To the Editor:
We read with much interest the study of Lai et al. published recently in the Journal (1). We were pleased to notice that our definition of premature luteinization (PL) employing the progesterone/estradiol (P/E2) ratio, on the day of hCG administration (hCG day), in controlled ovarian hyper-stimulation (COH) cycles (2, 3,) is continuing to trigger a debate.
In their retrospective study, Lai et al. evaluated the relation between PL, employing the P/E2 ratio, and clinical outcome in infertile women with normal ovarian reserve, undergoing IVF-ET with the long GnRH-a protocol. Their conclusion was that the increased P/E2 ratio, as an indicator of PL, has a poor predictive value for pregnancy outcome.
The authors present a major paradox in their study. At the beginning they conducted a receiver operator characteristic (ROC) analysis to search for the most efficient cutoff value for the P/E2 ratio associated with PL to discriminate between successful and unsuccessful IVF outcome. An optimal cutoff value of the P/E2 ratio for PL was >1.2 for “not achieving pregnancy”. Strangely, when they used this criterion to form two groups, the PL group and the non-PL group, they found that there was no difference in pregnancy rate between the two groups.
Although patients with low ovarian reserve were excluded a priori from this study, it is strange to notice that some women included needed a dosage of 450 IU/day of hMG/FSH to achieve adequate COH. Table 2 clearly shows that women with PL had clear manifestations of diminished ovarian response manifested by significantly lower E2 level, number of >15 mm follicles and number of oocytes retrieved. Taken together, the P/E2 ratio still differentiates between two populations of patients, low and good ovarian response and this was not adequately addressed in the manuscript.
Basal ovarian reserve parameters included in the study were FSH, LH and E2 on day 2 of the stimulated cycle following pituitary down-regulation. These parameters cannot be reliable at this stage of treatment, since there is a 40-50% drop of their levels following the long GnRH-agonist administration (4). It is therefore a common practice to perform these tests in a natural cycle before COH is initiated. Other ovarian reserve parameters should have been employed at this stage; the antral follicle count or the anti-mullerian hormone level (4). Unfortunately, this was not performed in the study.
For some reason, factors chosen to be included in the logistic regression analysis to examine which could affect pregnancy outcome were day 2 LH level, P level on hCG day, number of transferred embryos, endometrial thickness and total dosage of hMG (table 3). Significant factors related to pregnancy outcome such as, day 3 FSH level (natural cycle), E2 level and number of follicles on hCG day, total number of oocytes and embryos achieved and embryo grading, were not included. These factors, should they have been included, might have changed the final logistic regression analysis and results.
Beyond these key drawbacks, we are concerned for the inappropriate understanding of our previous definition of high P/E2 ratio as a predictor of poor outcome. Our study indicated that in the presence of low E2 levels and the concomitant low number of follicles, relatively high P values on hCG day reflects low fertility potential. Due to the above problematic methodology, the paradoxical lack of difference in conception rate, between women with high versus low P/E2, should be regarded with extreme caution.
Johnny S. Younis, M.D.a,b
Izhar Ben-Shlomo, M.D.a
Moshe Ben-Ami, M.D.a,b
aDepartment of Obstetrics & Gynecology
Poriya Medical Center
bRappaport Faculty of Medicine
The Technion – Israel Institute of Technology
1. Lai TH, Lee FK, Lin TK, Horng SG, Chen SC, Chen YH, Wang PC. An increased serum progesterone-to-estradiol ratio on the day of human chorionic gonadotropin administration does not have a negative impact on clinical pregnancy rate in women with normal ovarian reserve treated with a long gonadotropin releasing hormone agonist protocol. Fertil Steril 2009;92:508–14.
2. Younis JS, Haddad S, Matilsky M, Ben-Ami M. Premature luteinization: could it be an early manifestation of low ovarian reserve? Fertil Steril 1998;69:461–5.
3. Younis JS, Matilsky M, Radin O, Ben-Ami M. Increased progesterone/estradiol ratio in the late follicular phase could be related to low ovarian reserve in in vitro fertilization–embryo transfer cycles with a long gonadotropin-releasing hormone agonist. Fertil Sterilt 2001;76:294 –99.
