➢ The complex anatomy of the distal aspect of the humerus, in combination with the challenge of treating the traumatized soft-tissue envelope, addressing bone and articular cartilage loss, and minimizing postoperative complications, often leads to suboptimal outcomes following the treatment of open distal humeral fractures.
➢ The overall goals of treatment should focus on rigid fixation, maintaining or restoring a viable soft-tissue envelope, restoring functional range of motion, and limiting complications.
➢ Although uncommon, some open distal humeral fractures are associated with concomitant injuries and are associated with complex fracture patterns involving the entire articular surface.
➢ The initial treatment of an open distal humeral fracture includes tetanus prophylaxis and the administration of antibiotics followed by excisional debridement to reduce the risk of infection.
➢ Multiple operative fixation strategies have been used, including external fixation, internal fixation, and total elbow arthroplasty, each of which may be better suited for particular patients and fracture patterns.
➢ For large soft-tissue defects, the early use of soft-tissue procedures to provide adequate and stable wound coverage can result in improved outcomes and fewer complications.
Open distal humeral fractures remain an especially difficult injury pattern to treat. The complex anatomy of the distal part of the humerus in combination with the challenge of treating the traumatized and relatively thin soft-tissue envelope, addressing bone and articular cartilage loss, and minimizing postoperative complications often compromises functional outcomes. Even in the setting of appropriate soft-tissue handling, anatomic reduction, and early rehabilitation, substantial long-term functional upper-extremity morbidity can occur, including stiffness, infection, nonunion, malunion, heterotopic ossification, soft-tissue breakdown, and posttraumatic arthritis. Many aspects of open distal humeral fracture treatment demand attention, including expedient antibiotic administration, adequate debridement, meticulous soft-tissue treatment, acute and definitive fracture stabilization, and postoperative physiotherapy.
The elbow joint is made up of 3 distinct articulations: the ulnohumeral joint, the radiocapitellar joint, and the proximal radioulnar joint. The ulnohumeral joint serves as a hinge (ginglymus) joint that allows for flexion and extension of the elbow in the range of 0° of extension to 130° of flexion. The radiocapitellar and radioulnar joints serve as rotational (trochoid) joints that allow for supination and pronation of the forearm in the range of 80° from neutral in either direction. The axis of hinge rotation of the ulnohumeral joint is known as the trochlea; on average, the trochlea lies 6° valgus and 30° anterior to the axis of the humeral shaft. The valgus orientation contributes to the carrying angle of the elbow, allowing the arm to clear the pelvis during gait. The anterior angulation allows the elbow to achieve its large range-of-motion arc while still being in plane with the body when in full extension. The trochlea is covered with a 300° arch of cartilage and houses a groove that articulates intimately with the trochlear notch of the ulna. The result is a highly congruent, constrained joint that allows for a large range-of-motion arc.
The trochlea is supported by the medial and lateral columns of the distal part of the humerus. The medial column flares 45° from the axis of the humeral shaft in the coronal plane and terminates as the medial epicondyle. Similarly, the lateral column flares 20° from the humeral shaft in the coronal plane and terminates as the lateral epicondyle and in the anterior plane as the capitellum. The epicondyles serve as the origin sites for the ligamentous stabilizers of the elbow joint: the medial epicondyle for the anterior and posterior bundles of the medial collateral ligament, and the lateral epicondyle for the lateral collateral ligament. They also serve as the respective origins for the flexor-pronator and extensor muscle masses, which act as important dynamic stabilizers. Posteriorly, the medial and lateral columns are separated by the olecranon fossa, which is congruent with the olecranon during extension; anteriorly, the columns are separated by the coronoid fossa, which is congruent with the coronoid process during flexion. This intricate articulation does not tolerate displacement and must be anatomically maintained to ensure proper biomechanics.
