Journal of the American Academy of Orthopaedic Surgeons 266 Olecranon fractures, which are rela- tively common in adults, are a di- verse group of injuries ranging from simple nondisplaced fractures to complex fracture-dislocations of the elbow. They are all intra-articular injuries requiring anatomic restora- tion of the articular surface. Sev- eral methods of internal fixation are commonly utilized, including tension-band wiring, plate fixation, intramedullary screw fixation, and fragment excision with triceps ad- vancement. Fixation must be se- cure enough to permit early motion in order to avoid significant stiff- ness of the elbow joint. Because of the variability of fracture pattern and associated injuries, no single treatment method is appropriate for all fractures. Anatomy The elbow is a complex hinge joint in which the major stabilizers to valgus stress are the anterior band of the ulnar collateral ligament and the radial head. The major stabilizer against varus stress is the lateral col- lateral ligament complex, including the ulnohumeral ligament. 1,2 The coronoid process stabilizes the humerus against distal translation on the proximal ulna. Injury to the coronoid process may result in instability of the elbow, which greatly increases the complexity of the injury and adversely affects the prognosis. The olecranon prevents anterior translation of the ulna with respect to the distal humerus. 3 The articular portions of the olec- ranon and coronoid process are covered by hyaline cartilage. There may be a transverse bare area de- void of cartilage midway between the olecranon and the coronoid process. 4,5 Overcompression of this region during fracture reduction in an attempt to appose the articular cartilage is a technical error that will result in narrowing of the olec- ranon fossa and an incongruous reduction. The olecranon articu- lates with the trochlea of the hu- merus. The triceps inserts into the posterior third of the olecranon and proximal ulna, blending through a broad expansion with the aponeu- Dr. Hak is Assistant Professor of Orthopaedic Surgery, University of California - Davis School of Medicine, Sacramento. Dr. Golladay is Chief Resident, Section of Orthopaedic Surgery, University of Michigan Medical School, Ann Arbor. Reprint requests: Dr. Hak, Department of Orthopaedic Surgery, University of California - Davis, Suite 3800, 4860 Y Street, Sacramento, CA 95817. Copyright 2000 by the American Academy of Orthopaedic Surgeons. Abstract Fractures of the olecranon process of the ulna typically occur as a result of a motor-vehicle or motorcycle accident, a fall, or assault. Nondisplaced fractures can be treated with a short period of immobilization followed by gradually increasing range of motion. Open reduction and internal fixation is the stan- dard treatment for displaced intra-articular fractures. Stable internal fixation with figure-of-eight tension-band wire fixation for simple transverse fractures allows early motion to minimize stiffness. Use of two knots produces symmet- ric tension at the fracture site and provides more rigid fixation than a single knot. Care should be taken to ensure that the tension-band wire and the proxi- mal ends of the Kirschner wires are positioned deep to the triceps fibers to pre- vent wire migration. If the anterior cortex is engaged, overpenetration of the wires into the soft tissues should be avoided. Plate fixation is appropriate for severely comminuted fractures, distal fractures involving the coronoid process, oblique fractures distal to the midpoint of the trochlear notch, Monteggia frac- ture-dislocations of the elbow, and nonunions. For comminuted fractures and nonunions, a dorsally applied limited-contact dynamic-compression plate with supplemental bone graft should be utilized to support comminuted depressed articular fragments. A one-third tubular hook-plate can be used for fractures with a small proximal fragment for which additional fixation of the olecranon tip is desired. Fragment excision and triceps advancement is appropriate in selected cases in which open reduction seems unlikely to be successful, such as in osteoporotic elderly patients with severely comminuted fractures. J Am Acad Orthop Surg 2000;8:266-275 Olecranon Fractures: Treatment Options David J. Hak, MD, and Gregory J. Golladay, MD David J. Hak, MD, and Gregory J. Golladay, MD Vol 8, No 4, July/August 2000 267 rosis of the anconeus muscle and the common extensor origin. The periosteum of the olecranon is inti- mately associated with the triceps tendon. The ulnar nerve lies on the me- dial aspect of the elbow, posterior to the ulnar collateral ligament, and sweeps anteriorly to join the ulnar artery. The brachialis inserts broadly on the midportion of the anterior coronoid and the proximal ulnar metaphysis. 6 The ulnar neurovascu- lar bundle may be at risk for anterior cortical penetration by Kirschner wires used during tension-band wiring. Mechanism of Injury Olecranon fractures may occur as a result of direct trauma, indirect trauma, or a combination of both. 7 The subcutaneous location of the olecranon renders it susceptible to direct trauma, in which the olecra- non is impacted against the distal humerus, often resulting in com- minuted fractures with depression of a portion of the joint surface. Indirect trauma results from force- ful contraction of the triceps muscle during a fall on an outstretched hand and usually produces a trans- verse or short oblique fracture. The most common causes of injury include motor vehicle and motorcy- cle accidents, falls, and assaults. 8-11 Open fractures have been reported to occur in 2% to 31% of cases. 