Table 1.
Data about general features of the manuscripts evaluated in this systematic review and meta-analysis including the dose and the days of transdermal testosterone treatment
| Authors & year | Type of study | Patients (n) | Definition of poor responder | COS protocol | Intervention | Main findings |
|---|---|---|---|---|---|---|
| Massin et al. (2006) | - Type: RCT - Randomization: allocation sequence generated by a random permutation table (blocks of four) - Blinding: double blind - Treatment vs placebo |
Intention-to-treat: - Total: 53 - Treatment: 27 - Control: 26 Per protocol: - Total: 49 - Treatment: 24 - Control: 25 |
i) Previous poor response: E2 < 1200 pg/ml at day of hCG; ≤ 5 COCs retrieved - Basal FSH > 12 IU/l ii) Decreased ovarian reserve: day-3 of spontaneous cycle with FSH > 12 IU/l, E2 > 70 pg/ml and inhibin B < 45 pg/ml - Exclusion: > 42 years |
- COS protocol: long GnRH agonist (T: 22; C: 19), other ns - Gonadotropins: rFSH starting dose 300–450 IU |
Treatment group - Treatment: Testosterone gel 1% transdermal 10 mg/day. - Timing: for 15–20 days preceding ovarian stimulation (then stopped) Control group: - Treatment: placebo gel - Timing: the same |
No significant beneficial effects. |
| Fabregues et al. (2009) | - Type: RCT - Randomization: computer-generated randomization list - Blinding: ns - Treatment vs no treatment. |
Intention-to-treat: - Total: 62 - Treatment: 31 - Control: 31 Per protocol: - Total: 62 - Treatment: 31 - Control: 31 |
First IVF cycle canceled due to poor follicular response - Exclusion: > 39 years |
Treatment group: - Long GnRH agonist protocol - Gonadotropins: rFSH 450–300–150–150 IU for the first 4 days (then individualized) Control group: -Long GnRH agonist protocol - Gonadotropins: 300 IU per day of r-hFSH plus 150 hMG (then individualized) |
Treatment group - Treatment: Testosterone transdermal patch 2.5 mg/day - Timing: 5 days before COS (when pituitary-ovarian suppression was confirmed) Control group: No treatment |
Significant improvement of ovarian sensitivity to FSH and follicular response to gonadotropin treatment in previous low-responder IVF patients |
| Kim et al. (2011) | - Type: RCT - Randomization: Computer-generated list - Blinding: ns - Treatment vs no treatment. |
Intention-to-treat: - Total: 110 - Treatment: 55 - Control: 55 Per protocol: - Total: 110 - Treatment: 55 - Control: 55 |
-≤ 3 COCs retrieved despite the use of a high gonadotropin dose (> 2500 IU) in a previous failed IVF/ICSI cycle | - COS protocol: short GnRH antagonist - Gonadotropins: rFSH starting dose 300 IU |
Treatment group - Treatment: Testosterone gel 1% transdermal 12.5 mg/day - Timing: for 21 days preceding ovarian stimulation (then stopped) Control group: No treatment |
In testosterone pre-treatment group: - Significantly lower total dose/days of gonadotropin administration - Significantly higher oocytes (total/MII) retrieved, fertilized oocytes, good-quality embryos, implantation/clinical pregnancy rate. |
| Kim et al. (2014) | - Type: RCT - Randomization: sealed envelopes/computer-generated list - Blinding: ns - Treatment vs no treatment. |
- Total: 120a - Treatment: T2 week: 30 T3 week: 30 T4 week: 30 - Control: 30 (no treatment) |
-≤ 3 COCs retrieved despite the use of a high gonadotropin dose (> 2500 IU) in a previous failed IVF/ICSI cycle - Bologna criteria |
- COS protocol: short GnRH antagonist protocol - Gonadotropins: rFSH starting dose 300 IU |
Treatment group - Treatment: Testosterone gel 1% transdermal 12.5 mg/day - Timing: for 2, 3, or 4 weeks preceding ovarian stimulation, stopped during COS Control group: No treatment |
- After 3–4 weeks treatment: significant decrease of MFD, RI of ovarian stromal artery, total dose of rFSH. Significant increase of AFC, COC, and MII oocytes. - Only after 4 weeks treatment: significant increase of clinical pregnancy and live birth rate |
| Bosdou et al. (2016) | - Type: RCT - Randomization: Computer-generated randomization list - Blinding: single blinded - Treatment vs no treatment. |
Intention-to-treat: - Total: 50 - Treatment: 26 - Control: 24 Per protocol: - Total: 47 - Treatment: 24 - Control: 23 |
- Bologna criteria | - COS protocol: long GnRH agonist - Gonadotropins: rFSH nr |
Treatment group - Treatment: Testosterone gel 2% transdermal 10 mg/day - Timing: for 21 days starting from the GnRH agonist initiation stopped one day before COS |
Testosterone pre-treatment failed to increase the number of COCs by more than 1.5 as compared with no pre-treatment in poor responders undergoing ICSI |
| Mskhalaya et al. (2016)b | - Type: RCT - Randomization: ns - Blinding: ns - Treatment vs no treatment. |
- Total: 128 - Treatment: 77 - Control: 51 |
- Bologna criteria | - COS protocol: ns - Gonadotropins: ns |
- Treatment: Testosterone undecanoate per os, 40 mg/day—timing: at least 40 days (median—51 days) preceding COS | CPR (per cycle) was significantly higher in the treatment group (24.7%) than in the control group (7.8%), p = 0.018. Live birth rate (LBR) was higher in the treatment group than in the control group (18 vs 5.9%); difference was not statistically significant (p = 0.06) |
| Saharkhiz et al. (2018) | - Type: RCT - Randomization: ns - Blinding: single blinded - Treatment vs placebo. |
Intention-to-treat: - Total: 50 - Treatment: 25 - Control: 25 (placebo) Per protocol: - Total: 50 - Treatment: 25 - Control: 23 (placebo) |
- Bologna criteria | - COS protocol: short GnRH antagonist - Gonadotropins: rFSH starting dose 300 IU |
Treatment group -Treatment: Testosterone gel 25 mg/daily transdermal - Timing: started with rFSH until hCG. Control group: - Treatment: placebo gel - Timing: the same |
The mean oocyte retrieved (2.48 ± 1.64 versus 1.17 ± 1.27) and embryo (1.60 ± 1.58 versus 0.39 ± 0.58) was significantly higher in the intervention group (p < 0.01). CPR (16% versus 0%) and embryo of quality was significantly higher in the intervention group (p < 0.05) |
IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; RCT, randomized controlled trial; nr, not reported; ns, not specified; COCs, cumulus oocyte complexes; AFC, antral follicle count; RI, resistance index; MFD, mean follicle diameter; E2, estradiol; FSH, follicle-stimulating hormone; rFSH, recombinant follicle-stimulating hormone; hMG, human menopausal gonadotropin; hCG, human chorionic gonadotropin; GnRH, gonadotropin-releasing hormone
aKim et al. enrolled three treated groups using testosterone for 2 weeks (T2), 3 weeks (T3), or 4 weeks (T4) in preceding period of study stimulation cycle
bMeeting abstract