The elbow joint is traversed by the median, ulnar, and radial nerves as well as by a robust collateral blood supply derived from the brachial artery. In general, the ulnar and radial nerves are the neurovascular structures that are most often injured in association with adult distal humeral fractures. The ulnar nerve travels along the medial aspect of the triceps after piercing through the medial intermuscular septum of the arm and distally through the arcade of Struthers, about 8 cm proximal to the medial epicondyle, before continuing posterior to the medial epicondyle in the cubital tunnel. It then travels between the 2 heads of the flexor carpi ulnaris before entering the anterior aspects of the forearm and hand. The radial nerve traverses posteriorly around the humerus in the spiral groove to emerge approximately 14 cm proximal to the lateral epicondyle. From there, it continues through the lateral intermuscular septum and travels between the brachialis and brachioradialis before bifurcating into the posterior interosseous nerve and the radial sensory nerve. The median nerve and brachial artery, which typically are not injured or encountered during standard surgical exposures, travel between the brachialis and biceps muscles, anterior to the medial intermuscular septum, and then continue distally between the pronator teres and the biceps tendon.
After a proper trauma evaluation according to the Advanced Trauma Life Support protocol1, the upper extremity should be examined for readily apparent soft-tissue injuries, osseous deformities, and joint dislocations. Assessment of the radial, ulnar, and median nerves as well as distal pulses is paramount.
High-quality traction radiographs, including anteroposterior and lateral views of the injured elbow (Fig. 1) along with orthogonal views of the ipsilateral shoulder and wrist, should be made to rule out concomitant fractures. Traction radiographs take advantage of ligamentotaxis and are helpful for elucidating the fracture pattern as it looks under reduction. Traditional computed tomography (CT) scans have limited benefit as compared with traction radiographs; however, CT scans may be valuable for assessing injuries of the capitellum and/or trochlea. Three-dimensional images may provide additional information, but their utility has not been determined. Orthogonal views of the contralateral, uninjured elbow may be used to define the patient’s normal anatomy and to help in preoperative planning.
Multiple classification systems have been used to describe distal humeral fractures. The Riseborough and Radin classification system describes 4 types of intercondylar T-type fractures: type-1 fractures (nondisplaced), type-2 fractures (characterized by separate trochlear and capitellar fragments that are not substantially rotated), type-3 fractures (characterized by separate trochlear and capitellar fragments with substantial fragment rotation), and type-4 fractures (severely comminuted articular fractures with substantial separation of the humeral condyles)2. The classification system described by Jupiter and Mehne divides distal humeral fractures into 3 grades based on fracture location in relation to the capsule and articular surface: grade 1 (intra-articular), grade 2 (extra-articular but intracapsular), and grade 3 (extracapsular)3. Intracapsular fractures are further divided into single-column (A), bicolumnar (B), capitellar (C), and trochlear (D)3. Last, the OTA/AO classification system4 characterizes fractures on the basis of the bone involved, the location of the fracture, articular involvement, and the complexity of the fracture.
Traditionally, open fractures are classified with use of the Gustilo and Anderson classification system5 (Table I). Type-I and II open injuries are lower-energy injuries associated with limited soft-tissue disruption. Type-III open injuries are higher-energy injuries associated with extensive soft-tissue involvement. Gustilo et al. further classified type-III injuries into subtypes A, B, and C, corresponding with segmental fractures associated with extensive soft-tissue lacerations but with adequate bone coverage, fractures requiring soft-tissue transfer, and fractures associated with arterial injuries requiring vascular repair, respectively6.
After the identification of an open fracture, antibiotics and tetanus prophylaxis should be administered immediately. Patients with open fractures should receive a first-generation cephalosporin, with the addition of an aminoglycoside for those with Gustilo type-III open fractures. Penicillin should be added for all wounds with gross soil or organic-source contamination; however, patients with a penicillin allergy should be managed with vancomycin, gentamicin, and metronidazole. Patzakis and Wilkins reviewed 77 infections associated with 1,104 fractures and showed an increased risk of infection related to the failure to receive antibiotics, the resistance of wound-contaminant organisms, increased time from injury to the initiation of antibiotics, and the extent of soft-tissue damage7.