11,12 Associated Injuries Although olecranon fractures are usually isolated injuries, a high index of suspicion for associated injuries should be present in the evaluation of the polytrauma pa- tient. Wolfgang et al 11 reported a 20% incidence of associated injuries, including long-bone fractures, skull fracture, splenic injury, pulmonary contusion, and axillary artery rup- ture. Ipsilateral extremity injuries should be carefully assessed, as fractures of the coronoid process or radial head and Monteggia fracture- dislocations have a significant im- pact on elbow stability. Occasion- ally, articulated external fixation may be required to treat an unstable fracture, so as to provide adequate stability and allow early range-of- motion. 13 When a supracondylar humerus fracture occurs in conjunc- tion with an olecranon fracture, exposure of the humerus can be obtained by utilizing the olecranon fracture site. Classification Systems Although numerous classification systems have been described for olecranon fractures, none has been universally accepted. Classifica- tions serve several purposes, in- cluding improving communication among surgeons, determining treat- ment, and predicting prognosis. Some classification systems have incorporated associated injuries to the radial head and supracondylar humerus, which may have a signifi- cant impact on prognosis. 11,13 The AO classification system divides fractures of the proximal ra- dius and ulna into three broad cate- gories. Type A are extra-articular fractures involving the metaphysis of either the radius or the ulna. Type B fractures are intra-articular fractures of either the radius or the ulna, with type B1 being an intra- articular fracture of the olecranon alone. Type C fractures are intra- articular fractures of both the radial head and the olecranon. 14 The Or- thopaedic Trauma Association clas- sification system for olecranon frac- tures follows the AO system. Morrey 13 reported the Mayo classification of olecranon fractures, which is based on degree of dis- placement, elbow joint stability, and comminution. Type I fractures are nondisplaced with minimal or no comminution. Type II fractures are displaced, but the elbow joint remains stable; sufficient anterior joint surface remains to maintain stability, and the anterior portion of the medial collateral ligament also remains intact. Type III fractures render the elbow unstable and involve a large portion of the olec- ranon. They are frequently com- minuted and may have an associated radial head fracture. Type II and type III fractures are subclassified as noncomminuted (subtype A) and comminuted (subtype B). In a review of 100 consecutive fractures at the Mayo Clinic, 12 were nondis- placed (type I), 82 were displaced with a stable elbow joint (type II), and 6 were displaced with an un- stable elbow joint (type III). Schatzker’s classification of olec- ranon fractures 15 includes mechan- ical considerations related to the type of internal fixation required (Fig. 1). There are six types: type A is a simple transverse fracture; type B, a complex transverse fracture with impaction of the central por- tion of the articular surface; type C, a simple oblique fracture; type D, a comminuted fracture; type E, an oblique fracture distal to the mid- point of the trochlear notch (Schatz- ker states that this pattern requires one or two interfragmentary lag screws and a 3.5-mm dynamic- compression plate rather than a one-third tubular plate, which is not strong enough to resist the tor- sional forces); type F, a fracture of the olecranon with associated radial head fracture, which is frequently associated with a rupture of the medial collateral ligament. No single classification system is universally applicable, and any clas- sification is subject to interobserver variability. However, a working knowledge of the existing classifica- tion systems is essential in assessing fractures radiographically and se- Olecranon Fractures Journal of the American Academy of Orthopaedic Surgeons 268 lecting appropriate treatment. For these purposes, the Schatzker classi- fication may be the most useful to the practicing orthopaedist. Diagnostic Evaluation Most olecranon fractures are isolated injuries. When present, concomi- tant injury most often involves the ipsilateral extremity. A careful ex- amination of the upper extremity, including the clavicle, shoulder, hu- merus, elbow, forearm, wrist, and hand, is essential. The elbow typi- cally has both soft-tissue swelling and an effusion. The subcutaneous location of the fracture often makes it easily palpable, with a depression present when the fracture is signifi- cantly displaced. The skin should be carefully inspected for an open fracture. Function of the median, ulnar, radial, and posterior interos- seous nerves should be examined. The presence of radial and ulnar pulses should be documented. Standard anteroposterior and lateral radiographs of the elbow are sufficient for evaluation of isolated olecranon fractures. Direct supervi- sion of the radiographs may be nec- essary to ensure that true antero- posterior and lateral radiographs are obtained. A radiocapitellar view may be helpful for delineation of radial head or capitellar shear fractures. Treatment Options The goals of olecranon fracture treatment include anatomic recon- struction of the articular surface, preservation of motor power, res- toration of stability, prevention of joint stiffness, and minimization of morbidity. 