Combination of in-situ collagen injection and rehabilitative treatment in long-lasting facial nerve palsy: a pilot randomized controlled trial

Combination of in-situ collagen injection and rehabilitative treatment in long-lasting facial nerve palsy: a pilot randomized controlled trial

Alessandro Micarelli 1,2; Andrea Viziano 3; Ivan Granito 2; Giuseppe Antonuccio 2; Alessio Felicioni 2; Marco Loberti 4; Pasquale Carlino 2; Riccardo Xavier Micarelli 3; Marco Alessandrini 3.

(1) Institute of Mountain Emergency Medicine, Eurac Research, Bolzano, Italy

(2) ITER Center for Balance and Rehabilitation Research (ICBRR); Rome, Italy

(3) Department of Clinical Sciences and Translational Medicine, University of Rome ‘Tor Vergata’, Italy

(4) DOPSITERE; Rome, Italy

ABSTRACT

Background: Many rehabilitative attempts have been made to prevent or reduce residual deficits in patients with established and long-term facial palsy (FP). In many clinical settings in-situ injection of collagen-based medical devices (MDs) have been demonstrated to provide nutritional support for tissues.

Aim: To test the effectiveness of a collagen-based treatment for patients complaining of long standing FP, who are following a proprioceptive neuromuscular facilitation protocol (Kabat method) (Group A), compared to a FP group only undergoing the Kabat method (Group B).

Design: Randomised controlled trial

Setting: Tertiary referral outpatient center and University Hospital

Population: 41 patients with a medical diagnosis of long-term unilateral peripheral FP

Methods: Twenty-one Group A patients were compared, after randomization, to nineteen matched Group B patients after 8 weeks of treatment. The outcomes were electromyographic findings, validated questionnaires (Facial Disability Index, FDI and General health-related quality of life assessment, QOL) and clinical grading (House-Brackmann, HB, and synkinesis grading scale). A correlation analysis was performed between pre-/post-treatment differences (Δ) in outcome and clinical-demographic measures.

Results: A significant within-subjects improvement, both in electrophysiological and questionnaire scores, was found in both groups. When compared with Group B, Group A patients exhibited a significant reduction of post-treatment polyphasic potentials of voluntary activity of orbicularis oculi (p = 0.017) and oris (p = 0.015) and a significant increase in post-treatment duration of voluntary activity of orbicularis oris (p = 0.018). Group A subjects demonstrated a significant improvement in questionnaire subscales regarding overall disease perception. Although positive correlations between the ∆FDI and ∆percentage of polyphasic potentials of voluntary activity were found in both groups, negative correlations in Group A were found between disease duration and ∆duration of voluntary activity of orbicularis oculi and oris.

Conclusions: The combination of physical rehabilitative procedures with in-situ collagen injections, possibly acting in redirecting the phenomena of reinnervation/reorganization, demonstrated encouraging results in patients affected by long term FP.

Clnical Rehabilitation Impact: In-situ collagen injection could be a safe option enlarging the ‘window of opportunity’ to improve the voluntary muscle contraction pattern and general and specific disability referred by patients affected by long standing FP.

Keywords: Facial nerve paralysis; Rehabilitation; Collagen; Electromyography

INTRODUCTION

Facial nerve palsy (FP) is a neurological condition in which function of the facial nerve is partially or completely lost. [1,2] According to previous studies, common causes of facial nerve paralysis include Bell’s palsy (51%), trauma (22%), herpes zoster (7%), tumors (6%), and infection (4%). [2,3] FP weakens the facial muscles, altering facial symmetry and functions. Patients with established (axonotmesis) and long-term facial injury have several problems: hampered verbal communication, closure of the eyes, psychological distress, disturbed facial self-identification, [4] and difficulty eating and drinking liquid and solid foods. [5] Facial symmetry determines facial attractiveness, it is a marker of good health, it influences interpersonal attraction and is a vehicle of the expression of feelings. [6] Together with idiopathic palsy, representing the most common type (50%), in which deficits of varying magnitude persist in about 30% of patients, severe and long-lasting sequelae (i.e. residual high grade paralysis, facial contractions, synkinesis, or spasm) are complained in cases caused by injury from infectious, traumatic (temporal bone fractures), neoplastic, or iatrogenic noxae (ear and parotid surgery), affecting the facial nerve along its intra- and extra-cranial course. [4,5]

