Contemporary systems for the classification of olecranon fractures consistently include a category described as fracture-dislocations of the olecranon,7,17,22 but few published reports specifically address these injuries.1,2,19 Consequently, the characteristics of these injuries, the difficulties in treatment, and the results of treatment remain poorly defined. Studies describing fracture-dislocations do not consistently include these injuries,5,8,11,12,21,24 and some studies specifically exclude them.5,12 Some authors have described complex fractures of the proximal ulna,2,13,19,23 but these may not be readily recognized or accepted as olecranon fractures or fracture-dislocations.
We applied a specific definition of olecranon fracture-dislocations based on accepted classification schemes to consecutive series of patients with traumatic elbow instability to better characterize these injuries, the complications and pitfalls of treatment, and the functional results according to the Broberg and Morrey4 rating system and the American Shoulder and Elbow Surgeons elbow evaluation method.14
MATERIALS AND METHODS
Billing records were used to identify all patients with a fracture or dislocation of the elbow treated at two Level 1 trauma centers (Massachusetts General Hospital and Beth Israel-Deaconess Medical Center, Boston, MA) during a 7-year period. Retrospective review of medical records and radiographs identified all patients with disruption of the normal alignment between the distal humerus and the proximal radius and ulna. Loss of this alignment was defined on the radiographs taken at the time of injury as either a change in the normal alignment of the articular surfaces of the trochlea of the distal humerus and the coronoid process of the ulna, or the radial head and the capitellum, or both. Thirty consecutive skeletally mature patients meeting these criteria also had a fracture of the olecranon process of the ulna. Twenty-six patients were contacted and returned for a free examination and radiographs under a protocol approved by the Human Research Committee (Tables 1, 2). Two additional patients had died, one declined to participate, and one could not be located.
There were 15 men and 11 women with an average age of 49 years (range, 20-82 years). The right arm was involved in 17 patients and the left arm was involved in nine patients. The dominant arm was injured in 17 patients.
Seven patients were injured in a fall from a standing height and 19 were injured in a higher-energy accident: seven were injured in a fall from a substantial height, six were injured in motor vehicle accidents, four were injured in a fall down stairs, one was injured in a bicycle accident, and one was injured in a sports accident.
Four patients had open wounds. These were rated according to the system of Gustilo and Anderson9 and Gustilo et al10 as Type 2 in two patients and Type 3A in two patients. Three patients had ipsilateral skeletal injuries including a fracture of the distal radius in two patients and a shoulder dislocation in one patient. Two patients had contralateral upper limb injuries (one fracture each of the distal humerus and radial head). Eight patients had multiple injuries involving other organ systems. Two patients had ipsilateral nerve injuries at the time of the accident. These involved the brachial plexus in one patient and the ulnar nerve in the other patient.
There were two types of olecranon fracture-dislocations: 10 patients had an anterior, or transolecranon2,19 fracture-dislocation of the elbow (Table 1; Figs 1-3), and 16 patients had a posterior olecranon fracture-dislocation (Table 2; Figs 4, 5), which has been included in the spectrum of posterior Monteggia fractures.11,13,20
Ten patients had a fracture of the olecranon with anterior displacement of the radius and ulna from their normal relationships with the trochlea: an anterior, or transolecranon fracture-dislocation of the elbow.2,19 There were seven men and three women with an average age of 42 years (range, 20-72 years). Two injuries were the result of a fall from a standing height and the remaining eight injuries involved a higher-energy injury mechanism: four motor vehicle accidents, two falls from a substantial height, one snowmobile accident, and one bicycle accident. Three patients had open wounds: one Grade 3A and two Grade 2 according to the system of Gustilo and Anderson9 and Gustilo et al.10 These Thesthree patients and two additional patients had multiple traumatic injuries. One patient had ulnar nerve palsy.