16 Nonoperative Treatment Nondisplaced fractures in which the elbow extensor mechanism is in- tact may be treated nonoperatively. Controversy exists about the amount of acceptable articular displacement. Although immobilization in full extension may improve fracture reduction, it often results in dimin- ished flexion. Immobilization of the elbow in 45 to 90 degrees of flexion for approximately 3 weeks has been recommended for nondisplaced fractures. 7 Motion is then begun, limiting flexion to 90 degrees until there is radiographic evidence of fracture healing. Operative Treatment The ideal construct for fixation of olecranon fractures has been the subject of considerable research. Tension-band wiring, as recom- mended in the AO manual, is designed to convert the tensile dis- traction force of the triceps into a compressive force at the articular surface. 13 Rowland and Burkhart 17 recom- mended modification of the stan- dard AO technique to minimize the possibility that the articular fracture surface may not be adequately com- pressed. They argued, on the basis of free-body analysis, that the distal drill hole for the figure-of-eight wire should be placed anterior to the long axis of the ulna rather than through its subcutaneous border, to increase static compression at the articular surface. Roe 18 challenged the math- ematical validity of this technical modification, and Paremain et al 19 failed to demonstrate an increase in static resistance to gap formation at the fracture site when the proposed modification was used. Several studies have tested fixa- tion strength of olecranon fractures in vitro. Prayson et al 20 tested four different tension-band constructs in simulated transverse fractures. They demonstrated that bicortical Kirschner-wire purchase and braided cable reduced fracture displacement more than traditional intramedul- lary Kirschner wires and monofila- ment figure-of-eight wire. Horner et al 21 reported on a ca- daveric study of 10 oblique distal olecranon fractures. They found that fixation with a one-third tubu- lar plate was approximately three times more rigid than tension-band wiring in resisting the deforming forces of the biceps and brachialis. Fyfe et al 22 assessed movement at the fracture site in cadaveric elbows with transverse, oblique, and commi- nuted olecranon osteotomies tested by slow loading with the elbow in 90 degrees of flexion. Transverse osteotomies were most rigidly fixed with a tension-band wire construct with two tightening knots. Oblique Figure 1 Schatzker classification of olecranon fractures. (Adapted with permission from Browner BD, Jupiter JB, Levine AM, Trafton PG [eds]: Skeletal Trauma. Philadelphia: WB Saunders, 1992, p 1137.) A: Transverse B: Transverse-impacted C: Oblique D: Comminuted E: Oblique-distal F: Fracture-dislocation David J. Hak, MD, and Gregory J. Golladay, MD Vol 8, No 4, July/August 2000 269 osteotomies were fixed equally well with either a tension-band plate or a one-third tubular plate. In commi- nuted osteotomies, plate fixation was found to be slightly more rigid than fixation with a tension band. Fixa- tion with a cancellous screw was found to be erratic, depending on the match between screw diameter and medullary canal size. Augmen- tation with a single figure-of-eight wire improved screw fixation. The validity of this study is compro- mised by the fact that specimens were tested with more than one technique, and slow loading (rather than the more physiologic rapid loading) was used. Murphy et al 23 tested the fixation strength of transverse olecranon osteotomy at the midpoint of the semilunar notch in fresh cadaver specimens by rapid loading to fail- ure. An intramedullary screw plus a tension-band wire was found to have the greatest energy to failure. The authors described the modes of failure of the four methods of fixa- tion they tested. The figure-of-eight wire failed by breakage at the tight- ening loop. The cancellous screw pulled out or bent. AO tension-band wiring failed because of pullout or breakage of the Steinmann pins. The screw-and-wire combination failed by wire displacement and screw breakage. Surgical Techniques The patient is commonly posi- tioned supine with the arm draped across the chest or supported on an arm holder placed across the chest. Alternatively, a lateral decubitus or prone position may be used with the arm draped over a well-padded support. Either general or regional anesthesia (Bier block or axillary block) may be utilized. With the tourniquet applied high on the upper arm, the olecranon is approached through a posterior incision. Some authors recommend a curvilinear incision to avoid plac- ing a scar over the tip of the olecra- non. On the medial side, the mus- cular origin of the flexor digitorum profundus, flexor digitorum super- ficialis, and deep head of the prona- tor teres may be elevated if neces- sary. The location of the ulnar nerve can usually be identified by palpation. Rarely is it necessary to isolate or transpose the ulnar nerve. The fracture site is cleared of hema- toma, and the periosteum is elevated approximately 2 mm from the edges of the fracture. The fracture is reduced and held with a tenacu- lum. Placement of a small oblique drill hole in the ulnar shaft distal to the fracture will aid in anchoring the distal tine of the tenaculum. Fixation alternatives include ten- sion-band wire fixation with Kirsch- ner wires or in combination with an intramedullary screw, intramed- ullary screw fixation alone, or plate fixation. Separate interfragmentary compression screws may be re- quired for certain fracture patterns. Occasionally, excision of the frag- ments and advancement of the tri- ceps may be indicated. After internal fixation is com- pleted, the elbow should be taken through a range of motion to con- firm stability and guide postopera- tive rehabilitation. Pronation and supination should be examined to confirm that there is no blockage due to malpositioned hardware. Tension-Band Wiring Tension-band wire fixation can be effectively utilized for most sim- ple noncomminuted transverse olec- ranon fractures. The tension-band technique converts the extensor force of the triceps to a dynamic compressive force along the articu- lar surface (Fig. 2). 