This article discusses the causes and symptoms, evaluation, and management of unilateral vocal fold paralysis (UVFP). Cross-sectional imaging is appropriate in the work-up of idiopathic UVFP, but the routine use of serology is not well supported. The usefulness of laryngeal electromyography has remained controversial. Predictors of poor prognosis for functionally meaningful recovery include fibrillation potentials, positive sharp waves, and absent/reduced voluntary motor unit potentials. Voice therapy may be helpful. Injection and laryngeal framework surgery (medialization thyroplasty) improve vocal quality. The vocal impact of laryngeal reinnervation is comparable with that of medialization. Some patients may benefit from multiple procedures.
The following points list the level of evidence as based on Oxford Center for Evidence-Based Medicine.
- •
Unilateral vocal fold paralysis (UVFP) has a broad range of causes, including postsurgical, idiopathic, and neoplasm-related (evidence grade C).
- •
Work-up for UVFP should include computed tomography imaging, but not serology. Electromyography is useful for predicting poor prognosis (evidence grade C).
- •
Voice therapy can be beneficial, but is not sufficient for many patients with UVFP. In the short term, injection medialization can achieve comparable clinical results with medialization thyroplasty. Thyroplasty and reinnervation also achieve comparable voice outcomes (evidence grade B).
Overview
Unilateral vocal fold paralysis (UVFP) continues to command attention as a fundamental clinical problem in otolaryngology. Its impact on voice, swallowing, and even airway function is notable. Patients and their physicians have many helpful diagnostic and therapeutic options for treatment, but at times this clinical decision making must occur without the luxury of scientifically established principles and practices. This article provides an evidence-based overview of (1) the causes and symptoms, (2) evaluation, and (3) management of UVFP. Publications addressing aspects of this topic number in the thousands, and therefore selected articles are presented to provide a sense of how the evidence has been developed and assessed. The discussion focuses primarily on UVFP rather than on the broader topic of unilateral vocal fold immobility. For each topic, the Oxford Center for Evidence-Based Medicine levels of evidence are listed.
| CT | Computed tomography |
| CXR | Chest x-ray/radiograph |
| LEMG | Laryngeal electromyography |
| UVFP | Unilateral vocal fold paralysis |
Causes of UVFP (Evidence Level 3–4)
The list of potential causes of UVFP is broad, and includes:
Iatrogenic
Iatrogenic injury is commonly related to retraction and/or dissection along the route of the recurrent laryngeal nerve or even the vagus itself. Procedures associated with risk of postoperative vocal fold paralysis include thyroidectomy (0.8%–2.3% rate of permanent UVFP), anterior cervical spine surgery (less than 1% risk of permanent UVFP per recent data), esophagectomy (∼11% risk), cardiac/aortic surgery (∼2% risk), mediastinoscopy (0.2%–6% risk), and carotid endarterectomy (∼4% risk).
Interpretation of the risks of UVFP associated with surgical procedures can be confusing because of the difficulty of determining the contribution of underlying disorders to the postoperative outcome. Without systematic preoperative and postoperative assessment of laryngeal function, which may not be clinically feasible, precise risk of UVFP associated with a given surgical procedure can be difficult to estimate. The picture is further clouded by the association of endotracheal intubation or laryngeal mask airway with UVFP, although some work has shown that the risk of iatrogenic UVFP related to retraction may be decreased by ongoing monitoring and adjustment of endotracheal tube pressure, particularly when retractors are placed or repositioned.
Traumatic
Traumatic causes associated with UVFP include high vagal nerve injury caused by direct trauma, although vagal nerve injuries more commonly result from surgical removal of masses involving the vagus itself. Arytenoid dislocation has been proposed as a cause of unilateral vocal fold immobility but this topic remains controversial; a recent review suggests that the diagnosis cannot be made by laryngoscopy alone and that there is insufficient evidence in the literature to characterize arytenoid dislocation as a unique entity.
Neoplasms and thoracic diseases
Tumors and thoracic problems have also been implicated in the pathophysiology of UVFP, including lung cancer, thoracic aortic aneurysm, metastases, pulmonary/mediastinal tuberculosis, esophageal cancer, patent ductus arteriosus, and laryngeal chondrosarcoma. Direct infiltration of the recurrent laryngeal nerve can also occur with thyroid carcinomas in addition to lung carcinomas as mentioned earlier. UVFP has also been reported after iodine-131 treatment of thyrotoxicosis or radiation of other types to the head and neck or upper chest. Tumors in the central nervous system can also cause UVFP, but typically have a constellation of associated symptoms.