In all 10 patients the articular surfaces, although disrupted, remained apposed; there was not a true dislocation of the ulnohumeral joint. The fracture was complex and multifragmented in nine patients and was a simple transverse fracture in one patient. Among the nine patients with multifragmented fractures, five had an associated fracture of the coronoid process. In each of these five patients the coronoid process was separated from the proximal ulna as one large fragment representing nearly the entire coronoid process (Type 3 according to the system of Regan and Morrey18). All nine of the multifragmented fractures were characterized by fragmentation in the depths of the trochlear notch between the olecranon and coronoid articular facets. This probably reflects impaction of the trochlea through the trochlear notch of the ulna. In three patients, this fragmentation extended into the ulnar diaphysis. One patient had a small (less than 20% of the articular surface) marginal fracture of the radial head (Type 1 according to Mason15).
Operative treatment was done through a dorsal longitudinal skin incision in each case. The fracture was secured with a 3.5-mm plate and screws (Synthes Ltd, Paoli, PA) in seven patients (Fig 1), a tension band-wire construct in two patients, and a tension suture construct in one patient. A limited-contact dynamic compression plate was used in five patients, a dynamic compression plate was used in one patient, and two stacked ⅓ tubular plates with a supplementary tension band-wire construct was used in one patient. The sizes of the plates used averaged 10 holes (range, 8-12 holes). Loosening of the tension band-wire construct occurred in one patient and was revised to fixation with two adjacent 3.5-mm pelvic reconstruction plates (each seven holes in length) 17 days after the initial operation (Fig 2).
Four of the five fractures of the coronoid were realigned and secured with one or two screws entering through the dorsal aspect of the ulna, either through the plate or independent of it. The one concomitant radial head fracture was treated by excision of the fragment. All patients were treated with a postoperative dressing that incorporated a posterior plaster splint with the ulnohumeral joint held in 90° flexion and the forearm in neutral rotation.
Sixteen patients had an apex posterior fracture of the olecranon with posterior displacement of the radial head, a posterior olecranon-fracture dislocation. Only one patient had dislocation of the ulnohumeral joint (loss of apposition of the articular surfaces) seen on the initial radiographs taken after the injury (Fig 4). In the other 15 patients, the articular surfaces remained apposed (Fig 5).
There were eight men and eight women with an average age of 53 years (range, 21-82 years). The average age of the eight women was 67 years, compared with 39 years in men. Five of these injuries were the result of a fall from a standing height and 11 were caused by a higher-energy accident, including five falls from a substantial height, four falls down stairs, and one motor vehicle accident. One patient had an open wound (Type 3A according to Gustilo and Anderson9 and Gustilo et al.10 Two patients had an ipsilateral fracture of the distal radius and one had an ipsilateral dislocation of the shoulder. The patient with a shoulder dislocation had a brachial plexus injury. Two patients had skeletal injuries of the contralateral upper extremity; including one radial head fracture and one distal humerus fracture. Five patients had multiple traumatic injuries.
All 16 patients had a fracture of the coronoid process. According to the system of Regan and Morrey,18 fractures of the coronoid process were classified as Type 3 (a fragment involving more than 50% of the coronoid process) in 15 patients and Type 2 (a fragment involving 50% of the process or less) in one patient. The radial head was fractured in 13 of 16 patients. The fractures of the radial head were classified as Type 2 (involving part of the head) in three patients, and Type 3 (a comminuted fracture involving the entire head) in 10 patients according to the system of Mason.15
A dorsal midline skin incision was used. Exposure of the proximal ulna and radius were achieved through separate muscular intervals in all but one patient in whom a Boyd-type simultaneous exposure of the proximal ulna and radius was done.3 In 15 patients, the fractures of the ulna were secured with 3.5-mm plates and screws (Synthes Ltd), and in one patient a tension band construct was used. Limited-contact dynamic compression plates were used in 11 patients, dynamic compression plates were used in two patients, and a reconstruction plate was used in one patient. The average length of the plates was nine holes (range, seven-11 holes). All but one of the plates were applied to the dorsal surface of the proximal ulna, and the reconstruction plate was applied to the lateral surface. All of the Regan and Morrey18 Type 3 fractures were realigned and secured using one or two screws entering the dorsal surface of the proximal ulna either through a hole in the plate or independent of it. The Type 2 fracture of the coronoid was not secured. Fractures of the radial head were treated with open reduction and internal fixation with plates and/or screws in four patients, excision or partial excision of the radial head in five patients, and were not specifically addressed in four patients. The lateral collateral ligament was reattached to the lateral epicondyle in two patients.
