The author examined 1,362 patients with VFP over 37 years, between October 1976 and December 2013. In this report, the 37 years are divided into three periods of 10 years and one period (the last) of 5 years; the periods are: October 1976 to December 1986, January 1987 to December 1997, October 1998 to December 2008, and January 2009 to December 2013. Data on patients with VFP, who were examined by the author at Ehime University Hospital (Ehime, Japan), were collected between October 1976 and December 1997; data on patients seen at Kumamoto University Hospital (Kumamoto, Japan) were collected from October 1998 to December 2013. Since the author moved from Ehime University to Kumamoto University in October 1998, data from the interval between January and September 1988 were not included. Patients with vocal fold immobility attributable to invasive laryngeal or hypopharyngeal cancer, subluxation of the cricoarytenoid joint, adhesion of the posterior glottis, or rheumatoid arthritis-derived cricoarytenoid joint fixation were excluded.

Figure 2.1 presents a flow chart indicating how the cause of VFP should be investigated. History taking and physical examination, including endoscopy of the head-and-neck region, are essential in this context. Other modalities, including neck ultrasonography, chest radiography, chest computed tomography (CT), and magnetic resonance imaging of the cranial cavity and the skull base to the parapharyngeal space, should be requested when indicated. If such investigations fail to define the cause of VFP, the patient should be followed up for at least 3 months, at which time a thorough examination should be repeated. An idiopathic diagnosis should only be made when the underlying cause of VFP cannot be established at the end of the follow-up period.

Viral infections can cause VFP, but such infections are not easy to confirm. Antiviral antibody titers are not routinely measured, because patients do not visit hospitals immediately after symptom onset. The presence of vesicles in the laryngopharyngeal region on the same side as the VFP may suggest that viral infection is in play. Herpes zoster virus, herpes simplex virus, and other viruses have been reported to cause VFP [5–10]. Thus, VFP patients in idiopathic categories may include those with virus-related VFP.

Table 2.1 shows the numbers of male and female patients seen at the author’s institutions, by decade of patient age. The patients included 775 males (56.9 %) and 587 females (43.1 %), ranging in age from 2 months to 93 years, with a mean age of 60.3 years (61.1 years in males and 59.1 years in females). The numbers of male patients increased from the fifth decade of patient age, peaking in the seventh decade, and the numbers of female patients increased from the fourth decade, peaking in the seventh decade. The numbers of female patients in the second, fourth, and fifth decades of life were greater than those of male patients in the corresponding decades. The female preponderance in these decades of age may be associated with the higher incidence of thyroid disease in females. Statistical analysis was performed using the chi-squared test. A difference was considered significant when the P value was less than 0.05. Unilateral VFP (UVFP) was evident in 1,221 patients (89.6 %), whereas 141 patients (10.4 %) presented with bilateral VFP. Of the 1,221 patients with UVFP, 835 patients (68.4 %) had left-side paralysis and 386 patients (31.6 %) had right-side paralysis; the incidence of UVFP on the left side was twice that on the right. Twenty-three patients were lost to follow-up, and their data were excluded from analysis.

Table 2.2 shows the numbers of patients who developed VFP not associated with and associated with surgery in the four study periods. The numbers of patients who developed VFP after surgery were significantly higher in Groups 2–4 than in Group 1 (P < 0.0001).

In Tables 2.3 and 2.4, patients are classified by cause of VFP; the former table shows the VFP etiologies of 747 patients with diseases unrelated to surgery, and the latter table shows the etiologies of surgery-associated disease in 592 patients. Among patients with surgery-unrelated paralysis, 227 (30.4 %) had upper thorax disorders, including malignant neoplasms of the lung and esophagus. The numbers of patients who developed VFP after diagnosis of tuberculosis fell from eight in Group 1 to zero or one in the other groups. Thyroid cancer was the most frequent cause of VFP among patients with neck lesions, and the incidence of such disease was significantly higher in Group 1 than in Group 3 (P = 0.0400). Idiopathic VFP was evident in 265 of the total of 1,339 patients (19.8 %) and in 35.4 % of the 747 patients in whom VFP was not associated with a surgical procedure.