People suffering from the sequelae of FP may be referred to physiotherapists since the medical practitioner’s options are limited to invasive treatments such as Botulinum toxin A injections which – inhibiting the release of acetylcholine [7] – were found to reduce long-lasting sequelae such as synkinesis and hypercontraction [7,8] and surgical reconstruction. [9] Although a variety of physiotherapy interventions have been used to treat FP, [9] randomised controlled trials have found no evidence in favour of one intervention over another, [9] apart from proprioceptive neuromuscular facilitation (Kabat method), which has recently demonstrated significant improvements even in long standing, severe facial palsy. [7,10] In an attempt to prevent or reduce these residual deficits, continuous efforts have been made to identify the most effective therapeutic approaches, including medication, physical therapy, or a combination of both. [6,11]

Accordingly, biomaterials provide a promising alternative for injury repair: [12] ideally, they should possess good tissue compatibility, sufficient mechanical strength for sustaining muscle and nerve regeneration, [12] give orientation guidance for the regenerated muscle and nerve fibers [12] and carry out bio-activities that can efficiently improve regeneration. [12-14] In this scenario, experimental models using collagen have been proposed because of its favorable biocompatibility, biodegradability, weak immunogenic reactions [15] and capability to drive neofibrillogenesis of the matrix. [16,17] Indeed, it has been routinely introduced in many clinical settings, given the availability on the market of safe collagen-based medical devices (MDs), and in-situ injections have been demonstrated to act as effective natural bio-scaffolding, since the degradation of the collagen into its constituent aminoacids has proven to provide nutritional support for tissues. [18,19]

However, given the absence of previous experiences with collagen-based MDs in the field of FP rehabilitation, the aim of this pilot randomized study was to test the short-term effectiveness – in terms of electrophysiological and clinical findings – of a collagen-based treatment on patients complaining of long standing facial nerve axonotmesis. This was performed by in-situ injections of a solution of commonly used MDs in a group of patients following an 8 week proprioceptive neuromuscular facilitation protocol (Kabat method) (Group A), compared with a matched group of patients only treated with the same rehabilitative procedure and not receiving the injection treatment (Group B), serving as the control group.

MATERIALS AND METHODS

Population in study

The study was conducted on patients with a medical diagnosis of long-term unilateral peripheral axonotmesis of the facial nerve, screened in the outpatient service of ITER Balance and Rehabiliation Research Center (ICBRR) during the period from July 2016 to March 2019. Patients of either sex, between 18 and 65 years old who did not improve with conventional treatment were included in our study. Two clinical-otoneurological researchers (M.A. and A.V., who were blinded to the treatment allocation) carried out a detailed history and clinical examination, pure tone audiometry (PTA), and impedance audiometry in all patients. Routine investigations such as hemogram, electrolytes, chest X-ray, brain Magnetic Resonance Imaging and electrocardiography were also performed in each case. [20] They further assessed static and dynamic facial asymmetry, and manually tested muscle strength of the frontalis, corrugator, orbicularis oculi, zigomaticus, caninus, platysma and orbicularis oris muscles. These clinical findings were classified and recorded according to the House-Brackmann (HB) scale from 1 (normal) to 6 (severe dysfunction). Particular attention was also given to investigating the presence of synkinesis in the orbicularis oculi and orbicularis oris muscles and in the orbicularis oculi-zigomaticus and platysma muscles. According to previous studies, when present, synkinesis was defined as either mild, moderate or severe. [20] According to the same protocols the exclusion criteria were as follows: (1) patients had facial palsy caused by central nerve disease; (2) patients had a history of recurrent FP; (3) patients had a facial skin disease hampering the possibility of injecting the drug; (4) patients had difficulty in moving their face due to previous plastic surgery or facial surgery; (5) patients had systemic diseases that can affect facial electromyiography (EMG); (6) patients were deemed inappropriate by the researchers. Furthermore, as per producer disclosure, patients with a history of hypersensitivity to any collagen solution constituent or showing pregnancy, lactation, neuromuscular junction disorders (myastenia gravis), peripheral motor neuropathies, or active infections were also excluded.

