10.1 Introduction

From a neurolaryngologist’s perspective, glottic insufficiency due to lower motor neuron-type injury of the recurrent or vagal nerve is the most frequent and most important application for phonosurgical techniques. There are two main strategies to improve this condition: restitution of vocal fold mobility and closure of the glottic gap by bringing the paralysed fold into a more median position. Theoretically, reinnervation techniques as well as laryngeal pacing are treatment options also for cases of unilateral vocal fold immobility of neurogenic origin. Reinnervation is a complex procedure which is mastered by only few surgeons, therefore lacking reproducibility so far. As a consequence, it is reserved for cases of bilateral paralysis. Laryngeal pacing as an innovative tool is under investigation only for the treatment of bilateral vocal fold palsy for the time being. It may be speculated that there is a role for treating unilateral paresis in the future, once there is convincing evidence of efficiency and efficacy for the device as a modality for bilateral vocal fold paresis, which should be transferable to a significant extent. Both approaches are discussed in detail elsewhere in the book.

Surgical options aiming at closing a glottic gap can be divided in open (thyroplasty, arytenoid adduction) and transoral endoscopic techniques. As a general rule, transoral vocal fold augmentation is working best in smaller gaps of up to 3 mm, while thyroplasty is more suitable for marked glottic insufficiency and, in combination with arytenoid rotation, posterior gaps. A combination of techniques is an option for severe cases, using thyroplasty to close the glottic gap as much as possible and to bring the arytenoid into an appropriate position, followed by injection laryngoplasty to correct a persisting insufficiency in a sense of “fine-tuning” for an optimal result. With indications overlapping both approaches can accomplish convincing results and should be considered as complementary more than competitive [1]. Neither injection laryngoplasty nor thyroplasty should be labelled as “superior” in general compared to each other. Nevertheless, a variety of parameters should be analysed in detail prior to making a choice in order to tailor surgical voice rehabilitation to the specific patient’s needs.

The term “injection laryngoplasty” comprises all techniques aiming at adding bulk to one vocal fold by injecting some kind of augmenting substance in order to correct a glottic insufficiency. Since its first description by Brünings [2] well over 100 years ago, a great variety of different materials have been used for this purpose, with some showing unfavourable results or significant complications [3, 4].

Over the decades, only minor achievements have been made in finding a perfect material for vocal fold augmentation. In recent years, the application of injection laryngoplasty techniques for managing unilateral vocal fold paralysis has regained popularity. Nowadays a number of injectable substances are available, which have been tested in the clinical setting of glottic insufficiency. Basically almost every filler substance having been tested and used in facial plastic surgery has also been utilized in laryngology at some point.

10.2 Application Techniques

In contrast to thyroplasty requiring an external incision, injection laryngoplasty as a transoral endoscopic procedure may be considered as minimally invasive. There are several factors influencing the setting in which the injection can be performed, like surgeon’s and patient’s preference, individual anatomy, exact goal of the procedure and type of material, which may require overcorrection or, in contrast, needs to be injected meticulously to fill the defect precisely. It is the laryngologist’s responsibility to choose from these multiple options, which requires comprehensive knowledge of these technical details as well as personal experience and expertise.

Suspension microlaryngoscopy is the classic approach. By using an operation microscope, the view of the glottis is magnified. The patient can be positioned optimally and is not moving. Exposure can be modified repeatedly and with ample time. Multiple injections in every vocal fold segment are possible, and instruments can be used to adjust the final localization of materials like Vox Implants. Subglottic overfilling (Video 10.2) as complication of particular importance, which can be almost impossible to correct, can still occur in the hand of less experienced surgeons but can be controlled more effectively in comparison to other techniques. For all these reasons, suspension microlaryngoscopy definitively remains the most precise, accurate and versatile way to perform vocal fold injection. This is of paramount importance for materials which do not resorb, since malpositioning cannot be corrected. Typically, these non-resorbable substances need to be deposited in the lateral paraglottic space, which can be controlled effectively under suspension microlaryngoscopy only. It seems to be safe to say that all materials with these characteristics have to be injected under suspension microlaryngoscopy without exception. On the other hand, also for resorbable materials, the highest standards of precision and accuracy should be respected for medical purposes. Even though due to dissipation over time malpositioning has less drastic consequences, it still may seem odd that a less meticulous approach may be sufficient for these substances, as it is put by some authors [5].

