Necrosis of the Long Process of the Incus Following Stapes Surgery: New Anatomical Observations

The Laryngoscope VC 2009 The American Laryngological, Rhinological and Otological Society, Inc.

Imre Gerlinger, MD, PhD; Miklo´sTo´th, MD, PhD; La´szlo´ Lujber, MD, PhD; Istva´n Szanyi, MD; Pe´ter Mo´ricz, MD; Krisztina Somogyva´ri, MD; Adrienn Ne´meth, MD; Ga´bor Ra´th, MD; Jo´zsef Pytel, MD, PhD; Wolfgang Mann, MD, PhD, FACS

Objectives/Hypothesis: The most frequent incudes revealed 1-4 nutritive foramina on the long complication (generally recognized during revision processes of 48% (24) of the left-sided incudes and procedures) following seemingly successful stapedoto- 56% (28) of the right-sided incudes. The positions of mies and stapedectomies is necrosis of the long pro- these foramina differed, however, from those previ- cess of the incus. This is currently ascribed to a mal- ously described in the literature. They were mostly crimped stapes prosthesis or to a compromised blood located not on the medial side of the incus body or on supply of the incus. The two-point fixation can cause the short process or on the cranial third of the long a mucosal injury with a resulting toxic reaction, and processes, but antero-medially, mostly on the middle also osteoclastic activity. An important aspect in the or cranial third. In 48% of all the incudes examined, engineering of ideal stapes prostheses is that they an obvious foramen was not observed either in the should be fixed circularly to the long process of the body or in the long process of the incus. No relation- incus with appropriate strength. The objective of this ship was discerned between the chronological age of study was to compare current knowledge relating to the incus specimens and the numbers and/or locations the blood supply of the ossicular chain with the pres- of the nutritive foramina. In each case, the opening of ent authors’ observations on cadaver incudes. Most of the foramen was the beginning of a tunnel running the papers dealing with this issue appeared in the obliquely and medially upward through the corticalis mid-20th century. of the long process of the incus. The foramina are Methods: The published data were compared thought to be capable of ensuring a richer blood sup- with the authors’ findings gained from photodocumen- ply between the surface and the inside of the long tation on 100 cadaver incudes. The photos were taken process, allowing the arteries to run in and out. with a Canon EOS 20 digital camera (Canon, Inc., Conclusions: These observations indicate that Lake Success, NY) with a 5:1 macro-objective. The long conclusions drawn from classical anatomical works processes of the incudes were examined from four appear to need reconsideration. The present authors directions under a Leica surface-analyzing microscope consider that the reason for the necrosis of the long (Leica Microsystems GmbH, Wetzlar, Germany). process of the incus is not a compromised blood sup- Results: Analysis of the positions of the entran- ply, except in some exceptional anatomical situations. ces of the feeding arteries (nutritive foramina) on the They discuss the possible reasons why malcrimping may lead to necrosis of the long process of the incus. To prevent such malcrimping, attention is paid to the From the Department of Otorhinolaryngology and Head Neck new generation of prostheses. Surgery, University of Pećs, Pećs, Hungary (I.G., L.L., I.S., P.M., K.S., A.N., Key Words: Incus, necrosis, piston, long process G.R., J.P.) and the Department of Otorhinolaryngology and Head and Neck of the incus, stapes, stapes surgery. Surgery, University of Mainz, Germany (M.T., W.M.). Laryngoscope, 119:721–726, 2009 Editor’s Note: This Manuscript was received on July 26, 2008 and accepted for publication on October 28, 2008. Presented during the meeting of the 8th Conference on Cholestea- toma and Ear Surgery in Antalya, Turkey, June 15–20, 2008. INTRODUCTION This study is supported by grants from the HSR Fund (OTKA After an initial hearing improvement, stapedecto- NKTH3 68476). mies or stapedotomies sometimes fail, the main reasons Conflict of interest: none. being erosion or necrosis of the long process of the incus, Send correspondence to Imre Gerlinger, MD, PhD, Department of 1–7 Otorhinolaryngology and Head Neck Surgery, Medical School, University caused by the different prostheses. The shape of the of Pećs, Pećs, Munkaćy Miha´ly utca 2, H-7621, Hungary. E-mail: i.gerlin- piston and its attachment to the incus are additional im- [email protected]. portant factors. Crimping that is too strong causes DOI: 10.1002/lary.20166 damage to the mucosal lining or to the cortical bone of

