The Treatment of Displaced Supracondylar Humerus Fractures

Evidence-based Guideline

Abstract

Background: 

Supracondylar humerus fractures are widely considered the most common fracture of the elbow in children. Fractures can range from a less severe, nondisplaced type I fracture to a more severe, displaced type III fracture with no cortical contact. Type III fractures can lead to adverse physical, social, and emotional consequences if they are not treated effectively. The American Academy of Orthopaedic Surgeons recently carried out a systematic review of the literature to develop a clinical practice guideline. The guidelines provided answers for the following questions regarding the treatment for type III supracondylar fractures (1) which is the preferred treatment for displaced supracondylar fractures of the humerus: reduction and casting versus closed reduction and percutaneous pinning; (2) which is the preferred method for fixing displaced supracondylar fractures of the humerus: medial (crossed) versus lateral pinning; and lastly, (3) does open reduction cause increased stiffness or have a high rate of complication? The purpose of this paper is to summarize and highlight the major findings from this systematic review.

Methods: 

PubMed, EMBASE, CINAHL, and The Cochrane Central Register of Controlled Trials were searched to locate 1726 relevant articles published from January 1966 to July 29, 2010. Of these, 44 met our criteria for inclusion and were reviewed systematically.

Results: 

On the basis of the results from the systematic review: (1) we suggest closed reduction with pin fixation for patients with displaced (eg, Wilkins type II and III and displaced flexion) pediatric supracondylar fractures of the humerus. (2) The practitioner might use 2 or 3 laterally introduced pins to stabilize the reduction of displaced pediatric supracondylar fractures of the humerus. Considerations of potential harm indicate that the physician might avoid the use of a medial pin. (3) The practitioner might perform open reduction for displaced pediatric supracondylar fractures of the humerus after closed reduction if varus or other malposition of the bone occurs.

Conclusions: 

Clearly, controversy exists regarding the best treatments for pediatric supracondylar humerus fractures. Properly designed randomized controlled trials comparing treatment options are necessary to determine optimal treatments.

Level of Evidence: 

Level II.

Supracondylar fractures of the humerus in children are the result of trauma to the elbow, most often sustained as a result of a fall from a height or during sports and leisure activities.¹ Supracondylar humerus fractures are widely considered the most common fracture of the elbow in children. The annual incidence of supracondylar fractures has been estimated at 177.3 per 100,000.¹

Supracondylar fractures are categorized as extension or flexion injuries. The extension type is the most common caused by a fall onto an outstretched hand with the elbow hyperextended. Of the many supracondylar fractures classification systems created, Wilkins modification of Gartland classification is the simplest and is widely used to describe extension-type fractures.² In this system, fractures are classified as type I fractures if the humerus is not displaced, type II if there is displacement with an intact posterior cortex and a variable amount of angulation, and type III if the humerus is displaced with no cortical contact.

The treatment options for these fractures vary based on the severity of the injury. A particular concern when managing supracondylar humerus fractures is the potential for this fracture to cause vascular compromise of the limb, which can lead to long-term loss of nerve and/or muscle function. Most treatments are associated with some known risks, especially invasive and operative treatments. Contraindications vary widely based on the treatment administered. The complications and burden of disease associated with these fractures, including the potential for a cubitus varus deformity, prolonged loss of mobility, and absence from school, can have adverse physical, social, and emotional consequences for the child and the child’s family. Treatments that minimize these concerns are, therefore, desirable.

Contemporary evidence-based medicine standards are for physicians to use the best available evidence from the literature to inform their clinical decision-making. American Academy of Orthopaedic Surgeons (AAOS) recently undertook an extensive systematic review of the literature to develop a clinical practice guideline to assist in treatment decision-making for displaced supracondylar fractures of the humerus in children. The purpose of the clinical practice guideline was to answer the following questions:

1. Which is the preferred treatment for displaced supracondylar fractures of the humerus: closed reduction and casting versus closed reduction and percutaneous pinning;
2. Which is the preferred method for fixing displaced supracondylar fractures of the humerus: medial (crossed) versus lateral pinning; and
3. Does open reduction cause increased stiffness or have a high rate of complication?

The purpose of this paper is to summarize and highlight the major findings from this systematic review. The complete AAOS Clinical Practice Guideline: Treatment of Pediatric Supracondylar Humerus Fractures is available at: http://www.aaos.org/research/guidelines/SupracondylarFracture/SupConFullGuideline.pdf.

