 pregnancies: 447,510 fertile, 8,778 subfertile, and 13,687 ART; and 459,623 singleton and 10,352 twin pregnancies. The risks of adverse perinatal and infant outcomes by fertility status and plurality are shown below. The risks were significantly increased for singleton and twin infants born to both subfertile and ARTtreated women for low birthweight and prematurity, for singletons born to ART women for small-for-gestational age and infant death, and for singletons and twins born to subfertile women for infant death. CONCLUSIONS: The risks for adverse perinatal and infant outcomes are significantly increased for both subfertile and ART-treated women, even after stratifying by plurality and adjusting for confounding factors. Supported by: NIH grant R01 HD067270. P-170 Tuesday, October 18, 2016 FMR1 CGG REPEAT LENGTH AND OVARIAN FUNCTION IN A DONOR POPULATION. N. Banks,a G. Patounakis,a C. M. Owen,a M. W. Healy,a B. W. Whitcomb,b M. E. Hartman,c K. Devine,c K. S. Richter,c A. DeCherney,a E. Levens,c M. J. Hill.a aNICHD, Bethesda, MD; bUniversity of Massachusetts Amherst, Amherst, MA; cShady Grove Fertility Reproductive Science Center, Rockville, MD. OBJECTIVE: Previous studies have suggested that low FMR1 CGG repeat length (<26) may be a predictor of decreased functional ovarian reserve, but this has not been confirmed. Our objective is to study FMR1 CGG repeat length in a fertile donor population. DESIGN: Retrospective cohort study. MATERIALS AND METHODS: From 2011-2015, a total of 326 egg donors had available FMR1 testing and underwent 514 stimulation cycles. The primary outcome was mature oocyte (MII) yield. Analysis was performed separately using patient maximum and minimum repeat length alleles. Generalized estimating equations (GEE) modeled the association of repeat length with MII yield while adjusting for female age and multiple cycles. A separate analysis with GEE modeling was performed grouping patients by FMR1 allele combinations using the definitions low repeats <26, normal 26-34 and high 34-55 and allele combinations designated by low-low, lownormal, et cetera. RESULTS: FMR1 CGG repeat length ranged from 19-44. Repeat length was not associated with MII yield for either the maximum or minimum allele. FMR1 groups included 31 low-low cycles, 115 low-normal, 11 low-high, 308 Repeat length by allele combinations
Allele Difference in number of Combination MIIs retrieved (unadjusted) Low-low Low-normal Low-high Normalnormal Normal-high
Difference in number of MIIs retrieved*
-0.05 (-2.13-2.03), p¼0.96 2.71 (-1.25-6.67), p¼0.18 -1.3 (-6.25-3.65), p¼0.6 Reference
0.22 (-2.07-2.52), p¼0.84 0.09 (-5.24-5.42), p¼0.97 -0.97 (-6.32-4.38), p¼0.72 Reference
-0.9 (-3.73-1.93), p¼0.53
-0.78 (-4.1-2.53), p¼0.64
*Adjusted for age and antral follicle count
FERTILITY & STERILITYÒ
normal-normal, 54 normal-high and 0 high-high. Comparison between the groups showed no difference in MII yield (table 1). Low-high repeat group was associated with 882 additional IUs total gonadotropin use (p¼0.038) and 116 additional IUs gonadotropin use per MII retrieved (p¼0.04) compared to reference. CONCLUSIONS: As more couples have access to FMR1 CGG repeat length through pre-conception genetic screening, data to guide patient counseling are increasingly important. We do not find a significant correlation between FMR1 repeat length and MII yield in a fertile donor population. Our data are reassuring that ovarian response is comparable in young fertile women with low repeat length alleles, but more studies are needed to follow these women as they age. Supported by: Work supported in part by the NICHD intramural research program. P-171 Tuesday, October 18, 2016 LACK OF PROGESTERONE RISE POST-TRIGGER CORRELATES WITH SUB-OPTIMAL CLINICAL OUTCOMES DESPITE NORMAL L. Sekhon,a,b FERTILIZATION RATE. J. Rodriguez-Purata,a J. A. Lee,a M. Luna,a T. Mukherjee,a,b A. B. Copperman,a,b B. Sandler.a,b aReproductive Medicine Associates of New York, New York, NY; bObstetrics, Gynecology and Reproductive Science, Icahn School of Medicine at Mount Sinai, New York, NY. OBJECTIVE: A rise in LH and P4 after an r-hCG trigger, which mimics an endogenous LH flare, indicates oocyte maturation initiation. P4 level posttrigger has been characterized as an independent predictor of number of total and mature oocytes retrieved. This study sought to analyze clinical outcomes in patients with differing levels luteinization. DESIGN: Retrospective. MATERIALS AND METHODS: Patients underwent an IVF cycle with a hCG trigger between 2002-2015. Patients post-trigger b-hCG<40 mIU/mL were excluded. P4 (ng/mL) rise post-trigger was considered absent (<1), low (1.0-1.5), intermediate (1.6-2.3) or normal rise (>2.3). Cohorts were segregated by fertilization rate (<25%, 25-50%, 51-75%, >75%) and P4 level was compared. Pearson correlations were used to analyze correlation between age, BMI, FSH, AMH and total GND dosage with post-trigger P4 levels. A binary logistic regression analysis was used to model the influence of P4 level on pregnancy rate (PR), clinical PR, loss rate and live birth rate (LBR). RESULTS: A total of 16087 cycles were studied. Patient’s demographic, stimulation, laboratory parameters and clinical outcomes are shown in Table 1. Normal (n¼10797; P4: 4.42.0), intermediate rise (n¼3535; P4: 2.00.3), low rise (n¼1321; P4: 1.30.1) and absent luteinization (n¼434; P4: 0.80.2) cycles were identified. Age (r¼ -0.17), BMI (r¼ -0.09), day 3 FSH (r¼ -0.19) and total GND dose (r¼ -0.23) had a negative correlation and AMH (r¼0.18) had a positive correlation with the level of P4. Fertilization rate was observed similar between <2.3 or >2.3 ng/mL: 51.7% vs. 53.6% vs. 54.2% vs. 55.8%, respectively (p<0.05). P4 level was similar when compared according to fertilization rate (<25, 25-50, 51-75 and >75): 2.9 vs. 3.4 vs. 3.6 vs. 3.2 ng/mL, respectively. Raw data analysis showed PR (34.4% vs. 41.6% vs. 49.6% vs. 56.5%, respectively; p<0.05), clinical PR (26.4% vs. 32.5% vs. 39.7% vs. 46.5%, respectively; p<0.05), loss rate (16.2% vs. 17.2% vs. 18.1% vs.18.2%, respectively; NS) and LBR (17.4% vs. 23.8% vs. 30.9% vs. 38.0%, respectively; p<0.05) increased with increasing P4 level. After adjusting for age, FSH, AMH, BMI and total GND dose in a logistic regression model, elevated P4 was still associated with increased odds of pregnancy (OR 1.074 (95% CI 1.056 - 1.092), p<0.05), clinical pregnancy (OR 1.068 (95% CI 1.051 - 1.086), p<0.05); and live birth (OR 1.074 (95% CI 1.057 - 1.093), p<0.05), with no influence on loss rate (OR 1.009 (95% CI 0.988 - 1.030), NS).
e169