Laryngoscope 119: April 2009 Gerlinger et al.: Long Process of the Incus Necrosis 721 Two main arteries can be observed along the tendon of the stapedial muscle, one on the lower surface, and the other on the lateral surface of the tendon (Figs. 1 and 2). These arteries are branches of the stylomastoid artery leaving the facial canal in the region of the pyramidal process. Occasionally, they originate from the depth of the oval window, supplying the surrounding mucosa as branches of the superficial petrosal artery. The arteries running along the stapedial tendon partici- pate exclusively in the blood supply of the incus. An artery arriving directly from the pyramidal process has never been observed. The blood supply of the incus mucosa originates from a posterior branch of the anterior tympanic artery.17 An arc-shaped artery, running in a considerable mucosal fold, can frequently be observed on the medial side of the incudostapedial joint, running on the long process of the incus to the lenticular process and the Fig. 1. The blood supply of the surroundings of the incudostape- head of the stapes (Fig. 1). Vessels can also be seen run- dial joint as described by Alberti in 1963 (reproduced with kind ning distally in the mucosa of the long process of the permission from The Laryngoscope 1963;73:605–628). incus, similarly to the vessels inside the long process. In the mid-20th century, the blood supply of the the incus; this adverse consequence is less common long process and the body of the incus was intensively when bucket-type or self-crimping pistons are used.8 investigated.9,10,17,18 A cavernous vascular network Whereas some authors accentuate the likely damage inside the body of the incus may reach as low as the to the blood supply of the long process of the incus during middle third of the long process. A supplying artery gen- surgery,9–11 caused by too strong over-crimping or circum- erally enters the long process anterolaterally through ferential fixation of the piston, others do not consider the the nutritive foramen, immediately under the body of mucosal damage of the incus to be an important factor, the incus, at the beginning of the long process (in some pointing out the anastomosing vessels between the stapes cases a little more distally).17 The nutritive foramen can andtheincus,andbetweenthedeepandsuperficialparts be located laterally, or less frequently medially, on the of the incus.12,13 A relevant and interesting clinical obser- body of the incus, close to the surface of the incudomal- vation is that the incus deprived of its blood supply can be leolar joint, or occasionally on the short process of the used as an ossicular replacement prosthesis immediately, or even a year later in staged tympanoplasties, with excel- lent long-term results.14 Moreover, cortical bone columella was utilized for ossicular chain reconstruction for a period of 34 years, and necrosis was experienced in only one case.15 Relatively frequent necrosis of the distal end of the long process of the incus might be anticipated after disar- ticulation of the incudostapedial joint, due to damage to the vessels coming along the stapedius tendon, but this is not observed clinically.16–18 We felt it important to review the available relevant publications from the mid-20th century that described the blood supply of the tympanic cavity and the osssicular chain, in order to highlight the possible con- tradictions resulting from more recent research findings in stapes surgery.7,9–13,17–19 It has previously been reported that small vessels arriving from the medial wall of the tympanic cavity run on the crura of the stapes toward the incudostapedial joint, cross the capsule of the joint, and then make an incomplete vascular ring around the capsule of the joint, creating anastomoses with vessels running on the surface and inside the incus in a distal direction (Fig. 1). Fig. 2. Endoscopic photograph of the blood supply of a cadaver The locations of these anastomoses are always the part incudostapedial joint. Note the two arteries along the stapedial ten- of the incudostapedial joint adjacent to the incus. Its don (thin black arrow), the artery on the lower surface of the lenticular process (thick black arrow), the artery supplying the head of the blood supply is ensured across the hiatus in the lenticu- stapes (white arrow), and the artery on the anterior stem of the sta- lar process, but an artery supplying the neck of the pes (white pentagon). [Color figure can be viewed in the online stapes can also be found here (Figs. 1 and 2). issue, which is available at www.interscience.wiley.com.]

