Bacterial Cellulose as Laryngeal Medialization Material: An Experimental Study *Flavia Coelho de Souza, †Henrique Olival-Costa, *Leonardo da Silva, ‡Paulo A. Pontes, and *,§Carmen Lu´cia Penteado Lancellotti, *yzxSa˜o Paulo, Brazil Summary: Hypothesis. The use of a material made of bacterial cellulose with the aim of obtaining vocal fold medialization has not hitherto been fully investigated. Although the material has been tested in other animal models, the evaluation did not include the larynx; hence, situations, such as tissue reaction, material absorption, and extrusion, need to be addressed to evaluate its usefulness as a material for laryngeal reconstruction. Objective. To evaluate the medialization, tissue response, and healing of rabbit vocal folds, after the implantation of a membrane of bacterial cellulose. Study Design. Experimental study. Methods. A total of 32 rabbits were used, two of which were used to check out the adequacy of the implant location. The animals were followed for 4 months and grouped according to follow-up times of 2, 4, and 16 weeks. All test animals received an implant of bacterial cellulose in one vocal fold and the injection of distilled water in the other, both performed by videoendoscopic cervicotomy. At the end of the follow-up, the presence of inflammatory and medial displacement was evaluated. Results. No statistically significant difference in the inflammatory parameters between the study and control vocal folds or among follow-up times was found. All animals receiving cellulose presented medial displacement of vocal folds, and all retained this material at the implant site up to study endpoint. Conclusion. Bacterial cellulose is a useful material for laryngeal medialization, showing no signs of rejection or absorption. Key Words: Implant–Cellulose–Vocal fold–Rabbits. INTRODUCTION Surgical treatment of glottic incompetence caused by paralysis, atrophy, or loss of tissue has as its goals to medially displace the affected vocal fold to enable glottic closure. The displacement can be accomplished using substances inserted lateral to the thyroarytenoid muscle, either by injection or by laryngotomy and material implantation.1 There are also techniques in which the material is introduced directly into the Reinke space, increasing its volume and closing the gap.2 Since Brunings3 first described the method of narrowing of the laryngeal gap using paraffin injections in 1911, an array of materials have been used, with Teflon (Polytef Paste Coloplast Manufacturing US/LLC, Minneapolis, MN)4–7 being the most popular for a long time. Teflon has fallen into disfavor because of concerns about the possible tissue reaction with formation of granulomas, encapsulation, and extrusion. More recently, biological or semibiological materials, such as collagen,8,9 autologous fat,10–12 fascia,13,14 hyaluronic acid,15 and Alloderm (Life Cell, Branchburg, NJ)16 have been tried in a bid to change the physiological vibration of the vocal folds. Possible complications of injection procedures include over-infiltration, immune reaction to the injected substance,
Accepted for publication July 9, 2010. From the *Faculty of Medical Sciences, Doctor of Health Science, Santa Casa de Sa˜o Paulo (FCMSCSP), Sa˜o Paulo, Brazil; yDepartment of Otorhinolaryngology, FCMSCSP, Sa˜o Paulo, Brazil; zDepartment of Otorhinolaryngology, UNIFESP, Sa˜o Paulo, Brazil; and the xDepartment of Pathologic Sciences, FCMSCSP, Sa˜o Paulo, Brazil. Address correspondence and reprint requests to Henrique Olival-Costa, Department of Otorhinolaryngology, Santa Casa Medical School, FCMSCSP, Sa˜o Paulo 01220-020, Sa˜o Paulo, Brazil. E-mail:
[email protected] Journal of Voice, Vol. 25, No. 6, pp. 765-769 0892-1997/$36.00 Ó 2011 The Voice Foundation doi:10.1016/j.jvoice.2010.07.005
granuloma formation, rapid absorption of the material, and migration or extrusion of the injected material.17 From the biological perspective, the ideal implant would be one that integrates and incorporates itself with the surrounding tissues, while adjusting to the form and volume necessary for optimal surgical outcome. Numbering among the commercially available materials offering the aforementioned characteristics is bacterial cellulose obtained from the bacteria Gluconacetobacter hansenii. Bacterial cellulose possesses a fibrillar nanostructure. It is a linear polymer of glucose (C6H10O5)n, with n ranging from 500 to 5000 and is the most predominant polymer found in nature. It is water insoluble and is a polysaccharide excreted extracellularly in the form of long nonaggregated nanofibers by many bacteria, including the bacteria G. hansenii, formerly known as Acetobacter xylinum.18–24 The cellulose