The aim of this retrospective study was to investigate the impact of endogenous and exogenous luteinizing hormone (LH) activity on treatment outcome, when taking into consideration potential confounding variables. Data were derived from IVF patients (n = 358) stimulated with highly purified menotrophin (HP-hMG) in a long gonadotrophin-releasing hormone (GnRH) agonist protocol. Simple retrospective logistic regression analysis showed that the mid-follicular exogenous concentrations of human chorionic gonadotrophin (hCG) (p = 0.027), provided by the HP-hMG preparation, and female age (p = 0.009) were significantly associated with live-birth rate, while the mid-follicular progesterone concentration (p = 0.075), the estradiol concentration on last stimulation day (p = 0.075) and number of embryos transferred (p = 0.071) were borderline significant. Endogenous LH was not associated with live-birth rate; neither at start of stimulation (p = 0.123), nor in the mid-follicular phase (p = 0.933) or on the last day of stimulation (p = 0.589). In the multiple regression analysis of life birth, mid-follicular hCG (p = 0.016) was identified as a positive predictor, and age (p = 0.004) and mid-follicular progesterone (p = 0.029) as negative predictors. In conclusion, mid-follicular concentrations of exogenous hCG and progesterone, but not endogenous LH, are associated with live-birth rate in IVF patients treated with HP-hMG in a long GnRH agonist cycle.