Abstract
Introduction:
Fixing the two-part Mason II radial head fracture using screws is becoming a popular practice. However, the screw fixation efficacy for three-part Mason III and IV fractures is controversial. The purpose of this study is to determine the effectiveness of using a uniform technique of headless compression screw fixation in simple, isolated Mason II and complex three-part Mason III and IV radial head fractures in terms of functional outcome, treatment efficiency and assessment of complications with the procedure.
Methods:
A prospective evaluation were performed on 31 adult patients with closed, non-pathological Mason II, III and IV radial head fractures sustained due to trauma and who underwent fixation using either two or three cannulated headless compression screws of 2.0 to 2.5 mm, and all patients were followed up for 2 years after the injury. They were divided into simple Mason II fracture group and complex three-part Mason III–IV fracture group. Operation time, time to discharge and radiological union were used as parameters for assessment of clinical outcome, while Mayo Elbow Performance Score, range of motion and complications were used to assess the functional outcomes.
Results:
Twelve cases of two-part simple fracture group and 18 cases of complex fracture group were identified. The mean age of 39 years is comparable between the two groups. Both groups had comparable days to union, mean hospital stay and operation time. In the simple fracture group, the mean Mayo Elbow Score was 97 (80–100), which is better than the complex fracture group score of 89 (75–100), p = 0.035. Both groups had no statistical difference in complication rates. All fractures united in our series. The mean range of motion for the simple fracture group was significant, with 133° ± 17.0° for flexion–extension arc, 85° ± 5° in pronation and supination as compared to the complex fracture group with 120° ± 20° flexion–extension arc, 69° ± 10° in pronation and 70° ± 8° in supination, p = 0.068.
Conclusion:
Overall clinical and functional outcomes of this technique are satisfactory in both simple and complex fracture groups, with simple Mason II fracture group doing better than the complex three-part Mason III and IV fractures in terms of Mayo Elbow Score and range of motion. Screw fixation has the advantage of less periosteal stripping and less impingement compared to other fixation methods and also allows for flexible fixation in constrained areas. Headless compression screw fixation can be considered as a method of fracture fixation for both simple and complex three-part radial head fractures.
Introduction
Traditionally, open reduction and internal fixation methods of the radial head involved T-plate fixation, Kirschner wire and screw fixation within a safe zone. Screws enable fragment compression and maintain position when physiological load is applied 1 to promote healing. Fracture compression increases the contact area across the fracture and stabilizes the fracture, allowing for faster union. 2
Initially, non-cannulated, AO screws were used in fixation of radial head fractures. However, they often posed problems of exposed hardware and impingement of cartilage and soft tissue, 3 specifically when placed in constrained areas due to prominence of the screw head that protruded out of the articular surface. Countersinking the screw head was done to prevent impingement, but it is still an imperfect solution.
In 1984, Herbert’s headless compression screw (HS) was developed in Australia, which was initially used in the fixation of scaphoid fractures. 4 It had a lower inter-fragmentary compression than the AO screw, but it became popular because of its other redeeming qualities of less periosteal stripping and impingement due to its absent screw head that allowed it to be buried within the articular surface. The Herbert screw was later used in the fixation of small bones in the hand and eventually used for the fixation of the radial head. 5,6 Studies using the Herbert’s screw for fixation of the Mason II fractures often yielded excellent results, while that of Mason III gave satisfactory results. 7 In addition to that, a study by Duckworth et al. 8 on the fixation of a specific type of Mason III fracture using headless screws also suggested a positive outcome of screw fixation technique on comminuted fractures. Recent publication by Demiroglu et al. 9 found type II Mason fracture fixed by HS showed favourable results, and Wu et al. 10 found that headless screws are as effective as plates with less complication in their series.
In addition to these findings, a method using HSs as an alternative technique to fix two-part Mason II and three-part Mason III and IV radial head fractures is being assessed in this study. According to this study, it is a minimally invasive technique using solely 2.0–2.5 mm SBi Autofix Morrisville PA HSs for the fixation of Mason II–IV fractures. The primary purpose of this study is to determine the outcomes of using headless screws to fix radial head fractures. Secondary purpose is to determine any difference in simple, isolated two-part Mason II and complex three-part Mason III and IV radial head fractures with regard to their functional outcome and treatment efficiency.
Materials and methods
We designed a prospective study to assess the fixation of radial head fractures using HSs. After the approval of the study by the local medical ethics committee, and informed consent was obtained from the patient with agreement for data usage for research purposes. Data were collated prospectively of patients undergoing operations for fixing radial head fractures using the headless screw technique, from 1 January 2009, to 31 March 2014, in a single-centre tertiary referral hospital. Patients included in the study were above age of 18, with closed two-part Mason II and three-part Mason III and IV radial head fractures sustained due to trauma. All study patients underwent fixation with either two or three headless cannulated SBi Autofix Morrisville PA fixation. Exclusion criteria are open fracture, chronic radial head fracture and radial head fracture fixed with combination of techniques, for example HSs and plating and four or more radial head fracture fragments were excluded. All patients are followed up continuously for 2 years.
