Difference Between Posterior Monteggia Fractures and Posterior Fracture‐Dislocation of Proximal Ulna in Adults

Objective To figure out the difference between patients with posterior Monteggia fractures which were concomitant with proximal radioulnar joint (PRUJ) dislocation and posterior fracture‐dislocation of the proximal ulna that were not concomitant with PRUJ. Methods From January 2016 to January 2019, 37 consecutive adult patients who had posterior fracture‐dislocation of proximal ulna (no PRUJ dislocation, n = 16) and posterior Monteggia fractures (PRUJ dislocation, n = 21) were included. All patients had intraoperative fluoroscopy, computed tomography (CT) scans, and standard radiography (anteroposterior view and lateral view). The mechanism of injury, the cases with open fracture, sustained multiple injuries and classification of fracture was recorded. The clinical details of the patients such as the final range of motion (ROM) and the Broberg–Morrey scores were described. Results Patients with PRUJ dislocation (ten type A, five type B, and six type D) and those without concomitant PRUJ dislocation (fifteen type A and one type C) exhibited an obvious difference according to the classifications of Jupiter et al. (P = 0.010). Ninety‐five percent of patients who had PRUJ dislocation were accompanied by a metaphyseal fracture, while only 50% of the patients who did not have PRUJ dislocation were accompanied by a metaphyseal fracture (P = 0.002). Meanwhile, 16 of 20 metaphyseal fractures had more than one fragment in the group of dislocations, but five of eight metaphyseal fractures were comminuted in the control group. The two groups exhibited an obvious difference (P = 0.009). The 21 patients who sustained a radioulnar dislocation had less mean arc of flexion, pronation, and Broberg–Morrey scores were significantly less than the patients of the control group (flexion: 117.38 ± 14.46 vs 127.50 ± 13.416, P = 0.035; pronation: 59.76 ± 11.88 vs 67.50 ± 6.58, P = 0.017; Broberg–Morrey: 80.48 ± 12.17 vs 88.19 ± 10.28, P = 0.040). Conclusions Patients suffering posterior Monteggia fractures had more metaphyseal fractures, more comminuted fractures of the metaphysis, and worse ultimate ulnohumeral motion than patients of posterior fracture‐dislocation of proximal ulna.


Introduction
M onteggia fracture is defined as proximal ulnar fracture accompanied by proximal radioulnar joint (PRUJ) disruption 1 . It includes an apex ulna fracture, PRUJ dislocation, and radiocapitellar dislocation. According to Giovanni Battista Monteggia's description in 1814, it referred to a fracture of the proximal third of the ulna accompanied by radial head anterior dislocation from PRUJ and radiocapitellar joints 2 . Monteggia fracture dislocation was classified by Bado into four different types. Type I, II, and III was explained as a dislocation of the radial head dislocation in anterior, posterior, and lateral directions, respectively. Type IV referred to fractures of double bones of the forearm with radial head dislocation 2 . Jupiter and his colleagues subclassified type II posterior Monteggia fractures into four types considering position of ulnar fracture 3 . The ulna fracture has four types: type A, the distal part of the coronoid process and olecranon; type B, metaphysis fracture; type C, diaphysis fracture; and type D, a complicated multi-fragmented fracture combining some of above levels. Type C of Jupier Classificationis a classical posterior Monteggia fracture which is always accompanied with PRUJ dislocation and radiocapitellar dislocation. But other types of Jupiter Classification are not described whether a PRUJ dislocation exists.
Many authors concluded that the apex posterior fractures of the ulna accompanied by posterior dislocation of radiocapitellar joint, with or without dislocation of PRUJ, are posterior Monteggia fractures or Monteggia-like lesions [4][5][6][7] . Application of these eponyms to all injuries accompanied by radiocapitellar dislocation and subluxation has brought about many confusions 8 , such as the anterior olecranon fracture-dislocation with anterior Monteggia fractures, and the posterior olecranon fracture-dislocation (posterior fracturedislocation of the proximal ulna) with posterior Monteggia fractures. Based on the study by Giannicola 9 . these injuries involve six basic lesions which shall be recognized and well-treated: (i) ulnar fracture; (ii) radio-humeral dislocation; (iii) ulnohumeral dislocation; (iv) proximal radio-ulnar dislocation; (v) radial fracture; and (vi) distal radio-ulnar joint dislocation or interosseus membrane lesion. This theory was accepted and used by other authors 10-12 . We think that if there is an apex posterior ulna fracture, dislocation of PRUJ and radiocapitellar dislocation should be called posterior Monteggia fractures, and an apex posterior ulna fracture and radiocapitellar dislocation without dislocation of PRUJ should be called posterior fracture-dislocation of proximal ulna. This retrospective study aimed to: (i) evaluate injury patterns, demographics, injury mechanisms of patients and (ii) the short-run operative treatment results of patients suffering posterior fracture-dislocation of proximal ulna (with PRUJ dislocation) and posterior Monteggia fractures (without PRUJ dislocation), as well as (iii) confirming prognostic factors impacting the functional result. It is hypothesized in the study that patients with PRUJ dislocation will present a worse final motion and functional score.