Systemic
Systemic causes can be divided into a variety of categories. Infectious causes include West Nile, varicella and herpes, Lyme, and syphilis, while inflammatory processes can include sarcoidosis, lupus, amyloidosis, polyarteritis nodosa, and silicosis. Neurologic diagnoses associated with vocal fold paralysis include myasthenia gravis, severe degenerative spine disease, multiple sclerosis, amyotrophic lateral sclerosis, Guillain-Barré (although typically bilateral), Parkinson (also more commonly described as bilateral), Charcot-Marie-Tooth, and familial hypokalemic periodic paralysis. Diabetes or malnutrition, such as B12 deficiency, can contribute, as can medications such as vinca alkaloids. Idiopathic UVFP is a diagnosis of exclusion and, in those cases, the pathogenesis remains poorly understood.
Evolving Distribution of Causes (Levels 3–4)
Both in the United States and elsewhere, numerous studies have focused on the relative distribution of causes for UVFP. Neoplasm, trauma, and surgery were the most consistently cited causes in the 1970s, but the distribution has evolved over time. In 1998, a retrospective review by Ramadan and colleagues examined causes in 98 patients with UVFP; they were categorized as neoplastic in 32%, surgical in 30%, idiopathic in 16%, traumatic in 11%, central in 8%, and infectious in 3%. Several years later, a large comparative retrospective analysis spanning a 20-year period was presented by Rosenthal and colleagues, comprising 827 patients who were seen with vocal fold immobility. In the first decade, spanning 1985 to 1995, the most common cause was malignancy (mostly lung). By contrast, in the second decade, 1996 to 2005, the most common cause was nonthyroid surgery (including anterior cervical spine, carotid). Consistent with these findings, a 1-year retrospective study published in 2006 reported a greater proportion of anterior cervical spine surgery than thyroid, thoracic, or cranial procedures among patients with iatrogenic vocal fold motion impairment.
Thus, although the distribution of causes has evolved over time and may vary depending on geographic location, commonly reported causes include neoplasms, particularly lung and thyroid neoplasms, and postsurgical causes (commonly spine surgery, carotid surgery, and thyroidectomy), with a persistent minority remaining idiopathic after thorough evaluation.
Natural History of Idiopathic UVFP (Level 4)
Observational studies have provided some insight into the natural history of UVFP. A retrospective review of 633 patients with vocal fold paralysis diagnosed between 1940 and1949 included 181 of unknown cause. Of those, 31 had respiratory infection before onset of symptoms, and 29 had incidental findings such as goiter or pharyngoesophageal diverticulum, but most had no apparent predisposing factors. Long-term survival data suggested that patients with truly idiopathic UVFP seemed to have normal life spans, with 33% chance of vocal improvement over time. More recently, Sulica performed a review of the literature and identified 20 articles reporting 717 cases of idiopathic vocal fold paralysis. He reported that idiopathic vocal fold paralysis comprises 24% ± 10% of UVFP. When findings from all of the studies were summarized, complete recovery of motion was observed in 36% ± 22% and some recovery (complete and partial) in 39% ± 20%. Complete recovery of voice was reported in 52% ± 17%, and some recovery in 61% ± 22%. Most recovered in less than a year, but a small minority (5/717) described recovery after more than a year. As noted in the review, the variable recovery rates reported in different studies likely relates to heterogeneity of timeframe as well as criteria for defining recovery, but most patients with idiopathic UVFP showed some vocal improvement, typically in less than a year.
Symptoms (Levels 3–4)
Although evaluation of UVFP is frequently focused on voice complaints (discussed later), dysphagia and other complaints appear fairly common. In a survey of 63 patients with UVFP secondary to a variety of causes, all patients with UVFP reported voice problems, 60% reported swallowing problems, and 75% reported subjective dyspnea. In patients with dysphagia and UVFP who underwent flexible endoscopic evaluation of swallowing studies, liquid bolus retention and penetration were associated with aspiration in nearly half, and the pharyngeal residues were noted at the base of tongue, valleculae, and piriform sinuses. These findings may suggest that difficulty with swallowing in patients with UVFP is not solely explained by the vocal fold problem and may reflect associated problems, possibly secondary to the underlying cause of the UVFP.
