Meta-analysis on luteal phase support: reply

Elizabeth A. Pritts

Department of Obstetrics and Gynecology, University of Wisconsin, Madison, Wisconsin, USA e-mail: eapritts{at}wisc.edu

Dear Sir,

This is in response to the letter by Dr Claire Bourgain regarding our recent meta-analysis (Pritts and Atwood, 2002Go).

Dr Bourgain requested the addition of data by Smitz et al. (1992Go) to the meta-analysis. This investigation was a randomized trial of 262 women comparing vaginal versus i.m. progesterone treatment in the luteal phase of assisted reproductive technology cycles (Smitz et al., 1992Go).

We were aware of this excellent study, however we were unable to include it in our data analysis due to the choice of treatments in both groups. In our meta-analysis, we combined trials in which vaginal progesterone was compared with i.m. progesterone. However, none of these trials added additional hormonal support in the luteal phase. Smitz et al. used either intravaginal or i.m. progesterone, but added oral estrogen treatment in both groups (Smitz et al., 1992Go). The addition of Smitz’s data would confound these particular statistics, rendering the analysis biased.

Interestingly, if the data from Smitz et al. (1992Go) are added to our meta-analysis, the outcome is virtually unchanged (see Table I). In our publication, the relative risk (RR) for clinical pregnancy rate favoured i.m. progesterone at 1.33, with a 95% confidence interval of 1.02–2.19. The addition of data from Smitz et al. brings the RR to 1.2 with a 95% confidence interval of 1.01–1.45. There was also enough data in the paper by Smitz et al. to add the ongoing pregnancy and miscarriage data to the original statistics. These outcomes were essentially unchanged when the new data were added.


View this table:
[in this window]
[in a new window]
 
Table I. Original meta-analysis outcomes compared with those after addition of data by Smitz et al.
 
When taken with the RR of delivery per transfer of 2.06 with a 95% confidence interval of 1.48–2.88, I again must assert that based upon the data, it is more beneficial to use the i.m. route for administration of progesterone in the luteal phase.

Prior to this meta-analysis, I was an avid user of vaginal progesterone. Since our findings, I have changed my practice and utilize only i.m. progesterone for luteal support. I have changed my practice to the dismay of my patients, however. I await further trials comparing new and improved vaginal progesterone preparations that confer fertility outcomes similar to i.m. progesterone preparations.

References

Pritts, E.A. and Atwood, A.K. (2002) Luteal phase support in infertility treatment: a meta-analysis of the randomized trials. Hum. Reprod., 17, 2287–2299.[Abstract/Free Full Text]

Smitz, J., Devroey, P., Faguer, B., Bourgain, C., Camus, M. and Van Steirteghem, A.C. (1992) A prospective randomized comparison of intramuscular or intravaginal natural progesterone as a luteal phase and early pregnany supplementation. Hum. Reprod., 7, 168–175.[Abstract]





This Article
Extract
FREE Full Text (PDF )
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by Pritts, E. A.
PubMed
PubMed Citation
Articles by Pritts, E. A.