14 Figure 2 Tension-band wire fixation of a transverse olecranon fracture. Static compres- sion is achieved dorsally (paired thin arrows). The extensor force of the triceps (single thick arrow) is converted into dynamic compression just below the articular surface (paired thick arrows). (Adapted with permission from Müller ME, Allgöwer M, Schneider R, Willenegger H [eds]: Manual of Internal Fixation: Techniques Recommended by the AO-ASIF Group, 3rd ed. Berlin: Springer-Verlag, 1991, p 19.) Olecranon Fractures Journal of the American Academy of Orthopaedic Surgeons 270 Several technical tips are helpful in achieving optimal results with the tension-band wire technique. One-point 6-mm Kirschner wires are utilized, as their ends can be easily bent. Some surgeons prefer to place the Kirschner wires in the intramedullary canal; others prefer to angle the wires volarly, engaging the anterior cortex to provide greater resistance to wire migration. The most important factor in pre- venting wire migration is ensuring that the bent proximal end of the wire is buried beneath the fibers of the triceps. If the anterior cortex is engaged, care should be taken to avoid overpenetration of the wires, as they may cause neurovascular damage, restrict forearm rotation, or incite heterotopic ossification or radioulnar synostosis. Full prona- tion and supination should be ensured after the wires have been inserted. The length of the wire should be noted at the point where it engages the second cortex. Once the wire penetrates the far cortex, it should be partially backed out and bent 180 degrees at the previously measured position. The excess wire should then be cut off. The fibers of the triceps tendon should be split sharply with a scalpel at the site of the Kirschner wires to allow the cut and bent ends to be impacted against the cortex (Fig. 3, A). If the bent end of the Kirschner wire is left superficial to the triceps fibers, routine postoperative elbow exten- sion may cause the Kirschner wire to back out (Fig. 3, B). An intravenous catheter is uti- lized to pass an 18-gauge wire beneath the fibers of the triceps. The needle and plastic cannula are inserted deep to the triceps tendon, adjacent to the bone, beneath the two Kirschner wires (Fig. 3, C). The insertion needle is removed, leaving the plastic cannula in place. The 18- gauge wire can then be inserted into the end of the plastic cannula, and both cannula and wire are pulled back, passing the wire deep to the triceps fibers. The wire is passed through a transverse drill hole placed distal to the fracture. Two twisted knots are placed in the wire, one radial and one ulnar, and each is tightened to produce symmetric ten- sion at the fracture site. Placing two knots results in more rigid fixation than using a single knot. 22 The ends of the twisted wires are then cut and bent down against the cortex. Following fixation, the elbow should be examined to confirm full range of motion, including prona- tion and supination, and to confirm fixation stability. Plain radio- graphs should be obtained in the operating room. It is important to confirm that the tension-band wire is properly looped proximally around the Kirschner wires, as oc- casionally the wire may be passed dorsal to one or both of the wires and may engage only the triceps tendon. Wolfgang et al 11 treated 45 frac- tures with tension-band wiring with or without supplemental fixa- tion, depending on the fracture configuration. Excellent or good results were reported in 98% of cases. Tension-band wiring both with Kirschner wires engaging the anterior cortex and with use of a double-loop 18-gauge figure-of- eight wire is adequate for all sim- ple transverse fractures for which internal fixation is chosen. Braided cable has been shown in a cadaveric model to be stronger than mono- filament wire; however, it may fray and increase the risk of sympto- matic hardware prominence. 20 An interfragmentary lag screw is use- Figure 3 Technique for tension-band wiring. A, Fibers of the triceps tendon should be split to allow the bent end of the Kirschner wires to be impacted firmly against bone. B, If the ends of the Kirschner wires are left superficial to the triceps tendon, elbow extension may cause migration or fatigue failure of the Kirschner wires. C, A 16-gauge or larger intravenous catheter is used to pass the tension-band wire deep to the triceps fibers. A B C David J. Hak, MD, and Gregory J. Golladay, MD Vol 8, No 4, July/August 2000 271 ful when an oblique fracture plane is present. Intramedullary Screw Fixation The use of a single large-diameter intramedullary cancellous screw has also been advocated. 13 In one study, a higher rate of fixation loss was reported after intramedullary screw fixation alone compared with tension-band wiring. 