Secondary Logo

Share this article on:

Reconstruction of the Ossicular Chain with a 3D-Printed Prosthesis

Eisenman, David J., MD; Hirsch, Jeffrey, MD

doi: 10.1097/01.HJ.0000544487.88138.e8
3D-Printed Prosthesis

Dr. Eisenman is an associate professor and the vice chairman of the department of otorhinolaryngology-head and neck surgery at the University of Maryland School of Medicine, where Dr. Hirsch is an assistant professor and the section chief of community radiology of the department of diagnostic radiology and nuclear medicine.

Ossicular chain disruption from inflammatory middle ear disease or trauma is a common cause of conductive hearing loss. Ossicular chain reconstruction (OCR) is an accepted procedure for re-establishing ossicular continuity to improve the hearing of affected patients. The presence of so many different types and designs of ossicular prosthesis is itself a testimony to the fact that no single optimal prosthesis has been developed. Impediments to the optimal design of a standard, universal prosthesis include (but are not limited to) normal anatomic variation, effects of the pathologic process on natural anatomy, prosthesis stability, and biocompatibility. Intraoperative measurements of prosthesis size are imprecise, and small inaccuracies can compromise the efficiency of sound transmission and long-term stability.

Although many factors contribute to poor hearing results after OCR, including the extent of disease, ongoing Eustachian tube dysfunction, and alterations in middle ear size, studies suggest that inaccurate sizing, which can lead to unstable or inaccurate placement, may play a significant role (J Int Adv Otol. 2016 Dec;12(3):231; Otolaryngol Head Neck Surg. 2007 Nov;137(5):757; Otol Neurotol. 2001 May;22(3):299). In fact, results of one study suggest that up to 40 percent of OCR failures can be attributed to issues of surgical technique, which are often related to inaccurate sizing (Otol Neurotol. 2006 Jan;27(1):20).

Back to Top | Article Outline


The technology of 3D printing has the potential to solve many design challenges. This technique has been applied successfully to a variety of surgical problems, and has shown promising results in improving outcomes, decreasing morbidity, and shortening surgical time (Trends Mol Med. 2016;22(3):254; Biomed. 2016;15(1):115). However, the extremely small size of and limited accessibility to the middle ear pose unique challenges compared with other anatomic regions. In particular, accurate in vivo measurements, on one hand, and faithful reproduction of those measurements, on the other, are harder to obtain and possibly less reliable.

Back to Top | Article Outline


To assess the feasibility of creating a 3D-printed, customized prosthesis for OCR, we performed a study designed to answer three fundamental questions (3D Print Med. 2017; 3(1):7):

  1. Can measurements that reflect variations in middle ear anatomy be obtained from routine, clinically available CT scanners?
  2. Can a current, commercially available 3D printer produce a prosthesis that accurately reflects those variations?
  3. Can otologic surgeons detect the differences in the printed prosthesis?

Since absence of the incus is one of the most common issues that require OCR, we designed a simple incus replacement prosthesis, analogous to the typical columellar design of commercially available models. This consisted of a trough in which the manubrium (long process) of the malleus sits, connected by a straight shaft to a cup on the stapes capitulum (head; Fig. 1). Unlike a commercially available, off-the-shelf prosthesis, a custom-designed prosthesis based on CT measurements has variations that reflect not only differences in the distance from the manubrium to the capitulum but also differences in angulation of the manubrium relative to the footplate and anterior-posterior offset between the manubrium and capitulum. Quantitative measurements of the 3D-printed prostheses showed that each was unique, with measurable variation in all axes. For example, the length of the connecting shaft ranged from 2.09 to 2.50 millimeters.

Three human cadaveric temporal bones were chosen from donors with no known history of ear disease. The incudes were removed, and CT scans were performed. A prosthesis was individually designed for each bone. Surgeons, including two experienced attending physicians with practices dedicated to otology, and two chief residents, all of whom were blinded to the source bone from which each prosthesis was designed, were asked to insert each prosthesis to see if they could determine which prosthesis belonged to which bone. Each surgeon accurately matched the prosthesis to its parent bone. The chance of this happening randomly is 1:1,296.

Although there are still many challenges in the commercial reproduction of this approach, this study demonstrated that production of 3D-printed middle ear ossicular prostheses is feasible. One day, we can hope that a custom-designed prosthesis will be ready for patients before they undergo an OCR. This would significantly shorten surgical time and increase the likelihood of a stable and accurate fit that will improve long-term outcomes.

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.