After the administration of antibiotics, the next priority is excisional debridement and irrigation. The goals of operative debridement and irrigation are to reduce bacterial load, to clear gross contamination, and to excise devitalized tissue. Articular fragments should be preserved as they are vitally important for restoring the congruency of the elbow joint. Cleaning the articular fragments for 5 minutes with use of a 10% povidone-iodine solution and normal saline solution rinse affords the best chance of decontamination and of maximizing the cellular vitality of osteoarticular fragments8. Wounds and underlying articular fragments can be protected by negative-pressure wound dressings if delayed definitive treatment is deemed appropriate. Marsupialization of the articular fracture in healthy soft tissue (e.g., abdominal wall, thigh) also may be considered. The use of antibiotics in saline solution has not been shown to be of any benefit9. Normal saline solution with low-pressure gravity-driven irrigation should be used; high-pressure pulse lavage is traumatizing to both compromised and healthy tissue. Current recommendations call for low-pressure irrigation with use of normal saline solution and limited antiseptics for patients with open fractures9.
Traditionally, operative debridement and irrigation within 6 hours after the initial injury has been the recommended treatment. A systematic review by Schenker et al. demonstrated no association between delayed debridement and higher infection rates; however, elective delay of operative debridement is still not advocated10. Pollak et al. also found that time from injury to operative debridement was not an independent predictor of infection11. Other factors that contributed to reduced infection risk included early antibiotic administration, the quality of debridement, patient comorbidities, admission to a trauma center, and smoking status11.
Nonoperative treatment has a very limited role in cases of open intra-articular distal humeral fractures. Srinivasan et al., in a study involving a mixed group of closed and open osteoporotic distal humeral fractures in elderly patients, reported worse functional results and higher pain scores for patients who were managed nonoperatively12. Nonoperative treatment should be reserved for patients with stable, minimally displaced fractures for which the risks of surgery outweigh the benefits.
The use of external fixators in the acute setting provides early stabilization and allows for the treatment of neurovascular or soft-tissue injuries prior to definitive reconstruction. In rare instances, external fixation can be used for definitive stabilization. Chaudhary et al. reported on a series of 8 open intra-articular distal humeral fractures that were treated with open reduction and the application of medial and lateral uniplanar mini-external fixators13. All fractures were classified as type 13-C according to the OTA/AO system. According to the system of Gustilo et al., 1 fracture was classified as type I; 3, as type II; 3, as type IIIA; and 1, as type IIIC. After a mean duration of follow-up of 11.4 months, the mean range of motion was 20° to 120° and 6 of the 8 patients had a good to excellent outcome. Kömürcü et al. reported on 20 open intra-articular distal humeral fractures that occurred secondary to gunshot wounds14. According to the system of Gustilo et al., there were 9 type-IIIA, 8 type-IIIB, and 3 type-IIIC fractures. Nineteen fractures were treated with Ilizarov ring fixation; of those, 8 had a good result, 7 had a moderate result, and 4 had an unsatisfactory result as defined by Morrey et al.15. The mean time to removal of the fixator was 4.6 months, and the mean duration of follow-up was 34.3 months. No fractures resulted in nonunion. It should be noted that both studies had relatively large numbers of anatomic reductions, suggesting the importance of an adequate reduction in the treatment of these fractures.
Multiple operative exposures of the distal part of the humerus can be utilized for open reduction and internal fixation of open fractures, including the triceps-splitting, paratricipital, triceps-reflecting, triceps-reflecting anconeus pedicle (TRAP), and olecranon-osteotomy approaches. In the setting of an open fracture, the direct posterior incision that is typically used to perform the exposure may be compromised. One option is to incorporate the open wound as part of the incision, raising large, thick flaps in order to perform the deep dissection. An additional option is to perform the standard incision and to address the open wound via an inside-out technique during which the traumatic wound is irrigated through the surgical incision rather than enlarging and debriding directly through the traumatic wound.
Wilkinson and Stanley, in a cadaveric study, reported that the triceps-splitting, triceps-reflecting, and olecranon-osteotomy approaches revealed 35%, 46%, and 57% of the articular surface, respectively16. The olecranon osteotomy failed to provide visualization of >40% of the distal humeral articular surface. The olecranon osteotomy also reduced compression forces across the articular surface through the release of tension between the bone and soft tissue, increasing both visualization and ease of reduction. Hewins et al. and Ring et al. both reported that olecranon osteotomy was associated with low rates of nonunion and implant removal17,18. Complications associated with olecranon osteotomies included increased operative time, osteotomy nonunion, prominent implants associated with symptoms. If total elbow arthroplasty is the ultimate treatment course, olecranon osteotomy is contraindicated.