24 Some authors recommend supplementa- tion of intramedullary screw fixa- tion with a tension-band wire around the screw head. 23,25 In the frontal plane, there is approximately 4 degrees of valgus angulation of the ulnar shaft with respect to the sigmoid notch. If an intramedullary screw is used, care must be exer- cised to properly place the screw along the intramedullary shaft axis and thus avoid displacement of the fracture (Fig. 4). Johnson et al 26 reported good re- sults in 24 patients treated with a 6.5-mm cancellous screw with or without supplementary tension banding. Sixteen patients (67%) had motion within 15 degrees of full range by postoperative week 9. The authors reported four operative complications: one poor reduction, one loss of compression, one fixa- tion failure, and one bent screw. For large displaced fractures or osteotomy fixation, Wadsworth 27 recommended use of a partially threaded intramedullary screw with or without a washer. He reported 100% union in six patients and no complications with this technique and emphasized the importance of early motion. The indications for intramedul- lary screw placement mirror those for tension-band wiring (e.g., sim- ple noncomminuted transverse fractures). An intramedullary screw may be best suited for fixa- tion of an olecranon osteotomy, as predrilling the screw prior to osteotomy helps guide anatomic reduction later. Plate Fixation Plate fixation is most commonly recommended for comminuted frac- tures in which tension-band wire fixation is not feasible. It is also indi- cated for oblique fractures distal to the midpoint of the trochlear notch, fractures that involve the coronoid process, and those associated with Monteggia fracture-dislocations of the elbow. 15,28 Oblique fractures are best treated with one or two inter- fragmentary compression screws in conjunction with plate fixation to resist torsional forces. 15 Some authors have reported use of one-third tubular, dynamic com- pression, and pelvic reconstruction plates for fixation of comminuted olecranon fractures. The proximal end of the one-third tubular plate can be modified to make a hook- plate that will provide additional fixation for small proximal fracture fragments. In severely comminuted fractures, one-third tubular plates may not provide sufficient strength, leading to hardware fatigue failure. 28 While the subcutaneous location of the hardware raises concern about prominence necessitating subsequent removal, the frequency of hardware prominence may actu- ally be higher after tension-band wiring than after plate fixation. 8 In most cases, the plate is placed along the dorsal surface of the olec- ranon and contoured around the tip of the olecranon (Fig. 5). The dorsal ulna is the tension side of the bone and as such is biomechan- ically best suited to plating. A screw placed in the most proximal hole may either engage the coro- noid process or be inserted down the intramedullary canal. If this screw is intramedullary, the other screws in the plate must be angled slightly radially or ulnarly. In a prospective, randomized study of 41 patients with displaced fractures treated with either tension- band wiring or one-third tubular plates, tension-band wiring was more likely to result in either articu- lar incongruity greater than or equal to 2 mm or loss of reduction. The authors concluded that strong con- sideration should be given to plating of olecranon fractures. 8 Simpson et al 28 reported 73% good or excellent results in a retrospective study of the use of a dorsally applied 3.5-mm limited-contact dynamic-compression (LC-DC) plate for fixation of 13 com- plex proximal ulna fractures and 24 Figure 4 A, Proper placement of an intramedullary screw. B, Placement of an intramedullary screw slightly off the intramedullary axis results in fracture malreduction. A B Figure 5 Plate fixation of a comminuted olecranon fracture. (Adapted with permis- sion from Mast JW, Jakob R, Ganz R: Planning and Reduction Techniques in Fracture Surgery. Berlin: Springer-Verlag, 1989.) Olecranon Fractures Journal of the American Academy of Orthopaedic Surgeons 272 Monteggia fracture-dislocations. Only one patient had problems with hardware prominence. They recom- mend plate fixation if the fracture ex- tends to the metaphyseal-diaphyseal junction or if the coronoid process is involved. The LC-DC plate is lower in profile and easier to contour, and its screw holes allow greater screw angulation than those of the standard dynamic-compression plate. In severely comminuted frac- tures, care must be taken not to narrow the olecranon-to-coronoid distance. 16 Because there is no ar- ticular cartilage in the midportion of the sigmoid notch, aligning the remaining articular surfaces in comminuted fractures will result in narrowing of the olecranon-to- coronoid distance. 4,5 The dorsal cortical fragments may serve as a guide to reconstruct the correct anatomic alignment. Use of an AO universal distractor may aid in re- duction and provisional stabiliza- tion. 28 The congruency of the artic- ular surface should be meticulously restored. Bone graft should be uti- lized to support the articular sur- face after elevation of depressed fragments. Supplemental Kirschner- wire fixation may also be required. Plate fixation is appropriate for severely comminuted fractures, dis- tal fractures involving the coronoid process, oblique fractures distal to the midpoint of the trochlear notch, Monteggia fracture-dislocations of the elbow, and nonunions. In com- minuted fractures and nonunions, a dorsally applied LC-DC plate with supplemental bone graft should be utilized to support comminuted depressed articular fragments that have been elevated. A one-third tubular hook-plate can be used to achieve additional fixation of the olecranon tip for fractures with a small proximal fragment. A portion of the triceps insertion may need to be incised to allow apposition of the plate to the bone. Excision of Fragment and Triceps Advancement Excision of the fracture fragment and reattachment of the triceps ten- don may be indicated in a select group of elderly patients with os- teoporotic bone in whom the olec- ranon fracture fragments involve less than 50% of the joint surface 13,29 and are too small or too comminuted for successful internal fixation. The integrity of the medial collateral lig- ament, the interosseous membrane, and the distal radioulnar joint must be established before consideration is given to excision; otherwise, instability will result. 