OBJECTIVE To determine whether transcranial motor-evoked potential monitoring of the facial nerve (FNMEP) during eloquent tumor resection is feasible in children and can predict both immediate and postoperative facial nerve (FN) function. METHODS We included 24 consecutive procedures involving 21 patients (median age 5.5 years, range 5 months to 15 years, 8 female) who were operated on with FNMEP monitoring by the first author in 2013 and 2014. During surgery, we maintained a constant response amplitude by increasing the stimulation intensity and aimed to establish a warning criterion based on the "threshold-level" method. A threshold increase of greater than 20 mA for eliciting the FNMEP in the most reliable facial nerve target muscle was considered to be a prediction of reduced postoperative facial nerve function and consequently, a warning was given to the surgeon. The preoperative and early postoperative function was documented with the House-Brackmann grading system. RESULTS Monitoring of the FNMEP was feasible in all the surgeries in at least one facial nerve target muscle. The orbicularis oris muscle yielded the best result (95% of the trials), followed by the mentalis (87%) and orbicularis oculi muscles (86%). The median stimulation threshold was initially 65 mA (range 40-110 mA) for the FNMEP and 60 mA (15-220 mA) for the motor-evoked potential of the thenar muscles. The FNMEP deterioration showed a sensitivity of 100% for House-Brackmann deterioration and specificity of 74%. CONCLUSIONS Intraoperative FNMEP monitoring is feasible and safe in infants and children. We found no evidence that the procedures and thresholds should differ from FNMEP monitoring in adults. FNMEP monitoring provides valid evidence for FN function in pediatric eloquent area surgery; its use is complementary to direct electrical FN stimulation and continuous EMG monitoring of FN target muscles.