Continual rehabilitation motivation of patients with postparalytic facial nerve syndrome

This study was part of a cross-sectional survey of adult patients with facial palsy (International Statistical Classification of Diseases and Related Health Problems, German Modification (ICD-10-GM) code: G51.0). These patients had visited the Department of Otolaryngology, Jena University Hospital, Germany, between 2006 and 2016 and had given written consent to be contacted of research purposes. 300 patients counting backwards from last contact with the department were contacted by post or directly in the department (Flowchart, see Supplement Fig. 1). 81 patients (response rate: 27%) answered the questionnaire. The present study analyzed the data of the patients with (PFS) at the time of the survey. The criteria were as follows: (a) a unilateral peripheral facial palsy; (b) incomplete recovery, (c) interval between onset and assessment at least 6 months; (d) at least one facial electromyography (EMG) confirming a PFS including synkinetic activity between periocular and perioral facial muscles [1, 13]. 204 of the contacted 300 patients fulfilled these PFS criteria (69 answered, 135 did not answer; response rate of PFS patients: 33.8%). Hence, these 69 patients with PFS constituted the opportunity sample. All facial palsy-related data were prospectively collected in the department and were complete for all 69 patients. The institutional review board of the Jena University Hospital approved the cross-sectional survey and the study protocol for the additional retrospective data analysis.

Socio-demographic data were collected. Furthermore, questions were asked about initial treatment during the acute phase of the disease, prior treatment in the chronic phase of the disease and current most burden complaints. A list of therapy options was offered with the question to select therapy forms the patient would use like to continue or in be interested in. This list included standard therapy forms (Home mirror training, Facial training with therapist, biofeedback training, facial electrostimulation, acupuncture, light/heat/cold therapy), drugs (supportive eye protection with drops/ointment, botulinum toxin), surgery (nerve surgery, eye lid surgery, angle of the mouth surgery), and also the newer technology allowing a computer-based home training [14]. Finally, we offered the implantation of a potentially in the future available facial pace maker [15]. All therapy forms were shortly explained. User preferences regarding a described concept of an emotion-sensitive training system and sleep quality of the patients were also recorded and published elsewhere [12]. The survey covered several validated patient-reported outcome measures (PROMs). The patient questionnaire for assessment of rehabilitation motivation (PAREMO-20) was used as primary outcome parameter. PAREMO-20 is a multidimensional instrument to determine the general rehabilitation-related motivation of the patients [16, 17]. PAREMO-20 consists of 20 items forming six subscales: “psychological burden”, “physical burden”, “social support”,”readiness to change”, “knowledge”, and “skepticism”. Answers are given on a 4‐point Likert scale. Higher values on subscales indicate higher rehabilitation motivation except for the subscale “skepticism”. Here, higher scores indicate lower rehabilitation motivation. General quality of life was measured using the 36-item SF-36 Health Survey [18]. Higher scores indicate higher quality of life. The Facial Clinimetric Evaluation (FaCE) scale was used to measure the facial palsy-related quality of life [19, 20]. The FaCE has six independent domains: social function, facial movement, facial comfort, oral function, eye comfort, lacrimal control, and a total core incorporating all domains. Using a specific formula, a score from 0 (worst) to 100 (best) is calculated. The 24-item Liebowitz Social Anxiety Scale (LSAS), uses two subscales that address social interaction (11 items) and performance (13 items), measuring an individual's fear and avoidance of social situations over the past week. Answers are given on a 4‐point Likert scale [21]. The German version of patients’ health questionnaire (PHQ-D) was used to record depression symptoms [22]. With nine items, the DSM-IV criteria for depression are asked on a 4-point Likert scale. Higher LSAS or PHQ-D scores indicate higher social anxiety or higher depression levels, respectively. Finally, the survey included the questionnaire for technical commitment [23]. The questionnaire consists of 12 items using a 5-point Likert scale to record in subitems “technology acceptance”, “technology competence”, “technology control beliefs”, and “technology willingness”. Finally, the Technology Affinity questionnaire—attitude to and handling of electronic devices (TA-EG), was used to understand the patient’s interest, experience and trust in technology [24]. The instrument comprised 19 items covering 4 subscales and uses a 5-point Likert scale: “enthusiasm for technology”, “competence in dealing with technology”, its “positive consequences”, and “negative consequences”. Higher technical commitment subscore and higher TA-EG subscores, respectively, indicate a higher agreement.

Additionally to self-ratings of facial palsy-related quality of life (with FaCE), facial motor function grading was performed using the Stennert index and the Sunnybrook Facial Grading Scale [25, 26]. The Stennert index was used because of its popularity in Germany. It was also used to classify the initial facial function at first presentation in the hospital. The observer judges facial symmetry at rest in four regional categories (0 = normal resting tone/symmetry up to 4 = no resting tone/gross asymmetry) and the motility of the facial muscles in six regional categories (0 = normal motility up to 6 = complete paralysis). The total score of the Stennert index summarizes both subscores. The Sunnybrook Facial Grading Scale was used because of its international recognition and because it allowed a separate classification of the degree of synkinesis. The Sunnybrook Facial Grading Scale is a regional weighted system that rates three subscores: resting symmetry, the degree of voluntary facial muscle movement, and involuntary muscle contraction (synkinesis). The three subscores are used to calculate a composite score (0 = total paralysis; 100 = normal function).