Inclusion Criteria and Exclusion Criteria
The patients were eligible for inclusion in this study if they: (i) had proximal ulna fracture, radiocapitellar posterior dislocation with or without proximal radio-ulnar dislocation; (ii) were surgically treated between January 2016 and January 2019 at our hospital; (iii) equaled to or over the age of 18, and were followed for more than 1 year; and (iv) the related outcomes were completely recorded.
Exclusion criteria were: (i) patients who were followed up for less than 1 year; (ii) patients who did not take preoperative radiographs and (iii) patients whose initial operation were conducted in another hospital.
All of the 37 adult patients with 37 elbow fractures and posterior dislocations met the above criteria. All patients had standard radiograph (anteroposterior radiograph and lateral radiograph), computed tomography (CT) scans and intraoperative fluoroscopy, and the mechanism of injury, the patient with open fracture and sustained multiple injuries were recorded.

Classification
Classification of all injuries was performed based on the system proposed by Jupiter et al 3 . Type A fracture was at the olecranon level and entered the ulnohumeral joint. Type B fracture was at the junction of the diaphysis and the metaphysis and in the distal end of the coronoid process. Type C referred to the ulna diaphyseal fracture. Type D involved a fragmented ulna fracture extending from the olecranon to the diaphysis. Some fractures were classified as type D in the paper because the fractures had some fragmentations of coronoid and metaphysis/diaphysis. They could not be classified into any other types. Mason's classification was used to classify radial head fractures into three types 13 : type 1 is the nondisplaced fracture; type 2 is the displaced partial fracture; and type 3 is the displaced fracture that involves the entire radial head. The classification of Regan and Morrey divided the coronoid process fractures into three grades: Grade I denoted the avulsion of the process tip; grade II denoted a fragment which involved 50% of the process or less; and grade III denoted a fragment that involved over 50% of the process 14 .
The presence of radioulnar joint dislocation, ulnohumeral dislocation, comminution of a coronoid fracture, metaphyseal fracture, comminution of metaphyseal fracture, repairing to lateral collateral ligament, and other informations were also recorded.

Treatment
Patients went into the emergency room. Doctors should check the ipsilateral shoulder, elbow, and wrist. The X-ray images were taken, which included the full length of the forearm. When treating patients with ulnohumeral joint dislocation and minor coronoid fracture, closed reduction should be performed under general anesthesia, and it would be beneficial to those patients. But, When treating patients with severe fracture injuries, closed reduction should not be performed. Only sling or plaster fixation should be used.
In order to eliminate anesthesia factors, operations under general anesthesia were anesthetized by the same anesthesiologist. Furthermore, patients assigned to collect data were blinded to the anesthesiologist. In the operating room, all patients were monitored with electrocardiogram, pulse oxygen saturation, non-invasive blood pressure and bispectral index (BIS).
CT examination was completed to evaluate the injury of the bone structure, and the ligament lesions should be assessed after the fractures were fixed in the surgery.
The treatment process of these fractures was performed by five different attending surgeons. All 37 injuries were treated operatively. All patients were placed in the supine position, and posterior approach allowing most of ulnar fractures was used. The Kocher approach was used to fix or reconstruct the radial head. The anteromedial elbow over-the-top approach was used if the coronoid could not be fixed from the posterior approach or Kocher approach (when radial head prosthesis was used). There was no transpose or release of the ulnar nerve in situ (Fig. 1).

Evaluation Methods
The clinical and radiological data was collected during the follow-up period.

