Monteggia-like lesions

Institutional review board approval was granted, and informed consent was obtained from each patient.

Searching our trauma database from January 2014 to December 2016, we identified 14 patients who were admitted and surgically treated due to a Monteggia-like lesion. We defined a Monteggia-like lesion as a proximal ulna fracture together with dislocation of the radial head and in combination with additional lesions such as radiohumeral dislocation, ulnohumeral dislocation, proximal radioulnar dislocation, and radial head fracture.

In all patients, preoperative CT scans of the injured elbow joint were obtained to rule out associated injuries and to improve preoperative classification and planning. According to the Bado classification, all 14 patients had a type II injury. Only 1 patient (7%) presented with an open fracture, which was classified according to Gustilo and Anderson as a type I injury [11]. Bado type II lesions were subclassified according to Jupiter et al. [15]. There were four type IIa, one type IIb, four type IIc, and five type IId injuries. No associated nerve or vascular injuries were observed. Osteosynthesis of the ulna was performed with a variable angle locking compression plate (VA-LCP; Synthes, West Chester, PA, USA) in nine cases, with a double plate (Aptus, Medartis, Basel, Switzerland) in four cases, and with a tension plate (Aptus, Medartis, Basel, Switzerland) in one case. Use of the different plating systems depended on the surgeon’s preference; the tension plate was used in case of a very proximal small single-fragment ulna fracture in a Bado/Jupiter type IIa lesion. Although all radial head fractures were dislocation fractures per definition, we classified radial head fractures according to the Mason classification depending on the actual fragment dislocation in the preoperative imaging. These consisted of three Mason type II, four Mason type III, and seven Mason type IV fractures [5, 14, 20]. Depending on fracture pattern, fragment dislocation, and patient age, osteosynthesis, radial head replacement, or radial head resection was performed. Radial head fractures were either addressed “through the ulna” or by dissecting subcutaneously from the ulna to the posterolateral border and opening the radiocapitellar joint with the Boyd approach. A fractured coronoid process could be observed in nine cases. These fractures were classified according to O’Driscoll [21] and consisted of two O’Driscoll type 1 (one 1.1 and one 1.2), two O’Driscoll type 2 (two 2.2), and five O’Driscoll type 3 (five 3.2) lesions. In case of intraoperative ulnohumeral instability, these lesions were addressed either with indirect transosseous refixation according to the so called lasso technique described by Garrigues et al. [9], or with retrograde screw fixation through the olecranon plate. Intraoperative testing revealed 5 patients with lateral instability due to rupture of the lateral collateral ligament. In these patients, refixation of the lateral ligament complex was performed using suture anchors.

In all cases, surgical treatment was performed within the first 24 h after injury. Surgical therapy was provided by three different surgeons specialized in upper extremity and trauma surgery. Patients received standardized postoperative treatment with early immediate continuous active and passive motion. Therefore, 2 to 5 days after surgery, an elbow orthosis for mobilization with an extension and flexion limitation of 0°–30°–120° was adjusted and worn for 6 weeks after surgery [12].

For follow-up examination, patients were asked to grade pain on a visual analogue scale (VAS). Moreover, elbow flexion, extension, pronation, and supination were measured with a goniometer and compared to the contralateral side. In addition, the distal radioulnar joint was examined for tenderness, and valgus and varus instability of the elbow joint were tested in full extension, if possible, as well as at 30° of flexion. The overall functional outcome was assessed using the Broberg and Morrey Score (BMS) and the Mayo Elbow Performance Score (MEPS) as elbow-specific scores. Furthermore, the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire as well as the subjective elbow value (SEV) were used as patient-focused outcome tools. Additionally, a neurologic examination of the operated side was performed to detect sensory or motoric deficits. Postoperative radiographs of the elbow were routinely obtained 2 days after surgery. If medically indicated (pain/loss of function), additional radiographs of the elbow were performed during follow-up.

Due to the low case number, only a descriptive statistical analysis was performed using Excel (Microsoft, Redmond, WA, USA).

A total of 13 patients were available for follow-up (93%) at an average of 21.9 months (range 7–40 months) after surgery, while 1 patient had died of unrelated causes. Patients’ main characteristics and the outcome parameters are reported in Table 1. In Table 2, the individual results of each patient can be seen. According to the BMS [4], 2 (15%) patients had excellent, 7 (54%) good, 2 (15%) fair, and 2 (15%) poor results. Mean BMS was 79 (23 to 100), mean MEPS was 82 (35 to 100), mean DASH Score was 23.6 (0 to 61), and mean SEV was 66% (20 to 90). Average pain level on the VAS was 2.6 out of 10 points (0 to 8). Ulnohumeral ROM was 116° (70 to 155°) on the injured side and 146° (110 to 160°) on the contralateral side. Mean forearm rotation was 138° (70 to 180°) on the fractured side and 164° (140–180°) on the contralateral side. At follow-up, none of the patients demonstrated instability of the distal radioulnar joint or medial or lateral instability of the elbow.

In total, 3 patients underwent revision surgery other than implant removal (overall revision rate 23%). In one case, postoperative redislocation occurred due to persisting instability after failed transosseous coronoid refixation and primary radial head resection. Therefore, in the first revision, direct screw fixation of the coronoid process and implantation of a radial head prosthesis was performed; however, re-redislocation occurred. Because of extreme noncompliance and reduced general health, re-refixation of the coronoid process as well as a temporary transfixation of the elbow joint with an external fixator was realized in order to stabilize the joint. After removing the external fixator, joint stability was achieved; nevertheless, elbow function was severely limited. The prosthesis had to be removed after 1.5 years due to chronic subluxation causing severe elbow pain and a limited ROM. In another case, failure of radial head osteosynthesis was observed; thus, a secondary radial head prosthesis was implanted. In the third case, postoperatively persistent sensory irritation of the ulnar nerve occurred; consequently, revision surgery with neurolysis of the ulnar nerve was realized.