30 The triceps tendon is reattached with nonab- sorbable sutures that are passed through the drill holes in the proxi- mal ulna. Cabanela and Morrey 16 recommend that the triceps be re- attached adjacent to the remaining articular surface, creating a sling for the trochlea (Fig. 6). Triceps reat- tachment in this manner creates a smooth, congruent transition from the triceps tendon to the articular cartilage of the olecranon but de- creases the moment arm and may result in greater extensor weakness. McKeever and Buck 29 stated that as much as 80% of the trochlear notch can be excised without com- promising elbow stability, provided the coronoid and distal trochlea are preserved. Gartsman et al 12 re- ported one case of anterior instabil- ity in a patient in whom approxi- mately 75% of the articular surface had been excised. An et al 3 evalu- ated elbow stability with varying degrees of proximal ulnar resection in vitro. They found linear de- creases in elbow constraint with increasing amounts of resection and suggested that resection of more than 50% of the articular sur- face may result in instability. Inhofe and Howard 31 reported good or excellent results in 11 of 12 patients with adequate follow-up after excision of as much as 70% of the articular surface. Gartsman et al 12 reported lower complication and reoperation rates following excision compared with internal fixation and concluded that excision is the pre- ferred treatment alternative provided the coronoid process remains intact. Although weakening of the extensor apparatus has been a criticism of the technique of fragment excision and triceps advancement, Gartsman et al found no differences in isometric strength between patients treated by excision and those treated with inter- nal fixation. Although that series did provide some comparison between excision and internal fixation, the treatment was not randomized, and selection bias requires cautious inter- pretation of the conclusions. Other authors have recommended excision only as a last resort in cases in which open reduction and internal fixation is not possible. 10,11 Excision should be reserved for selected cases in which open reduc- tion seems unlikely to be successful. Open reduction and internal fixa- tion should be attempted in most cases, as it permits early motion and allows bone-to-bone healing. Excision and triceps advancement can still be performed as a salvage procedure if internal fixation fails. Figure 6 When excision and triceps advancement is performed, the triceps should be attached adjacent to the articular surface. (Adapted with permission from Cabanela ME, Morrey BF: Fractures of the proximal ulna and olecranon, in Morrey BF [ed]: The Elbow and Its Disorders, 2nd ed. Philadelphia: WB Saunders, 1993, p 416.) David J. Hak, MD, and Gregory J. Golladay, MD Vol 8, No 4, July/August 2000 273 Postoperative Rehabilitation Operative management of olecra- non fractures should provide suffi- ciently stable fixation to allow early motion. The ideal time to start motion has not been addressed in any prospective study; therefore, the surgeon must consider fixation stability, patient compliance, and wound healing. Patients typically are placed in a posterior splint or sling, and active motion is instituted as early as postoperative day 1. 11 Immediate supervised gravity- assisted range-of-motion exercises are effective for all fractures with stable internal fixation. 28 Unless there are wound-healing problems, a removable posterior splint is ap- plied, and the patient is instructed in range-of-motion exercises on the first postoperative day. More com- plex or comminuted fractures may require longer periods of immobi- lization, and more stiffness can be anticipated. Distraction devices may be helpful in the postoperative regimen for unstable, comminuted fractures, followed by the use of adjustable splints to help regain motion. 13 Muscle strengthening is begun when bone healing is ade- quate, generally 6 weeks from surgery. 28 Patients may return to work involving rigorous use of the extremity at 3 to 4 months. Complications Hardware prominence requiring removal is one of the most frequent complications after internal fixation of olecranon fractures. Symptoms due to hardware prominence have been reported in 3% to 80% of cases. 25,28 The wide range of symp- tomatic hardware prominence is likely related both to technical fac- tors and to varying definitions of prominence. Wire migration, soft- tissue irritation, olecranon bursitis, wire breakage, or fracture displace- ment may occur with tension-band wiring. 32 Kirschner wires should be firmly seated against the olecra- non through slits in the triceps ten- don, and the wire knots should be positioned away from the subcuta- neous border of the ulna. Preopera- tive patient counseling should include the possibility of sympto- matic hardware prominence and the potential need for hardware re- moval. Hume and Wiss 8 reported a higher incidence of painful hard- ware prominence after tension- band wiring than after compression plating. No cases of symptomatic hardware prominence were reported by Simpson et al 28 after LC-DC plat- ing. Hardware failure or loss of fix- ation occurs more commonly in comminuted fractures and in pa- tients with poor bone stock. Loss of motion is a common problem after fractures about the elbow but is rarely significant in patients with isolated olecranon fractures. 