Main argument against the use of suspension microlaryngoscopy is the lack of feedback from the patient, who cannot phonate during the intervention. Hence, there is no functional control available guiding the surgeon through the injection process. While admittedly there is some truth to this objection, it should not be overestimated. Basically all resorbable substances are requiring substantial overcorrection, which is questioning the concept of functional feedback. For the reasons discussed above, possible risks clearly outweigh the benefits of office-based procedures whenever using non-resorbable materials.

Depending on healthcare peculiarities in different countries, there may be contrasting incentives to avoid or to prefer interventions requiring general anaesthesia. In many instances, this may seem to be a biasing factor, which is not discussed openly, but has a potential to influence surgeon’s preference. Without any doubt the need for general anaesthesia is increasing total costs, requires a more sophisticated setup and is more time-consuming [6].

Ventilation in suspension microlaryngoscopy can be maintained in several ways: infraglottic jet ventilation using a small-bore catheter is offering a good compromise between technical effort and undistorted view and is therefore the option being used most frequently, probably. More sophisticated jet ventilation techniques require special equipment: the twin-stream device (Carl Reiner, Austria) delivers a combination of high- and low-frequency jet streams over customized laryngoscopes allowing for a completely unobstructed view. In combination with air humidification, interventions of more than 1 h length with unmatched exposure of the glottis are possible.

More recently, a completely tubeless system of apnoea oxygenation has become available (Optiflow, Fisher & Paykel Healthcare Ltd., New Zealand), which is delivering high-flow oxygen at up to 70 l/min via a nasal mask. With the laryngoscope in place, apnoeic periods of well over 30 min can be accomplished (personal experience). CO2 retention is usually the limiting factor in this setting.

If none of these options should be available, an endotracheal ventilation tube can be used as well. It should be chosen as small as possible to remain in the posterior aspect of the glottis without blocking the view on the vocal folds. Even for male adults, a short-term ventilation using an ET tube of not more than 5.5 mm outer diameter is usually possible. As an alternative the laryngoscope can be positioned in apnoea first, and then the ET tube is positioned through the channel of the laryngoscope into the trachea. After preoxygenation usually an apnoeic period of sufficient length for injection laryngoplasty is available before reinserting the ET tube becomes necessary.

Office-based transoral techniques in the sitting, awake patient have become more and more common in the recent years. This development has been fuelled primarily by technical innovations like chip-on-the-tip flexible endoscopes allowing for excellent visualization of the larynx with less discomfort compared to conventional indirect laryngoscopy. Obviously there is a benefit in having a patient being able to phonate during the procedure, thus giving an immediate functional feedback. This may allow for more precise placement of smaller amounts of injection material [7]. Voice outcomes, as measured by standardized patient-based voice surveys, seem to be similar to injection performed under general anaesthesia when performed by an experienced laryngologist [8, 9]. Even in patients tolerating the procedure well, some movement is inevitable, which may lead to inaccurate substance delivery. There are a significant number of patients in which any awake endolaryngeal intervention is not possible. This is the main limitation of the concept. The technical challenge is remarkable and requires laryngological expertise and training. Actual time requiring the presence of the surgeon may well be longer than for suspension microlaryngoscopy in general anaesthesia. In summary, office-based injection laryngoplasty deserves an important role but is not right away superior to conventional suspension microlaryngoscopy. Many laryngologists including the author still find there are more and better arguments for performing injection laryngoplasty in suspension microlaryngoscopy than office-based in the awake patient. With general anaesthesia at nowadays’ standards being a safe routine approach very rarely causing any complications, it makes sense to leave the final decision to the patient. In many centres, however, there is only one approach established in clinical routine.

There are several descriptions on how to perform injection laryngoplasty in an awake patient [10,11,12]. Typically, surface anaesthesia is applied to the hypopharynx first. In a second step, a curved applicator is used to target the anaesthetic agent under direct vision to the epiglottis, the false vocal folds and the glottis. Sufficient anaesthesia of the laryngeal aspect of the epiglottis is necessary to allow for contact of the injection device in order to retract the epiglottis [13]. Uncritical overuse of anaesthetic agents in the whole upper respiratory tract is contraindicated, since it may cause oversecretion as well as aspiration and pooling of secretions, adding significantly to patient’s discomfort.