Implant removal was not counted as revision surgery. In total, 3 patients underwent elective implant removal—twice due to the young patient age (24 and 32 years) and in one case because of cosmetic demands.

Postoperatively, 3 patients (23%) presented an associated neurological sensory deficit of the ulnar nerve. Two cases resolved without surgical intervention until follow-up; however, in one case, neurolysis of the ulnar nerve had to be performed.

No postoperative wound infection occurred in this case series.

In our case series from January 2014 to December 2016, exclusively Monteggia-like lesions Bado type II were registered. Several other authors also observed the predominance of Bado type II fractures in adults [8, 16, 22, 23, 29], while in children and in cases of high-energy trauma, Bado type I injuries are more common [28, 32]. Bado type I lesions show excellent or good functional outcomes in most cases, probably due to a low incidence of concomitant fractures of the radial head or the coronoid process [16, 28, 32]. In contrast, Bado type II lesions are reported to show significantly poorer outcomes [16, 26, 28]. Fractures of the radial head or the coronoid process have been reported to correlate with poor functional outcome scores and occur more frequently in older patients with associated bone weakness [16, 32]. Konrad et al. described Bado type IIa and IId lesions as negative prognostic factors [16]. In this context, our case series revealed similar results.

Our case series is too small to enable establishment of a data-based treatment algorithm. However, several pitfalls could be identified retrospectively, and the following treatment strategies can thus be recommended. In addition, two exemplary cases are illustrated in Figs. 1 and 2.

First, in patients with concomitant radial head fractures, open reduction and internal fixation (ORIF) should be strived for, as patients undergoing radial head replacement showed worse clinical results in various studies examining the treatment of comminuted isolated radial head fractures [3, 27, 31, 33]. In our experience, these findings can be transferred to Monteggia-like lesions. Minimally displaced radial head fractures (Mason type I/II) should be treated with screw fixation, plating represents a valid treatment option for radial neck fractures or comminuted fractures (Mason type III). Lindenhovius et al. reported that open reduction and internal fixation of unstable displaced fractures of the radial head occasionally fail, but could reduce the risk of subsequent elbow dislocation and protect against long-term arthrosis [19]. In severely comminuted and displaced fractures (Mason type IV) or in case of bad bone quality, ORIF becomes impossible and a radial head prosthesis should be implanted [27]. Ring et al. reported that ORIF of displaced fractures with more than three fragments is associated with early failure, nonunion, and loss of forearm rotation [27]. Therefore, a radial head prosthesis should be available when operating on a Monteggia-like lesion with a comminuted radial head fracture. The literature shows that radial head excision is linked to inferior clinical results compared to ORIF and radial head prosthesis, as radiocapitellar contact is important for elbow and forearm stability [27]. Due to a subsequent loss of stability, radial head excision is contraindicated in acute situations and should only be considered as a salvage procedure or for elderly patients with low functional demands in order minimize operating time [3, 13, 19, 27]. However, to date, this has only been examined for isolated radial head fractures and not in Monteggia-like lesions. As mentioned above, in our cohort, 1 patient underwent primary radial head resection due to severe intraoperative problems caused by morbid obesity. Postoperatively, this patient presented with persistent instability; however, it remains unclear whether this was caused by radial head resection or failed osteosynthesis of the coronoid fracture.

Second, olecranon fractures can be addressed with different techniques, as long as exact anatomical reposition and fixation is achieved. To date, it remains unclear whether any one of the available plating systems (VA-LCP vs. double plate vs. tension plate) is superior to the others; thus, we recommend using the system the surgeon is most familiar with.

Third, fractures of the coronoid process can result in ulnohumeral instability [7, 24]. For better evaluation of the coronoid fracture, the O’Driscoll classification should be used [21]. In this context, all O’Driscoll type 2 and 3 lesions must be addressed, as they potentially result in an unstable ulnohumeral joint. Special care should be taken in case of O’Driscoll type 2.3 lesions, which include fracture of the sublime tubercle, the ulnar insertion of the medial collateral ligament (MCL). Therefore, in an unstable posterior Monteggia-like lesion with fracture of the coronoid process, stable reduction of the coronoid fracture is necessary to restore ulnohumeral articulation, thereby providing elbow stability and minimizing the risk of future ulnohumeral arthritis due to chronic instability [10, 28]. Depending on the fracture pattern, repair of the coronoid fracture can be realized either directly by ORIF with screws, volar plating, or suture anchors [6, 30], or with the lasso technique by passing a suture through the anterior capsular attachment, shuttling it through the ulna, and tying it down on the subcutaneous posterior border of the ulna [9, 30]. Another possible technique is indirect osteosynthesis from the posterior ulna through the olecranon plate. In most O’Driscoll type I lesions the ulnohumeral joint remains stable and it is not necessary to address these fractures of the coronoid process; however, ulnohumeral joint stability has to be tested intraoperatively [7, 30].

Forth, our case series revealed that Monteggia-like lesions can be associated with concomitant lateral elbow instability. This has not yet been reported in the literature, although several case series of Monteggia-like lesions can be found [16, 18, 32]. In our study, four cases presented lateral elbow instability intraoperatively; therefore, refixation of the lateral ligament complex with suture anchors was performed. During the operation, lateral stability should be checked and in case of instability, refixation of the lateral capsule–ligament complex should be performed with a suture anchor.

Our study has strengths and weaknesses. Among the strengths is the relatively large population of adult patients having this uncommon injury with a very low loss to follow-up. Our study has the inherent limitations of a retrospective series. Moreover, the study cohort is quite inhomogeneous due to many influencing factors and the average follow-up of 21 months is rather short.