has been used successfully as a dressing in bedsores and skin ulcers, burns, abrasions, and in donor skin areas. It has also been used as a substitute for meninges, dorsum of the nose, tympanic membrane, as a lining for intravascular prostheses, and in the trachea to avoid circumferential stenosis.25–28 Although the material has been tested in other animal models, the evaluation did not included the larynx; hence, situations, such as tissue reaction, material absorption, and extrusion, need to be addressed to evaluate its usefulness as a material for laryngeal reconstruction. The aim of this study was to evaluate the medialization, inflammatory response, and healing of rabbit vocal folds after implantation of a membrane of bacterial cellulose. METHODS After approval by the Ethic Committee of Experimental Research in animals (number 06/07), the rabbits were housed
766
Journal of Voice, Vol. 25, No. 6, 2011
in individual cages, with free access to water and standard commercial feed, with a preoperative water withdrawal 4 hours before surgery and overseen by a veterinarian. Surgical procedures were carried out according to the norms established by the ethical experimentation principles of the Brazilian Society for Science in Laboratory Animals. The number of animals used followed the recommendation of the National Institute of Health (http://grants.nih.gov/grants/olaw/ faqs.htm), which considers a sample size of 6–10 animals per group to be sufficient for a preliminary pilot study. Selection and sample size A total of 30 6-month-old New Zealand white male rabbits, weighing between 2400 and 2800 g, were used. Rabbits were chosen for study, because their laryngeal structure is similar to humans. They are also easier to handle during follow-up than other animals. The rabbits were followed for a maximum of 4 months and grouped according to a range of follow-up periods to detect changes in any of the parameters studied over different time points. Volumes of the substances inserted were 0.3 cc of distilled water and a membrane of 0.25 mm3 of bacterial cellulose (Bionext, Colombia, Brazil). Animals were divided into three groups and allocated to receive 14, 28, or 120 days of follow-up. Two animals were used to assure that the surgical technique would accomplish what was intended. The morphological analysis after the extraction of the larynx showed that both the animals had the cellulose implanted lateral to the thyroarytenoid muscle and medial to the thyroid lamina, the way we wished; hence, the method was considered adequately done. Surgical procedure Surgical procedures were performed using intramuscular anesthesia. The animals were placed in supine position, with spontaneous respiration using an open oxygen mask. A tracheotomy was performed in the anterior cervical region. A transverse incision measuring 0.5 cm was made in midline at the level of the cricoid bone. Subsequently, a subcutaneous pouch was established by blunt dissection. A 11-mm trocar was placed in the subcutaneous space. With the trocar in position, the area was insufflated with CO2 at a maximum pressure of 0.5 mm Hg. An optic of 10 mm and 30 was placed in the trocar, and another two openings were made on each side of the first, each measuring 0.2 mm, located 2 cm laterally and cranially to the initial incision (Figure 1). Two 5-mm trocars were placed. A Merilan forceps was introduced in the left trocar and scissors in the right trocar. The subcutaneous tissue dissection was completed, and the larynx was exposed. Once the cricoid and thyroid cartilages were exposed, scissors were used to make a 2-mm opening in the crycothyroid membrane and a 2-mm opening to the left of the medial line, extending this longitudinally through the thyroid cartilage. Using this opening, we created a tunnel in the longitudinal direction using forceps. The endolaryngeal musculature was exposed and dissected along its internal subperichondrial plane. The membrane of cellulose measuring 0.25 cm3 was inserted lateral to the vocal muscle (Figure 2). A 30 3 8-mm needle was used to inject 0.3 cc of distilled water at the
