12.2 Situations with No Bony Deficiency

Fracture of the stem is a rare cause for revision typically without loss of host bone and can occur at both the humerus and the ulna. There is scarce information, but Athwal et al. described their experience in 24 patients presenting with 27 TEA in a single institution for a prevalence of 0.65% of humeral component fracture and 1.2% of ulnar component fracture [22]. Of note, the ulnar component fracture was seen mostly with the porous-coated stem of the Coonrad-Morrey prostheses and since the design change to the plasma sprayed this complication virtually disappeared. The mean time between the index operation and the revision was 8.2 years for the humeral component and 4.6 years for the ulnar component. The authors described a technique to progressively expand the cortical cement mantle after extraction of the fracture stem and a revision with a cement-in-cement technique in 14 cases. In the remainder, all the cement was extracted with or without the use of cortical windows (three cases). The bushings were exchanged for wear in five cases. The clinical results at 5 years follow-up were similar for both techniques (traditional vs. cement-in-cement) with comparable MEPS (78 vs. 82), corresponding to excellent results in eight patients, good in five, fair for six, and poor for two. Nineteen complications occurred in 14 patients, 7 of which were intraoperative and included intraoperative cortical perforations, 5 nerve injuries (2 permanent), 3 triceps avulsions, and 1 deep infection. Most cortical perforations were small and were treated with strut allograft and in one case with cancellous bone chips. Three transient ulnar neuropathies resolved postoperatively with one additional persistent sensory ulnar neuropathy and one postoperative radial nerve palsy in a patient that was lost to follow-up. Of the three triceps ruptures two underwent repair, and one refused further surgery. One patient sustained an olecranon fracture for the treatment of early stiffness while being manipulated during rehabilitation that required internal fixation. Another suffered a stable periprosthetic fracture that healed with bracing. One patient had bushing wear that required revision bushing exchange 41 months after the revision, another patient underwent revision of the humeral component for loosening at 51 months after the operation, and the third patient had a failure of both implants due to infection despite surgical debridement and suppressive antibiotics.

12.3 Situations with Loss of Host Bone

Surgical options for failed TEA with loss of host bone include resection, allograft, standard TEA, semi-constrained long-flanged prosthetic component, the use of a custom-made component, or use of a TEA with allograft. Specific considerations learned from revision hip surgery also apply to the elbow, including that a successful revision requires a stem that bypasses any cortical weakness or fracture, adequate distal humeral or ulnar fixation, and a viable articulation, and although on certain situations, and if bony reconstruction is successful, one may choose a short stem. Obviously, the cause of failure of the failed TEA must be addressed. Malone et al. showed that bone loss negatively affects the longevity of a semiconstrained TEA, so different strategies have been developed to reconstruct bony deficiency [11].

Using an unlinked revision system, Ehrendorfert et al. reported on the results of 15 revision arthroplasties with bone loss of less than 4 cm without any use of bone augmentation [23]. The treatment included the use of longer Souter-Strathclyde cemented implants. The authors found a mean arc of motion of 85° with five patients showing less than 90° of arc of motion and a mean elbow performance score of 75. Five patients experienced ulnar nerve paresthesia, one had numbness, and three of those had a weak motor function, with an average score of postoperative pain of 6.9 (10 being no pain). Complications included ulnar and humeral perforation, fracture at the tip of the prosthesis and two patients having poor results due to residual instability. Curiously the authors note that the chief complaint of the patients was the impaired ability to carry.

12.4 Use of Impaction Grafting

Lobenberg et al. reviewed the results of impaction grafting performed before 1997 in 12 patients with a mean follow up of 72 months [24]. Seven were rheumatoid arthritis patients, and five were posttraumatic patients. Impaction grafting was performed with the new component in three cases, and in nine cases there had been a prior surgery where revision of the prior TEA had been performed. Four patients had bone grafting at the ulna, six at the humerus, and two at both sides. Additional strut allografts were placed to span cortical effects in five patients. Eight of the 12 prostheses were in place at last follow-up. Two patients were revised for loosening, one for fracture of the ulnar component, and one patient underwent resection arthroplasty due to infection. The patients with the implant in place had an improvement in bone quality without signs of loosening. There were three more revisions at final follow-up with five excellent, four good, and three fair results.

Rhee et al. described the results of impaction grafting in 16 patients with a mean age of 58.4 years [25]. Fourteen elbows had loosening of both the humeral and ulnar component, and two elbows only had humeral loosening. Two elbows had a perforation of the humeral cortex, and one had a perforation of the ulnar cortex. Bone loss was King grade IV in seven cases, grade III in six, and grade II in three elbows. Impaction was performed in all cases with allograft and additional autograft from the iliac crest in three cases. Pain and total arc of motion improved with an improvement in MEPS from 41 points preoperatively to 82.8 points postoperatively (p = 0.001). The results were good or excellent in 15 cases and fair in 1. Mild graft resorption (grade I or II) was observed in all cases, and incomplete radiolucent lines were observed in 12 cases, complete radiolucent lines in 3, and probable loosening in 1 case. Additional surgery was needed in two cases.

12.5 Use of Strut Allograft

The use of struts in the femur for periprosthetic fractures and revision hip surgery has been successful because they provide similar support to metal plates and may unite to host bone augmenting resulting bone stock. Struts are usually fixed by circumferential wire or cables, and an anterior and a posterior strut are commonly used in the humerus. Of particular importance is the engagement of the anterior flange with the anterior strut graft for added stability of the construct. The use of a flange and an anterior strut graft can make up for significant distal humerus defects. The results of this technique are summarized in Table 12.5 [7, 24, 26]. However, cost, availability, disease transmission, and the need for long exposures and surgical time for proper contouring can limit the use of the allograft [27]. Typically a strut allograft has been used with one or two struts in the humerus (anterior or anterior and posterior) and typically with a single posterior strut in the ulna although there is an occasional report of its use in a posteromedial and posterolateral fashion. Struts have been used because it is a simple and yet effective way of dealing with bone loss until the development of allograft-prosthetic composite techniques was refined. The use of strut allograft has been seldom reported in the literature.

Table 12.5

Results of impaction grafting and cortical strut grafting

Author	Type of reconstruction	Patients (N)	Age at surgery	Follow-up (months)	MEPS	Survival rate	Complications/observations
Sanchez-Sotelo et al. [7]	Periprosthetic fracture around loose humeral component using strut allograft augmentation	5	NA	36	Functional ROM	10/11 united grafts 7/8 functional ROM and slight or no pain. One had moderate pain and limited motion	1 required revision 1 postoperative periprosthetic humeral fracture 1 olecranon fracture 1 triceps failure 1 permanent ulnar nerve injury
Kamineni et al. [26]	Strut allograft for ulnar osteolysis	22	NA	4 years	79		8/22 had a complication (36%)
Loebenberg et al. [24]	Severe osteolysis of distal humerus and proximal ulna treated with impaction grafting	12	57	72 (minimum 2 years)	9/12 good or excellent result	2 revisions for loosening 1 revision for fracture of ulnar component	1 infection with resection arthroplasty

N number, MEPS Mayo Elbow Performance score, ROM range of motion

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Total Shoulder Arthroplasty: Complications and Results Total Hip Arthroplasty: Complications and Results Total Shoulder Arthroplasty: Surgical Technique Total Shoulder Arthroplasty: Epidemiology and Causes Total Hip Arthroplasty: Surgical Technique Total Elbow Arthroplasty: Surgical Technique

Stay updated, free articles. Join our Telegram channel

Join