A total of 34 patients underwent open reduction internal fixation of the radial head fractures, involving HSs, and 3 were excluded as 1 had open fracture, 1 had four parts radial head fractures which is fixed using HSs followed by T-plate fixation, and 1 had four-part fractures in whom and an initial attempt was made to fix the fracture using HSs but had to be abandoned as the fracture was too comminuted to be fixed with HSs and therefore underwent a radial head replacement. After exclusion, 31 patients were finally included in the study, which involved 12 two-part Mason II radial head fractures, 13 three-part Mason III fractures and 6 three-part Mason IV fractures patients. None of the patients had bilateral radial head fractures, and no two-part Mason IV fractures were included in the study.
Preoperative evaluation
During preoperative evaluation, all the patients were assessed for associated injuries. Clinical outcomes were assessed using the data collated regarding the post-operative range of motion, roentgenogram and union assessments. Functional outcomes were assessed using the Mayo Elbow Performance Score and the range of motion. For the purpose of analysis, we compared simple two-part Mason II fracture with more complex three-part Mason III and IV fractures. Clinical, radiological and functional parameters were analysed in these two subgroups.
Technique
All patients were evaluated preoperatively with clinical and radiological assessment (Figure 1). General anaesthesia were administered to the patients, with tourniquet on the upper arm. Patients were asked to lie in a supine position, with the arm positioned on a bolster to support the extremity over the chest. Alternative position is supine with the upper limb placed on a radiolucent hand table. The elbow is cleaned and draped.
Right elbow radiograph showing three-part radial head fracture superficial exposure.
Superficial exposure
Curvilinear incision was made on the skin from the lateral epicondyle to a point 3 cm distal to the tip of olecranon. This incision is centred on the head of the fractured radius. The total length of the incision is typically about 5 cm (Figure 2). Dissection is carried out between extensor digitorum communis (EDC) and extensor carpi radialis brevis (ECRB) to expose supinator muscle that is incised longitudinally under pronation to expose the annular ligament and capsule (Figure 3).
Skin marking for incision to the approach the radial head fracture.
Dissection is carried out between extensor digitorum communis (EDC) and extensor carpi radialis brevis (ECRB).
Deep exposure
Annular ligament is divided and tagged with braided 1/0 absorbable suture with the aim of repairing it on the closure of wound. The capsule should be incised anterior to the lateral ligamentous structures, avoiding injury to the lateral ulnar collateral ligament. If the exposure is inadequate, an ECRB split is made so that the radial head becomes more visible, without going more distal along the radial neck. Typically, the radial head that is less than 3 cm from radiocapitellar joint is exposed (Figure 4).
Exposure of radiocapitellar joint and stabilization by Kirschner wire.
Fracture fixation
Reduction of fragments and maintaining reduction and fixation of fragments
Radial head is exposed and its fracture fragments are reduced under direct vision and radiological guidance of an image intensifier. Provisional fixation of the radial head articular surface is facilitated by fine Kirschner wire (“joy stick”) manipulation and/or gentle application of reduction clamps (Figure 4). Surgeon can consider placing a second guide wire to stabilize the fragment thereby preventing spinning while tightening the screw. K wires can be placed to obtain satisfactory reduction during both clinical and radiological evaluation (Figure 5). The guide wires are placed in the centre of intended SBi screw implant path which lies in the middle of fracture fragment, with the tip of the wire just engaged to the subchondral bone at the far cortex of the fracture being repaired, perpendicular to the fracture line. The guide wire is placed under fluoroscopic guidance. The choice of 2.0 mm screw and 2.5 mm screw depends on the size of fracture fragments. The variability of screw sizes provides some flexibility for fracture management. We recommend the size of the screw should be less than 1/3 the size of the fracture fragment. The length of screw is determined by the depth gauge over the guide wire and 2–4 mm deducted from measured length to allow the implant head to be buried. All subsequent steps of screw insertion are guided by guide wire. Cannulated headless cortical screw (HS) implant is placed over the central guide wire, and with a cannulated screw driver, it is carefully inserted in a clockwise manner (Figure 6) . The image intensifier prevents overinsertion of the implant, achieving a fully flushed screw with or below cortical bone and articular surface. Two to three HSs are used to fix the fracture.
Intraoperative image intensifier showing reduced radial head fracture.
Cannulated headless cortical screw (HS) implant is placed over the central guide wire with cannulated screw driver.