The study was approved by the Regional Review Board (n. 650/CE Lazio 1, April 12th, 2018) and registered in the clinicaltrials.gov website with identification number NCT04353908, it adhered to the principles of the Declaration of Helsinki and all the participants provided written informed consent after receiving a detailed explanation of the study.

Outcome measures

Before the random allocation in terms of gender, age, disease duration (DD), cause and side of lesion performed with a computer-generated schedule and one month after the respective treatment protocols, an experienced neurologist/neurophysiologist and a physiotherapist (M.L. and A.F., blinded to the treatment allocation) performed the following evaluations in both groups of patients (See flow-chart in Figure 1):

ELECTROPHYSIOLOGICAL EVALUATION

An EMG coaxial needle examination was performed on the orbicularis oculi and orbicularis oris to evaluate the electrical alteration of these muscles. These muscles were examined at rest and during voluntary activity (amplitude, duration and percentage of polyphasic potentials). In accordance with previous procedures, [20] the EMG recordings were analyzed for spontaneous fibrillation potentials, the degree of voluntary polyphasic reinnervation potentials, and for synkinetic activity. [21] Following the same guidelines [20] and considering the nature of the study, the presence of denervation signs, alterations of EMG voluntary activity and the excitability of facial nerve motor responses were used to classify neurogenic findings during the EMG/electroneurography evaluation into two major groups that would benefit from rescue rehabilitative/injection treatment, as follows:

1) axonotmesis with moderate impairment: spontaneous activity present with fibrillations, positive sharp waves, reduced voluntary EMG activity, motor latency more than 5 ms, CMAP amplitude of nerve VII less than 1 mV. We used the inferior normal limit (at 2 standard deviations) of facial nerve motor response amplitude following the normative data from our laboratory (2.2±0.6 mV orbicularis oculi; 2±0.5 mV orbicularis oris);

2) axonotmesiswithsevereimpairment:spontaneousactivitypresentwithfibrillations,positive sharp waves, absence of voluntary EMG activity, absence of CMAP by stimulating motor axons of nerve VII.

VALIDATED QUESTIONNAIRES

Facial Disability Index

The FDI is a 10-item questionnaire with 2 subscale scores: 5 items concern the physical function subscale, and 5 items concern the social/well-being function subscale. Each item is rated on a 6-point scale, ranging from severe disability to absence of disability. Both subscales are transformed to a score on a 100 point scale, with 100 indicating unimpaired physical or social/wellbeing function. All questions referred to the preceding month. [22]

General health-related quality of life (QOL) assessment. The 36-item short form survey (SF-36) was used for general health-related QOL assessment. The SF-36 is an established, validated, generic instrument that measures health-related QOL in diverse patient populations [23]. It is based on a 36-item questionnaire with eight subscales and two component summaries of physical and mental health-related QOL. The Italian version of the SF36 was used in this study. [23]

TREATMENT PROCEDURES

Proprioceptive neuromuscular facilitation procedure

Rehabilitation started in both groups of patients after the diagnostic battery and it was carried out according to Kabat et al., i.e. a proprioceptive neuromuscular facilitation procedure, [24] twice a week for 8 weeks, by an experienced physical therapist (G.A., blinded to the treatment allocation). The mean rehabilitative session duration was 45 minutes for all study subjects. This method considers that harmony, coordination and optimal strength of body movements mainly depend upon the fact that they are performed following diagonal lines with respect to the sagittal axis of the body, thus implying a ‘rotational’ effect. It consists of facilitating the voluntary response of an impaired muscle through a global pattern of an entire muscular section which undergoes resistance. Further details have been thoroughly documented in previous experiences. [10,24]

