Introduction:

Being one of the most common childhood injuries, paediatric supracondylar humeral fractures have been frequently encountered in different specialties including Accidental and Emergency Department, paediatric and orthopaedics department. Treatment of supracondylar fractures are primarily based on fracture pattern, with conservative treatment for undisplaced and operation for displaced fractures. For Gartland II injuries, different authors have suggested different methods of management, from conservative casting, closed reduction, to operative fixation.

The anterior humeral line (AHL) is commonly used to identify subtle supracondylar fractures, but its relevance with treatment and clinical outcome has not been emphasized. Our study is a retrospective case-control study, to assess whether there is any correlation between the final clinical outcome and different anatomical variations after fracture union, using the anterior humeral line as reference.

Methods:

In the past 5 years, 51 patients with supracondylar fractures were treated in our hospital. Fracture pattern was evaluated in anteroposterior and lateral x-ray films. The lateral capitello-humeral angle and Baumann’s angle were measured. Plaster casts were removed after radiological evidence of fracture union. Clinically, elbow range of motion, presence of elbow deformity, pain, and parental anxiety documented in the final outpatient clinic follow-up were retrospectively reviewed. Either self or early physiotherapy was arranged after the cast was removed.

Patients were evaluated on interval follow-ups to determine whether Morrey’s functional arc of motion could be achieved. Stiffness was defined when elbow motion could not reach an elbow flexion of 130 degrees or extension of 30 degrees.

Fisher’s exact test was used to identify any clinical difference between the four groups of patients with different final radiographic presentations, with the anterior humeral line intersecting with the middle 1/3, anterior 1/3, and anterior border of capitellum, which is regarded as the aligned group. Patients with the anterior humeral line touching or intersecting with the capitellum were regarded as the aligned group. Patients with anterior humeral line not touching the capitellum in the lateral view after fracture union are regarded as the malunion group (Figure 1).

Figure 1

Different anatomical variations in the four groups of patients. Group 4 is regarded as the malunion group.

Results:

There were 33 male and 18 female patients. The age ranged from 1.5 to 11 year-old at the time of injury. Gartland I fracture was present in 14, Gartland II in 23, and Gartland III in 14 patients. Fifty of them were extension type and only 1 of them was flexion type. Twenty-five of them were left sided fractures while twenty-six of them were right sided fractures. Twenty-nine patients were treated conservatively, including 14 Gartland I fractures treated with casting alone, and 15 Gartland II fractures treated with closed reduction and casting. Eight and 14 patients having Gartland II and III fractures respectively received operative reduction and K wire fixation. Mean immobilization angle in plaster is 97 degrees of elbow flexion. No compartment syndrome nor plaster complications were observed.

Patients had follow-ups ranging from 4 weeks to 24 months, with a mean follow-up of 15.1 weeks (9.74-20.48, 95% Confidence Interval). Radiological results included the presence of varus deformity in antero-posterior film, and fracture angulation in lateral film. Two cubitus varus deformity were observed so far.

Excluding cases with Gartland I fracture, there were 28 patients in the aligned group. These patients were followed up for an average period of 11.18 weeks (8.31-14.04, 95% Confidence Interval). The mean angle of immobilization in this group was 96.09 degrees (93.61-98.56, 95% Confidence Interval). The mean period of immobilization was 4.23 weeks. Mean Baumann angle was 74.1 degrees (71.63-76.56, 95% Confidence Interval). Mean lateral humeral capitellum angle was 31.07 degrees (28.5-33.64, 95% Confidence Interval). Average elbow flexion range in final follow-up was 137.1 degrees (133.82-140.46 degrees, 95% Confidence Interval), and was regarded as comparable when examined with contralateral side in 27 out of 28 patients. Parental anxiety was present in 1 patient.

Nine patients had radiographic features of malunion after fracture consolidation, including 5 patients with Gartland II fracture treated conservatively with plaster immobilization, and 4 patients with Gartland III fracture treated with operative reduction and K wire fixation. All of them complained of persistent stiffness in flexion in the latest follow-ups. (Figure 2) The mean angle of immobilization in plaster was 99.89 degrees (92.27-107.5 degrees, 95% Confidence Interval) degrees in flexion. The mean period of immobilization was 4.33 weeks. The follow-up period averaged 27.33 weeks (8.56-46.11 weeks, 95% Confidence Interval). Mean Bauman angle on healing was 71.78 degrees (60.57-82.98 degrees, 95% Confidence Interval). Mean lateral humeral capitellum angle was 28.33 degrees (14.39-42.27 degrees, 95% Confidence Interval). The elbow flexion ranged from 90 to 130 degrees, with an average of 115 degrees in early follow-up (2 to 4 months) and 118 degrees in final follow-up. No significant extension lag was observed. Parental anxiety was present in 6 patients. Table 1 shows the comparison between the aligned and mal-aligned groups.

The different anatomical variations of fracture union in relevance to elbow stiffness were further evaluated. Using Gartland I injury as the control group, there was no statistical significance for Group 1, 2 and 3 patients in terms of elbow stiffness. Group 4 patients had a statistically significant p value of 0.0018 in terms of elbow stiffness. (Table 2)

The effect of operative reduction and K wire fixation versus conservative treatment on elbow stiffness was also evaluated. Of the 22 patients who received operation (8 Gartland II, 14 Gartland III), 4 out of 18 patients in the aligned group developed elbow stiffness. The other 4 patients with fracture malunion all developed elbow stiffness, with a statistically significant p value of 0.0096. Of the fifteen patients (Gartland II) were treated conservatively, none out of 10 patients in the aligned group developed elbow stiffness. The other 5 patients with malunion all developed elbow stiffness, with a statistically significant p value of 0.0325. Thus, elbow stiffness appeared to be correlated with quality of reduction rather than the treatment received (conservative vs operation).