Closure
Wound is irrigated, and capsule and annular ligament are repaired with absorbable braided suture. This is followed by layered closure without drain.
Post-operative management
All patients had bulky dressing and an arm sling for 1 week. They are reviewed 1 week later for wound inspection, radiological evaluation (Figure 7) and referral to physiotherapy. Gentle active range of motion exercises are initiated at 1 week post-operatively and then progressed to active and passive range of motion exercises by 1 month. Strength training starts after fracture union. The patients are reviewed at 1 week, 2 weeks, 1 month, 3 months, 6 months, 1 year and 2 years post-operatively to assess clinical and radiological union, which in turn are determined by painless movement, nontender palpation of the original fracture site and roentgenogram assessment. Their respective operating consultants perform clinical and radiological assessment, during follow-up. Two orthopaedic consultants further assess bone union, callus formation and complications at 2 years.
The first-week postoperative radiograph.
Statistical analysis
SPSS v.22 was used for the data analysis. Patients’ demographic and baseline clinical data are summarized descriptively. Mann–Whitney U test was used to compare patients’ numerical clinical outcomes, such as range of motion, time to union and final Mayo Elbow Score, between patients with Mason Type II radial head factures versus those with Mason Type III and IV fractures. Chi-square test or Fisher’s exact test, whichever was appropriate, was used to compare the categorical clinical outcomes.
Results
This study had a total of 31 patients who had undergone radial head fixation using headless 2.0 to 2.5 mm SBi Autofix Morrisville PA screws. According to Mason–Johnston classification, 12 of the patients had simple Mason Type II, 13 with three-part Mason III and 6 with three-part Mason IV radial head fractures. The median age was 40 years (21–77 years), most of the patients were male, and seven patients had preoperative CT scan while the rest had only roentgenogram evaluation prior to surgery. The median hospital stay is 2 days (1–5 days). Median time to union from index operation is 56 days (33–89 days), with all patients achieving union at a mean Mayo Elbow Score of 92.3 across both groups.
Subgroup analysis
There are 12 cases of simple two-part Mason II radial head fractures and 19 cases of three-part Mason III and IV radial head fractures. Their epidemiological data yield comparable mean age, spread of gender, median hospital stay and associated injuries. The median follow-up period in both groups is comparable.
Details of the epidemiology and perioperative parameters are listed in Table 1. The median radiological union time is comparable between two groups, that is, it is 50 days (34–77 days) and 59 days (26–89 days) in simple and complex groups, respectively, p = 0.12. In both the groups, all patients had achieved union during the study period.
Epidemiology and perioperative parameters of simple Mason II in comparison to complex Mason III and IV radial head fractures.
Mayo Elbow Performance Score measured pain intensity, motion, stability and function related to the activities of daily living, wherein the scores greater than 90 is considered excellent, 75–89 as good, 60–74 as fair and less than 60 as poor. In both the groups, all patients achieved good to excellent scores. The mean Mayo Elbow Score in simple group was 97 (80–100) which is significantly higher than that of the complex group, that is, with score of 89 (75–100) (p value = 0.035).
In the simple Mason II group, 83% of the patients achieved full flexion–extension arc and pronosupination arc. In complex Mason III and IV group, only 53% had full range of flexion–extension and pronosupination arc. There is a trend of better flexion–extension arc and pronosupination arc in simple group than that of complex group, but it did not achieve statistical significance.
Details of post-operative clinical outcomes are listed in Table 2. The difference in complication rate is not statistically significant for both groups as listed in Table 3. Only one complication was observed in the simple fracture group compared to three in the complex fracture group. The complication observed in the simple fracture group was a case of heterotopic ossification, which led to a relatively stiff elbow; the patient had post-operative indomethacin medication prior to the occurrence of heterotrophic ossification. The patient opted for conservative management and managed to return to his work. In the complex fracture group, two patients required removal of all screws following fracture union for implant prominence and pain but did not require revision. One had persistent lateral-sided elbow pain and stiffness of elbow which was investigated by magnetic resonance imaging and did not find any significant abnormality after fracture union. He did not require any re-operation and the pain reduced with physiotherapy after a year.
Results of parameters to determine post-operative clinical and functional outcomes of each group.
Complications in each group.
Discussion
Open reduction and internal fixation has been a preferred treatment option for simple Mason II fractures and has given excellent results, 2,9 but its use in the treatment of type III and IV fractures remains controversial. In highly comminuted complex Mason III or IV fractures that cannot be reconstituted, the option of radial head arthroplasty usually remains the most viable option. 11
In this study, a fixation method using only HSs was used for the treatment of simple two-part Mason II and complex three-part Mason III and IV fractures. It showed excellent function with mean Mayo Elbow Score of 97 in simple Mason II group and good function with mean score of 89 in complex Mason III and IV fractures. The difference is statistically significant. This observation may be confounded by increased likelihood of more severe soft tissue injury sustained due to Mason III and IV fractures as compared to Mason II fracture, which may lead to stiffness in the elbow joint as there is a decrease in the range of motion in Mason III and IV group which does not show statistical significance.