Quality of Life at Presentation (Levels 3–4)
It is intuitive that UVFP would have a considerable effect on quality of life given the findings described earlier, and validated scales have shown this impact. In a study at Vanderbilt University, patients with UVFP prospectively completed The Medical Outcomes Study Short Form 36-Item Health Survey (SF-36), Voice Handicap Index (VHI), and Voice Outcome Survey (VOS) at presentation and at first postoperative visit after thyroplasty with or without arytenoid adduction. The SF-36 is a general health status measure, the VHI is a voice-specific handicap measure, and the VOS is a survey designed to assess vocal quality and life impact of voice-related problems specifically in patients with UVFP. At presentation, patients with UVFP scored significantly lower (worse) than normal on all domains of the SF-36 and on the VOS. After surgery, acoustic and aerodynamic measures of voice were improved. All domains of the SF-36 were observed to have a trend toward increased scores, with some domains showing a statistically significant increase. The VHI and its subscales, as well as the VOS, had significant improvement. Several other measures of voice-related patient-reported quality of life have been used, including the Voice-Related Quality of Life (VRQOL) and a variety of study-specific scales. These and other studies underscore the significant impact of UVFP on both voice-related and overall health-related quality of life.
Evidence-based clinical assessment
Examination for UVFP (Levels 3–5)
Examination of the patient who presents with suspected UVFP may include several components, including:
Auditory-perceptual evaluation
Auditory-perceptual evaluation using the GRBAS (grade, roughness, breathiness, asthenia, strain) scale shows that patients with UVFP are rated significantly worse than normal. More recently, the Consensus Auditory Perceptual Evaluation of Voice (CAPE-V) was developed for voice disorders in general. Little information is available in the literature about CAPE-V evaluation of UVFP, but work in postthyroidectomy patients at Walter Reed Hospital suggest that overall severity, habitual loudness, habitual pitch, and roughness are parameters that may be affected. The challenges with auditory-perceptual evaluation of voice are well documented and include issues of interrater and intrarater reliability as well as the impact of listener experience and knowledge of the patient’s history and/or diagnosis. In addition, patients’ perceptual self-ratings seemed to be distinct from those of trained listeners. Patient ratings of the impact of vocal problems on quality of life do not correlate well with auditory-perceptual judgments. Nonetheless, these judgments do allow raters to follow voice changes over time.
Acoustic and aerodynamic evaluation
Acoustic and aerodynamic evaluation in UVFP shows worse jitter, shimmer, noise/harmonic ratio, and maximum phonation time compared with normal voice. As noted by Behrman, acoustic and aerodynamic measures, although sometimes thought to be objective, are not truly so, because of the need for behavioral investment on the part of both patient and clinician to obtain representative phonatory samples and the challenge of performing some of these measurements when vocal fold vibration is irregular. The relevance and validity of measures such as maximum phonation time and s/z ratio is also questioned because, in some cases, suboptimal techniques such as excessive supraglottic recruitment can lead to apparently improved maximum vocal performance measures. Nonetheless, these measures are frequently reported. Development of nonlinear, random time-series analysis may provide further information, but is in its early stages. Another potentially promising technique is spectral moment analysis. Reduction of cepstral peak prominence has been observed in patients with UVFP compared with controls, but it is unclear whether this is diagnostic.
Intensity
Intensity has been used occasionally as a measure of vocal function, particularly habitual speaking intensity (loudness) and/or maximum physiologic dynamic range; these measures may be more closely related to the patient’s assessment of vocal impact of vocal fold paralysis.
