In intraoperative neuromonitoring (IONM), facial nerve motor function (FaNMF) is assessed by facial muscle corticobulbar motor evoked potentials (FMcoMEP). Mostly only amplitude decrease is used as warning criterion. We related a refined criterion for FMcoMEP consisting of a bilateral final-to-baseline motor threshold ratio with standard criteria and postoperative FaNMF. 79 patients (45 females; 48 ± 16 years) undergoing IONM-guided cerebellopontine angle tumor surgery were retrospectively analyzed. An intraoperative final-to-baseline motor threshold increase ≥ 20% ipsi- versus contralaterally (bFBMT20) was correlated to postoperative FaNMF at day 1 (D1), 7 (D7) and 3 months (3 M). An ipsilateral-only final-to-baseline motor threshold increase ≥ 20 mA (iAMT20) and amplitude decrement ≥ 50% (iAR50) served as reference. Tumors included vestibular schwannomas (68%), meningiomas (19%) and others (13%). Mean tumor diameter was 2.7 ± 1.1 cm. Postoperatively, HB-increase ≥ 2 was seen in 27% (D1), 17% (D7), and 6% (3 M) of patients, respectively. FMcoMEP were obtained in 75/79 cases. Pathological bFBMT20, iAMT20 and iAR50 were seen in 17, 17, and 46 cases, respectively. Area under the ROC curve for bFBMT20 (iAMT20) was 0.894 (0.868) at D1; 0.903 (0.822) at D7 and 0.941 (0.959) at 3 M. iAR50 performed worse at all time points. Diagnostic odds ratios were highest for bfBMT20 compared to iAMT20 and iAR50 for D1 (172.5 vs. 8.7 vs. 0.45) and D7 (51.4 vs. 6.1 vs. 0.8). The refined parameter bFBMT20 provides a valuable contribution to the prognostic assessment of FaNMF. Due to its bihemispheric character, it might thus circumvent false-positive events which affect FMcoMEP bilaterally. © 2019, Springer Nature B.V.