FIGURE 1. The endoscopic view of the larynx. same position of the right side of the thyroid cartilage, as a control. All rabbits received clindamycin 0.1 mL/kg and sodium dipyrone 0.1 mL/kg as analgesia in the postoperative period. At the end of the follow-up periods, the animals were once again anesthetized and received intravenous sodium thiopental (dose 40 mg/kg) and then 19.1% potassium chlorate (KCl) for euthanasia (Figures 1 and 2). After euthanasia, a longitudinal opening was made in the cervical region, and a total conventional laryngectomy with removal of the superior one-third of the trachea was done to obtain the study specimen, which was examined histopathologically under microscope with hematoxylin-eosin (HE) staining. The nature of the experimental animals we used avoided the possibility of laryngeal visualization through laryngoscopy, because the larynx is positioned in the rhinopharynx. Hence, we decided the adequacy of the medialization after removal of the larynx under direct visualization of the vocal folds. The specimens were named as with or without medial displacement compared with the contralateral vocal fold. The tissues were analyzed at the Santa Casa Pathology Department, and the tissues were included in paraffin and cut into 6-mm samples. The HE method was applied. Using optical visualization, the inflammatory reaction was compared among groups. The reactions were classified according to a severity scale: none, mild, moderate, and severe inflammation. The scores used to determine the degree of inflammation were based on previous studies.29,30 Biocompatibility was determined in accordance with the evaluation procedures described by International Organization for Standardization (ISO) 10993-6.31 The qualitative analysis of the histological sections was done considering the inflammation level and absence or presence and predominance of cell’s types. The images were submitted to microscopy (403) (Carl Zeiss AG, Jena, Germany) regarding polymorphonuclear (PMN) inflammatory infiltrate; blood vessels; hyperemia and edema (acute inflammation features); and mononuclear inflammatory infiltrate (lymphocytes, plasmocytes, and macrophages). The presence of fibroblasts and less vascularization and edema represented a chronic inflammatory reaction. Another feature analyzed was the presence of newly formed blood vessels,
Flavia Coelho de Souza, et al
Bacterial Cellulose as Laryngeal Medialization Material
767
FIGURE 2. Schematic drawing of the steps of cellulose implantation showing (A) the scissors opening the thyroid cartilage under direct endoscopic view; (B) the forceps grabbing the cellulose through the thyroid window; and (C) the cellulose positioned lateral to the thyroarytenoid muscle.
which suggests a healing process and the presence of giant cells and granulomas. Two independent and blind observers, professors of Pathology at our school of Medicine, with experience in tissue inflammation, were asked to decide over the aforementioned inflammatory parameters. Only the results with unanimity were taken. When there was no agreement, a third observer should be called as Minerva vote. Individual values and the mean of the results were used in the statistical analysis. The degree of healing and inflammation was evaluated in each specimen. The inflammatory parameters examined were vascular congestion (new capillaries and local vascular beds, excessive dilatation of existing vessels, and concentration with agglutination of red blood cells in their interior); cellular exudates (presence of dead phagocytes or microorganisms within phagocytes, PMN cells, monocytes, lymphocytes, and plasmocytes); and edema (leakage from opening of endothelial junctions). Evidence of angiogenesis and fibrovascular proliferation was considered to be healing by secondary intention. Each of the inflammatory parameters was categorized as follows—inflammatory transudate: absent (0); discrete (1)— extravascular edema; moderate (2)—presence of few PMN cells, namely, neutrophils, macrophages, lymphocytes, and
plasma cells intense; and (3)—presence of predominantly PMN cells, macrophages, lymphocytes, and plasma cells. Vascular congestion: absent (0); discrete (1)—vessel dilation; moderate (2)—agglutination of red blood cells; and intense (3)—opening of new capillaries and local vascular beds. Cellular exudates: absent (0) and present (1). Angiogenesis and connective tissue proliferation: absent (0), discrete (1), and moderate (2). The integrity of the cellulose plug was classified as integral (1), partially fragmented (2), and fragmented (3). Additionally, we classified the position of cellulose plug in the anatomic specimen as anterior, medial, and posterior, and the medial displacement caused at the free border of the vocal fold was classified as absent (0)—no bulging at the free border of the vocal fold—and present (1)—when there was a curving at the free border of the vocal fold. Statistical evaluation The Wilcoxon (Mann-Whitney) nonparametric test was used for the noncontinuous variables, such as inflammation, healing, and fibrosis. The Jonckheere test was used to evaluate differences in the following parameters: inflammation, vascular congestion, connective tissue proliferation, and cellulose implant integrity, according to the different follow-up periods.