Injection treatment

Experimental group subjects (GroupA) were treated while comfortably seated on a vertical backrest chair with their feet resting on a rubber carpet. Injections of an equally-balanced solution of MD Neural, MD Matrix and MD Muscle (Guna S.p.a., Milan-Italy), containing collagen of porcine origin, were administered subcutaneously after applying lidocaine/prilocaine cream (EMLA; AstraZeneca, Milan, Italy) on the affected side, by a skilled otonaryngologist in the field of injection treatments (A.M., blinded to the treatment allocation). With an insulin-type syringe and 30-gauge needle, a bolus of about 0.4 cc per point of the solution was injected in the orbicularis oculi muscle 1 cm outside the orbital rim and 0.5 cm superior and inferior to the first injection point, respectively, and in the orbicularis oris muscle, where the 2 injection points are on the border between the pars peripheralis and pars marginalis, located about 5 mm above the superior and inferior vermilion, respectively (Figure 2A). After the procedure, a gentle massage was applied on the injection points. The injection treatment was performed twice a week for 8 weeks, right after the rehabilitative session. Patients scored along a visuo-analogue scale (VAS) ranging from 0 to 10 the discomfort reported during the procedure and the average score per week was collected. Furthermore, a diary of possible immediate and/or delayed adverse reactions was filled in by patients at home. [25]

Data Handling and Statistical Analysis

The sample size was calculated to detect inter-group differences in the results for the electrphysiological variables. The sample size for the test hypothesis was calculated in accordance with the context (independent samples and continuous variables), using a statistical power of 80% (1−β) for an α error probability of 0.05. We used the t-test for independent samples and an effect size of 0.80. We thus aimed to include at least 13 participants per group. The Chi-square test was implemented to define associations between category factors and groups at rest and after treatment. Mean and standard deviations (SDs) of all variables were assessed in both groups. In order to define that the data followed a Gaussian distribution, D’Agostino K-squared normality and Levene’s homoscedasticity tests were performed (where the null hypothesis is that the data are distributed normally). A mixed analysis of variance (ANOVA) was performed with group as between factor and pre- and post-treatment variables as within-subject factors. Gender, age and DD (in months), between the two groups were used,where possible,as categorical and continuous predictors. The significant cutoff level (α) was set at a p value of 0.05. Bonferroni correction for multiple comparisons was applied to test the significant main effects post hoc. Then in accordance with previous protocols, [26] given the exploratory nature of the study, a two-tailed Spearman’s rank correlation was applied in both groups between pre-/post-treatment differences (Δ) in electrophysiological and clinical scores and clinical-demographic measures. Thus, considering the sample size of this group and the two-tailed nature of the test, a significant cut-off level (α) was set at a p value of 0.05 (STATISTICA 7 package for Windows).

RESULTS

Population in study

Among an initial sample of 41 enrolled patients, after the randomization procedure, one patient belonging to Group B left the rehabilitative procedure due to hospitalization for urological surgery. Finally, 21 patients belonging to Group A and 19 belonging to Group B completed the 8 week treatment and were followed up. In our study sample, adherence to both rehabilitation and injection treatment was 100%. Table I shows the clinical-demographic aspects of these patients as well as Table II and III the description – respectively – of HB and severity of synkinesis (which involved 12 and 10 patients in Group A and Group B, respectively), demonstrating a reduction, although not significant, of these parameters when comparing both groups of subjects affected by long standing FP and undergoing the respective treatments. Figures 2B and 2C represent the progressive reduction on discomfort VAS average scores per week and the recurrence of home diary adverse reactions respectively, which were referred – among a total of 336 injections – in 16 cases (4.6%). These included 7 (43.75%) and 5 (31.25%) cases of mild peri-ocular and peri-labial hematoma respectively, which lasted a maximum of 7 days, 2 (12.50%) cases of skin flushing limited to the peri-labial area that began 6 hours after treatment and vanished within 12 hours and 1 case of peri- labial (6.25%) and peri-ocular (6.25%) itching complained of from 12 to 24 hours after the treatment.

Electrophysiological findings

The within-subjects comparison demonstrated a significant reduction of the percentage of polyphasic potentials of voluntary activity of orbicularis oculi and orbicularis oris in both groups of FP patients (Table IV). On the other hand, the same analysis demonstrated a significant increase in duration of voluntary activity of orbicularis oculi and orbicularis oris in Group A (Figure 3A). Furthermore, the between-group analysis showed a significant reduction of post-treatment polyphasic potentials of voluntary activity of orbicularis oculi (p = 0.017) and orbicularis oris (p = 0.015) and a significant increase in post-treatment duration of voluntary activity of orbicularis oris in Group A when compared to Group B patients (p = 0.018) (Table IV, Figure 3A).