Laryngoscopy
Laryngoscopy is an essential part of the evaluation of UVFP. The most common laryngoscopic findings beyond vocal fold motion impairment include bowing, incomplete glottal closure, and phase asymmetry on videostroboscopy. The position of the vocal fold (eg, paramedian vs lateral) does not necessarily clarify the location of the lesion along the neurologic pathway from brain to motion of vocal fold. However, the paralyzed side does tend to be shortened and arytenoid is commonly anteriorly rotated. Passive gliding motion of arytenoid is seen in 91% patients with UVFP examined by three-dimensional (3D) computed tomography (CT), and caudal displacement in 100%. Some have suggested that the position and shape of the false vocal fold may be informative, but this is controversial. The specific value of stroboscopy compared with routine flexible fiberoptic laryngoscopy has also been debated, and its use is limited by challenges in capturing an adequate signal in profoundly dysphonic patients.
Imaging for UVFP (Levels 3–5)
A variety of imaging techniques have been used in the work-up of patients with suspected UVFP. In a survey of members of the American Broncho-Esophagological Association, respondents indicated that: Chest radiography (CXR) and/or neck/chest CT is always or often necessary (69%–72%). Magnetic resonance imaging (MRI) was thought to be always or often necessary by 39%, sometimes by 51%.
CXR and CT
An area of particular interest is the comparison of CXR versus neck/chest CT (typically with contrast) for evaluation of possible causes of UVFP given the considerable differences in cost and exposure to radiation. CXR can detect important diagnoses such as goiter and pulmonary fibrosis, but may miss findings detected by CT, particularly those in the left aortopulmonary window. It has also been suggested that MRI is more sensitive, but carries a higher rate of false-positives.
Because of the false-negative rate seen on CXR, several algorithms have been proposed for imaging used as part of the work-up of UVFP. Altman and Benninger described starting with CXR and proceeding with CT or MRI if the CXR is negative. The CT is performed from skull base to thoracic inlet for right UVFP, and skull base to aortic triangle for left UVFP. In contrast, El Badawey and colleagues described primary use of CT, without routine use of CXR.
Liu and colleagues described stratification of patients with newly diagnosed UVFP using clinical findings (such as a history of malignancy) to divide into high-suspicion and low-suspicion groups. They then examined costs associated with imaging for each group. The high-suspicion group work-up (which included MR and/or CT) cost $2304 per true-positive, whereas the low-suspicion group cost $10,849 per true-positive case. An implication of these findings is that imaging could be deferred for the low-suspicion group, but the associated risks and costs of delayed diagnosis need to be evaluated thoroughly before making such a recommendation.
Ultrasound
The use of ultrasound has attracted more attention in recent years ; neck ultrasonography identified subclinical tumors in 30% of 53 patients with UVFP, including papillary thyroid carcinoma and metastatic cervical lymph nodes from lung and other cancers. Some describe using ultrasound if physical examination suggests low right recurrent laryngeal nerve impairment.
Positron emission tomography
Although positron emission tomography (PET) scanning is not routinely used in the diagnosis of UVFP, it is important to be aware of the potential for misleading results on PET that are related to the presence of UVFP. Several studies have shown that, when UVFP is present, the contralateral normal side can have high fluorodeoxyglucose uptake thought to be secondary to attempted compensatory motion, potentially raising misleading concern for malignancy. These findings have most commonly been described in patients with primary lung malignancies who had secondary unilateral recurrent laryngeal nerve paralysis. Other potentially misleading findings arise from the treatment of UVFP; granulomas that arise from the use of Teflon (polytetrafluoroethylene, DuPont, Wilmington, DE) for injection medialization can lead to false-positive findings on PET, as can an elastomer suspension implant (trade name Vox, Uroplasty Inc., Minnetonka, MN).
Summary of evidence on recommended imaging
The ideal algorithm for imaging in the work-up of UVFP remains controversial. Evidence in the literature is inadequate to make a blanket recommendation, but numerous studies have reported the use of imaging to identify significant abnormalities in patients who present with idiopathic vocal fold paralysis, and cross-sectional imaging is likely indicated. It can be difficult to directly synthesize across studies given different recruitment and/or inclusion criteria. Prospective controlled studies are necessary for further evaluation. Other factors to consider include cost and exposure to radiation.
