An observational, matched, case-control study was carried out to compare the efficacy of recombinant human luteinizing hormone (r-hLH) supplementation with that of urinary human menopausal gonadotrophin (u-hMG)-based LH activity during controlled ovarian stimulation (COS) for assisted reproductive technology (ART) using a long gonadotrophin-releasing hormone (GnRH)-agonist protocol. A total of 4719 women, 1573 per group, matched by age, body mass index, indication and number of previous ART cycles, were treated with either recombinant human follicle-stimulating hormone (r-hFSH) and r-hLH in a fixed 2:1 ratio or u-hMG, either alone or in combination with r-hFSH, after down-regulation in a long GnRH-agonist protocol. Compared with the two u-hMG groups (u-hMG alone or in combination with r-hFSH, respectively), r-hFSH consumption was significantly lower (p < 0.001; p < 0.001), and pregnancy rates per cycle (p = 0.006; p = 0.022) and per embryo transfer (p = 0.025; p = 0.008), and implantation rate per embryo transferred (p < 0.001; p < 0.001) were significantly higher in the group treated with the fixed combination of r-hFSH and r-hLH. In COS protocols with r-hFSH, supplementation with r-hLH appears to be more effective than supplementation with u-hMG using the long GnRH-agonist protocol for ART.