In patients not lending themselves for the transoral approach, but still willing to avoid general anaesthesia, transcutaneous techniques may be a good alternative. This approach seems to be of increasing popularity, since it is possible even in the presence of a significant gag reflex. Therefore, patient selection is less critical. The procedure is described as being painless in general [5]. Major disadvantage is limited access to the vocal folds. Several anatomical trajectories have been described to direct the injection needle, either using a straight pathway through the thyroid cartilage or a more angled approach from below (through the cricothyroid membrane) or above (through the preepiglottic space) the level of the vocal folds [14].

10.3 Injection Materials

A basic distinction can be made between resorbable and non-resorbable substances, with the former having a temporary effect, though with considerable differences in longevity, and the latter providing a permanent effect. In the heterogeneous and bigger group of resorbable materials, a further classification between biological origins will be made. For the smaller group of non-resorbable substances, the term “injectable implants” has been coined; they are of non-biological origin.

10.4 Resorbable Substances

10.4.1 Xenogeneic Origin

These substances are derived from animals, with bovine collagen being the most popular due to its long record in facial plastic surgery. The first description on collagen use for vocal fold augmentation is dating back to 1986 [15, 16]. The variety of preparations is mainly differing in concentration of the solution that comes in prefilled syringes ready for injection. Collagen is providing a temporary augmentation between 3 and 12months, according to several authors. With about 3% of human beings carrying preformed antibodies against collagen, hypersensitivity skin testing is considered mandatory prior to its use. Collagen is a fluid of low viscosity, making application possible via small-bore injection devices with mild pressure. For these reasons it is an obvious choice for office-based interventions. Foreign body reactions seem to be minimal to non-existent. Although FDA approved, there is no formal approval in Europe for using collagen in the human larynx, but it has been probably the most widely used injection material for many years. For reasons not communicated, bovine collagen has not been available for medical use any more since a couple of years now.

Similar to collagen in many aspects, hyaluronic acid is a polysaccharide, which is not species specific and does not elicit humoral or cell-mediated immune response. It can be derived from different animals or can be produced by biological engineering techniques. Therefore, hypersensitivity skin testing is not necessary. Viscoelastic properties could be shown to be quite similar to vocal fold lamina propria [17]. This is making hyaluronic acid the material of choice for aligning vocal fold irregularities like in sulcus vocalis or vocal fold scarring. An optimal indication for hyaluronic acid seems to be a unilateral paresis with supposedly good prognosis in a patient wishing for transient voice improvement. Injection may and should be superficial within Reinke’s space. Deep injection into the paraglottic space seems to increase speed of resorption, leading to augmentation effects of some weeks only. This can be used intentionally, however, to offer a trial intervention to patients in doubt without any significant risk profile (“test-drive”). There is an increasing amount of publications on the use of hyaluronic acid in the human larynx [18,19,20,21,22,23,24,25].

10.4.2 Allogenic Origin

An alternative to bovine collagen is human collagen, made of cadaveric dermal tissue. It has been FDA approved for skin augmentation, but not injection laryngoplasty, so it is used off-label in the USA. It is not distributed in Europe. During the production process, all cellular components are removed without altering the collagen and elastin matrix. The final material is a micronized acellular compound which is freeze-dried into powder form (Life Cell Corp., Branchburgh, NJ, USA), which needs to be reconstituted for injection. So far, no infectious transmission has been documented. Viscosity and injection characteristics are similar to bovine collagen. Information on longevity after injection are quite contradictory: Pearl [26] reported stable parameters like laryngeal airflow and subjective voice assessment (VHI) at 1 and 3 months, although some absorption was noted on follow-up laryngoscopy. In sharp contrast Karpenko et al. report no significant improvement at 1 month in patient perceptual voice quality and aerodynamic measurements [27]. In addition, there are first reports on substance-associated complications like locoregional migration [28]. As a consequence, Cymetra seems to have lost ground to other materials in the USA over the recent years.