Serology in UVFP (Levels 3–5)
Use of serology in the evaluation of patients with UVFP has been described in a variety of studies. A survey of American Broncho-Esophagological Association members indicated that 54% of respondents indicated that serum tests could be considered as part of a work-up, but most (80%) of these indicated that the tests were appropriate only occasionally or rarely. The most commonly mentioned tests were rheumatoid factor (38%), Lyme titer (36%), erythrocyte sedimentation rate (34%), and antinuclear antibody (ANA) (33%). Review of the literature at that time showed mostly case reports, with 1 case-control study on diabetes ; sarcoidosis and ANA were also frequently addressed but there remains no population-based information. There remains no definite evidence to support routine serology in patients with UVFP who do not have signs/symptoms of underlying disease, and practitioners are likely best served by ordering serology only if they have a clinical index of suspicion for particular associated diseases.
Laryngeal Electromyography (Levels 2–5)
The inclusion of laryngeal electromyography (LEMG) in the evaluation of patients with UVFP has garnered attention as a technique that could evaluate the current neurologic status of the affected vocal fold and perhaps provide prognostic information. Although the increasing popularity of injection medialization has perhaps tempered the impact of LEMG because immediate and temporary intervention is now available, interest in electromyography (EMG) continues. In a 2005 survey of members of the American Broncho-Esophagological Association, 75% of respondents used EMG to evaluate UVFP. These evaluations were typically performed in an unblinded fashion (85%); 66% of respondents thought that having clinical information was helpful. Congruent results were generally reported, with some variability.
In 2009, the Neurolaryngology Study Group of the American Academy of Otolaryngology/Head and Neck Surgery convened a multidisciplinary panel to develop recommendations based on available evidence combined with expert opinion in areas in which evidence was not yet available. They summarized data that may suggest usefulness of EMG, but concluded that EMG was primarily a qualitative, not quantitative, examination. The study group advocated caution regarding the use of EMG data for early management of UVFP, and encouraged consideration of serial examinations. General recommendations included the need for prospective and blinded studies as well as standardized methods and interpretations. A 2012 meta-analysis by Rickert and colleagues examined the usefulness of LEMG for prognosis in vocal fold palsy using laryngoscopy as the gold standard; their analysis showed that, among patients with abnormal findings such as fibrillations, positive sharp waves, and absent or reduced voluntary motor unit potentials, 91% had no recovery of vocal fold mobility, although the length of follow-up was variable across the included studies. Other proposed methods for LEMG interpretation include interference pattern analysis, which allows description of motor unit recruitment in patients with UVFP, and the use of the ratio of mean peak-to-peak amplitude comparing motor unit amplitude on sniff versus sustained phonation for evaluation of synkinesis and associated poor prognosis for recovery.
The current evidence indicates that LEMG with negative prognostic factors is likely to predict a poor functional outcome, but the optimal timing of LEMG in relation to symptom onset remains unclear. Prospective blinded studies are needed to confirm these and other potential ways to use LEMG in a quantitative, objective, and reproducible fashion.
Evidence-based clinical assessment
Examination for UVFP (Levels 3–5)
Examination of the patient who presents with suspected UVFP may include several components, including:
Auditory-perceptual evaluation
Auditory-perceptual evaluation using the GRBAS (grade, roughness, breathiness, asthenia, strain) scale shows that patients with UVFP are rated significantly worse than normal. More recently, the Consensus Auditory Perceptual Evaluation of Voice (CAPE-V) was developed for voice disorders in general. Little information is available in the literature about CAPE-V evaluation of UVFP, but work in postthyroidectomy patients at Walter Reed Hospital suggest that overall severity, habitual loudness, habitual pitch, and roughness are parameters that may be affected. The challenges with auditory-perceptual evaluation of voice are well documented and include issues of interrater and intrarater reliability as well as the impact of listener experience and knowledge of the patient’s history and/or diagnosis. In addition, patients’ perceptual self-ratings seemed to be distinct from those of trained listeners. Patient ratings of the impact of vocal problems on quality of life do not correlate well with auditory-perceptual judgments. Nonetheless, these judgments do allow raters to follow voice changes over time.