Validated Questionnaires

Both groups demonstrated a significant improvement in Physical Function, Social Function, Mental Health, Change in Health, Energy/Vitality and Health Perception subscales of SF-36 after the respective treatments (Table V). However, Group A subjects also demonstrated a significant improvement in Role limitation – Physical and Role limitation – Mental subscales. When comparing post-treatment values, Group A demonstrated a significant improvement in Role limitation – Physical (p = 0.015), Role limitation – Mental subscales (p = 0.014) and Health Perceptions (p = 0.008) subscales with respect to Group B (Table V).

Finally, both groups’ post-treatment FDI-p and FDI-s values were found to significantly improve after treatment. However, Group A post-treatment FDI-p values were found to be significantly increased (p = 0.02) compared to Group B post-treatment scores (Table V, Figure 3B).

Correlation Analysis

Positive correlations between ∆FDI-p and ∆percentage of polyphasic potentials of the voluntary activity of orbicularis oculi and orbicularis oris were found in Group A (r = 0.76 and r = 0.83, respectively) (Figure 4A and 4B) and Group B (r = 0.61 and r = 0.72, respectively) (Figure 4C and 4D). Furthermore, negative correlations in Group A were found between DD and ∆duration of voluntary activity of orbicularis oculi (r = -0.69) and orbicularis oris (r = -0.74).

DISCUSSION

Many attempts have been made to investigate the efficacy of different types of rescue treatments in the case of long-term facial palsy, since the disfiguring outcome that can ensue after FP usually motivates physicians to achieve the best functional outcome for their patients. [10] In the clinical setting, however, the interpretation of recovery outcomes is further complicated by confounding factors such as the aetiology of the palsy, time from onset of the disease, patients’ age, association of medical treatment and the mismatch between clinical and electrophysiological recovery, possibly due to technical barriers, finally impacting on decision making and final results within one year after onset of the disease. [5,7,10,27]

The first interesting finding of the present study demonstrated a significant within-subjects improvement in the duration of voluntary activity in Group A patients and a decrease in the percentage of polyphasic potentials of voluntary activity of orbicularis oris and orbicularis oculi in both groups (Table IV, Figure 3A). Moreover, significant decreases in the percentage of polyphasic potentials of voluntary activity of orbicularis oris and oculi and increases in duration of voluntary activity of orbicularis oris were found in Group A when compared with Group B (Table IV). Indeed, on one hand, this evidence tends to corroborate, although larger randomized trials are needed, the Kabat method as a valid option in treating patients with severe long-lasting FP with different sources of disease [6,10,28] and thus a reasonable approach to all the patients recruited in this study, in order to propose a rescue treatment. On the other hand, considering that the consequences of FP related to muscles hypotony/hypertony are known to be due to a failed reinnervation or reorganization of the facial nucleus, [29] further medical options (i.e. botulinum toxin) have been proposed as a single treatment or in association with rehabilitative procedures, in order to reduce stiffness, synkinesis, tension and even pain in the area. [7,25,30] According to these notions, Group A patients exhibiting significant variations of duration and percentage of polyphasic potentials of voluntary activity when compared with Group B, could possbily have benefitted from the collagen in the MDs. The aminoacid content of the collagen-based MDs used in this study is Glycine 22.8 %, Proline 13.8%, Hydroxy-Proline 13 % and an average of 3% of other aminoacids (max Glutamic Acid 9.5 %; min Tyrosine 0.4 %). Thus, it is possible to hypothesize that degradation into its constituent aminoacids could constitute a nutritional support for muscles, as previously demonstrated also in other experiences, so that this tissue may react more actively to the rehabilitative treatment. [18,19] By introducing the device in-situ, structural and mechanical effects provided by the collagen possibly operated as effective natural scaffolding to support cell growth, also providing a framework for cell-to-cell interactions [16,31] by redirecting reinnervation/reorganization phenomena of the facial nucleus, thus facilitating processes of recognizing afferent stimuli from the fifth cranial nerve [29] and/or by redirecting cortical projections to non-synkinetic motoneurons alone. [7,13,14] This may have resulted in improving the electrophysiological parameters that represent phenomena of dispersion of compound action potentials and desynchronization of the discharges of motor units (Table IV). [18,19,32] Further evidence of these aspects could reside in the different correlation behaviour found between ∆FDI-p and ∆percentage of polyphasic potentials of voluntary activity of orbicularis oris and oculi. In fact – although a significant positive correlation in both groups may also suggest the Kabat method alone as rescue support for increasing functional activity related to muscle strength [7,10] – Group A demonstrated an higher degree of correlation slope (Figure 4). Although previous studies on patients not undergoing facial nerve rehabilitation failed to discover correlations between EMG and FDI, [33] this evidence may further suggest that the impact of the in-situ delivery of collagen-based MDs on the structural aspects of muscle and the greater decrease in dispersion of compound action potentials could promote functional recovery of facial activity more effectively, with a positive impact on the quality of life. [34]