10.4.3 Autogenic Origin

In the early 1990s of the last century, autogenic fat has been described as material for augmentation [29]. Harvesting of the graft material is being performed at the same time as injection laryngoplasty, usually by needle aspiration or through a periumbilical incision. Survival of adipocytes after transplantation is most likely depending on a correct harvesting procedure [30]. The obvious advantages of excellent biocompatibility, simple acquisition and low cost have made this approach rapidly popular throughout the world. Before the turn of the millennium, autogenic fat has probably been the substance used most frequently in many countries. Information on extent and time course of resorption is quite different. By using magnetic resonance tomography, permanent longevity as well as complete resorption could be found [31, 32]. The unpredictable rate and degree of resorption are the main disadvantage of fat injection, limiting significantly the predictability of long-term voice outcomes. Overcorrection is recommended in general, but it remains unclear at large to which extent. Repeat injection procedures may be necessary to maintain adequate vocal fold bulk. Lacking reproducibility in combination with the advent of new materials avoiding these shortcomings has continuously diminished the role of fat as injection material [33].

Originally described by a Finnish group [34, 35], fascia lata is another source of autogenic material. The tissue harvested is minced at the table by the surgeon in order to form an injectable preparation, requiring an injection cannula with great needle bore. Initial reports have suggested greater longevity compared to fat, which could not be reproduced by other authors [36]. Actual use in clinical routine seems to be limited to a few centres only.

10.5 Non-resorbable Substances

Polytetrafluoroethylene, better known as Teflon, has been the mainstay of treatment for vocal fold augmentation between 1960 and 1980. Apparently, it is still available in the USA, although it has fallen out of favour because of significant substance-associated complications. Teflon has a tendency to elicit foreign body reactions with occasional formation of hardly resectable and potentially airway-obstructing granuloma, even if applicated in flawless technique [3]. In addition, migration to distant organs could be shown [37]. In Europe, Teflon is being considered as obsolete for these reasons.

Polydimethylsiloxane elastomers (PDMS), introduced in 1989, have been used in plastic surgery for the treatment of soft tissue deficiencies [38] as well as in paediatric urology for the treatment of vesicoureteral reflux and in adults for the treatment of male and female urinary stress incontinence [39]. First systematic use in the human larynx has been published in 2000 [40].

PDMS particles used for injection purposes consist of textured elastomers, mean particle size being 200 μm approximately, with a minimum size of 100 μm. Earlier studies [41] suggest a particle size of less than 65 μm to be critical for the risk of local and distant migration via lymphatic pathways. Particle size combined with surface texture prevents mechanical dislodgement or migration, thus warranting greater anchorage and stability [42, 43]. PDMS elastomers have been in use for medical applications like insulation for pacemaker leads, catheters, hydrocephalic shunts and others for many years. It should be noted that controversies regarding the safety of silicone implants were aimed exclusively at silicone oil and gel, not silicone elastomers [44]. In contrast to silicone oil and gel, an elastomer is a rubbery solid made of fully polymerized PDMS molecules. The textured implants are held in place at the implantation site when host fibroblasts subsequently deposit collagen around the particles. PDMS are commercially available under the brand name Vox Implants™ (manufactured by Bioplasty BV, Hofkamp 2, 6161 DC Geleen, the Netherlands, distributed by Medtronic Europe, Route du Molliau, CH-1131 Tolochenaz, Switzerland). It is a thick yellow paste shipped in 1.0 cc polypropylene syringes that becomes tacky on exposure to air. In 2001 Vox Implants™ have been approved for injection laryngoplasty in humans within all member states of the European Union. In 1991, the FDA declared the use of injectable silicone illegal. Although this ban was aimed at silicone oil, a completely different material sharing almost no characteristics with PDMS particles, Vox Implants have no FDA approval so far.

