I was very interested in reading the article of Schwarze
named 'DHEA use to improve likelihood of
IVF/ICSI success in patients with diminished ovarian reserve: A systematic review and
meta-analysis'. In the article they combined the results of 5 different studies and
concluded that DHEA supplementation prior to IVF treatment has a positive effect on the
pregnancy rate and frequency of abortion, though no effect on the mean number of
retrieved oocytes was found. As I was reading the article closely, I discovered some
errors that should be checked and corrected.First of all, in the 'materials and methods' section, clinical pregnancy is defined as
"the identification of at least one embryo displaying cardiac activity". In figure 2,
showing the clinical pregnancy rate per initiated cycle, the number of patients in the
control group of the Barad study (Barad is listed as 18. However, when reading the article of
Barad the
clinical pregnancy rate is only 11, while the number of patients with positive hCG is
18. This does not match the earlier described definition of clinical pregnancy.Secondly, the study of Wiser performed 2 study cycles of IVF treatment with a total of 33 patients.
In figure 2, the results of the first treatment cycle are processed with an outcome of 4
and 2 clinical pregnancies in the treatment and control group respectively. However,
figure 3, showing the abortion rate, contains the results after the second treatment
cycle, containing more clinical pregnancies. Both figures should include the results of
the same treatment cycle to be able to make an accurate comparison. Additionally, figure
4, containing the mean number of oocytes recovered, states that the SD of the treatment
group is 2.6 and the SD of the control group is 1.7. However, in the original study
(Wiser et al., 2010), those numbers are being
described as the mean oocyte retrieval from 'the cycle prior to the stud y cycle'. After
the second cycle of IVF treatment those numbers are respectively 3.2 and 3.5 in de
treatment and control group with a total of 26 and 25 patients. These are the numbers
that should be taken into account in the results.At last, the study of Vlahos listed 6 live births out of 8 clinical pregnancies in the control
group. However, figure 3 states that the abortion rate in this group is 6/8, while this
is only 2/8.All together, these errors might affect the overall results and could influence the
conclusion. Therefore the errors should be checked and there should be determined
whether adjustment of the conclusion is necessary.