Ulnohumeral instability was seen intraoperatively in two patients and these two elbows were immobilized in a cast at 90° elbow flexion for 1 month (Fig 5). The remaining 15 patients were treated with a postoperative dressing that incorporated a posterior plaster splint with the ulnohumeral joint held in 90° flexion and the forearm in neutral rotation.
Twenty surgeons cared for these patients, and no standard postoperative protocols were used. The splint was removed sometime between the next day and 4 weeks later (average, 10 days), and active-assisted elbow motion was initiated. In 10 patients, a hinged elbow brace was used for 2-6 weeks (average, 4 weeks). Resistive exercises were delayed until healing was established. Passive manipulation of the elbow was not allowed. We encourage active-assisted and functional mobilization of the elbow as soon as possible after surgery-usually the next day. No external immobilization or passive manipulation by the therapist was used, and resistive exercises were delayed until healing was established.
The patients and radiographs were evaluated by one of the authors (DR) at final followup. The authors were not involved in the care of the patients. The final overall results of all patients were rated according to the system of Broberg and Morrey,4 and with the American Shoulder and Elbow Surgeons elbow evaluationmethod.14 According to the 100-point scale of Broberg and Morrey, 95-100 points were considered excellent results; 80-94 points were good results; 60-79 points were fair results; and 0-59 points were poor results. Radiographic signs of arthrosis also were rated according to the system of Broberg and Morrey.4 According to this system, a normal elbow is Grade 0, an elbow with slight joint-space narrowing with minimum osteophyte formation is Grade 1, an elbow with moderate joint-space narrowing and moderate osteophyte formation is Grade 2, and an elbow with severe degenerative change and gross destruction of the joint is Grade 3. No formal evaluation of the reliability of this grading system was done.
Continuous variables, such as degrees of motion, were compared using Student’s t test. Dichotomous variables (eg, complications) were compared using Fisher’s exact test. All statistical analyses were done with SPSS (Chicago, IL).
The patients were evaluated at an average of 6 years (range, 3-10 years) after the injury. Functional motion was obtained on average, but some patients had poor mobility. The flexion and extension arc averaged 103° (range, 40°-40°) with an average arc of flexion of 125° (range, 90°-145°) and an average flexion contracture of 25° (range, 0°-60°). The arc of forearm rotation averaged 130° (range, 0°-180°) with supination averaging 65° (range, 0°-90°) and pronation averaging 65° (range, 0°-90°).
The functional evaluation systems showed a high rate of restoration of good elbow function. The average score according to the system of Broberg and Morrey was 83 points (range, 35-100 points). There were nine excellent, 12 good, one fair, and four poor results. The average score on the American Shoulder and Elbow Surgeons elbow evaluation method was 83 points (range, 28.5-100 points).
Ulnohumeral arthrosis was fairly common and in some cases moderate or severe. Ten patients had radiographic signs of ulnohumeral arthrosis. These were Grade 1 in six patients, Grade 2 in two patients, and Grade 3 in two patients. The two patients with nerve injuries had only partial recovery of motor and sensory functions.
Few complications were encountered among patients with anterior olecranon fracture-dislocations (Table 1). One patient had a capsular contracture of the elbow treated with manipulation under anesthesia 6 months after the injury. The patient with an unfixed large coronoid fracture had malunion, arthrosis, and a stiff elbow (Fig 3). Subsequent elbow capsulectomy was complicated by infection, and the patient had a poor functional result and Grade 3 arthrosis. One patient had heterotopic ossification develop anteriorly in the brachialis that restricted flexion to 110°, but he declined operative intervention.
Patients with anterior olecranon fracture-dislocations were followed up for an average of 7 years (range, 3-10 years). All of the fractures healed within 12 weeks.