All outcome variables were analyzed with IBM SPSS statistics software (Version 25; IBM. New York) for medical statistics. Data are presented as frequencies or mean ± standard deviation (SD) if not otherwise indicated. To investigate selection bias we compared the data of patients who answered the questionnaire with the patients not responding. Pearson’s chi-square test was used for nominal data and the Mann–Whitney U test for metric data. McNemar test was used to compare binominal data of facial therapy types received in the past to future interests. Predictors for future interest in computer-based home therapy were explored using chi-square test for nominal data and Mann–Whitney U test for metric data. Predictors for the rehabilitation motivation were explored using Spearman’s correlation of PAREMO-20 subscales and all other parameters, e.g. social anxiety score. Comparisons of subgroups were only performed if a subgroup contained ≥ 10 patients. Linear regression analyses including parameters from univariate analysis and p < 0.05 were performed to evaluate associations related to the PAREMO-20 subscales. The significance level was set at p < 0.05.

Sixty-nine patients with PFS answered the survey. 135 patients did not answer. The comparison of both groups is summarized in Supplement Table 1. The study participants performed more frequently facial exercises at home (p < 0.0001), had more therapy with a therapist (p < 0.0001), and received more frequently botulinum toxin injections (p = 0.003). Gender distribution, age, and initial severity of the palsy were not different between both groups (all p > 0.05). Related to the Stennert index in motion and total index, improvement of the palsy from onset to time of the survey was better for non-participants (p = 0.008; p = 0.023, respectively). The composite score of the Sunnybrook grading revealed a lower score (worse function) for study participants (p = 0.022).

Median age of the participants was 54 years. Most patients were females (72.5%). Median onset of facial palsy was 2.6 years ago. More details are shown in Table 1. Nearly all patients had performed mirror training at home (94.2%) or reported a prior specific facial training with a therapist (82.6%). All patients had a therapy form in direct contact with a therapist. The median number of different types of facial therapy (eye protection excluded) was 4 (range: 1–7). Other frequent therapies in the chronic phase of the disease were as follows: continued eye protection (85.5%), acupuncture (49.3%), electrostimulation (44.0%), and physical therapy with cold, heat or light (40.6%). The three most often mentioned still disturbing symptoms were as follows: Impaired eye closure (66.7%), asymmetric face (63.8%), and impaired smiling (52.2%). Stennert index and Sunnybrook grading confirmed the still affected facial motor function: Median total Stennert index and composite score of the Sunnybrook index were 4 and 41, respectively. This corresponded to patient’s self-reported facial function (Table 2): The FaCE Facial movement subscore was the lowest with a median value of 33.0, followed by the FaCE Facial comfort subscore (median: 50.0), and the FaCE Eye comfort subscore (median: 62.5). From the SF-36 subscores, SF-36 Vitality (median: 55.0), and SF-36 General health (median: 57.0) showed the worst results. LSAS and PHQ-9 showed a significant impairment in the study group. Technical commitment and technology affinity to electronic devices was moderate.

All patients showed interest in further therapy (Supplement Table 2), mostly in the form of home mirror therapy (84.1%), continual eye protection (82.6%), facial training with therapist (72.5%), with a computer (55.1%), and biofeedback therapy (42.0%). Compared to the past, the future interest in home facial mirror training, acupuncture, electrostimulation, and any light/heat/cold therapy decreased significantly (all p < 0.05; Fig. 1), while the interest in computer-assisted home facial training increased significantly (p < 0.0001). Therefore, factors with association with this wish were further explored (Supplement Table 3). Patients with future interest in computer-assisted facial therapy showed a lower FaCE Eye comfort subscore (p = 0.044), lower SF-36 Vitality subscore (p = 0.008), lower SF-36 Social functioning (p = 0.009), lower SF-36 Role functioning (p = 0.010), lower SF-36 Mental health (p = 0.005, leading also to a lower SF-36 Mental summary score (p = 0.002).

Table 2 lists the results of the PAREMO-20 subscores. The two highest scores were the Physical burden subscore (median: 10) and the Knowledge subscore (median: 9). The PAREMO-20 subscores Psychological burden, Physical burden, and Readiness to change showed the highest bivariate correlation to each other (Supplement Table 4). The correlations between the six PAREMO-20 subscores and patients’ characteristics and the other assessments did not show a uniform picture (Supplement Table 5).

The results of the multivariate linear regression analyses for independent associations to the PAREMO-20 subscores are shown in Table 3. For the Psychological burden subscore, SF36 Emotional role was the highest negative correlative factor (beta = − 0.026; 95% confidence interval [95% CI] = − 0.040 to − 0.012; p < 0.0001). For the Physical burden subscore, SF-36 General health was the highest negative correlative factor (beta = − 0.048; 95% CI = − 0.088 to − 0.009; p = 0.018). Working (beta = 2.149; 95% CI = 0.184 to 4.114; p = 0.033) and a steady relationship (beta = 1.823; 95% CI = 0.192 to 3.453; p = 0.029) predicted the Social Support Subscore. A predictor for readiness for change was not found. Younger age was associated with a higher Knowledge subscore (beta = − 0.070; 95% CI = − 0.125 to − 0.015; p = 0.013). Higher positive impacts of technology affinity was correlated to lower Knowledge subscore (beta = − 1.700; 95% CI = − 3.082 to − 0.317; p = 0.017). Finally, a lower SF-36 Role physical subscore (beta = − 0.023; 95% CI = − 0.046 to − 0.001; p = 0.045) and a lower SF-36 General health (beta = − 0.038, 95% CI = − 0.067 to − 0.008; p = 0.013) were correlated to a higher Skepticism subscore.