Acoustic and aerodynamic evaluation
Acoustic and aerodynamic evaluation in UVFP shows worse jitter, shimmer, noise/harmonic ratio, and maximum phonation time compared with normal voice. As noted by Behrman, acoustic and aerodynamic measures, although sometimes thought to be objective, are not truly so, because of the need for behavioral investment on the part of both patient and clinician to obtain representative phonatory samples and the challenge of performing some of these measurements when vocal fold vibration is irregular. The relevance and validity of measures such as maximum phonation time and s/z ratio is also questioned because, in some cases, suboptimal techniques such as excessive supraglottic recruitment can lead to apparently improved maximum vocal performance measures. Nonetheless, these measures are frequently reported. Development of nonlinear, random time-series analysis may provide further information, but is in its early stages. Another potentially promising technique is spectral moment analysis. Reduction of cepstral peak prominence has been observed in patients with UVFP compared with controls, but it is unclear whether this is diagnostic.
Intensity
Intensity has been used occasionally as a measure of vocal function, particularly habitual speaking intensity (loudness) and/or maximum physiologic dynamic range; these measures may be more closely related to the patient’s assessment of vocal impact of vocal fold paralysis.
Laryngoscopy
Laryngoscopy is an essential part of the evaluation of UVFP. The most common laryngoscopic findings beyond vocal fold motion impairment include bowing, incomplete glottal closure, and phase asymmetry on videostroboscopy. The position of the vocal fold (eg, paramedian vs lateral) does not necessarily clarify the location of the lesion along the neurologic pathway from brain to motion of vocal fold. However, the paralyzed side does tend to be shortened and arytenoid is commonly anteriorly rotated. Passive gliding motion of arytenoid is seen in 91% patients with UVFP examined by three-dimensional (3D) computed tomography (CT), and caudal displacement in 100%. Some have suggested that the position and shape of the false vocal fold may be informative, but this is controversial. The specific value of stroboscopy compared with routine flexible fiberoptic laryngoscopy has also been debated, and its use is limited by challenges in capturing an adequate signal in profoundly dysphonic patients.
Imaging for UVFP (Levels 3–5)
A variety of imaging techniques have been used in the work-up of patients with suspected UVFP. In a survey of members of the American Broncho-Esophagological Association, respondents indicated that: Chest radiography (CXR) and/or neck/chest CT is always or often necessary (69%–72%). Magnetic resonance imaging (MRI) was thought to be always or often necessary by 39%, sometimes by 51%.
CXR and CT
An area of particular interest is the comparison of CXR versus neck/chest CT (typically with contrast) for evaluation of possible causes of UVFP given the considerable differences in cost and exposure to radiation. CXR can detect important diagnoses such as goiter and pulmonary fibrosis, but may miss findings detected by CT, particularly those in the left aortopulmonary window. It has also been suggested that MRI is more sensitive, but carries a higher rate of false-positives.
Because of the false-negative rate seen on CXR, several algorithms have been proposed for imaging used as part of the work-up of UVFP. Altman and Benninger described starting with CXR and proceeding with CT or MRI if the CXR is negative. The CT is performed from skull base to thoracic inlet for right UVFP, and skull base to aortic triangle for left UVFP. In contrast, El Badawey and colleagues described primary use of CT, without routine use of CXR.
Liu and colleagues described stratification of patients with newly diagnosed UVFP using clinical findings (such as a history of malignancy) to divide into high-suspicion and low-suspicion groups. They then examined costs associated with imaging for each group. The high-suspicion group work-up (which included MR and/or CT) cost $2304 per true-positive, whereas the low-suspicion group cost $10,849 per true-positive case. An implication of these findings is that imaging could be deferred for the low-suspicion group, but the associated risks and costs of delayed diagnosis need to be evaluated thoroughly before making such a recommendation.
Ultrasound
The use of ultrasound has attracted more attention in recent years ; neck ultrasonography identified subclinical tumors in 30% of 53 patients with UVFP, including papillary thyroid carcinoma and metastatic cervical lymph nodes from lung and other cancers. Some describe using ultrasound if physical examination suggests low right recurrent laryngeal nerve impairment.
Positron emission tomography
Although positron emission tomography (PET) scanning is not routinely used in the diagnosis of UVFP, it is important to be aware of the potential for misleading results on PET that are related to the presence of UVFP. Several studies have shown that, when UVFP is present, the contralateral normal side can have high fluorodeoxyglucose uptake thought to be secondary to attempted compensatory motion, potentially raising misleading concern for malignancy. These findings have most commonly been described in patients with primary lung malignancies who had secondary unilateral recurrent laryngeal nerve paralysis. Other potentially misleading findings arise from the treatment of UVFP; granulomas that arise from the use of Teflon (polytetrafluoroethylene, DuPont, Wilmington, DE) for injection medialization can lead to false-positive findings on PET, as can an elastomer suspension implant (trade name Vox, Uroplasty Inc., Minnetonka, MN).