METHODS

Systems of Classification

There are numerous fracture classification systems used by surgeons to help evaluate, plan, and standardize treatment. Classification systems communicate the displacement, comminution, and rotation of the fracture being treated, but no single classification system has perfect interobserver and intraobserver reliability. Further, no classification system can precisely classify all fractures without consideration of additional clinical factors including the mechanism of injury, time and duration since injury, soft tissue damage and swelling, and/or presence of neurovascular compromise. Hence, within the guideline, we reference the Wilkins Modification of the Gartland classification system² as a point of reference and not a standard for fracture classification.

The Wilkins Modification of the Gartland classification system² applies only to extension and not flexion fractures. However, within this paper, displaced fractures refer to both extension and flexion fractures. The ultimate goal of treatment is to achieve optimal outcomes for the patient.

Flynn et al’s³ criteria were used to grade the outcome of the displaced supracondylar fractures (Table 1). The criteria grades cosmetic and functional factors separately, based on the loss of carrying angle and loss of total range of motion. These measurements are made with a goniometer and compared with that of the normal opposite extremity.³

TABLE 1: Flynn Criteria for Grading Supracondylar Humerus Fractures³

Literature Searches

We searched 4 electronic databases: PubMed, EMBASE, CINAHL, and The Cochrane Central Register of Controlled Trials for articles published from January 1966 to July 29, 2010. Strategies for searching electronic databases were constructed by the AAOS Medical Librarian using previously published search strategies to identify relevant studies.^4–9 We supplemented searches of electronic databases with manual screening of the bibliographies of all retrieved publications. We also searched the bibliographies of recent systematic reviews and other review articles for potentially relevant citations. All articles identified were subject to the study selection criteria listed in the published clinical practice guidelines. The study attrition diagram (Fig. 1) provides details about the inclusion and exclusion of the studies considered for this guideline. Detailed methods that were used for extracting the data are outlined in the AAOS Clinical Practice Guideline.

FIGURE 1: Study attrition flowchart.

Quality

We evaluated the quality of the data for each outcome using a domain-based approach. Such an approach is used by the Cochrane Collaboration.¹⁰ Details for this can be found in the AAOS Clinical Practice Guideline: Treatment of Pediatric Supracondylar Humerus Fractures.

RESULTS

Question 1: Which is the preferred treatment for displaced supracondylar fractures of the humerus: closed reduction and casting versus closed reduction and percutaneous pinning?

Closed reduction and percutaneous Kirschner wire pinning is the preferred treatment for most displaced supracondylar fractures of the humerus. A meta-analysis demonstrates more favorable outcomes for closed reduction and pinning with regard to cubitus varus (Fig. 2) and Flynn elbow criteria (Fig. 3). The number needed to treat (NNT) for cubitus varus is 20 and the NNT for Flynn elbow criteria is 7. However, as may be expected, closed reduction and casting is favored when considering iatrogenic ulnar nerve injury as the outcome (Fig. 4). The number needed to harm (NNH) is 108.

FIGURE 2: Cubitus varus.

FIGURE 3: Flynn elbow criteria.

FIGURE 4: Iatrogenic ulnar nerve injury.

Question 2: Which is the preferred method for fixing displaced supracondylar fractures of the humerus: medial (crossed) versus lateral pinning?

When iatrogenic ulnar nerve injury is the outcome, meta-analysis data favor lateral pinning compared with medial (cross) pinning. The NNT for retrospective studies is 21 and the NNT for randomized controlled trials is 36 when the data are categorized by study type (Fig. 5). When the data are stratified by study quality (Fig. 6), the NNT for low-quality studies is 20, and an NNT cannot be calculated for the study of moderate quality. Using Flynn elbow criteria as the outcome (Fig. 7), medial pinning is favored with an NNT of 65. When change in Baumann angle (Fig. 8) is used as the outcome, there is almost no difference, with lateral pinning being slightly favored.

FIGURE 5: Iatrogenic ulnar nerve injury (by study design).

FIGURE 6: Iatrogenic ulnar nerve injury (by study quality).

FIGURE 7: Flynn elbow criteria.

FIGURE 8: Change in Baumann angle.