8,11 Patients with isolated injuries typically lose 10 to 15 de- grees of extension. 13 However, in patients with associated fractures of the radial head, capitellum, or coro- noid or with a Monteggia fracture- dislocation, the range of motion may be more severely compro- mised. 27,30 Comminuted fractures and open injuries are also more likely to result in stiffness. Some gains in motion may be achieved with ag- gressive physical therapy. Patients with a functional deficit related to stiffness may be treated with pro- gressive splinting, a turnbuckle- type brace, or capsulectomy. Heterotopic ossification may oc- cur after olecranon fractures, par- ticularly in patients with significant associated soft-tissue injuries and in those with closed head injuries. Wolfgang et al 11 reported a 13% rate of heterotopic ossification, mainly in patients with an associated radial head dislocation. Simpson et al 28 reported a 14% rate of heterotopic ossification in 27 complex fractures of the proximal ulna. Ilahi et al 33 found a 0% incidence of grade II, III, or IV heterotopic ossification about the elbow when unstable elbow fractures were operated on within 48 hours of injury, com- pared with a 33% incidence when the delay between injury and surgi- cal treatment exceeded 48 hours. Nonunion of olecranon fractures is infrequent, and patients typically present with pain, instability, or loss of motion. Papagelopoulos and Morrey 34 reported only two nonunions in 196 fractures initially treated at the Mayo Clinic over a 10-year period. Treatment options for nonunions include excision, osteosynthesis with a compression plate or lag screw, or elbow arthro- plasty in cases of severe posttrau- matic arthritis. Cancellous bone graft or a corticocancellous bone plate fixed with screws may be use- ful. Papagelopoulos and Morrey also reported on the treatment of 24 patients with olecranon nonunion, most of whom had been referred from other institutions. After use of one or more of the treatment options mentioned, the results were excellent in 12 patients (50%), good in 4 (17%), fair in 6 (25%), and poor in 2 (8%). In another study, Danziger and Healy 35 reported that union was achieved in all five cases treated by either tension-band plat- ing or wiring and bone graft. Poorer results have been reported with intra-articular step-off of more than 2 mm, but few studies have sufficient follow-up to document the long-term incidence of post- traumatic arthrosis. 25 Gartsman et al 12 reported a 20% rate of arthrosis following olecranon fractures regardless of whether they were treated by excision or internal fixa- tion. The reported rates of infection after operative olecranon fracture treatment range from 0% to 6%. 12,34 Papagelopoulos and Morrey 34 re- Olecranon Fractures Journal of the American Academy of Orthopaedic Surgeons 274 ported two cases of transient reflex sympathetic dystrophy in their series of 24 olecranon nonunions. Ulnar neurapraxia has been report- ed in 2% to 5% of cases. Ulnar neu- ritis may occasionally occur. 10 Symptoms usually resolve with conservative treatment, but late neurolysis or transposition may occasionally be required. Outcomes Generally good and excellent re- sults have been reported for all types of olecranon fracture treat- ment. However, prospective stud- ies in which validated outcomes measures were used to compare different forms of treatment have not yet been reported. Murphy et al 25 performed a ret- rospective study of patients with simple transverse or oblique olecra- non fractures, using a 19-point scale to evaluate pain, function, range of motion, and radiographic findings. Of the 33 patients, 10 were treated with an intramedullary screw plus wire, 13 with an intramedullary screw alone, and 10 with tension- band wire fixation. The average ratings for the three groups were similar: 17.7 for intramedullary screw plus wire treatment, 17.2 for intramedullary screw fixation, and 16.7 for tension-band wire fixation. Gartsman et al 12 reported a ret- rospective review of a series of 107 isolated olecranon fractures in 53 patients treated by excision and 54 treated by internal fixation (primar- ily screw or tension-band fixation). Pain, subjective function, isometric strength, isokinetic work, range of motion, stability, and the incidence of degenerative changes were simi- lar between the two groups at an average follow-up of 3.6 years. Thirteen patients who underwent internal fixation had complications, compared with only 2 in the exci- sion group. Thirteen patients in the internal fixation group required reoperation for hardware removal. Because of the high rates of compli- cations and reoperation after inter- nal fixation, the authors concluded that excision is the preferred treat- ment alternative provided the coro- noid process is intact. Summary Fractures of the olecranon process may present in isolation or in combi- nation with more complex injuries about the elbow. Generally good or excellent results have been reported with all forms of treatment of simple olecranon fractures. Nondisplaced fractures may be treated nonopera- tively. Excision and triceps advance- ment may be indicated for patients with small extra-articular fragments or severely comminuted fractures, as well as for elderly patients with os- teoporotic bone. Open reduction and internal fixation is preferred for displaced intra-articular fractures. Tension-band wire fixation is com- monly utilized for simple fractures. Plate fixation is recommended for comminuted or unstable oblique fractures. Intramedullary screw fixa- tion has also been recommended. Because of its subcutaneous location, hardware may need to be removed after union is achieved. Although minor losses of motion are common, most patients can be expected to have good results provided early controlled motion can be instituted. References 1. Cohen MS, Hastings H II: Rotatory instability of the elbow: The anatomy and role of the lateral stabilizers. J Bone Joint Surg Am 1997;79:225-233. 