Summary of evidence on recommended imaging
The ideal algorithm for imaging in the work-up of UVFP remains controversial. Evidence in the literature is inadequate to make a blanket recommendation, but numerous studies have reported the use of imaging to identify significant abnormalities in patients who present with idiopathic vocal fold paralysis, and cross-sectional imaging is likely indicated. It can be difficult to directly synthesize across studies given different recruitment and/or inclusion criteria. Prospective controlled studies are necessary for further evaluation. Other factors to consider include cost and exposure to radiation.
Serology in UVFP (Levels 3–5)
Use of serology in the evaluation of patients with UVFP has been described in a variety of studies. A survey of American Broncho-Esophagological Association members indicated that 54% of respondents indicated that serum tests could be considered as part of a work-up, but most (80%) of these indicated that the tests were appropriate only occasionally or rarely. The most commonly mentioned tests were rheumatoid factor (38%), Lyme titer (36%), erythrocyte sedimentation rate (34%), and antinuclear antibody (ANA) (33%). Review of the literature at that time showed mostly case reports, with 1 case-control study on diabetes ; sarcoidosis and ANA were also frequently addressed but there remains no population-based information. There remains no definite evidence to support routine serology in patients with UVFP who do not have signs/symptoms of underlying disease, and practitioners are likely best served by ordering serology only if they have a clinical index of suspicion for particular associated diseases.
Laryngeal Electromyography (Levels 2–5)
The inclusion of laryngeal electromyography (LEMG) in the evaluation of patients with UVFP has garnered attention as a technique that could evaluate the current neurologic status of the affected vocal fold and perhaps provide prognostic information. Although the increasing popularity of injection medialization has perhaps tempered the impact of LEMG because immediate and temporary intervention is now available, interest in electromyography (EMG) continues. In a 2005 survey of members of the American Broncho-Esophagological Association, 75% of respondents used EMG to evaluate UVFP. These evaluations were typically performed in an unblinded fashion (85%); 66% of respondents thought that having clinical information was helpful. Congruent results were generally reported, with some variability.
In 2009, the Neurolaryngology Study Group of the American Academy of Otolaryngology/Head and Neck Surgery convened a multidisciplinary panel to develop recommendations based on available evidence combined with expert opinion in areas in which evidence was not yet available. They summarized data that may suggest usefulness of EMG, but concluded that EMG was primarily a qualitative, not quantitative, examination. The study group advocated caution regarding the use of EMG data for early management of UVFP, and encouraged consideration of serial examinations. General recommendations included the need for prospective and blinded studies as well as standardized methods and interpretations. A 2012 meta-analysis by Rickert and colleagues examined the usefulness of LEMG for prognosis in vocal fold palsy using laryngoscopy as the gold standard; their analysis showed that, among patients with abnormal findings such as fibrillations, positive sharp waves, and absent or reduced voluntary motor unit potentials, 91% had no recovery of vocal fold mobility, although the length of follow-up was variable across the included studies. Other proposed methods for LEMG interpretation include interference pattern analysis, which allows description of motor unit recruitment in patients with UVFP, and the use of the ratio of mean peak-to-peak amplitude comparing motor unit amplitude on sniff versus sustained phonation for evaluation of synkinesis and associated poor prognosis for recovery.
The current evidence indicates that LEMG with negative prognostic factors is likely to predict a poor functional outcome, but the optimal timing of LEMG in relation to symptom onset remains unclear. Prospective blinded studies are needed to confirm these and other potential ways to use LEMG in a quantitative, objective, and reproducible fashion.
Evidence-based medical management and surgical technique
Speech Pathology
Voice therapy (Level 4)
Several studies have described the use of voice therapy in the management of UVFP, although there is an opportunity in the literature for further examination of this issue. Although swallowing therapy is outside the scope of this article, techniques may include chin tuck, neck extension, head turn, supraglottic and supersupraglottic swallow, and/or dietary modification. Some patients also benefit from oral motor exercises, vocal adduction exercises, Valsalva swallow, and Mendelsohn maneuvers.