Figure 2

Example of supracondylar fracture nonunion. The anterior humeral line was anterior to the capitellum. Patient complained of persistent elbow stiffness 6 months after the initial injury and was unable to touch his own shoulder.

Figure 3

Demonstration of relatively ‘undisplaced’ elbow fracture. Computed Tomography was performed showing both cortices in continuity. Conservative treatment was adopted. Resulting radiographic malunion and elbow stiffness was present.

Figure 4

Example of malunited humeral supracondylar fracture. At 1 year, elbow flexion was up to 120 degrees only. Subsequent regain of elbow range flexion up to 160 degrees took place after 2 years with bone remodeling, raising the problem of parental anxiety and prolonged follow-ups.

Table 1

Comparisons between aligned and mal-aligned groups.

Table 2

Clinical outcome with reference to anatomical variations after fracture union

Discussion:

Despite being the most common fracture in the paediatric population so far, currently there is no international consensus of treatment in Gartland II fractures. Some authors advocate cast immobilization without reduction, reduction by overflexion to 120 degrees followed by cast mobilization, while others suggested closed reduction with pinning and casting in all cases of Gartland II fractures. (1)

The fracture line may be difficult to identify owing to thick periosteum in the paediatric population. This leads to the impression of fracture being relatively undisplaced, especially when both cortices are in continuity, and raised the conception that whether these fractures can be treated by casting in situ without a means of closed reduction (Figure 3). As described by Luis, potential complications of closed reduction and pin fixation include pin-tack infection, neurovascular injury, pin migration, compartment syndrome after closed reduction, and chance of loss of reduction. In his 46 case series of Gartland II fractures treated conservatively without attempt of closed reduction, 80 percent of patients achieved good functional outcome with reference to Flynn’s criteria, although there may be an increased in mild cubitus varus deformity and elbow extension. (2)(3)

On the contrary, according to AAOS guidelines 2011, closed reduction with pin fixation for patients with displaced (Gartland Type II and III, and displaced flexion) pediatric supracondylar fractures was suggested, with moderate recommendation strength. (4)

We believe the anterior humeral line is a useful tool to evaluate the degree of posterior angulation which has the prognostic value. The anterior humeral line was first described by Roger in year 1978, as clues to detect obscure fracture of pediatric elbow injury (5). Traditionally, it should intersect with the middle third of the capitellum. In our study, whether the anterior humeral line just touches the capitellum or intersects with anterior 1/3 or middle 1/3 of the capitellum have no statistical significance with regard to final elbow flexion stiffness. As a reference, in a literature study of 60 normal children, variations of AHL with respect to capitellum were evaluated, with only 50 and 62 percent of normal elbows having AHL passing through middle 1/3 of capitellum, in children less and more than 4 years of age respectively. (6)

In the malunion group, i.e. AHL falls off the capitellum, a significant proportion of patients presented with problem of elbow stiffness and parental complaints or anxiety in our population. According to Lewis (7), time to return of “full elbow motion” lies within 6 months after surgery, which clearly is not met in the mal-aligned group from our current study. The main concern from parents is asymmetry of elbow flexion and the inability to touch shoulders using ipsilateral hand. Traditional parents even complained of their children being “handicapped”. Prolonged follow-ups and repeated parental counselling and physiotherapy were required, which led to increased financial burden and adverse consequences on patient-doctor relationship.

Although remodeling can be used as an argument for tolerating initial mal-reduction, the remodeling power and duration cannot be accurately predicted. (Figure 4) There is an increasing trend to believe that accurate reduction with minimum joint and soft-tissue trauma is required to achieve the best possible functional result. (8) (9) (10)

As described by Charnley, closed reduction and immobilization require 120 degrees of elbow flexion to maintain stable reduction. (11) There are two disadvantages to this method, described by McLaughlinas the “supracondylar dilemma”. (12) Over-flexion in a swollen elbow may compromise the circulation, and even leads to serious complications like compartment syndrome, but less flexion predisposes to loss of reduction.

We propose the utilization of anterior humeral line as guidance of treatment in Gartland II and III supracondylar fractures. In cases with AHL falling off capitellum, closed reduction and pinning fixation is the preferred method of treatment. The adequacy of reduction is guided by the restoration of AHL to meet the capitellum. Over-flexion is to be avoided in view of potential serious complications.

Our limitation is a relatively small number of patients, the diversity of patient groups in terms of age, injury pattern and treatment methods, lack of a prospective study, and inadequate duration of follow-up in assessment of potential remodeling.

Conclusion:

Supracondylar fractures are a common yet notorious problem to treat. The anterior humeral line serves not only as a guide to subtle fracture, but is also a prognostic indicator. Early alignment restoration is more cost-effective as it helps to restore elbow flexion range, relieve parental anxiety and shorten follow-up period in our locality. The adequacy of reduction should be assessed using the anterior humeral line, and can be accepted as long as the line touches the capitellum.