Blink Reflex IONM Predicts Corneal Outcomes in Skull Base Surgery

Blink-reflex intra-operative neurophysiological monitoring (BR IONM) detected post-operative corneal complications with 100% sensitivity and 96.7% overall accuracy across a 42-patient Hong Kong cohort operated between January 2021 and April 2024. The prospective study, published in Frontiers in Surgery, positions BR IONM as a practical addition to standard posterior fossa multimodality monitoring.

The technique elicits the electronic equivalent of the corneal reflex by surface-electrode stimulation of the supraorbital nerve and records the early R1 response with needle electrodes over the ipsilateral orbicularis oculi at the lower eyelid. R1 signal loss or an amplitude drop of 50% or more from baseline flagged significant change. Anaesthesia followed total intravenous propofol-remifentanil (10 mg/kg/h and 10 µg/kg/h respectively) with BIS maintained at 40-60 and no muscle relaxant beyond intubation.

What does the cohort look like?

Of 42 consecutive patients (17 women, 13 men; age 41-81, mean 62.0) with lesions near the brainstem, the team successfully monitored BR in 30 (71.4%). Ten underwent microvascular decompression for trigeminal neuralgia or hemi-facial spasm; 20 had tumour excisions ranging from 1.1 to 81.6 cm³ (mean 21.6 cm³). Mean R1 latency under general anaesthesia was 13.3 ms (range 11.9-14.4 ms), slightly longer than the 10.7 ms reported in conscious adults.

How accurate is the test?

Six of the 30 monitored patients (20%) met criteria for significant change:

• 4 patients (13.3%) with R1 amplitude drop ≥50%
• 2 patients (6.7%) with complete R1 loss

All four with significant decrease showed decreased ipsilateral corneal sensation; one developed transient ulceration. Both with BR loss developed corneal ulceration and exposure keratitis. Statistical performance:

• Sensitivity: 100% (95% CI 54.0-100.0)
• Specificity: 95.8% (95% CI 78.9-99.9)
• Positive predictive value: 85.7%
• Negative predictive value: 100%
• Accuracy: 96.7% (95% CI 82.8-99.9)

Why does this technique add value beyond facial-nerve monitoring?

Two patients with complete BR loss (cases 8 and 15) retained intact facial corticobulbar MEPs and normal House-Brackmann scores, yet still developed corneal ulceration. The dissociation shows that facial-nerve EMG alone misses trigeminal afferent injury that compromises corneal sensation. Authors Michael W. Y. Lee and S. Chau argue that BR IONM should function as a complementary modality rather than a duplicate facial-nerve test.

Why do tumour volume and midline shift matter?

Among the 20 tumour cases, Mann-Whitney U tests showed significantly higher pre-operative tumour volume (37.8 vs 13.0 cm³, p = 0.0016) and maximal brainstem midline shift (8.1 vs 4.8 mm, p = 0.0271) in patients with significant BR change. Midline shift was measured on T2-weighted MRI using the cerebral aqueduct, the mid-point of the interpeduncular fossa, and the median sulcus of the fourth ventricle. The pattern supports BR as a sentinel for mechanical stretch of brainstem neural tissue during dissection.

How does the stimulation protocol compare across centres?

The Hong Kong (PYNEH) protocol uses a train-of-four constant-current stimulus with 2 ms inter-stimulus interval, 300 µs pulse width, and intensity up to 40 mA through surface electrodes at the supraorbital notch. Recording parameters are a 50 ms time sweep, 25-50 µV amplification, and a 10-3000 Hz bandpass. The Istanbul group (Aydinlar et al., 2022) used subdermal needles with a double train, achieving 80% success and 96.8% accuracy. The Beijing group used biphasic constant-current pulses with halogenated agents and achieved 93.6% success with 94.4% accuracy. Surface electrodes at the supraorbital nerve, the authors argue, avoid needle-related nerve puncture and stabilise the R1 waveform against anaesthetic fluctuation.

What should clinical neurophysiologists do with the signal?

Tracings were obtained at baseline before bone work, again after dural opening, then continuously during microdissection, sometimes as often as every minute, with a final acquisition at dural closure. Control electrodes in the midline helped detect current spread and false-positive artefact. For significant BR change, the authors recommend immediate ophthalmology referral and tarsorrhaphy to prevent exposure keratitis, ulceration, and blindness.

What are the limitations?

The 30-patient monitored subset is small, and 12 (28.6%) could not be monitored because BR was absent or diminished from the start. Performance comparisons against larger Turkish and Beijing cohorts suggest the 96.7% accuracy figure should be replicated before BR IONM enters routine IONM protocols for every posterior skull base case. The authors call for larger multicentre verification of their 50% amplitude-cutoff criterion and surface-electrode montage.

The Hong Kong experience adds intra-operative blink-reflex monitoring to the neurophysiology toolbox for posterior fossa surgery. A multicentre validation of the 71.4% feasibility rate and the 96.7% accuracy in a larger cohort would determine whether this modality becomes standard practice for protecting corneal sensation when brainstem shift is anticipated.