McKee et al. reported on 26 patients who were managed with internal fixation for the treatment of open distal humeral fractures, with half of the patients being managed with an olecranon osteotomy and the other half being managed with a triceps-splitting approach19. The patients who were managed with the triceps-splitting approach had significantly better MEPS (Mayo Elbow Performance Scores) (p = 0.05), DASH (Disabilities of the Arm, Shoulder and Hand) scores (p = 0.05), and Short Form-36 (SF-36) bodily-pain scores (p = 0.002), as well as a trended toward increased range of motion, when compared with the patients who were managed with olecranon osteotomy. The mean arc of motion was 97°. Additionally, 2 patients who underwent olecranon osteotomy required removal of prominent implants. Patients who have been managed with an extensor mechanism-on, triceps-sparing approach have been shown to have comparable findings in terms of range of motion and strength on the involved and contralateral sides20,21.
Regardless of approach, when the medial side of an intra-articular distal humeral fracture is addressed, the ulnar nerve should be identified and released. There is debate about whether the nerve should be released in situ or a formal transposition should be performed. Recent literature has supported isolated ulnar nerve neurolysis without transposition because transposition has been associated with a high rate of postoperative ulnar nerve neuropathy and also because ulnar nerve neurolysis alone has not been associated with an increased rate of ulnar dysfunction at the time of long-term follow-up22,23. Transposition might be considered in various settings—for example, during revision procedures when transposition would facilitate localization, after neurolysis when the nerve is in close contact to implanted hardware, or following a traumatic wound when the medial soft-tissue envelope is compromised.
Although stable, anatomic, and bicolumnar internal fixation is the recognized goal of treatment for open intra-articular distal humeral fractures, the ideal fixation strategy continues to be debated. Several configurations of plates have been utilized over the years, with dorsal plates, 90°-90° orthogonal plates, and medial-lateral 180° (parallel) plates being the most popular configurations (Fig. 2). O’Driscoll advocated parallel plating with maximum fixation in the distal fragments and with all distal-fragment fixation contributing to stability between the distal fragments and the shaft24. In addition, he described 8 technical objectives that should be met in the fixation construct (Table II).
In a randomized prospective study of 35 patients who were managed with either orthogonal or parallel plating, Shin et al. found no significant difference between the groups in terms of clinical outcomes or complications25. After a minimum duration of follow-up of 2 years, 30 of the 35 patients had a good or excellent functional result and the mean arc of elbow motion was 106°. There is still no consensus with regard to which technique is more clinically advantageous.
In addition, with the increased utilization of mini-fragment plates and screws, additional stabilization of the distal part of the humerus with use of fragment-specific fixation may facilitate increased stability and also may decrease implant load and infection risk in patients with larger-surface implants.
Total elbow arthroplasty was originally indicated for patients with symptomatic rheumatoid arthritis of the elbow; however, Cobb and Morrey found total elbow arthroplasty to be a viable surgical option for comminuted distal humeral fractures in elderly patients in whom comminution and osteoporosis complicate and compromise internal fixation26. Ali et al., in a retrospective case series of 26 patients over the age of 65 years who underwent total elbow arthroplasty for the treatment of distal humeral fractures, reported a mean MEPS of 92 and a mean range-of-motion arc of 27° to 125° at 5 years27. Baksi et al., in a similar retrospective review of 41 distal humeral intercondylar fractures that were treated with total elbow arthroplasty either primarily or after nonunion, reported excellent results in 90.5% of the patients with acute fractures and 85% of those with a nonunion28.
McKee et al., in what we believe to be the only randomized controlled trial (RCT) comparing total elbow arthroplasty with open reduction and internal fixation (ORIF) for the treatment of distal humeral fractures in patients older than 65 years of age, reported a better MEPS at all follow-up intervals up to 2 years, with trends toward better range of motion and decreased revision rates, in the total elbow arthroplasty group29. Egol et al., in a subsequent retrospective series, found no difference between ORIF and total elbow arthroplasty with regard to range of motion, DASH scores, and MEPS30. The authors concluded that good results can be achieved with both options, depending on bone quality, expectant outcome, and surgeon experience (Fig. 3).