Regarding this aspect, both groups of patients demonstrated a significant post-treatment improvement in many sub-scales of SF-36 and FDI and demonstrated an increase, although not significant, in percentage of post-treatment reduction of synkinesis and HB severity (Table II, III, V). However, Group A subjects, when compared with Group B, demonstrated a significant improvement in SF-36 subscales investigating role limitation (physical and mental) and health perception. as well as in FDI-p (Table V, Figure 3B). This behaviour is in line with previous research protocols investigating the impact of rehabilitative and/or medical treatment on general and specific quality of life [20,35] and confirms similar results obtained in other studies implementing rehabilitation treatment with medical treatments (for example botulinum toxin injections). [7,25]

As previously highlighted in the literature, [35] in this study too, no evidence of correlation between general quality of life and specific facial disability scores was found. However, present between- group findings assume more relevance with regard to patients receiving MD injections, since, if on one hand there is still limited data available on the social function of patients suffering from long standing facial palsy or patients who underwent facial reanimation procedures, [35] on the other hand recent studies have underlined the importance of a general quality of life assessment in all facial palsy patients, as the severity of facial nerve impairment alone does not fully indicate patient distress and effects on social function. [34]

Finally, similarly to other rehabilitative protocols (for example mime therapy), the Kabat method alone (Group B) did not demonstrate an increase in outcome results in relation to elapsed time from the onset of disease. [6] On the other hand, although larger clinical trials are needed, Group A patients demonstrated a negative correlation between ET and ∆duration of voluntary activity of orbicularis oris and orbicularis oculi. This finding may suggest injection of MDs as a further and safe option to propose, together with an adequate rehabilitative protocol, to patients affected by long standing FP, not responding to previous resuscitation treatments and/or not eligible for surgical procedures. In fact, since this treatment, similarly to other combined procedures proposed in the literature, [7] has shown to increase the outcomes of rehabilitative protocols only in relation to the time elapsed from FP onset, regardless to its cause and degree, it may be proposed as soon as patients need further medical options. Furthermore, the injection procedure demonstrated a good level of satisfactory compliance referred by the patients, safety and tolerability as accounted by VAS and home diary (Figure 2B and 2C). Indeed, patients demonstrated to progressively comply with and to slightly refer adverse reactions to the injection procedure ‘per se’, most of them frequently referred also in other injections procedures experiences. [25] However, no adverse reaction was found with regard the content of the collagen-based MDs, representing an intriguing aspect when pointing attention to those possible, though transient, side effects referred by other substances (for example Botulinum toxin A). [7]

LIMITATIONS OF THE STUDY

Findings of the present study should be interpreted carefully since some limitations exist. In particular, given the nature of the disorder which limited the possibility of obtaining a larger sample size also in previous studies [7,28], only 41 FP patients have been enrolled. This aspect – together with the short follow-up – should suggest caution when evaluating the study results, which could be strengthened by further evaluation after a longer follow-up. Finally, no FP group undergoing placebo treatment – unethically achievable for these protocols – was investigated. This procedure could have tempered results of the present study. For future perspectives, the routinary implementation of collagen-based MDs is needed in order to perform research studies with sham and/or placebo treatments.