2. O’Driscoll SW, Bell DF, Morrey BF: Posterolateral rotatory instability of the elbow. J Bone Joint Surg Am 1991; 73:440-446. 3. An KN, Morrey BF, Chao EYS: The effect of partial removal of proximal ulna on elbow constraint. Clin Orthop 1986;209:270-279. 4. Shiba R, Sorbie C, Siu DW, Bryant JT, Cooke TDV, Wevers HW: Geometry of the humeroulnar joint. J Orthop Res 1988;6:897-906. 5. Stormont TJ, An KN, Morrey BF, Chao EY: Elbow joint contact study: Com- parison of techniques. J Biomech 1985; 18:329-336. 6. Cage DJN, Abrams RA, Callahan JJ, Botte MJ: Soft tissue attachments of the ulnar coronoid process: An ana- tomical study with radiographic corre- lation. Clin Orthop 1995;320:154-158. 7. Hotchkiss RN: Fractures of the olecra- non, in Rockwood CA Jr, Green DP, Bucholz RW, Heckman JD (eds): Rock- wood and Green’s Fractures in Adults, 4th ed. Philadelphia: JB Lippincott, 1996, pp 984-996. 8. Hume MC, Wiss DA: Olecranon frac- tures: A clinical and radiographic com- parison of tension band wiring and plate fixation. Clin Orthop 1992;285: 229-235. 9. Macko D, Szabo RM: Complications of tension-band wiring of olecranon fractures. J Bone Joint Surg Am 1985; 67:1396-1401. 10. Rettig AC, Waugh TR, Evanski PM: Fracture of the olecranon: A problem of management. J Trauma 1979;19:23-28. 11. Wolfgang G, Burke F, Bush D, et al: Surgical treatment of displaced olecra- non fractures by tension band wiring technique. Clin Orthop 1987;224:192-204. 12. Gartsman GM, Sculco TP, Otis JC: Operative treatment of olecranon frac- tures: Excision or open reduction with internal fixation. J Bone Joint Surg Am 1981;63:718-721. 13. Morrey BF: Current concepts in the treatment of fractures of the radial head, the olecranon, and the coronoid. J Bone Joint Surg Am 1995;77:316-327. 14. Müller ME, Allgöwer M, Schneider R, Willenegger H (eds): Manual of Inter- nal Fixation: Techniques Recommended by the AO-ASIF Group, 3rd ed. Berlin: Springer-Verlag, 1991. 15. Schatzker J: Fractures of the olecranon, in Schatzker J, Tile M (eds): The Ra- tionale of Operative Fracture Care. Berlin: Springer-Verlag, 1987, pp 89-95. 16. Cabanela ME, Morrey BF: Fractures of David J. Hak, MD, and Gregory J. Golladay, MD Vol 8, No 4, July/August 2000 275 the proximal ulna and olecranon, in Morrey BF (ed): The Elbow and Its Disorders, 2nd ed. Philadelphia: WB Saunders, 1993, pp 405-428. 17. Rowland SA, Burkhart SS: Tension band wiring of olecranon fractures: A modification of the AO technique. Clin Orthop 1992;277:238-242. 18. Roe SC: Tension band wiring of olec- ranon fractures: A modification of the AO technique [letter]. Clin Orthop 1994;308:284-286. 19. Paremain GP, Novak VP, Jinnah RH, Belkoff SM: Biomechanical evaluation of tension band placement for the repair of olecranon fractures. Clin Orthop 1997;335:325-330. 20. Prayson MJ, Williams JL, Marshall MP, Scilaris TA, Lingenfelter EJ: Bio- mechanical comparison of fixation methods in transverse olecranon frac- tures: A cadaveric study. J Orthop Trauma 1997;11:565-572. 21. Horner SR, Sadasivan KK, Lipka JM, Saha S: Analysis of mechanical factors affecting fixation of olecranon frac- tures. Orthopedics 1989;12:1469-1472. 22. Fyfe IS, Mossad MM, Holdsworth BJ: Methods of fixation of olecranon frac- tures: An experimental mechanical study. J Bone Joint Surg Br 1985;67:367-372. 23. Murphy DF, Greene WB, Gilbert JA, Dameron TB Jr: Displaced olecranon fractures in adults: Biomechanical analysis of fixation methods. Clin Orthop 1987;224:210-214. 24. Helm RH, Hornby R, Miller SWM: The complications of surgical treat- ment of displaced fractures of the olec- ranon. Injury 1987;18:48-50. 25. Murphy DF, Greene WB, Dameron TB Jr: Displaced olecranon fractures in adults: Clinical evaluation. Clin Orthop 1987;224:215-223. 26. Johnson RP, Roetker A, Schwab JP: Olecranon fractures treated with AO screw and tension bands. Orthopedics 1986;9:66-68. 27. Wadsworth TG: Screw fixation of the olecranon after fracture or osteotomy. Clin Orthop 1976;119:197-201. 28. Simpson NS, Goodman LA, Jupiter JB: Contoured LCDC plating of the proxi- mal ulna. Injury 1996;27:411-417. 29. McKeever FM, Buck RM: Fracture of the olecranon process of the ulna: Treatment by excision of fragment and repair of triceps tendon. JAMA 1947; 135:1-5. 30. Teasdall R, Savoie FH, Hughes JL: Comminuted fractures of the proximal radius and ulna. Clin Orthop 1993;292: 37-47. 31. Inhofe PD, Howard TC: The treatment of olecranon fractures by excision of fragments and repair of the extensor mechanism: Historical review and report of 12 fractures. Orthopedics 1993;16:1313-1317. 32. Horne JG, Tanzer TL: Olecranon frac- tures: A review of 100 cases. J Trauma 1981;21:469-472. 33. Ilahi OA, Strausser DW, Gabel GT: Post- traumatic heterotopic ossification about the elbow. Orthopedics 1998;21:265-268. 34. Papagelopoulos PJ, Morrey BF: Treat- ment of nonunion of olecranon fractures. J Bone Joint Surg Br 1994;76:627-635. 35. Danziger MB, Healy WL: Operative treatment of olecranon nonunion. J Orthop Trauma 1992;6:290-293.