Patients with anterior olecranon-fracture dislocations on average obtained greater ulnohumeral motion than patients with posterior olecranon fracture-dislocations, and lost very little forearm rotation. The flexion and extension arc averaged 110° (range, 40°-140°) with an average flexion of 130° (range, 100°-145°) and an average flexion contracture of 20° (range, 0°-60°). The arc of forearm rotation averaged 155° (range, 100°-180°) with an average supination of 75° (range, 40°-90°) and an average pronation of 80° (range, 60°-90°). None of the patients had symptoms or signs of weakness or instability.
Arthrosis was uncommon in patients with anterior olecranon fracture-dislocations and was associated with inadequate treatment of a coronoid fracture. The patient with a malunited coronoid fracture had Grade 3 arthrosis and one additional patient had Grade 1 arthrosis seen on final radiographs. The other eight patients had no radiographic signs of arthrosis.
The average score according to the system of Broberg and Morrey4 was 88 points (range, 42-100 points). There were four excellent and five good results, and one poor result. The average score on the American Shoulder and Elbow Surgeons elbow evaluation method was 89 points (range, 52-100 points).
In two patients, the coronoid healed with slight malposition and widening of the trochlear notch. One of these patients had Grade 2 arthrosis and the other had Grade 3 arthrosis. The only Type 2 coronoid fracture that was not repaired did not heal. Four patients had a proximal radioulnar synostosis develop (Fig 5). One patient had subsequent operative resection 6 months after the injury. This patient and two additional patients had subsequent surgery for elbow capsular release. One of these four patients also had nonunion of the ulnar fracture. The fracture healed after reoperation, but the patient had Grade 3 ulnohumeral arthrosis develop. One patient had a second fall on the surgically treated elbow 7 months after the index fracture-dislocation and had open reduction and internal fixation of a fracture of the distal humerus.
Patients with posterior olecranon fracture-dislocations were followed up for an average of 6 years (range, 3-10 years). The flexion and extension arc averaged 95° (range, 50°-125°) with an average flexion of 125° (range, 90°-145°) and an average flexion contracture of 30° (range, 15°-50°).
Patients with posterior olecranon fracture-dislocations obtained less (p = 0.02) forearm rotation than patients with anterior olecranon fracture-dislocations. The arc of forearm rotation averaged 115° (range, 0°-170°) with an average supination of 55° (range, 0°-90°) and an average pronation of 60° (range, 0°-80°).
One of the patients with malunion of the coronoid and the patient with an ununited fracture of the ulna had Grade 3 arthrosis. Five patients had Grade 1 arthrosis and two patients had Grade 2 arthrosis, including the other patient with malunion of the coronoid. Seven patients had no radiographic signs of arthrosis.
The average score according to the system of Broberg and Morrey4 was 79 points (range, 35-98 points). There were five excellent, seven good, one fair, and three poor results. The average score on the American Shoulder and Elbow Surgeons elbow evaluation method was 78 points (range, 28.5-100 points). The fair result occurred in the patient who required cast immobilization because of ulnohumeral instability and who subsequently had a fracture of the distal humerus of the same elbow. All three patients with poor results had proximal radioulnar synostosis and severe arthrosis (Grade 2 or Grade 3).
Because olecranon fracture-dislocations are relatively rare and not frequently discussed, a retrospective series such as this is useful to evaluate and extend our understanding of these injuries despite the inherent scientific weaknesses of the study method. The clinical and radiologic evaluation systems of Broberg and Morrey4 have not been tested for reliability. The data generated from precise definition and analysis of olecranon fracture-dislocations raise the question of whether these are best considered olecranon fractures because they often are complex fractures of the proximal ulna, and also whether they should be considered dislocations as the articular surfaces nearly always remain apposed.
Limitations of this study include a relatively small numbers of patients, it is a retrospective review, and there was a lack of consistent treatment algorithms. These are the inherent limitations of the study of uncommon traumatic injuries.