Perhaps the most challenging aspect of treating open distal humeral fractures is soft-tissue management. The distal aspect of the humerus has a limited soft-tissue envelope, with some of the osseous anatomy being subcutaneous. The ultimate choice for defect coverage depends on variables such as the size and complexity of the defect, the exposure of vital structures, and patient comorbidity31. Typically, defects of the elbow occur laterally. However, for posterior defects that are subject to weight-bearing, skin grafts are unstable and are subject to erosion and breakdown if placed over osseous prominences and in positions that are subject to shear forces, such as over the olecranon. As such, the use of flaps is often warranted to gain adequate soft-tissue coverage.
A common pedicled flap is the radial forearm flap. This flap is one of the most common and most versatile fasciocutaneous axial flaps used for coverage of the elbow and is capable of covering as much as a 15 × 25-cm defect. The flap is based off of the radial artery from its origin 10 cm distal to the antecubital fossa, allowing the flap to be safely rotated from its volar harvest site to the posterior aspect of the elbow joint (Fig. 4). The most important criterion for successful use of the radial forearm flap is a patent ulnar artery providing flow to the hand. Any failed Allen test or failed Doppler examination of the ulnar artery is an absolute contraindication. In addition, the medial or lateral antebrachial cutaneous nerve can be incorporated into the flap, allowing for sensory protection around the elbow joint32. The disadvantage of a radial forearm flap is donor-site morbidity. Richardson et al., in a series of 100 patients, noted that 28% of the patients reported poor aesthetic results, 19% had delayed wound-healing, 13% developed exposed tendons, and 32% developed decreased sensation in the radial nerve distribution33. The ulnar artery perforator flap is another option for soft-tissue coverage in suitable patients34.
The anconeus flap is a pedicled muscle flap that can also be reliably utilized for smaller soft-tissue defects. In the case of an open distal humeral fracture, the flap is useful for coverage of the lateral radiocapitellar joint, the insertion of the triceps muscle, and the olecranon. The anconeus functions at the elbow joint as a supinator of the forearm and provides the final 15° of elbow extension. As such, the harvest of the anconeus for soft-tissue coverage results in an almost negligible deficit of function. The blood supply to the anconeus is provided by 2 arterial pedicles that anastomose within the muscle belly itself, which allows for variability in its mobilization. Most authors have agreed that the coverage provided by the anconeus flap is relatively small, with 5 × 7 cm being the maximum35,36. Also, as the flap involves only muscle tissue, primary skin closure over the flap or skin-grafting is typically required. Fleager and Cheung, in a small series of 20 patients who sustained substantial soft-tissue defects at the site of total elbow arthroplasty or distal humeral plating, reported that the anconeus rotation flap healed completely and provided a satisfactory MEPS result (mean, 79.5)37.
For much larger defects, the latissimus dorsi muscle flap can be harvested from its origin off the 4 inferior ribs and 6 inferior vertebrae to be rotated about the axilla in order to provide coverage as far distally as the middle part of the forearm. This flap is based off of the thoracodorsal vessels and must be divided from its humeral insertion for the coverage of more distal defects as an island pedicled flap. It can be harvested as a myocutaneous flap or as a muscle flap alone with skin graft (Fig. 5). Stevanovic et al., in a study of 16 patients who underwent latissimus dorsi flap coverage for large elbow soft-tissue defects as large as 100 cm2, found that only 3 patients had partial flap necrosis38. Incidentally, those patients required flap coverage fairly distal to the olecranon, leading the authors to caution against such distal use. Choudry et al. noted similar concerns in a review of 99 flaps in which 38% of the pedicled latissimus dorsi flaps had distal necrosis, with the majority of those flaps extending beyond the olecranon39. Last, the latissimus dorsi muscle is an important adductor, extender, and internal rotator of the humerus. For this reason, it should not be harvested as a flap in patients requiring the regular use of a wheelchair or crutches as such patients will require the strength of the latissimus dorsi muscle to provide mobility32.