Broberg-Morrey Score and Index
Broberg-Morrey Score and Index were taken into account for performing the specific evaluation of the study at the last follow-up. The cohort containing 21 patients accompanied by radioulnar dislocation was regarded as a subgroup. A comparison was performed between the subgroup and the remaining group containing patients without radioulnar dislocation. The rating system of Broberg and Morrey is a 100-point system, which consists of four sections: motion (40 points), strength (20 points), stability (5 points), and pain (35 points). Scores are as followed: 95-100 points indicates an excellent outcome; 80-94 points, a good outcome; 60-79 points, a fair outcome; ≤60 points, a poor outcome. The outcome can be considered satisfactory if the result is rated as good or excellent, and unsatisfactory if it is fair or poor.

Varus and Valgus Stability
We tested the varus and valgus stability in the largest extension and 30 of flexion. The posterolateral rotatory stability was evaluated by the pivot shift test 15 , reflected in four grades: normal, mild, moderate, and severely unstable.

Measurement of Radiology
All patients received the anteroposterior and lateral radiographs on the injured elbow at follow-up. The radiographs of the degenerative change, capitellar osteopenia, as well as heterotopic ossification were evaluated by a blinded radiologist.
Lamas et al. 16 graded the capitellar osteopenia as severe, moderate, mild, and none. Considering the grading standard of Broberg and Morrey 17 , degenerative changes had four grades: grade 0 meant normal joint; grade 1 was slight narrowing of joint space and minimum formation of osteophyte; grade 2 represented moderate narrowing of joint space and moderate formation of osteophyte; and grade 3 was severe degenerative change with severe joint destruction. The bridging bone showed by the anteroposterior and lateral radiographs defined the union of fractures. Hastings and Graham 18 defined the heterotopic ossification into seven grades: 0, I, IIA, IIB, IIIA, IIIB, or IIIC.

Statistical Analysis
SPSS22.0 software (IBM Corporation, Chicago, USA) was used for statistical description and analysis of data. Student's t-test was performed on the age, flexion, extension, pronation, supination, and Broberg-Morrey Score. Fisher's exact test was conducted to assess the difference in gender, Jupiter classification, Mason classification, Morrey classification, metaphyseal fracture, and comminution of metaphyseal fracture of the two group. P < 0.05 was considered statistically significant.

General Results
The study involved 25 men and 12 women whose average age was 42.22 years (range, 20-79 years). The average follow-up time was 27.78 months (range, 12-48 months); 16 patients were injured due to falling from standing heights, 10 patients were injured due to falling from higher heights, eight due to motor vehicle collision, and three due to falling from electric bicycles. One fracture was accompanied by an open wound and grade II according to Gustilo and Anderson 19 .
Patients sustained multiple injuries: there were five patients suffering ipsilateral upper extremity injury (one distal radius fracture, one distal radius and humeral shaft fracture, and three capitellum fractures). There were three patients suffering contralateral upper extremity injury (one severe triad elbow injury-elbow dislocation accompanied by radial head and coronoid fracture, one forearm fracture, and one distal radius fracture). There were three lower extremity fractures (one femoral neck fracture, one patella fracture, and one tibial plateau fracture), and three sets of rib fractures. There was one abdominal injury. Nerve palsy was not found in any patients.

Results of Classifications
As classified by Jupiter et al., there were 25 type A fractures, five type B fractures, and seven type D fractures 3 . Thirty-six of the 37 injuries (97.3%) suffered radial head concomitant fracture and 22 fractures of type 2 (with partial articular displaced) and 14 fractures of type 3 (with complete articular displaced).
Based on the classification by Jupiter et al. 3 , patients who had radioulnar dislocation (10 type A, five type B, and six type D) were different from those patients without radioulnar dislocation (15 type A and one type C) (P = 0.010). When the classification was the Jupiter type B and type D, it was more likely to be the posterior Monteggia fracture which was concomitant radioulnar dislocation.