Most neck surgical causes of VFP were associated with surgery used to treat benign and malignant thyroid tumors. The numbers of VFP patients after thyroid surgery were 24 (36.9 %) of 65 in Group 1, 52 (41.9 %) of 124 in Group 2, 87 (37.2 %) of 234 in Group 3, and 50 (30.0 %) of 169 in Group 4. Thus, the incidence of VFP after thyroid surgery was consistent over the past four periods. The incidence of developing VFP after surgery to treat benign thyroid diseases was significantly lower in Groups 3 and 4 than in Groups 1 and 2 (please note the P values in Table 2.4). The incidence of VFP developing after surgery to treat lung cancer was higher in Group 4 compared to Groups 2 and 3 (the P values were 0.0123 and 0.00854, respectively). The incidence of postesophagectomy VFP was likewise higher in Groups 3 and 4 than in Groups 1 and 2 (the P value for the comparison between Groups 2 and 3 was 0.0363, thus statistically significant). Although statistical significance was not attained, the P values for comparisons between Groups 1 and 3 and Groups 2 and 4 were 0.0952 and 0.0597, respectively. Surgical intervention to treat aortic arch aneurysms was associated with an increased VFP incidence when Groups 2 and 4 were compared (P = 0.0252). In general, surgical treatments for patients suffering from previously inoperable life-threatening diseases have progressed over the years, and it is thus more important than ever to improve postoperative QOL via phonosurgical treatment of severe breathy dysphonia.

Figure 2.2 presents details of phonosurgical treatments for paralytic dysphonia, depending on the purpose of such treatment. Intracordal injections, insertion of fascia, and thyroplasty type I (Type I method) seek to increase the mass of the vocal fold. The type I method also effectively medializes the vocal fold, depending on the size and shape of the inserted material. Arytenoid adduction and various modifications thereof have been reported to move the affected vocal fold to the median position and to correct any vertical mismatch between vocal folds. Furthermore, reconstruction of laryngeal muscle innervation includes direct anastomosis of the severed ends of the recurrent laryngeal nerve (RLN), interpositioning of a free nerve graft between such severed ends, anastomosis between the proximal end of the ansa cervicalis nerve (ACN) and the distal end of the RLN (AC transfer), and implantation of a nerve–muscle pedicle flap (NMP) utilizing the ACN. The purposes of these reinnervation methods are recovery of vocal fold mass and tension and median localization of the affected vocal fold.

In 1911, Brunings first described intracordal injection of paraffin to treat paralytic dysphonia [11]. Unfortunately, adverse effects including tissue migration and granulation developed, and paraffin is no longer used. Many other intracordal materials have been described, but no widely accepted injection material has yet been defined. An ideal injection material should be biologically inert (i.e., incapable of stimulating a foreign body or allergic reaction), be stable in the injected location (i.e., not exhibiting tissue migration), be resistant to resorption (i.e., absorption is minimal after injection), and not inhibit development of a mucosal wave. Both Teflon [12–14] and silicon [15, 16] were used as injection materials after paraffin was abandoned. However, Teflon caused frequent granuloma formation because of foreign body reaction and tended to migrate from the injection site, being excreted into the airway [17–19]. Silicon also exhibited disadvantages, including a foreign body reaction (although less severe than that associated with Teflon) and migration from and diffusion within the tissue [20, 21]. Thus, these two materials are no longer popular.

In the time since Ford and Bless introduced atelocollagen (bovine collagen from which antigenic telopeptides have been removed) as an injection material in 1986 [22], many other materials have been considered. These include autologous fat [23, 24], minced autologous fascia [25, 26], extracellular components of human skin [27, 28], calcium hydroxyl apatite [29–31], and hyaluronic acid [32, 33]. In Japan, autologous fat injection has become popular [34, 35]. Autologous fat cannot cause an allergic reaction, and is not associated with any danger of transmittal of unknown infectious agents. However, injected fat tends to be absorbed quite quickly, and it is difficult to predict how much will be absorbed. To reduce resorption, Tamura et al. [36] included basic fibroblast growth factor with the fat injection. Other methods used to reduce resorption include purification of harvested fat [37–39] and the use of a buccal fat pad rather than subcutaneous abdominal fat [40]. To the best of the author’s knowledge, no ideal injectable material is yet available.

Tsunoda et al. [41–43] devised a method in which a sheet of fascia was inserted between the body and the cover of the vocal fold. Nishiyama et al. [44, 45] developed this method further by creating a roll of fascia to augment the effect thereof. Although the cited authors aimed to medialize the affected vocal fold, the long-term results (12 months after surgery) were not satisfactory, as exemplified by averages of only an 11 s maximum phonation time (MPT) and a 500 mL/s mean airflow rate (MFR) in eight patients with paralytic dysphonia. The injection laryngoplasties and fascia transplantation described above were performed through the laryngeal lumen, and thus, no skin incision was required [46].