Olecranon fracture-dislocations are included in all major classification systems,7,17,22 but little data have been presented, the majority of it reported as an olecranon fracture with anterior displacement of the forearm.1,2,21,24 Heim11 included olecranon fracture-dislocations in his review of elbow fracture-dislocations, whereas some of the most influential and widely quoted studies have specifically excluded them.5,12 Jupiter et al13 and Ring et al20 described posterior olecranon fracture-dislocations as a type of posterior Monteggia injury, whereas others specifically excluded this pattern of injury when considering Monteggia lesions.6 Finally, some of these injuries have been considered complex fractures of the proximal radius and ulna,23 without identifying them as olecranon fracture-dislocations.
The data from our series show that olecranon fractures are not commonly associated with complete loss of apposition of the ulnohumeral articular surfaces. In the current series, a dislocation was defined as a disruption of the alignment between the forearm bones and the humerus. It might be argued that the majority of these injuries represent a subluxation rather than a true dislocation because the articular surfaces of the ulna and distal humerus remain apposed; however, the definition presented is consistent with how olecranon fracture-dislocations have been described, classified, and drawn.7,17,22
An argument in favor of considering these injuries as olecranon fracture subluxations rather than olecranon fracture-dislocations is supported by the observation of relative sparing of the collateral ligament complexes. The instability of these injuries occurs more by virtue of disruption of the trochlear notch than by dislocation of the ulnotrochlear articulation. In anterior fracture-dislocations of the olecranon, neither Ring et al19 nor we, in our current series, saw radiographic or operative indications that the collateral ligaments were ruptured. Secure internal fixation of the fracture fragments, including the coronoid process, immediately restored elbow stability. In posterior olecranon fracture-dislocations, the lateral collateral ligament occasionally is injured, but the medial collateral ligament seems to remain intact. In the current series, the lateral collateral ligament was not evaluated routinely, and no conclusions can be made regarding the prevalence of injury to this ligament. Routine evaluation and repair of the lateral collateral ligament, which nearly always is avulsed from the lateral epicondyle,16 is recommended.
Fracture dislocations of the olecranon often are complex fractures of the proximal ulna or complex combined injuries of the radial head, coronoid, and collateral ligament complexes. Failure to recognize these complex injuries as fracture-dislocations of the olecranon23 may distort our understanding of these injuries. In addition, treatment of these injuries in the same manner as simple olecranon fractures may be problematic. Tension band wiring techniques with Kirschner wires or a screw may provide inadequate stability in the face of greater compromise to elbow stability. Ring et al recommended a plate and screws be applied to the dorsal surface of the proximal ulna and contoured to wrap around the olecranon process to increase the number of screws in the proximal fragments.19,20 Failure to address associated injuries to the coronoid, radial head, or lateral collateral ligament may lead to problems with instability, arthrosis, and loss of motion.20
There are two different patterns of olecranon fracture-dislocation with distinct injury characteristics, pitfalls, and prognoses: anterior and posterior. Anterior olecranon fracture-dislocations are characterized by relative sparing of the collateral ligaments and minimal involvement of the radioulnar articulation.1,2,19,21,24 Stability usually is restored by secure realignment of the fractures, and loss of forearm motion rarely is observed.1,2,19 In contrast, posterior olecranon fracture-dislocations often involve the forearm (by virtue of the radial head fracture), and true dislocation of the ulnohumeral joint can be observed if the fractures of the coronoid and radial head and the injury to the lateral collateral ligament are not adequately addressed.13,20
The preventable aspects of unsatisfactory results in patients with anterior or posterior injuries were mostly related to inadequate treatment of the coronoid fracture, and therefore inadequate restoration of the trochlear notch. This leads to malalignment, stiffness, and arthrosis. Realignment and stable fixation of the coronoid is challenging. The coronoid can be seen and aligned through the olecranon fracture by mobilizing the olecranon fracture proximally as one would do with an olecranon osteotomy. Stable fixation can be achieved either with screws inserted from the dorsal aspect of the coronoid, with a plate applied to the medial aspect of the coronoid after mobilizing the ulnar nerve and elevating part or all of the flexor pronator musculature off the proximal ulna, or with a suture capturing the capsular attachments to the coronoid. Very complex fractures may require a combination of techniques and also may benefit from temporary hinged external fixation.