Question 3: Does open reduction cause increased stiffness or have a high rate of complication?

There are mixed results for elbow stiffness in open reduction for treatment of displaced supracondylar fractures. There is no statistically significant difference in most outcomes. Closed reduction has fewer cases of avascular necrosis of the trochlea, decreased time to union, decreased wound dehiscence, and better Flynn functional result (Fig. 9).

FIGURE 9: Summary of evidence for open reduction.

DISCUSSION

Question 1: Which is the preferred treatment for displaced supracondylar fractures of the humerus: closed reduction and casting versus closed reduction and percutaneous pinning?

On the basis of the review, the AAOS guidelines suggested closed reduction with pin fixation for patients with displaced (eg, Gartland type II and III and displaced flexion) pediatric supracondylar fractures of the humerus. Strength of recommendation: moderate. Evidence is addressed in 12 level III studies and 2 level II studies.

Forty-eight outcomes from 14 studies formed the basis of this recommendation. Wilkins type II and III fractures were analyzed in aggregate, as many of the studies combined the results from the 2 types. Similarly, the less common flexion type pediatric supracondylar fracture was included in this group. The quality, applicability, and the strength of the evidence generated a preliminary strength of recommendation of “weak.” The work group upgraded the recommendation to “moderate” based on the potential for harm from nonoperative treatment of displaced pediatric supracondylar fractures. For example, casting the arm in hyperflexion may cause limb-threatening ischemia. The initial recommendation of “weak” was based on the lack of evidence addressing the 6 critical outcomes that the work group had identified. Pin fixation was shown to be statistically superior to nonoperative treatment for 2 critical outcomes, prevention of cubitus varus and loss of motion.

Among the noncritical outcomes, pin fixation was statistically superior to nonoperative treatment in a meta-analysis of Flynn criteria (Fig. 3). This outcome incorporates both range of motion and carrying angle. Two noncritical outcomes, infection and pin-track infection, favored nonoperative treatment, because they can only occur in patients who receive operative treatment. Although operative treatment introduces the risk of infection, the improved critical outcomes combined with the decreased risk of limb-threatening ischemic injury outweigh these risks. Closed reduction and percutaneous Kirschner wire pinning is the preferred treatment for most displaced supracondylar fractures of the humerus. A meta-analysis demonstrates more favorable outcomes for closed reduction and pinning with regard to cubitus varus (Fig. 2) and Flynn elbow criteria (Fig. 3). The NNT for cubitus varus is 20 and the NNT for Flynn elbow criteria is 7. However, as may be expected, closed reduction and casting is favored when considering iatrogenic ulnar nerve injury as the outcome. The NNH is 108 (Fig. 4).

Question 2: Which is the preferred method for fixing displaced supracondylar fractures of the humerus: medial (crossed) versus lateral pinning?

The practitioner might use 2 or 3 laterally introduced pins to stabilize the reduction of displaced pediatric supracondylar fractures of the humerus. Considerations of potential harm indicate that the physician might avoid the use of a medial pin. Strength of recommendation: weak. Evidence is addressed in 13 level III studies and 4 level II studies.

Pin configuration and the potential complications related to instability and iatrogenic ulnar nerve injury are recognized concerns in this population. Therefore, the work group deemed it important to examine the technique of pin stabilization. Critical outcomes investigated were iatrogenic ulnar nerve injury, loss of reduction, malunion, and reoperation rate. This recommendation is based on data on 65 outcomes from 15 studies comparing pinning technique using lateral-only pin entry to lateral and medial crossed pin technique.

One randomized, prospective study by Kocher et al,¹¹ examined loss of reduction and found a loss of reduction rate of 21% (6/28) in lateral-only pins. Medial and lateral pins had a statistically significant lower loss of reduction rate of 4% (1/24). This loss of reduction was not clinically significant enough to warrant reoperation in either group. Meta-analysis of low-quality and moderate-quality studies found no statistically significant difference between lateral and medial pin configurations with respect to Baumann angle, Baumann angle change, Flynn criteria, and infection.