CONCLUSIONS

Findings of the present study demonstrated the combination of physical rehabilitative procedures with in-situ collagen device injections as a safe option enlarging the ‘window of opportunity’ to improve the voluntary muscle contraction pattern and general and specific disability referred by patients affected by long standing FP. However, future studies – especially when validated by larger cohorts of subjects and longer follow-up – are needed to confirm these promising results.

CONFLICT OF INTERESTS AND SOURCE OF FUNDING

No potential conflict of interest nor source of funding were reported by the authors.

FIGURES LEGEND

Figure 1. Flow diagram of participants in the study, randomization procedure, subjects who dropped out and main outcomes. VAS, visuo-analogue scale; Group A, subjects undergoing Kabat treatment and in-situ collagen injections; Group B, subjects undergoing only Kabat treatment.

Figure 2. 2A: Sites of injection therapy for the treatment group (Group A). The solution was injected in the orbicularis oculi muscle 1 cm outside the orbital rim and 0.5 cm superior and inferior to the first injection point, respectively, and in the orbicularis oris muscle, where the 2 injection points reside into the border between the pars peripheralis and pars marginalis, located about 5 mm above the superior and inferior vermilion, respectively. 2B: Discomfort scores related to injection treatment as shown by a visuo-analog scale (VAS) along the 8-week time period. 2C: Adverse reactions related to the injection procedure (n=16).

Figure 3. Mean and standard deviations of pre- and post-treatment electromyogaphic (A) and Facial Disability Index (B) findings in within-subjects and between-groups comparisons of patients affected by facial palsy and undergoing Kabat rehabilitation treatment with (Group A) or without (Group B) in-situ collagen injection. Brackets indicate significant comparisons.

Figure 4. Significant positive correlations between Facial Disability Index (FDI) and pre/post- treatment differences (∆) in percentage (%) of polyphasic potentials of voluntary activity in orbicularis oris and oculi in patients affected by facial palsy and undergoing Kabat rehabilitation treatment with (Group A) or without (GroupB) in-situ collagen injection.

TABLES

Table I. Clinical-demographic aspects of facial palsy patients undergoing Kabat rehabilitation with (Group A) or without (Group B) in-situ collagen injections.

Clinical-demographic overview of patients affected by facial palsy and undergoing Kabat rehabilitation treatment with (GroupA) or without (GroupB) in-situ collagen injection. Disease duration, DD; Left, L; Right, R. Where necessary mean ± standard deviations are reported.

Table II. House-Brackmann grading scale of facial palsy patients undergoing Kabat rehabilitation with (Group A) or without (Group B) in- situ collagen injections before and after treatment

For each severity degree of House-Brackmann scale are reported the numbers of facial palsy patients undergoing Kabat rehabilitation with (Group A) or without (Group B) in-situ collagen injections before and after treatment.

Table III. Synkinesis grading scale of facial palsy patients undergoing Kabat rehabilitation with (Group A) or without (Group B) in-situ collagen injections before and after treatment

For each severity degree of synkinesis scale are reported the numbers of facial palsy patients undergoing Kabat rehabilitation with (Group A) or without (Group B) in-situ collagen injections before and after treatment.

Table IV. Changes of electophysiological parameters in patients affected by facial palsy and undergoing Kabat rehabilitation with (Group A) or without (Group B) in-situ collagen injections.

Changes in pre- and post-treatment main electromyographic measures of voluntary contraction in patients affected by facial palsy and undergoing Kabat rehabilitation treatment with (GroupA) or without (GroupB) in-situ collagen injections. Significant within-subjects comparisons are reported in bold. F-ratio for between-groups comparisons is reported. Microvolt, μV; milliseconds, msec; standard deviation, SD.

Table V. Changes of validated questionnaire scores in patients affected by facial palsy and undergoing Kabat rehabilitation with (Group A) or without (Group B) in-situ collagen injections.

Changes in pre- and post-treatment validated questionnaire scores in patients affected by facial palsy and undergoing Kabat rehabilitation treatment with (GroupA) or without (GroupB) in-situ collagen injections. Significant within-subjects comparisons are reported in bold. F-ratio for between-groups comparisons is reported. 36-item short form survey for general health-related quality of life assessment, SF-36; Facial Disability Index, FDI, social, s; physical, p; standard deviation; SD.