4. Jayaprakasan K, Campbell BK, Hopkisson JF, Clewes JS, Johnson IR, Raine-Fenning NJ. Effect of pituitary desensitization on the early growing follicular cohort estimated using anti-Mullerian hormone. Hum Reprod 2008;23:2577-83.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2010.01.002
The Authors Respond:
We thank Drs. Johnny S. Younis, Izhar Ben-Shlomo and Moshe Ben-Ami for their interest in our article (1).
Dr. Younis’ team defined the P/E2 ratio >1 on hCG day as premature luteinization (PL). They suggested that P/E2 > 1 on hCG day was a manifestation of low ovarian reserve and related to low pregnancy outcomes in women with unexplained infertility undergoing controlled ovarian hyperstimulation (COH) with GnRH-a protocol (2, 3). However, we do not know why they used P/E2 > 1 as the cutoff value. It is reasonable to conduct a receiver operator characteristic (ROC) analysis to search the most efficient cutoff value for the P/E2 ratio to discriminate between successful and unsuccessful IVF outcome in our study. The optimal P/E2 cutoff value was ≧ 1.2 for not achieving pregnancy, which had a sensitivity of 75.0%, a specificity of 32.0%, a positive predictive value of 37.0%, and a negative predictive value of 71.0%. It is clear that using P/E2 ratio as a PL indicator has poor sensitivity, specificity, positive predictive value and negative predictive value to predict pregnancy outcome.
Whether the unfavorable pregnancy outcome is caused by low ovarian reserve or PL is a debate issue and needs to be clarified. We think that the influence of increased P/E2 ratio on hCG day on the clinical outcomes can be studied more clearly by excluding women with poor ovarian reserve. Therefore, we set the exclusion criteria of poor ovarian reserve as follows: age > 40 years; previous COH with poor response; poor ovarian reserve with a day 2 basal FSH ≧10 IU/L in natural cycle. We apologize for lacking the data of antral follicle counts and AMH level as ovarian reserve parameters in the study because these parameters do not perform routinely in our unit.
A logistic regression analysis by backward elimination method was used to evaluate the clinical variables that could affect pregnancy outcome in our study. Actually, all the variables in table 1 and 2 were included for the analysis. However, the results showed the significant variables for pregnancy outcome were including LH level on day 2, P level on hCG day, the number of embryos transferred, endometrial thickness, and the total doses of hMG. Unfortunately, the P/E2 ratio was not included as a significant factor.
Finally, we would like to emphasize that our goal was to evaluate the relationship between the P/E2 ratio on hCG day and the clinical outcomes in infertile women with normal ovarian reserve treated with a long GnRH-a protocol. The study was not focus on the relationship of PL and decreased ovarian reserve. Our data demonstrated that in women with normal ovarian reserve, PL was associated with a low response and oocyte yield, but not with poor pregnancy outcome.
Tsung-Hsuan Lai, M.D.a,b
Tseng-Kai Lin, M.D.a,c
Fa-Kung Lee, M.D., M.P.H.a,c
aDepartment of Obstetrics and Gynecology
Cathay General Hospital
bFu Jen Catholic University School of Medicine
cDepartment of Medical Laboratory Science and Biotechnology
1. Lai TH, Lee FK, Lin TK, Horng SG, Chen SC, Chen YH et al. An increased serum progesterone-to-estradiol ratio on the day of human chorionic gonadotropin administration does not have a negative impact on clinical pregnancy rate in women with normal ovarian reserve treated with a long gonadotropin releasing hormone agonist protocol. Fertil Steril 2009;92:508-14.
2. Younis JS, Haddad S, Matilsky M, Ben-Ami M. Premature luteinization: could it be an early manifestation of low ovarian reserve? Fertil Steril 1998;69:461-5.
3. Younis JS, Matilsky M, Radin O, Ben-Ami M. Increased progesterone/estradiol ratio in the late follicular phase could be related to low ovarian reserve in in vitro fertilization-embryo transfer cycles with a long gonadotropin-releasing hormone agonist. Fertil Steril 2001;76:294-9.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2010.01.001