The ulnar nerve was injured in 3 of 557 (0.53%) cases with laterally introduced pins. Medially introduced pins resulted in 49 of 808 (6%) cases of ulnar nerve injury. Iatrogenic ulnar nerve injury was noted to be statistically significant in favor of lateral pinning in 6 of 11 studies. A meta-analysis of these studies and 3 additional underpowered studies (1 moderate quality and 13 low quality) also demonstrated a statistically significant effect in favor of lateral pinning (NNH=22, odds ratio=0.27). This suggests a 1 in 22 chance of harm resulting from the medial pinning techniques used in these studies. The NNH is similar to the results of a recently published meta-analysis on iatrogenic ulnar nerve injury after surgical treatment of displaced supracondylar fractures of the humerus, which had an NNH of 28.¹² On the basis of weak evidence, the practitioner might use 2 or 3 laterally introduced pins to stabilize the reduction of displaced pediatric supracondylar fractures of the humerus. The risk of potential harm from a medial pin must be weighed against the potential advantages.

Question 3: Does open reduction cause increased stiffness or have a high rate of complication?

The practitioner might perform open reduction for displaced pediatric supracondylar fractures of the humerus after closed reduction if varus or other malposition of the bone occurs. Strength of recommendation: weak. Evidence is addressed in 8 level III studies.

We recognized that a percentage of pediatric supracondylar fractures of the humerus cannot be reduced using a closed technique. Fracture pattern, soft tissue interposition, patient characteristics, and surgeon experience may contribute individually or in combination. In these more challenging cases, the surgeon may need to perform an open reduction.

Data on 28 outcomes from 8 studies were analyzed. Significant flaws in study design limited the strength of all the studies. The critical outcomes studied were cubitus varus, hyperextension, loss of reduction, malunion, pain, and elbow stiffness. Statistically significant data were found for only 2 of these outcomes. Aktekin et al¹³ reported that stiffness was greater in the patients treated with open reduction compared with patients treated with a closed reduction and pinning. Li et al¹⁴ reported that the fractures treated open had a lower incidence of loss of reduction compared with displaced fractures that could be managed successfully with closed reduction and pinning. Sibly et al¹⁵ found no statistically significant difference between groups for cubitus varus or elbow stiffness.

These nonrandomized retrospective studies are prone to selection bias. More severe injuries may have been selected for open reduction, potentially confounding the comparative data. We could not determine whether adverse outcomes in the open reduction group were due to the severity of injury or to the intervention. Furthermore, the literature lacks clear definitions for an acceptable reduction.

Future Research

Despite being the most common fracture of the elbow in children, high-quality scientific data regarding the treatment of pediatric supracondylar humerus fractures are lacking.

Clearly, controversy exists regarding the best treatments for pediatric supracondylar humerus fractures. Properly designed randomized controlled trials comparing treatment options are necessary to determine optimal treatments. These trials should focus on patient-oriented outcomes using psychometrically validated instruments and also consider adverse events that commonly occur during treatment of these fractures. They should be subject to a priori power analysis to ensure clinically important improvements (ie, improvements that matter to the patients). Consideration may also be given to validated family-based outcomes, because their inclusion may improve recommendations for younger patients. Future studies would also benefit from attempts to increase the applicability of study results (ie, generalizability) as described by the PRECIS instrument.¹⁶

Specific trials that would be helpful to improve recommendations include:

• Prospective investigation of the adequacy of the initial reduction against outcome, with a focus on establishing criteria for accepting a closed reduction.
• Prospective randomized studies comparing medial with lateral entry pin fixation focusing on patient-centered outcomes and adverse events (eg, iatrogenic ulnar nerve injuries) along with maintenance and quality of reduction.
• Prospective investigation of the treatment options for fractures that cannot be reduced by closed reduction.
• Prospective investigation of the patient-centered outcomes and adverse events of treatment of vascular compromise.
• Prospective cohort investigation of the optimal time threshold for surgery.
• Prospective investigation comparing timing for removal of pins, timing for resumption of activities, and results of physical therapy.

ACKNOWLEDGMENTS

AAOS POSNA Collaboration Treatment of Supracondylar Fractures Work Group: Andrew Howard MD, Mark F. Abel MD, Stuart Braun MD, Matthew Bueche MD, Howard Epps MD, Harish Hosalkar MD, Charles T. Mehlman DO, MPH, Susan Scherl MD; Guidelines Oversight Chair Michael Goldberg MD; and the AAOS Staff: Charles M. Turkelson PhD, Janet L. Wies MPH, and Kevin Boyer.