Neuromotor Speech Recovery Across Different Behavioral Speech Modifications in Individuals Following Facial Transplantation

Abstract

Despite signs of facial nerve recovery within a few months following face transplantation, speech deficits persist for years. Behavioral speech modifications (e.g., slower-than-normal speaking rate and increased loudness) have shown promising potential to enhance speech intelligibility in populations with dysarthric speech. However, such evidence-based practice approach is lacking in clinical management of speech in individuals with facial transplantation. Because facial transplantation involves complex craniofacial reconstruction and facial nerve coaptation, it is unknown to what extent individuals with face transplant are capable of adapting their motor system to task-specific articulatory demands. The purpose of this study was to identify the underlying articulatory mechanisms employed by individuals with face transplantation in response to speech modification cues at early and late stages of neuromotor recovery. In addition, we aimed to identify speech modifications that conferred improved speech clarity. Participants were seven individuals who underwent full or partial facial vascularized composite allografts that included lips and muscles of facial animation and were in early (~2 months) or late (~42 months) stages of recovery. Participants produced repetitions of the sentence “Buy Bobby a puppy” in normal, fast, loud, and slow speech modifications. Articulatory movement traces were recorded using a 3D optical motion capture system. Kinematic measures of average speed (mm/s) and range of movement (mm3) were extracted from the lower lip (± jaw) marker. Two speech language pathologists rated speech clarity for each speaker using a visual analog scale (VAS) approach. Results demonstrated that facial motor capacity increased from early to late stages of recovery. While individuals in the early group exhibited restricted capabilities to adjust their motor system based on the articulatory demands of each speech modification, individuals in the late group demonstrated faster speed and larger-than-normal range of movement for loud speech, and slower speed and larger-than-normal range of movement for slow speech. In addition, subjects in both groups showed overreliance on jaw rather than lip articulatory function across all speech modifications, perhaps as a compensatory strategy to optimize articulatory stability and maximize speech function. Finally, improved speech clarity was associated with loud speech in both stages of recovery.

Correlation of 3D Morphometric Changes, Kinematics, and Muscle Activity During Smile

“…Furthermore, it has been shown that the velocity of oral commissure after facial transplantation increases with the clarity of articulation in the later stages of recovery. 23 In a study comparing hemi hypoglossal nerve transposition and cross-facial nerve graft for reanimation of facial palsy, it was determined that the velocity of the oral commissure was more asymmetric in the transposition group compared to the healthy side (48.7 AE 19 vs. 35.3 AE 16.7 mm/s). 24,25 Understanding the time-normalized relationship between muscle activity and kinematic data, including parameters such as displacement, velocity, and acceleration, can provide valuable insight into interpreting electromyographic and electroneurographic data.…”

“…It has been reported that the change in velocity and duration of oral commissure movement 21 or activity of related muscle, that is, zygomaticus majors, 22 have different meanings in transmitting emotions. Furthermore, it has been shown that the velocity of oral commissure after facial transplantation increases with the clarity of articulation in the later stages of recovery 23 . In a study comparing hemi hypoglossal nerve transposition and cross‐facial nerve graft for reanimation of facial palsy, it was determined that the velocity of the oral commissure was more asymmetric in the transposition group compared to the healthy side (48.7 ± 19 vs. 35.3 ± 16.7 mm/s) 24,25 .…”

“Facial Expression After Face Transplant: The First International Face Transplant Cohort Comparison”

“…23 This has proven to have no major influence on oral competence and speech intelligibility, both satisfactorily improving after transplantation. 53,54 Unfortunately, FDI data are not available for all cohorts, as it is not validated in all languages. However, the FDI might not be sensitive enough to measure all aspects of FT-related QoL, such as side effects of immunosuppression.…”

Longitudinal Recovery of Speech Motor Function Following Facial Transplantation: A Prospective Observational Study

“…Threedimensional motion capture analysis is commonly used to describe motor speech kinematics in both healthy and impaired populations. [8][9][10] Procedures used for kinematic data collection, marker placement, movement subtraction, and extraction of kinematic measures from the "nose top center," the "center lower lip," and the "virtual jaw center" markers were the same as those described in detail in Eshghi et al 11 Lower lip movement was represented in two ways: (1) movements of the underlying jaw included (lower lip + jaw) and ( 2) independent from the movements of the jaw (lower lip À jaw). 11 Each participant performed 10 repetitions of the phrase "buy Bobby a puppy" at his or her typical speaking rate and volume.…”

“…[8][9][10] Procedures used for kinematic data collection, marker placement, movement subtraction, and extraction of kinematic measures from the "nose top center," the "center lower lip," and the "virtual jaw center" markers were the same as those described in detail in Eshghi et al 11 Lower lip movement was represented in two ways: (1) movements of the underlying jaw included (lower lip + jaw) and ( 2) independent from the movements of the jaw (lower lip À jaw). 11 Each participant performed 10 repetitions of the phrase "buy Bobby a puppy" at his or her typical speaking rate and volume. The average speed of movement (mm/s) and range of motion (mm) from lower lip AE jaw movement time-series were calculated from speech samples using a customized MATLAB algorithm.…”