Fracture Patterns
Regarding the coronoid fracture, 33 (89.2%) had coronoid fractures. Twelve patients had fracture of Regan-Morrey type 1, four had fracture of type 2 (<50% of the height of coronoid), and 17 had fracture of type 3 (>50% of the height of coronoid). There were over one fragment in 22 of all coronoid fractures. Fifteen patients had concomitant ulnohumeral dislocation and the two groups exhibited no obvious difference.
All of the patients with dislocation of PRUJ had an associated radial head fracture. The two groups exhibited no obvious difference (P = 0.390). But with the radial head fracture in Mason III, the internal fixation and open reduction were worse than the arthroplasty in the function and Broberg-Morrey score. Most patients had coronoid fracture, and both of the groups had no significant difference (P = 0.180).
Ninety-five percent of patients who had PRUJ dislocation were accompanied by metaphyseal fracture, while only 50% of the patients who did not have PRUJ dislocation were accompanied by a metaphyseal fracture (P = 0.002). Meanwhile, 16 of 20 metaphyseal fractures were comminuted in the dislocation group, but 5 of 8 metaphyseal fractures were comminuted in the no dislocation group. The two groups exhibited an obvious difference (P = 0.009) ( Table 1).

Complications
In the cohort containing 37 patients, seven (18.9%) patients with injured elbows received more than one subsequent treating procedure, four patients were in the group of PRUJ dislocation (two rupture of the plates, one recurrent dislocation, one radial head non-union), and three patients were in the group of no PRUJ dislocation (one deep infection, one ulnar neuropathy, and one heterotopic ossification) (P = 0.357). The patients with complications had worse function and scores than the patients without complication

Broberg-Morrey Score
Clinical assessment showed significant difference between the two groups (PRUJ dislocation group 80.48 AE 12.17 vs no dislocation group 88.19 AE 10.28, P = 0.040). The scores of PRUJ dislocation group were significantly less than the scores of no dislocation group.
All patients showed a normal pivot shift and the elbows were stabilized ( Table 2).

Results of Imageology
Based on the radiographs, no significant difference existed between the two groups (capitellar osteopenia grade 1, nine patients vs grade 1, five patients, P = 0.515; degenerative changes grade 1, 17 patients, grade 2, two patients vs grade 1, 14 patients, grade 2, zero patient, P = 0.643 and heterotopic ossification class I, three patients, class IIA, two patients, class IIIA, zero patient vs class I, one patient, class IIA, zero patient, class IIIA, one patient, P = 0.887). fracture was defined as additional traumatic pathologies surrounding the elbow (such as additional radial head fracture, coronoid fracture, or humeroulnar joint dislocation) 20 . But the PRUJ dislocation, as an important characteristic, helps to differ Monteggia fracture from fracture-dislocation of the proximal ulna 21 .

R esearchers
As is well known, the study was the first one which had patients with clearly defined posterior Monteggia   Patients with dislocation of PRUJ had a significantly worse outcome in function (flexion and pronation), and Broberg-Morrey score than patients of the control group. In the dislocation group, two patients had plate fracture and one patient had ulna posterior angulation and recurrent radiocapitellar joint dislocation, because the metaphysis was comminuted and the fractures were unstable. This never happened in the control group.
Posterior Monteggia fractures and posterior fracturedislocations of the proximal ulna are complex fractures and dislocations of the elbow joint. There were similarities between them regarding the injury degree of soft issue and bone structure. But PRUJ dislocation, as an essential characteristic, differentiated posterior Monteggia fractures from posterior fracture-dislocations of the proximal ulna. If ulnar fractures were involved in the metaphysis or diaphysis, the fractures were usually comminuted, and the dislocations of PRUJ more likely to occur. Restoring the anatomical shape of ulna, stable fixation was more difficult, but more important. Finally, when treating complex elbow fractures and dislocations, a definite diagnosis is needed to accurately treat and evaluate the prognosis.
The retrospective nature of the study limited its outcomes. Patients were operated on by different doctors and had relatively short follow-up time. Besides, results of statistical analysis may present a bias, because of the small number of patients in two subgroups. It limited the ability to draw conclusions. Studies with multiple centers and large sample size should be performed to further confirm the research results.

Conclusion
In conclusion, there are similarities between posterior Monteggia fractures and posterior fracture-dislocations of the proximal ulna in fracture of proximal ulnar and dislocation of radiocapitellar joint. But posterior Monteggia fractures have dislocation of PRUJ and have more comminuted fracture of metaphysis than patients of posterior fracturedislocation of proximal ulna. Accurate diagnoses of these complex elbow fracture-dislocation help doctors to provide effective treatments and evaluate prognosis.