Laryngoscope 119: April 2009 Gerlinger et al.: Long Process of the Incus Necrosis 722 Fig. 3. Variable blood supply of the long process of the incus according to Alberti in 1965 (reproduced with kind pemission from The Journal of Laryngology & Otol- ogy 1965;79:966–970.). incus. Besides this branch of the anterior tympanic ar- min: 1 year, max: 89 years, SD ¼ 17.72), we set out to establish tery running inside and downward in the long process whether the positions of the nutritive foramina (which provide (Fig. 3),10 the blood supply of the lenticular process can the entrances for the main feeding arteries) on the body and vary, sometimes provided by vessels running in the the long and/or short procesess of the incudes truly correspond incus, and sometimes by vessels running only in the mu- to those reported earlier. The maceration, which lasted for 1 week on average, was performed in a thermostat at 56C; the cosa of the incus. The blood supply is occasionally water was changed every day. The ossicles were then bleached ensured from the crura of the stapes or the stapedial in 3% H2O2 solution for approximately 1 hour. The remaining tendon, which may explain why the lenticular process H2O2 was thoroughly rinsed off with water, and the ossicles remains intact after eradication of a cholesteatoma; in were subsequently dried at room temperature. In some cases these cases, the blood supply coming from above is com- the ossicles needed degreasing, for which we used household promised by the pressure of the cholesteatoma. It has benzine as solvent. been claimed that the long process of the incus (a rela- A Canon EOS 20 digital camera with a Canon 5:1 macro- tively long distance of 3–4 mm needs to be bridged) can objective (Canon, Inc., Lake Success, NY) were used to take the be comparatively injured.16 photos. The long processes of the incudes were studied from The blood supply of the distal part of the incus is four directions with a Leica surface-examining microscope (Leica Microsystems GmbH, Wetzlar, Germany). In the determi- provided through the mucosal plexus around the incu- nation of the position of a nutritive foramen, the criteria were dostapedial joint, through one or more tiny channels that the foramen (defined as the opening of an upward-running located mainly on the lower surface of the lenticular pro- tunnel) broke through the cortical bone obliquely, and proceeded cess. The artery entering through the most expressed deeply toward the medulla, and that the diameter of the fora- hiatus often makes an anastomosis with a main supply- men was approximately the same throughout its whole length. ing artery in the incus (Figs. 1–3). It was earlier With this technique, we excluded tiny or deeper superficial cav- considered that the vessels running along the upper ities or impressions on the bone. The method applied allowed a two-thirds of the long process of the incus do not create clear visualization of the fossa of the marrow. anastomoses with vessels running inside the bone,17 but almost 30 years later, electronmicroscopic investigations RESULTS 13 led to just the opposite view. The locations of the nutritive foramina examined in In the present cadaver study, our aim was to under- the 100 cadaver incudes demonstrated the following dis- line the increasingly more widely accepted opinion that tribution: the upper third of the long process of the from the viewpoint of successful stapes surgery, the com- incus on the left side in 15 cases (30%), and on the right promised blood supply of the long process of the incus is side in 13 cases (26%) (Fig. 4), with the openings mostly not as tenuous as previously thought. antero-medially; on the left side of the middle third in nine cases (18%), and on the right side in 14 cases MATERIALS AND METHODS (28%), again mostly antero-medially; and on the right The study was approved in advance by the local research side in the lower third in only one case (2%). The aver- and ethics committee of the University of Pe´cs, Hungary. age diameter of the long processes of the left-sided Through our photodocumentation on 50 pairs of macerated incudes, measured at the infero-medial border, was 0.74 cadaver incudes (28 male, 22 female; average age ¼ 68.8 years, mm (min: 0.39 mm; max: 1 mm; SD ¼ 0.09). During

