To assess the relationship of preinsertion vaginal ultrasound assessment and menstrual and gynecologic history as predictors of difficult or painful intrauterine device insertion in nulligravid women. Nulligravid women seeking contraception were invited to participate in this nonrandomized study and given the choice between the levonorgestrel-releasing intrauterine system or a copper-releasing intrauterine device. All 165 enrolled women were interviewed and a pelvic examination, including vaginal ultrasonography, was performed before insertion. Insertion difficulties and pain intensity were recorded and assessed against uterine measurements and background characteristics. Most insertions were assessed as easy (n=144 89.4%) and only two (1.2%) failed. Most women had uterine measurements smaller than the studied devices. Odds for difficulties at insertion decreased with every increasing millimeter in total uterine length (odds ratio OR 0.86, 95% confidence interval CI 0.78-0.96, P=.006) and cervical length (OR 0.85, 95% CI 0.74-0.97, P=.02) and similarly with every decreasing degree of (straighter) flexion angle (OR 0.96, 95% CI 0.94-0.99, P=.005). No absolute threshold measurements could be determined. Still, the majority of insertions in small and flexed uteri were uneventful. Severe insertion pain was common (n=94 58.4%). Severe dysmenorrhea was the only predictor of insertion pain (OR 8.16 95% CI 2.56-26.02, P<.001). Ultrasonographic evaluation does not give additional information compared with clinical pelvic examination and sound measure. Although smaller uterine length measurements and steeper flexion angle more often predicted difficulties, the majority of insertions were uneventful in women with small measures. Dysmenorrhea was the only predictor of pain. ClinicalTrials.gov, www.clinicaltrials.gov, NCT01685164. II.