The purpose of the type I method is to facilitate glottal closure during phonation by increasing the mass of the affected vocal fold. Payr first developed this idea in 1915 [47], and many modifications have since been proposed [48–51]. These methods are effective in reducing glottal insufficiency, but it is not possible to adjust the medial pressure finely during operation. In 1974, Isshiki et al. [52] proposed the type I method for augmenting the affected vocal fold and shifting it medially by pressing the fold through a window in the thyroid ala set at the level of the fold. Custom-designed silicon blocks were used as insertion materials. Surgery was performed under local anesthesia so that the patient’s voice could be monitored during the operation. The method was used to treat paralytic dysphonia, and it was possible to adjust the pressure intraoperatively by varying the thickness of the silicon block [53].

It is necessary to construct a window at the level of the vocal fold during phonation. Isshiki et al. [52, 53] developed a method for predicting the vertical level of the vocal fold. Misplacement of the window, often cranially to the vocal fold, is one of the most frequent causes of treatment failure. Methods used to confirm correct window positioning are (in addition to precise choice of window location using the Isshiki approach): [52, 53] piercing of the laryngeal lumen with a 30 G needle [54] and pressing the tissue through the window using a lacrimal duct bougie [55] under endoscopic guidance from inside the airway. Isshiki et al. [56–58] adjusted silicon thickness intraoperatively, thereby prolonging the operative time. To address this problem, prefabricated materials are currently available in several sizes. In 1998, McCulloch et al. [59] reported that polytetrafluoroethylene (Gore-Tex^®; PTFE) sheet could serve as an insertion material. PTFE sheet affords several advantages compared to other insertion materials as it is easily adjustable and biologically inert and only a relatively small window is required. However, the adductory force produced upon insertion of a large portion of stiff material cannot be attained by employing folded Gore-Tex^® sheet; the material is too pliable. Figure 2.3 shows the author’s modification of the type I method using a PTFE sheet.

Although the type I method and intracordal injection can reduce the glottal gap, any effect is usually limited to the anterior part of the glottis. Arytenoid adduction (AA) is performed to close the glottal gap between the vocal processes (the posterior glottal gap) during phonation. In 1948, Morrison developed a method by which the affected vocal fold could be moved toward the midline; the arytenoid cartilage was glided medially along the upper border of the cricoid cartilage after separating the two cartilages [60]. The procedure was considered invasive, and it was technically difficult to locate and fix the arytenoid cartilage at the right position. In 1966, Montgomery developed the cricoarytenoid arthrodesis technique, in which the arytenoid was fixed onto the cricoid cartilage using a custom-made pin via a laryngofissure approach [61]. The vocal fold mucosa inevitably became cicatricial after mucosal incision (a consequence of the laryngofissure approach), and the procedure was rather invasive. In 1978, Isshiki et al. [62] developed a method by which the affected vocal fold is rotated at the median position by pulling on the muscular process of the arytenoid cartilage (Fig. 2.4). This method affords several advantages compared to earlier approaches; the laryngeal lumen is not opened (i.e., the vocal fold mucosa is not injured), the operation can be performed under local anesthesia, the patient’s voice can be monitored during surgery, and the arytenoid cartilage is readily fixed at the median position. AA has therefore become popular and is now the treatment of choice for paralytic dysphonia. Several key features of AA are described in the following paragraphs.

The type I method or intracordal injection is usually indicated if the glottal gap during phonation is small, whereas AA is performed to close a larger gap, especially a posterior glottal gap. Because of the easy accessibility of the thyroid ala, the type I method is frequently used to treat paralytic dysphonia. Figure 2.5 shows laryngeal images obtained after AA (left) and the use of the type I method (right) on a single excised larynx. Arrows indicate the location of the tip of the vocal process. A silicon block (the insert in the left picture of Fig. 2.5) was used to push both the membranous and cartilaginous parts of the vocal fold medially. As indicated by the arrows, AA caused adduction of the vocal process, and the membranous part became concave (like a bow). On the other hand, the use of the type I method caused the vocal fold to bulge, but the vocal process was not sufficiently adducted, although the posterior part was pushed medially by a custom-designed silicon block [75]. Therefore, a combination of AA and the type I method is often applied [65, 76–78]. Similarly, AA can also be combined with intracordal injection to enhance medialization of the membranous part [79].