Voice therapy in UVFP is typically directed at abdominal breathing and humming/resonant voice to improve closure of the glottis, encourage abdominal breath support, and improve vocal fold function while avoiding supraglottic hyperfunction.
Depending on the study, significant numbers of patients with UVFP who opted for voice therapy reported vocal improvement subjectively or as measured by glottal closure, acoustic measurements, pitch range, and/or patient-reported voice handicap. Interpretation of the impact of voice therapy in UVFP may be obscured by returning neurologic function, and it is unknown whether there is a relationship between voice therapy and neurologic recovery.
Other studies have also suggested the usefulness of voice therapy in the management of UVFP, but there is often no comparison group, making it difficult to assess whether voice therapy affected the likelihood of these improvements. Randomized controlled studies may be difficult to perform in this area because patients who desire surgery may not be receptive to randomization into voice therapy, and vice versa, but controlled studies may be possible using alternate study designs. An additional challenge to studying this topic is the variability of therapy techniques across institutions or across individual therapists.
Surgical Techniques for UVFP
Medialization
One of the mainstays of surgical treatment of UVFP is the concept of medialization, in which the paralyzed vocal fold is displaced toward the midline to facilitate glottal closure. The ideal timing for medialization remains unclear. Some have postulated that the immobile cricoarytenoid joint may become fixed over time, whereas other studies suggest that joint mobility may remain intact even many years after onset of paralysis. Other studies examine voice and other functional outcomes after early medialization versus late medialization, and these are discussed later.
Injection medialization (levels 1–4)
Injection medialization of the vocal fold was first performed in 1911 by Brunings via peroral injection using paraffin. This technique lost popularity until the development of other injectables that were thought to be less reactogenic. A recent retrospective review from multiple institutions summarized characteristics and complications of 460 injections for augmentation of the vocal folds of which 54% were performed for vocal fold paralysis. There was an even split between awake injections performed in clinic versus those performed under general anesthesia. Most awake injections (47%) were performed via transcricothyroid approach. Also frequent were: Perioral: 23% and Transthyrohyoid: 21%.
Reported technical success rates were 97% or greater and complication rates were 3% or less, with no difference between awake and asleep techniques. Use of injection augmentation in awake patients is increasing; over the 5-year period from 2003 to 2008, the rate increased from 11% to 43%. The goal of injection medialization is to reposition the vocal fold medially, allowing contact between the affected side and the normal side ( Fig. 1 ). Recent data suggest that injection medialization causes passive medial rotation and translation of the arytenoid cartilage.
Injection medialization materials
A variety of materials is now available for use in injection medialization. The literature was recently summarized by Paniello, who examined findings from 42 articles describing up to 30 patients each with follow-up time up to 1 year, describing injection medialization and voice outcomes as reflected by a variety of measures.
- •
All studies showed vocal improvement after injection medialization.
- •
There were 2 level I studies, both by Hertegard and colleagues, describing the use of hyaluronan versus collagen for injection; their findings suggested better vibratory function and less resorption of hyaluron over time.
- •
The remainder of the studies described injection with fat, collagen, acellular dermis, fascia, Teflon, silicone, and other materials.
- •
The largest proportion of articles described the use of fat injection.
- •
Most did not compare different types of injectables, and longer clinical follow-up was not available.
The multi-institutional retrospective review by Sulica and colleagues gave a sense of current practice patterns:
Used most commonly for awake injections:
- •
Methylcellulose: 35%
- •
Bovine collagen: 28%
- •
Calcium hydroxylapatite: 26%
Used most commonly in the operating room:
- •
Calcium hydroxylapatite: 36%
- •
Methylcellulose: 35%
Though Teflon was previously popular because of its long duration and because it is easy to inject, Teflon injection is associated with giant cell granulomas that persist decades after injection and are challenging to address surgically. Some problems may have been related to technique, and some investigators describe vocal rehabilitation with multiple surgical procedures, but the potential disadvantages render Teflon difficult to support except in rare cases, particularly when other alternatives exist.
Individual surgeon preferences for use of a given injectable may also depend on characteristics of each material, including duration, ease of use, cost, and rheologic properties. Several of the commonly used injectable materials are summarized in Table 1 .
