Abstract
We describe a muscle sparing approach in which the triceps is elevated without injuring the muscle or disturbing its insertion. The entire extensor mechanism is preserved in continuity, thus preventing any extensor weakness. This can be used preferentially in cases of non-union intraarticular distal humerus fractures planned for Total Elbow Arthroplasty.
Introduction
The posterior approach to the elbow is the most commonly used approach for total elbow arthroplasty (TEA). 1 –3 In this approach, the joint is exposed either by an olecranon osteotomy or a triceps splitting or reflecting technique. Olecranon osteotomy approach cannot be used for TEA as it requires an intact ulna for the fixation of the distal component of the prosthesis. 4 Triceps splitting approach can be used for TEA but gives a less satisfactory exposure to the articular region of the ulna. The triceps reflecting approaches give excellent exposure but lead to weakness of the muscle. We describe a muscle sparing approach in which the triceps is elevated without injuring the muscle or disturbing its insertion. This can be used preferentially in cases of non-union intraarticular distal humerus fractures planned for TEA.
Operative technique
The patient is taken under general anaesthesia and placed in the lateral decubitus position with the arm supported and horizontal and the forearm hanging freely. A tourniquet is applied high on the arm. A posterior midline skin incision is made beginning 6 cm proximal to the olecranon, curved gently to the ulnar side of the olecranon and extending around 6 cm further along the subcutaneous border of the ulna. The ulnar nerve is identified along the medial border of the triceps, isolated, mobilized carefully along with its soft tissue envelope and blood vessels and retracted with a rubber tape.
The medial border of the triceps is identified and incised with a no. 15 knife up to a distance of 6–7 cm above the tip of the olecranon. The muscle is gently elevated subperiosteally with ribbon gauze and periosteum elevator up to half the breadth of the humerus. The lateral border of the triceps is similarly incised up to a distance of 5 cm from lateral epicondyle and the muscle gently elevated from the bone medially. Hence, ribbon gauze is passed under the belly of triceps to lift it from the bone taking care to preserve the continuity of the extensor mechanism.
The distal humerus is well exposed at this juncture with access possible through both the lateral and medial windows (Figure 1) so created by elevating the triceps. The fractured fragments of the distal humerus are removed allowing access to the coronoid fossa and the proximal ulna. The medial and lateral collateral ligaments are cut and the elbow supinated to expose the entire elbow (Figure 2).

Clinical photograph and line diagram showing: (A) lateral window, (B) medial window, (C) location of the tip of the olecranon and (D) ulnar nerve isolated and retracted from the field. Note the intact triceps mechanism all throughout the length of the field.

Clinical photograph and line diagram illustrating (A) Coonrad–Moorey prosthesis visible through the lateral window (C). The ulnar nerve (B) is retracted with a nerve tape.
The standard technique for implanting the Coonrad–Morrey total elbow prosthesis (Zimmer) is carried out. The lateral and medial soft tissue windows are meticulously repaired with absorbable vicryl sutures (Figure 3) and the ulnar nerve is reposited in its anatomical location. The suture tension is kept just adequate to allow for full mobilization and yet prevent any instability. The wound is closed over a drain and padded dressing applied.

Clinical photograph and line diagram illustrating closure of the lateral (A) and medial windows (C). The tip of the olecranon and the attachment of the triceps remains undisturbed (B).
Discussion
We have undertaken three TEAs (Figure 4) in cases of comminuted intercondylar distal humeral fractures not amenable to fixation. Each case was completed under a standard tourniquet time of 90 min (average 63 min) with minimal blood loss intraoperatively and no cases of ulnar neuropraxia. The Mayo Elbow Performance Score was optimal in all instances (average 90/100) and all patients achieved good range of elbow motion (0° to 120°) with good triceps strength at 1-year follow-up.

(a and b) Preoperative and postoperative X-rays of index case, highlighting the intercondylar fracture and the prosthesis in situ, respectively. (c and d) Preoperative and postoperative X-rays of another patient treated the same way.
The midline triceps splitting approach described by Campbell 5 was modified by Stieger 2 who raised osteoperiosteal flaps from the olecranon. Boyd 6 described a technique wherein the entire triceps mechanism is reflected from the lateral to the medial side. Bryan and Morrey 7 described a similar approach wherein the triceps is reflected from the medial to the lateral side. This technique had the advantage of preserving the entire muscle in continuity but carries the risk of ulnar neuropathies due to extensive mobilization and traction on the nerve during surgery (10 of the 80 TEA carried out by Morrey et al. had ulnar neuropathies). 1 Oizumi et al. 8 reported a satisfactory 42-month follow-up of 25 elbows with a triceps sparing ulnar approach similar to Morrey et al. Phadnis et al. 9 have reported excellent results of the similar triceps-on technique for hemiarthroplasty of the elbow in humeral fractures. The difference between our technique and that of Celli 10 is that the author raised the lateral half of the insertion of triceps, which was not needed in our cases.
Our technique precludes damage to the ulnar nerve as it is carefully isolated, mobilized and retracted without any undue traction or pressure and is reposited into its anatomical position at the end of the surgery, where it can glide freely as before. The need for anterior transposition of the ulnar nerve was assessed in each case at the end of the surgery. However, it was found that the nerve did not subluxate. The entire extensor mechanism is preserved in continuity, thus preventing any extensor weakness.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