Laryngoscope 119: April 2009 Gerlinger et al.: Long Process of the Incus Necrosis 723 Fig. 4. Positions of nutritive foramina on the upper, middle and lower thirds of the long process of the incus (A ¼ anterior; P ¼ posterior; M ¼ medial; L ¼ lateral). stapes surgery, this is the position where the prosthesis One of the critical aspects of stapes surgery is the is generally crimped around the long process. The corre- adequate attachment of the prosthesis to the long pro- sponding diameter for the right-sided incudes was 0.75 cess of the incus.1,8,20–22 Crimping of the prosthesis mm (min: 0.41 mm; max: 0.92 mm; SD ¼ 0.09). No rela- requires appreciable surgical experience if the intact tionship was discerned between the chronological age of blood supply of the incus is to be preserved, and injuries the incus specimens and the numbers and/or locations of to the mucous membrane and the bone surface are to be the nutritive foramina. Thus, an obvious nutritive fora- avoided. An excessively tight attachment causes pro- men was observed in the long process in 52% of all these nounced effects at two points and can easily result in incudes. It is noteworthy that in some cases several fo- damage to the bony periosteum. ramina were found in the same third or in two Surgeons must be aware of the fact that remodeling neighboring thirds of the same incus (Figs. 5 and 6). Ta- may occur in time, as a consequence of osteoclast-osteo- ble I presents data on the distribution of all foramina blast activity in the lower third of the long found in the 100 incudes. Foramina on the body of the process.20,21,23,24 One cause of incus erosion must be that incus, close to the incudomalleolar joint or on the short there is no organic content (i.e., osteocytes) in human 9,10,17 process, as reported in classical anatomical studies, ossicles after the age of 50 years.25 An important ques- were not observed at all in the present work. tion that demands further study is whether or not compromise of the ossicular microcirculation by aging DISCUSSION plays any role in the development of incus necrosis. It is generally accepted that the auditory ossicles There are four possibilities whereby problems aris- are similar in size at birth to those in the adult. The ing from crimping of the prosthesis may be avoided.8 blood supply of the distal part of the incus was revealed First, the surgeon may use a tissue-compatible, self- by earlier morphological studies to be ensured by many crimping plastic prosthesis, which can be opened briefly, small arteries arriving from various directions and creat- and closes on the basis of its biocompatible memory 9,10,12,17–19 ing anastomoses with one another (Fig. 1). (Shah’s teflon piston). These plastic implants have never

Fig. 5. Two nutritive foramina on the middle third of the long process of a right-sided incus (on the upper third, merely slight Fig. 6. Nutritive foramina on the upper and middle thirds of the long impressions can be observed). [Color figure can be viewed in the process of a right-sided incus. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] online issue, which is available at www.interscience.wiley.com.]

Laryngoscope 119: April 2009 Gerlinger et al.: Long Process of the Incus Necrosis 724 TABLE I. incudes; the question remains open as to the explanation Percentage Distribution of the Nutritive Foramina in the Different of this, which contrasts with earlier reports. We presume Thirds of the Long Processes of the Incudes, that, without an appropriate digital camera and macro- and the Minimum and Maximum Numbers of Foramina in the Individual Thirds. objective and surface analysing microscope, the authors of anatomical studies in the mid-20th century probably Left Incus Right Incus could not differentiate clearly between nutritive foram- Number (%) Min. Max. Number (%) Min. Max. ina and deep bony impressions, and overestimated the Upper third 15 (30) 1 4 13 (26) 1 3 former. It is conceivable that in the rare, coincident anatomical situations listed below, a seriously compromised Middle third 9 (18) 1 1 14 (28) 1 3 blood supply could develop in the long process of the Lower third 0 0 0 1 (2) 1 1 incus, with resulting necrosis: 1) when only one feeding Min. ¼ minimum; Max. ¼ maximum. artery ensures the blood supply of the long process; 2) when both the nutritive foramen and the superficial artery are compressed by malcrimping if they are not in gained broad popularity. A second possibility is to use a ventromedial position at the border of the middle–infe- the popular bucket handle prosthesis, which does not rior third of the long process; 3) when the lenticular require crimping. The third way involves the use of a process receives its blood supply only from the prosthesis similar to that designed recently