FIGURE 3. Histological examinations of the vocal folds showing the relationship of the cellulose with the subjacent tissue, epithelium, and musculature. In (A), we see the arrows pointing to the thyroid cartilage opening. The star shows the inflammatory reaction after 14 days of follow-up. In (B), we see the inflammation surrendering the cellulose, bringing the possibility of capsule formation, after 28 days. In (C), we see the lack of inflammatory reaction after 120 days, without encapsulation of the cellulose. BC, bacterial cellulose; TC, thyroid cartilage; TA, thyroarytenoid muscle—hematoxylin-eosin, 403.
768 RESULTS Analysis of results for both vocal folds (study and control) revealed no statistically significant difference for any of the inflammatory parameters studied. According to our findings, the inflammatory response was similar for the vocal folds that received the implant (study) and those that did not (water control). This similarity was maintained throughout the study, where inflammation remained constant for all periods. We found no change in the integrity of the cellulose implant in the three periods of the study, including the 4-month time period. All implants were present at time of euthanasia. The position of the implant in all cases was virtually the same; only three animals had cellulose in the medial third of the vocal folds, with the rest presenting cellulose in the anterior part. Medial displacement was evaluated visually, and every time, there was a difference of the implanted side from the midline compared with the other side the medial displacement was considered to have occurred. It was so in 20 animals, which was considered a satisfactory result (Figure 3). The statistical analysis showed no differences among the vocal folds that received cellulose and distilled water (transudate P ¼ 0.082944, vascular congestion P ¼ 0.158779, and angiogenesis and connective tissue proliferation P ¼ 0.371757) in overall and according to the follow-up period (transudate P ¼ 0.6377340023, vascular congestion P ¼ 1, and angiogenesis and connective tissue proliferation P ¼ 0.06976836568). The integrity of the cellulose did not change after implantation (P ¼ 0.13566609). DISCUSSION The use of biomaterials in procedures to augment the volume of the vocal folds and cause their medial displacement is a common procedure in laryngology today.6,15–17 However, the ideal material, which should be one that causes the least inflammation at the site, integrates well with the surrounding tissues, and does not degrade over time, is yet to be found. Based on the studies on the physiopathology of the tissue response to biomaterials,28,32,33 an initial acute inflammatory process around the implant during the first few hours or days is expected. This inflammation is characterized by exudation of liquid and plasma proteins and migration of lymphocytes. The distribution and number of these cells can be used as an indicator of the intensity of the inflammatory reaction. This process decreases over time, giving way to chronic inflammation, which is characterized by the formation of a thick fibrous capsule, granulation, and rupture of tissue. This may lead to abscess or fistula formation, causing implant extrusion and the possible occurrence of neoplastic alterations. From the beginning of the planning of our study, we did not think that the immunohistochemistry (IHC) would be essential to our evaluation, because the inflammatory process that could result if the cellulose eventually turned out to be not biocompatible would be a foreign-body inflammatory reaction. IHC is a method of molecular analysis of tissues, which aims to identify molecular characteristics of a disease. Among its many applications are the diagnoses of inflammatory diseases,
Journal of Voice, Vol. 25, No. 6, 2011
infectious diseases, and cancer. The technique is also very important to determine the predictive and prognostic factors in cancer. Thus, the technique could, perhaps, offer a better documentation of the process but would not give any more objective data, and its absence does not invalidate our results in our understanding. The biocompatibility of the implanted material can be evaluated by morphological methods (analysis of the connective tissue, collagen, and glycosaminoglycans around the tissue). These results depend on the physical properties of the material, such as porosity (also the influence of inflammatory cells), surface texture consistency, and chemical properties. The concept of biocompatibility is based on the interaction of the material to be implanted and the biological environment of the recipient. Failure of biocompatibility is generally shown by a breakdown in the properties of material, formation of a capsule around the implant, or an unsatisfactory biological response. Thus, the principal aspect of biocompatibility is the local tissue response.33 Interestingly, we could not find any capsule around the cellulose, and furthermore, there was the penetration of the cellulose into the neighboring tissue, which may signify a sort of incorporation of the implant. Cellulose produced by the bacteria G hansenii is highly pure and has a homogenous composition and an arrangement of fibers, which is very similar to the collagen of the superficial parts of the vocal folds.18 These characteristics make this material almost ideal for use in volumetric augmentation procedures involving the vocal folds. In our study, no statistically significant difference was found between the study and control groups in terms of acute inflammatory parameters. The slight differences that occurred over time also did not reach statistical significance. The lack of statistically significant differences shows that the material is relatively stable from the second week after surgery. The results of our study show that cellulose is a useful material in medial displacement procedures of the vocal folds, as it causes minimal inflammatory reaction and does not extrude. Although there was an initial loss of integrity in the implanted material, it remained in place for 4 months with no major modifications, suggesting that it is stable over a long period.