Major trauma centers have largely adopted the concept of an incremental approach for wound closure. Microvascular free tissue transfers have become an earlier treatment choice at centers with facile microsurgeons, and free tissue transfer affords large skin flaps with the ability to achieve skin-to-skin closure of defects. The rates of flap loss and wound-healing complications are much lower when the procedure is performed by an experienced surgeon. The evolution to free-tissue transfer for definitive soft-tissue coverage of elbow and humeral defects with free perforator flaps reflects a modern, microsurgical approach to these defects. Perforator flaps include skin and fat and can include fascia as well. The most common perforator flap is the anterolateral thigh flap (Fig. 6). Authors of the present study (S.M., S.K., and L.S.L.) previously showed the utility of the anterolateral thigh flap for soft-tissue coverage in orthopaedic trauma patients with lower-extremity injuries40,41. The flap is harvested based on 1 or more perforators that are traced back to the supplying pedicle (the descending branch of the lateral femoral circumflex artery) to allow for microvascular free tissue transfer. The pedicle is anastomosed to recipient vessels in the upper extremity to reestablish the blood supply, and the brachial or forearm vessels may be used, depending on the location of the defect and the ease of the anastomoses. Harvest of an anterolateral thigh flap results in the sparing of functional muscle, and very large flaps can be harvested with primary closure of the donor site. However, if the donor site cannot be closed primarily, the donor site will need to be treated with skin-grafting, which results in a poorer aesthetic outcome. The anterolateral thigh flap has the advantage of supplying a supple skin envelope to the upper extremity and elbow. It also allows for easy re-elevation to allow access to the elbow and humerus if orthopaedic revision is needed. It remains the authors’ flap of choice for the coverage of elbow defects.
Many studies have demonstrated favorable outcomes following the treatment of closed distal humeral fractures with ORIF or arthroplasty; however, few have analyzed outcomes with the added insult of an open injury. Min et al. recently performed a case-control study comparing the functional outcomes for open and closed OTA/AO type-C distal humeral fractures that were treated with ORIF42. Patients with open injuries had a substantially decreased range of motion (mean, 82.5° versus 108.7°) and worse SF-36 outcome scores (mean, 57.9 versus 79.0) compared with patients with closed injuries. The same group of authors examined staged and acute definitive treatment of open distal humeral fractures and found that staged treatment with initial external fixator placement resulted in lower SF-36 scores (mean, 50.2 versus 68.2) and Mayo Elbow Performance Index scores (mean, 55.6 versus 84.2). Range of motion was also slightly less for the staged group (mean, 73.8° versus 94.2°)43. Overall, the outcomes for open distal humeral fractures are likely to be less favorable than those for closed fractures, and patients should be counseled regarding this difference.
Complications following open distal humeral fractures are not uncommon and are the result of complex soft-tissue and osseous injuries. The most common complications include elbow stiffness, heterotopic ossification, ulnar neuropathy, infection, nonunion, malunion, residual instability, or even osteonecrosis44-48. Of these, elbow stiffness is the most common. Anatomically, stiffness has been categorized as either intra-articular or extra-articular, with most cases presenting as a combination of the two. Initially, stiffness is addressed nonoperatively with physical therapy and dynamic splinting. If range of motion does not improve, capsular release and heterotopic bone excision can be of some benefit. Ehsan et al. reported on a year-long follow-up study of 177 patients who underwent excision of heterotopic ossification, circumferential capsulectomy, implant removal, and ulnar nerve transposition, with 69% of the patients achieving an acceptable functional elbow range-of-motion arc of 100°48. Very few patients in that series required any additional operative intervention. Prevention of this common complication is often best achieved through early range-of-motion therapy to prevent adhesions and fibrosis. The use of radiation is contraindicated. The use of indomethacin is an option, although caution should be taken with respect to gastrointestinal bleeding and increased risk of nonunion. The correct balance between stiffness prevention and soft-tissue breakdown remains unknown. In addition, ulnar neuropathy has been proposed as having a higher occurrence in the setting of a distal ulnar columnar fracture44.
The authors would like to acknowledge G. Russell Huffman, MD, for providing radiographs (Fig. 3) for this manuscript.
Investigation performed at the University of Pennsylvania, Philadelphia, Pennsylvania
Disclosure: There were no costs associated with the production of this manuscript and thus no source of funding. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arena outside the submitted work.
- Copyright © 2017 by The Journal of Bone and Joint Surgery, Incorporated