by Wangen; superficial periosteal plexus, or the predominant sources its special clip maintains its position on the long process of feeding are the two arteries running along the stape- as a consequence of its shape and the memory of tita- dial tendon; and 4) when the crimping of the piston nium.26 The fourth possibility is the application of a results in a significant mucosal injury and osteoclastic recently developed metallic alloy prosthesis that has a activity. biomaterial memory allowing it to crimp itself, such as It must be emphasized that the ventromedial quar- the Nitinol shape memory prosthesis (SMart Piston, ter of the long process of the incus is an inaccessible Gyrus, Tuttlingen, Germany).27–32 The earlier assump- field for the surgeon. No clinical data are available con- tion that the concentric occlusion of the loop of the cerning the frequency of injuries of the arteries running prosthesis can cause ischemic incus necrosis is currently in and out through the foramina and those running in a losing its validity, in view of the recent findings regard- mucous membrane fold mentioned in classical descrip- ing the incus vasculature and its bone structure.8,13 It tions.16–18 should be emphasized that until recently no mention It appears to be an important new morphological had been made of the important role of the vascular observation that the nutritive foramen is the entrance of periosteal plexus (arterioles) that takes part in the blood an oblique tunnel passing through the bony cortex to- supply of the distal part of the long process of the ward the bone marrow, always in an infero-superior incus.13 direction. This finding supports the theory that some Our own investigations appear to confirm the arteries running downward in the long process may exit increasingly more widely accepted opinion that a com- to the surface, whereas others, running upward on the promised blood supply of the long process of the incus bone surface, can enter the long process. The positions can not be the reason for erosion and resulting necrosis. of the foramen and the oblique running canal are in con- Following the crimping of the piston, there is little likeli- trast with earlier data17 that a main branch of the hood that the seemingly important artery running along anterior tympanic artery entering the foramen always the ventromedial surface of the long process should be runs downward in the long process. It may be presumed injured, because this quarter of the loop remains that the anastomoses between the plexus inside the unclosed. Occlusion of this prominent artery was body of the incus and the vessels entering at the distal assumed in the background of the relatively frequent end of the long process of the incus serve as the artery incus necroses observed following the brief period of use or arteries running in the long process, and the nutritive of the tube-shaped polyethylene prosthesis introduced by foramen (if there is one on the incus in question) Shah.9 provides the connection between the vessels running on A further important conclusion of our study is that the surface and in the depth. Otherwise, the blood sup- the vast majority of the nutritive foramina were located ply of the long process of the incus is ensured by the ventromedially on the long processes of the incudes; this superficial vascular plexus, the role of which was under- means that the arteries running in and out at this point estimated for a long time. can not be occluded as a result of the crimping proce- dure. It was earlier reported that the foramina are located mostly on the body of the incus, on the upper CONCLUSION third of the long processes and occasionally on the short We consider that our findings tend to indicate that, processes of the incudes17; however, we could not con- from the aspect of incus necrosis, the blood supply to the firm those findings. In 52% of all the examined incudes, long process is not as tenuous as previously thought. the foramina were located on the long processes, and in Conclusions drawn from the classical anatomical works some cases several foramina were found on one incus. In appear to need correction and reconsideration, and that the remaining 48% of the ossicles, we could not detect incus necrosis after stapes surgery is unlikely to be a any foramina on the bodies or short processes of the consequence of an impaired local blood supply.