The headless design of the HS allows it to be buried beneath the subarticular surface of the radial head. Thus, proper radiocapitellar contact is maintained without obstruction or impingement upon fixation, which is helpful in achieving full elbow rotation. 12,13 It is especially important in radial head fractures where dislocation is involved or in unstable elbows with associated ligamentous injuries and in the terrible triad with associated stiff elbow post-surgical treatment.
The HS also prevents soft tissue impingement especially when placed in a constrained space, thus minimizing the need for implant removal that has an associated problem of open reduction and internal fixation. 13 In addition, less periosteal stripping involved in performing the HS fixation ensures that vascular supply to the bone is not disrupted, reducing scar formation and promoting healthy union of bone. As shown in our series, all radial head fractures treated with HS achieved union. A smaller exposure is required for placement of HS; hence a lower risk of injury to posterior interosseous nerve that is dangerously close to the area of surgery. 14
The SBi screw used in HS fixation possesses self-drilling and self-tapping properties that reduce the risk of displacement during drilling, tapping and insertion, specifically for comminuted fracture fragments that may be easily displaced during fixation.
As shown in earlier study, stiff fixation of the head to the shaft is not required in order to achieve stability and union. 15 In our technique, the radial head is only fixed to the shaft with two to three obliquely placed screws. Therefore, this technique has a less stiff construct and less metalwork in a constrained area, yet allowing for stable fixation enabling union.
A further advantage of this technique is that the screw may be placed beyond the safe zone of the radial head for maximum reduction, which provides the flexibility required especially in fixing three-part Mason III and IV fractures.
Mason III and IV fracture, as a group is inherently difficult to fix with poor midterm result using various internal fixation implants leading to a move to radial head arthroplasty with no quality long-term evidence of its success. 16
There is steep learning curve in this technique of radial head fixation especially in the complex fracture group. Firstly, there is a limited working space adjacent to the radial head, with the variation in the proximity of radial nerve limiting distal dissection. 14 Secondly, radial head fractures are intra-articular fractures that are best fixed with good articular conformity to prevent block in motion and more than 2 mm step off is associated with osteoarthritis. 17 Thirdly, the guide wire used in the placement of screw is of small calibre and has moderate risk of breakage when using cannulated drill, especially if the guide wire is bent while drilling. In the event of breakage, it is safe to leave the broken guide wire in the radial head, provided it is completely buried inside the bone.
Finally, there is limited bone stock and metaphyseal area for wire reduction and screw placement. Hence, precise placement of guide wire and subsequent placement of screw are crucial to prevent fragmentation of bone and reinsertion of screw. Misplacement and subsequent removal and reinsertion of screw would lead to a loss of interfragmentary compression strength. 18
The complication rate is higher in the complex Mason III and IV fracture group as compared to simple Mason II fracture group. It does not show statistical significance due to the small numbers in each arm. In our series, two patients in the complex fracture group had loosening of screws, which became prominent, requiring subsequent removal in the immediate post-operative period, despite the screw heads being buried. In both patients, there was adjustment and reinsertion of screw during the index surgery.
The authors felt that despite the technical difficulties in fixation of radial head fractures in among other techniques in treatment of radial head fractures. 9,18 –21 Headless screw fixation has a role in the treatment of simple and complex radial head fractures in view of its inherent attractive advantage of biological preservation and soft tissue preservation.
Limitations
This is a prospective case series study that has inherent bias in terms of the profile of patients and selection bias in terms of suitability of fixation using HSs for specific patients by different consultants. Long-term results of patients treated with open-reduction internal fixation by the HS method are unknown due to the medium follow-up period of our study. Several surgeons performed the operations in the two groups, leading to possible performance bias. Having a small number of patients in each group means some of the data do not show statistically significant results. We did not include four or more parts radial head fracture as we felt they were too comminuted to be fixed with HSs, hence we were unable to comment whether HSs can be used for more comminuted Mason III and IV fractures. Although qualitative assessment of the associated soft tissue injuries is done in our study, quantitative assessment of the severity of soft tissue injuries is not done.
Conclusion
Headless screw fixation technique provides a minimal invasive surgical option with less dissection and soft tissue stripping, thus reducing the risk of injury to the associated ligaments and nerves. While there is a trend that this technique gives better functional outcome in simple fracture, it provides good functional outcome in both simple and complex fractures, advocating its versatile use in even complex three-part radial head fractures, especially as an alternative to plate fixation or radial head replacements.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflict of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.