This open label randomized study aims to define the best protocol to be used with growth hormone in poor responders, with comparison performed to delineate which protocol offers the best cycle outcomes. Two-hundred eighty-seven poor responders were included. The patients were randomly allocated into four groups receiving growth hormone (GH) as an adjuvant therapy added to either long or short agonist protocol, miniflare or antagonist protocols. The short/GH gave significantly lower mean number of oocytes when compared with the long/GH, antagonist/GH and miniflare/GH (4 ± 1.69 versus 5.06 ± 1.83, 4.95 + / = 1.90 and4.98 ± 2.51, respectively p = 0.005). Considering the number of fertilized oocytes, the long/GH showed significantly higher levels than short/GH and antagonist/GH (3.73 ± 1.47 versus 3.02 ± 1.52 and 2.89 ± 1.14, respectively). The main drawback is that it required significantly higher HMG dose and longer duration of stimulation. The long/GH was superior when compared with the three protocols regarding the number of oocytes retrieved and fertilized. But, when considering the clinical pregnancy rates, there was a difference in favor of the long/GH but not reaching a statistically significant value (ClinicalTrials.gov Identifier: NCT01897324).
Keywords: Antagonist protocol; IVF/ICSI; growth hormone; long agonist protocol; miniflare protocol; poor responders; short agonist protocol.