CONCLUSION Bacterial cellulose produced by G. hansenii does not cause a significant foreign-body reaction when placed in the larynx of rabbits. The membrane produced medial displacement and remained stable throughout 4 months of follow-up.
REFERENCES 1. Rontal E, Rontal M. Vocal cord injection techniques. Otolaryngol Clin North Am. 1991;24:1141–1149. 2. Arnold GE. Vocal rehabilitation of paralytic dysphonia: IX. Technique of intracordal injection. Arch Otolaryngol. 1962;76:358–368. 3. Brunings W. A new treatment method of the laryngeal paralysis. Verh Dtsch Laryngol. 1911;18:93–151. 4. Koufman JA. Laryngoplasty for vocal cord medialization: an alternative to Teflon. Laryngoscope. 1986;96:726–731.
Flavia Coelho de Souza, et al
Bacterial Cellulose as Laryngeal Medialization Material
5. Alessi DM, Berke GS, Trapp TK, Gerratt BR, Hanson DG. An accurate method of Teflon injection using functional phonosurgery. Arch Otolaryngol Head Neck Surg. 1988;114:1321–1323. 6. Trapp TK, Berke GS, Bell TS, Hanson DG, Ward PH. Effect of vocal fold augmentation on laryngeal vibration in simulated recurrent laryngeal nerve paralysis: a study of Teflon and Phonogel. Ann Otol Rhinol Laryngol. 1989; 98:220–227. 7. Heman-Ackah YD. How I do it: miniplate reconstruction of the lateral thyroid lamina: one-stage restoration of voice after teflon granuloma resection. J Voice. 2005;19:504–509. 8. Belafsky PC, Postma GN. Vocal fold augmentation with calcium hydroxylapatite. Otolaryngol Head Neck Surg. 2004;131:351–354. 9. Rosen CA, Thekdi AA. Vocal fold augmentation with injectable calcium hydroxylapatite: short-term results. J Voice. 2004;18:387–391. 10. Sato K, Umeno H, Nakashima T. Histological investigation of liposuctioned fat for injection laryngoplasty. Am J Otolaryngol. 2005;26:219–225. 11. Sasai H, Watanabe Y, Muta H, Yoshida J, Hayashi I, Ogawa M, Kubo T. Long-term histological outcomes of injected autologous fat into human vocal folds after secondary laryngectomy. Otolaryngol Head Neck Surg. 2005;132:685–688. 12. Cantarella G, Mazzola RF, Domenichini E, Arnone F, Maraschi B. Vocal fold augmentation by autologous fat injection with lipostructure procedure. Otolaryngol Head Neck Surg. 2005;132:239–243. 13. Tamura E, Kitahara S, Kohno N, Ogura M, Hiroi S. Use of freeze-dried autologous fascia to augment the vocal fold: an experimental study in dogs. Acta Otolaryngol. 2002;122:537–540. 14. Reijonen P, Lehikoinen-Soderlund S, Rihkanen H. Results of fascial augmentation in unilateral vocal fold paralysis. Ann Otol Rhinol Laryngol. 2002;111:523–529. 15. Hallen L, Dahlqvist A, Laurent C. Dextranomeres in hyaluronan (DiHA): a promising substance in treating vocal cord insufficiency. Laryngoscope. 1998;108:393–397. 16. Milstein CF, Akst LM, Hicks MD, Abelson TI, Strome M. Long-term effects of micronized Alloderm injection for unilateral vocal fold paralysis. Laryngoscope. 2005;115:1691–1696. 17. Chan RW, Titze IR. Viscosities of implantable biomaterials in vocal fold augmentation surgery. Laryngoscope. 1998;108:725–731. 18. Czaja WK, Young DJ, Kawecki M, Brown RM. The future prospects of microbial cellulose in biomedical applications. Biomacromolecules. 2007;8:1–12. 19. Czaja W, Krystynowicz A, Bielecki S, Brown RM Jr. Microbial cellulose— the natural power to heal wounds. Biomaterials. 2006;27:145–151.