For injection a malleable needle with sufficient needle bore is supplied with the injection material as a set to make sure that also bigger particles of up to 500 μ size are passing through. A considerable pressure needs to be applied during injection, since the material has a high viscosity. To ensure high accuracy and perfect control during the injection process, the use of an administration gun incorporating a ratchet mechanism allowing for precise placement of the microimplants in portions of 0.03 cc is mandatory. The correct application site for PDMS particles is deep lateral to the thyroarytaenoid muscle (Figs. 10.1 and 10.2). This is basically the region in which thyroplasty implants are localized; hence, the term “injectable microimplants” seems to be justified. It is important to keep in mind that Vox Implants are put into motion as chain of particles following the principle of elastic coupling. As a result it takes some seconds for the row of particles to start actually moving as well as to come to a complete halt, making it necessary to keep the injection needle steady at the puncture site for about 10 s after having released the pressure on the syringe. Typically, a total volume of 0.5–0.7 cc in 2–3 injections is sufficient to augment a paralysed vocal fold (Video 10.1). Subglottic overfilling needs to be avoided carefully (Video 10.2). Building up of a sufficient bulk is usually well possible (Figs. 10.3 and 10.4). Injecting too superficial into Reinke’s space leads to stiffening of the vocal fold due to the high viscoelasticity and the corpuscular character of the material. Positioning of the material can be difficult with injecting alone, but immediate correction is well possible: using a closed cup forceps the mass of injected PDMS can be distributed in the paraglottic space, allowing for precise remodelling of the augmented vocal fold (Video 10.3). Taking into account these characteristics, it may seem self-explaining why suspension microlaryngoscopy is highly advocated for PDMS injection into the human larynx.

Fig. 10.1
figure 1

The optimal injection site for Vox Implants is the deep paraglottic space, between the thyroarytenoid muscle and the inner perichondrium of the thyroid cartilage

Full size image
Fig. 10.2
figure 2

Microlaryngoscopic view of the augmentation process (Vox Implants)

Full size image
Fig. 10.3
figure 3

Left recurrent nerve palsy with glottic gap, preoperative videostroboscopic view

Full size image
Fig. 10.4
figure 4

Same patient as in Fig. 10.3, 6 weeks post injection: note complete glottic closure

Full size image

Animal studies show an excellent biocompatibility of PDMS particles [45]. Hypersensitivity is not an issue for subcutaneously located PDMS elastomers, neither in animals nor in humans [46], so skin testing is not required prior to injection. Particle size prevents the material from being phagocytosed. Hence, it does not serve as an antigen.

Several studies show favourable results in terms of voice restoration (Videos 10.4, 10.5, 10.6, and 10.7) similar to thyroplasty [1, 47,48,49,50]. PDMS particles provide a long-term augmentation with documented follow-up of 6 years in average [51]. Substance-associated complications seem to be a rare exception and can be difficult to separate from adverse findings as a consequence of malpositioning, i.e. by injecting to superficial, which may happen more frequently. As a last resort, surgical removal of displaced PDMS particles is technically feasible to a certain degree, though a marked stiffness of the erroneously injected subepithelial space cannot be restored completely [52, 53].

In the group of permanent, non-resorbable substances, longevity, safety and efficacy in terms of voice improvement are best documented for PDMS particles. Therefore, Vox Implants™ are most suitable for deep injection laryngoplasty addressing mild to moderate glottic insufficiency in unilateral vocal fold paralysis. Thus, PDMS particles are offering a true alternative to open thyroplasty for patients preferring an endoscopic approach.

Calcium hydroxylapatite (CaHA) is a calcium phosphorus compound found as basic component in the bone and teeth. There is a commercially available material consisting of CaHA microspheres and a carrier gel marketed under the brand name Radiesse that can be injected through a fine needle (25–27 gauge). Once injected, the carrier gel is replaced by fibrosis with the CaHA microspheres remaining in place.

The substance has been studied in both animal and human studies. In an in vivo canine vocal fold model, it has been shown that CaHA injection provided adequate medialization of the canine vocal fold up to 12 months follow-up without migration or resorption and a giant cell reaction without appreciable chronic inflammation [54]. A recent multi-institutional clinical trial revealed excellent results of 80% improvement at 12-month follow-up Rosen 2009 [55]. It is intended as a permanent implant, but resorption occurs to some degree. There are data [56] on longevity showing a loss of benefit in 64% of patients, with an average length of benefit of 18, 6 months (ranging from 8 to 36 months). Therefore, Radiesse is more appropriately labelled as slowly resorbable than non-resorbable. In many aspects similar to collagen, CaHA can be used for the same indications; also the technical process of injection is basically identical. Versatility and ease of use have made it a very popular substance for injection laryngoplasty in the office as well as in the operating room.

There is FDA approval for potentially long-term vocal fold injection. Nevertheless, some substance-associated complications have been described [57, 58]. Voice results seem to be well comparable to those accomplished using thyroplasty type I [59].