OBJECTIVE To evaluate the effect of changed criteria for sperm count and morphology when performing fallopian tube sperm perfusion (FSP) in unexplained infertility. METHODS We report on a longitudinal study in which two periods of insemination cycles are compared. One hundred and eighty-three couples who started 361 stimulation cycles intended for FSP were included in the first period and 190 couples starting 303 cycles in the second period. The sperm requirement in the first study period was at least 7 million sperm in the prepared test sample. The FSP cycle included downregulation, controlled ovarian hyperstimulation (COH), ovulation induction and intrauterine and intrafallopian tube insemination using a sperm suspension of 3 mL volume. To avoid high-order multiple pregnancies, cycles at risk were converted to in vitro fertilization (IVF). Based on the results from the first period, sperm requirements were lowered to a sperm count of 3 million and at least 5% normal morphology according to strict criteria. RESULTS FSP was performed in 51% and 60% of started cycles during the two periods with similar pregnancy (24% and 23%) and birth (15% and 19%) rates. Neither sperm count nor morphology above the requirements affected pregnancy rates. IVF-converted, completed cycles reached equal pregnancy (41%) and delivery (32%) rates in both periods. Multiple pregnancy rates were higher after IVF (38%) than after FSP (24%). CONCLUSIONS The lowered requirements for sperm count and morphology have not impaired pregnancy results after FSP. Patients who either completed an FSP cycle or had their cycle converted to IVF were all at high risk of multiple pregnancy.