769
20. Hoenich N. Cellulose for medical applications: past, present, future. BioResources. 2006;1:270–280. 21. Martson M, Viljanto J, Hurme T, Laippala P, Saukko P. Is cellulose sponge degradable or stable as implantation material? An in vivo subcutaneous study in the rat. Biomaterials. 1999;20:1989–1995. 22. Muller FA, Muller L, Hofmann I, Greil P, Wenzel MM, Staudenmaier R. Cellulose-based scaffold materials for cartilage tissue engineering. Biomaterials. 2006;27:3955–3963. 23. Backdahl H, Helenius G, Bodin A, Nannmark U, Johansson BR, Risberg B, Gatenhlm P. Mechanical properties of bacterial cellulose and interactions with smooth muscle cells. Biomaterials. 2006;27:2141–2149. 24. Brown RM Jr, Willison JH, Richardson CL. Cellulose biosynthesis in Acetobacter xylinum: visualization of the site of synthesis and direct measurement of the in vivo process. Proc Natl Acad Sci USA. 1976;73: 4565–4569. 25. Costa HO, Souza FC. Evaluation of tissue regeneration of the burned pig’s skin followed by biotissue grafting. Acta Otorrinolaringologia. 2005;23: 23–27. 26. Panerari AD, Costa HO, Souza FC, Castro M, Silva L, Neto OMS. Tracheal inflammatory response to bacterial cellulose dressing after surgical scarification in rabbits. Braz J Otorhinolaryngol. 2008;74:512–522. 27. Amorim WL, Costa HO, Souza FC, Castro MG, Silva L. Experimental study of the tissue reaction caused by the presence of cellulose produced by Acetobacter xylinum in the nasal dorsum of rabbits. Braz J Otorhinolaryngol. 2009;75:200–207. 28. Kato M, Neto OMS, Silva L, Souza FC, Castro M, Costa HO. Tympanic membrane reconstruction with the use of cellulose membrane in guinea pigs. Acta ORL/Tecnicas em Otorrinolaringologia. 2008;26:7–12. 29. Perazzo PS, Duprat AC, Lancelotti C. Histological behavior of the vocal fold after hyaluronic acid injection. J Voice. 2009;23:95–98. 30. Duprat AC, Costa HO, Lancelotti C, Almeida R, Caron R. Histologic behavior of the inflammatory process in autologous fat implantation in rabbit vocal folds. Ann Otol Rhinol Laryngol. 2004;113: 636–640. 31. ISO 10993-6. Biological Evaluation of Medical Devices—Part 6: Tests for Local Effects After Implantation. International Standards for Business, Government and Society. International Organization for Standardization FDA Food and Drug Administration. USA; 2007. 32. Cotran RS, Kumar V, Robbins SL. Structural and functional pathology. 4th ed. Rio de Janeiro, Brazil: Ed. Guanabara-Koogan; 1991. 33. Chen GQ, Wu Q. The application of polyhydroxyalkanoates as tissue engineering materials. Biomaterials. 2005;26:6565–6578.