Laryngoscope 119: April 2009 Gerlinger et al.: Long Process of the Incus Necrosis 725 BIBLIOGRAPHY 17. Nager GT, Nager M. The arteries of the human middle ear with particular regard to the blood supply of the auditory 1. McGee TM. The loose wire syndrome. Laryngoscope 1981; ossicles. Ann Otol 1953;62:923–949. 91:1478–1483. 18. Anson BJ, Harper DG, Winch TR. Intraosseous blood sup- 2. Lesinski SG. Causes of conductive hearing loss after stape- ply of the auditory ossicles in man. Ann Otol Rhinol Lar- dectomy or stapedotomy: a prospective study of 279 con- yngol 1964;73:645–650. secutive surgical revisions. Otol Neurotol 2002;23:281– 19. Blunt MJ. The blood supply of the facial nerve. J Anat 288. 1988;88:520–526. 3. Prasad S, Kamerer DB. Results of revision stapedectomy 20. Kwok P, Fisch U, Strutz J, May J. Stapes surgery: how pre- for conductive hearing loss. Otolaryngol Head Neck Surg cisely do different prostheses attach to the long process of 1993;109:742–747. the incus with different instruments and different sur- 4. Schimanski G. Erosion and necrosis of the long process of geons? Otol Neurotol 2002;23:289–295. the incus after otosclerosis surgery. HNO 1997;45:682– 21. Kwok P, Fisch U, Gleich O, Achhammer K, Strutz J. Stapes 689. surgery: the diameter of the long process of the incus. 5. Hausler R. Advances in stapes surgery. In: Jahnke K, ed. Otol Neurotol 2006;27:469–477. Middle Ear Surgery—Recent Advances and Future Direc- 22. Huber AM, Ma F, Felix H, Linder T. Stapes prosthesis tions. Stuttgart, Germany: Thieme-Verlag; 2004:95–140. attachment: the effect of crimping on sound transfer in 6. Jahnke K, Solzbacher D, Dost P. Revision stapes surgery. otosclerosis surgery. Laryngoscope 2003;113:853–858. Adv Otorhinolaryngol 2007;65:314–319. 23. Lannigan FJ, O’Higgins P, McPhie P. Remodelling of the 7. Gibbin KP. The histopathology of the incus after stapedec- normal incus. Clin Otolaryngol 1993;18:155–160. tomy. Clin Otolaryngol 1979;4:343–354. 24. Lannigan FJ, O’Higgins P, Oxnard CE, McPhie P. Age- 8. Rajan GP, Diaz J, Blackham R, et al. Eliminating the limi- related bone resorption in the normal incus: a case of tations of manual crimping in stapes surgery: mid-term results of 90 patients in the Nitinol stapes piston multi- maladaptive remodelling? J Anat 1995;186:651–655. center trial. Laryngoscope 2007;117:1236–1239. 25. Z. Szabo L. Ep, egeszseges hallocsontok szoveti felepitese 9. Alberti PW. The blood supply of the incudostapedial joint (szovettani tanulmany). Ful-orr-gegegyogy 2004;50:107– and the lenticular process. Laryngoscope 1963;73:605– 127. 628. 26. Grolman W, Tange RA. First experience with a new stapes 10. Alberti PW. The blood supply of the long process of the clip piston in stapedotomy. Otol Neurotol 2005;26:595–598. incus and the head and neck of the stapes. J Laryngol 27. Rajan GP, Atlas MD, Subramaniam K, Eikelboom RH. Otol 1965;79:966–970. Eliminating the limitations of manual crimping in stapes 11. Lisonek P. The intraosseous canals and vessels of the incus surgery? A preliminary trial with the shape memory niti- and head of the stapes. Acta Univ Palacki Olomuc Fac nol stapes piston. Laryngoscope 2005;115:366–369. Med 1985;111:89–100. 28. Gerlinger I, Toth M, Bako P, Nemeth A, Pytel J. KTP-laser 12. Morgenstein KM, Manace ED. Incus necrosis following sta- stapedotomy with a self-crimping, thermal shape memory pedectomy. Laryngoscope 1968;78:600–619. Nitinol SMart piston. Clin Otolaryngol 2008;33:475–480. 13. Lannigan FJ, O’Higgins P, McPhie P. The vascular supply 29. Harris JP, Gong S. Comparison of hearing results of nitinol of the lenticular and long process of the incus. Clin Oto- smart stapes piston prosthesis with conventional piston laryngol 1993;18:387–389. prostheses. Otol Neurotol 2007;28:692–695. 14. Ferekidis E, Nikolopoulos TP, Yiotakis J, et al. Should we 30. Knox GW, Reitan H. Shape-memory tapes prosthesis for use ossicular remnants in ossicular reconstruction follow- otosclerosis surgery. Laryngoscope 2005;115:1340–1347. ing cholesteatoma removal? ORL J Otorhinolaryngol 31. Brown KD, Gantz BJ. Hearing results after stapedotomy Relat Spec 2006;68:243–246. with a nitinol piston prosthesis. Arch Otolaryngol Head 15. Bauer M. Ossiculoplasty: autogenous bone grafts, 34 years Neck Surg 2007;113:758–762. experience. Clin Otolaryngol 2000;25:257–263. 32. Kasano F, Morimitsu T. Utilisation of nickel-titanium shape 16. Hough JVD. Incudostapedial joint separation: etiology, trea- memory alloy for stapes prosthesis. Auris Nasus Larynx ment, and significance. Laryngoscope 1959;69:644–648. 1997;24:137–142.

Laryngoscope 119: April 2009 Gerlinger et al.: Long Process of the Incus Necrosis 726