1. Balakim G, Wippula E: Fractures of the olecranon complicated by forward dislocation of the forearm. Ann Chir Gynaecol Fenn 60:105-108, 1971.
2. Biga N, Thomine JM: La luxation trans-olecranienne du coude. Rev Chir Orthop 60:557-567, 1974.
3. Boyd HB: Surgical exposure of the ulna and proximal third of the radius through one incision. Surg Gynecol Obstet 71:86-88, 1940.
4. Broberg MA, Morrey BF: Results of delayed excision of the radial head after fracture. J Bone Joint Surg 68A:669-674, 1986.
5. Broberg MA, Morrey BF: Results of treatment of fracture-dislocations of the elbow. Clin Orthop 216:109-119, 1987.
6. Bruce HE, Harvey JP, Wilson JC: Monteggia fractures. J Bone Joint Surg 56A:1563-1576, 1974.
7. Colton CL: Fractures of the olecranon in adults: Classification and management. Injury 5:121-129, 1973.
8. Frankle MA, Koval KJ, Sanders RW, Zuckerman JD: Radial head fractures associated with elbow dislocations treated by immediate stabilization and early motion. J Should Elbow Surg 8:355-360, 1999.
9. Gustilo RB, Anderson JT: Prevention of infection in the treatment of one thousand and twenty-five open fractures of the long bones: Retrospective and prospective analysis. J Bone Joint Surg 58A:453-458, 1976.
10. Gustilo RB, Mendoza RM, Williams DN: Problems in the managment of type III (severe) open fractures: A new classification of type III open fractures. J Trauma 24:742-746, 1984.
11. Heim U: Combined fractures of the upper end of the ulna and the radius in adults: A series of 120 cases. Rev Chir Orthop 84:142-153, 1998.
12. Josefsson PO, Gentz CF, Johnell O, Wendeberg B: Dislocations of the elbow and intraarticular fractures. Clin Orthop 246:126-130, 1989.
13. Jupiter JB, Leibovic SJ, Ribbans W, Wilk RM: The posterior Monteggia lesion. J Orthop Trauma 5:395-402, 1991.
14. King GJ, Richards RR, Zuckerman JD, et al: A standardized method for assessment of elbow function. J Should Elbow Surg 8:351-354, 1999.
15. Mason ML: Some observations on fractures of the head of the radius with a review of one hundred cases. Br J Surg 42:123-132, 1959.
16. McKee MD, Schemitsch EH, Sala MJ, O’Driscoll SW: The pathoanatomy of lateral ligamentous disruption in complex elbow instability. J Should Elbow Surg 12:391-396, 2003.
17. Morrey BF: Current concepts in the treatment of fractures of the radial head, the olecranon, and the coronoid. J Bone Joint Surg 77A:316-327, 1995.
18. Regan W, Morrey BF: Fractures of the coronoid process of the ulna. J Bone Joint Surg 71A:1348-1354, 1990.
19. Ring D, Jupiter JB, Sanders RW, Mast J, Simpson NS: Trans-olecranon fracture-dislocation of the elbow. J Orthop Trauma 11:545-550, 1997.
20. Ring D, Jupiter JB, Simpson NS: Monteggia fractures in adults. J Bone Joint Surg 80A:1733-1744, 1998.
21. Scharplatz D, Allgower M: Fracture-dislocations of the elbow. Injury 7:143-159, 1975.
22. Schatzker J, Tile M: The Rationale of Operative Fracture Care. Ed 2. New York, Springer-Verlag 1996.
23. Teasdale R, Savoie FH, Hughes JL: Comminuted fractures of the proximal radius and ulna. Clin Orthop 50:129-133, 1993.
24. Wheeler DK, Linscheid RL: Fracture-dislocation of the elbow. Clin Orthop 50:95-106, 1967.