Measurement of Facial Nerve Function
The House-Brackmann (HB) facial nerve grading scale was adopted as the official metric of the International Facial Nerve Study Group in Bordeaux in 1984 and the American Academy of Otolaryngology-Head and Neck Surgery in 1985. The HB remains the most widely accepted scale in otolaryngology literature since its introduction.1 Its simplicity, quickness, and general descriptive power are its strengths, but the scale does not discriminate regional facial weakness or secondary deficits such as synkinesis, gustatory tearing, or hemifacial spasm. These secondary deficits are generally associated with the degree of facial weakness, however. It remains a fast and highly reliable assessment of clinical facial nerve function and has been proven predictive of multiple clinical outcomes. It does not require special equipment, can be performed anywhere, and has provided a common language for literature concerning facial nerve function.
The utility of a common standard for clinical tracking and reporting of facial nerve function has long been recognized in facial nerve surgery. As recently as 1980, the otolaryngology literature supported multiple facial nerve systems based on subjective or measured assessment of gross motor or regional motor function.2–7 Many scientific articles discussing facial nerve function would begin with a lengthy explanation of their specific metric, prior to discussion of results. This lack of standardization led to significant heterogeneity in the literature and made meta-analysis impossible. Several facial nerve scales suggested high levels of accuracy by assignment of a percentile from 0–100%, but poor interrater reliability and complexity rendered them impractical for widespread clinical usage. To aid with prediction of functional impairment, other scales allocated additional numerical weight to specific regions of the face. These scales were likewise challenging to administer, time-consuming, and had poor agreement among clinicians. In general, as systems attempted to improve descriptive accuracy, they became vulnerable to increased subjective interpretation. Of all the early facial nerve grading systems, few remain in use, except the Yanagihara scale, which still remains in widespread usage in Japan.8 It was developed in 1976 and is a regional system that measures 10 separate aspects of function on a scale of 0 to 4. It is easy to administer but does not address synkinesis or secondary deficits ( Table 5.1 ).
While developing the HB scale, the authors acknowledged that a system that assessed gross motor function would reduce variability in scoring at the expense of fine motor detail. The scale was then designed incorporating these clinical observations ( Table 5.2 ). Completely normal facial nerve function is HB grade I and complete absence of facial motion is HB grade VI. Minimal weakness and dynamic asymmetry is HB grade II, whereas minimal function and tone is considered HB grade V. Patients in these four grades are rarely afflicted by synkinesis or hemifacial spasm to a degree requiring therapy. The ability to activate musculature of the forehead is the dividing line between moderate (HB grade III) and moderately severe (HB grade IV) paralysis. Forehead function, although considered the least important functionally and cosmetically, suggests that the facial nerve has not degenerated, and portends excellent facial nerve recovery.9 Along this line, eye closure has significant clinical implications and was seen as a useful dividing line for patients with moderately severe paralysis. In this way, certain regional deficits are used to distinguish moderate to moderately severe forms of facial weakness.
In its original format, patients with synkinesis and hemifacial spasm severe enough to interfere with function were considered to have moderately severe dysfunction regardless of motor activity. The stipulations regarding synkinesis were subsequently removed prior to its acceptance by the American Academy of Otolaryngology-Head and Neck Surgery, and a modified HB scale is most commonly accepted in the literature. Although the scale was initially designed to record long-term facial nerve function, subsequent studies found that these minor modifications made it applicable for immediate postoperative facial nerve function.10
There is a degree of subjectiveness involved in the modified HB scale, which is most apparent for patients with moderate to moderately severe facial paralysis.11 For example, a patient with no brow movement who can completely close his/her eye with effort may be graded HB III or HB IV depending on the observer. Modifications recommended by the authors to improve scoring accuracy in moderately severe paralysis with objective measurement of brow elevation and commissure excursion have been largely overlooked, and the HB scale is most commonly used in the format excluding these measurements.1
A self-assessment facial nerve function questionnaire, based on the HB scale, has been validated for patients after acoustic neuroma treatment ( Table 5.3 ).12 High interrater reliability coefficients were found between clinicians and patients at 1 year after therapy. The HB scale has been found to have important implications for facial nerve recovery after acoustic neuroma surgery. Tumor size, intraoperative facial nerve stimulation, and severity of immediate postoperative paralysis can be used with the HB scale to make predictions regarding facial nerve recovery.13–15
Additional Scoring Metrics
Since its introduction, the modified HB Facial Nerve Grading Scale was found to be more efficient and reliable than the existing grading systems of that time.16 Multiple grading systems have developed in the past 25 years that have been suggested as refinements or replacement systems to further describe patients with moderate to moderately severe paralysis and their secondary deficits.
Attempting to reduce subjective influence on scoring, the Burres-Fisch system was developed in 1986, in which the observer compares a subject′s facial expressions against seven standard facial expressions. Although it demonstrated excellent correlation with the modified HB scale, this scale takes over 20 minutes to administer and does not account for secondary defects.
The Nottingham system was later proposed as a replacement system to address the subjectivity of the modified HB scale.17 At rest and at maximal effort, measurements are obtained from above the brow and below the orbit, as well as from the lateral canthus to the lateral oral commissure. These values are then averaged and compared between the two sides. Synkinesis and hemifacial spasm are considered modifiers, as are dry eyes, dysgeusia, and gustatory tears ( Fig. 5.1 ). The scale cannot be used in patients with bilateral weakness and requires more time than the HB scale, although it can still be administered in less than 5 minutes.
The Sunnybrook scale was introduced as an alternative to the HB scale in 1996 ( Table 5.4 ). An observer subjectively scores regional symmetry at rest on a scale of 0 to 2. Next, the observer rates facial movement in five different expressions from 1 to 5. Synkinesis is judged on a four-point scale in five basic facial expressions. These values are combined on weighted scales to become a single numerical value. The scale is relatively easy to administer and is sensitive to fine changes in facial nerve function. However, there is a significant time commitment to administering it, and it retains a subjective component, which limits its precision. Measurements for this scale can be obtained in approximately 5 minutes.
The Facial Nerve Grading Scale 2.0 is a recently proposed substitute for the modified HB scale.18 This adaptation has been developed to describe regional weakness, synkinesis, and other secondary deficits ( Table 5.5 ). In a series of facial movements, the examiner grades the function of the brow, eye, nasolabial fold, and oral commisure on a scale of 1 to 6. Synkinesis is given a score of 0 to 3. The cumulative score, from 4 to 24, is then ranked on a scale of I to VI. Using this scale, interobserver reliability in cases of moderately severe facial paralysis made modest improvements over the modified HB scale, from 57.5% to 64%.16 Additional work may demonstrate that this is a viable alternative to the modified HB scale that is quick, conveys a significant amount of information, and is easy to administer.
Computer-assisted grading has been suggested to be a rapid, reliable scoring method for the paralyzed face. Several programs have been developed that use image subtraction, luminescence, or moiré topography to generate a composite score evaluating dynamic facial function.19–21 While promising, computer-assisted techniques will always be limited by the requirements of specialized, proprietary software, and additional personnel training required to operate the equipment.