Reliability of the modified Gartland classification in pediatric supracondylar humeral fractures

Ethan Choo; Jie Er Janice Soo; Rehena Sultana; Kenneth Pak Leung Wong; Neeraj Mishra; Summaiyya Hanum Ahamed; Kar Yee Catrin Kong; Arjandas Mahadev; Sashikumar Ganapathy; Khai Pin Lee; Shu-Ling Chong

doi:10.1016/j.jorep.2023.100267

View/Download PDF

Buy Reprints

PDF

Translate this page into:

Research Article

2024

:3;

100267

doi:

10.1016/j.jorep.2023.100267

Reliability of the modified Gartland classification in pediatric supracondylar humeral fractures

Ethan Choo^a,b, Jie Er Janice Soo^c, Rehena Sultana^d, Kenneth Pak Leung Wong^e,h, Neeraj Mishra^e, Summaiyya Hanum Ahamed^f, Kar Yee Catrin Kong^g,, Arjandas Mahadev^e,h, Sashikumar Ganapathy^d,h, Khai Pin Lee^c,h, Shu-Ling Chong^c,h

a Eastern Health Clinical School, Box Hill Hospital, Victoria, Australia

b Faculty of Nursing and Health Sciences, Monash University, Victoria, Australia

c Department of Emergency Medicine, KK Women's and Children's Hospital, Singapore

d Centre for Quantitative Medicine, Duke-NUS Medical School, Singapore

e Department of Paediatrics Orthopaedics, KK Women's and Children's Hospital, Singapore

f Department of Radiology, KK Women's and Children's Hospital, Singapore

g Department of Paediatric Medicine, KK Women's and Children's Hospital, Singapore

h Duke-NUS Medical School, Singapore

∗Corresponding author: Kar Yee Catrin Kong. Kenneth.wong.p.l@singhealth.com.sg

Received: 2023-11-3, Accepted: 2023-11-19, Published: 2023-11-22

The Authors

Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

To assess the inter-rater reliability of the modified Gartland classification together with other radiological parameters for extension type supracondylar humerus fractures (SCHF) in children.

A retrospective cohort study was conducted at a tertiary children's hospital in Singapore on children less than 12 years old with a diagnosis of extension-type SCHF between January 2017 and October 2017. Anteroposterior and lateral radiographs were assessed by fellowship-trained radiologist and orthopaedic surgeon. The radiographs were assessed for any breach of anterior cortex, breach of posterior cortex, presence of medial comminution, disrupted hourglass configuration and abnormal Baumann's angle. CorrespondingCohen's κ values were used to assess the level of inter-rater agreement.

We found an inter-rater agreement for the modified Gartland classification of κ = 0.141 (95 % CI, 0.049 to 0.233). Inter-rater agreement was κ = 0.654 (95 % CI, 0.493–0.815) for assessing a breach of anterior cortex was, κ = 0.357 (95 % CI, 0.135–0.578) for Baumann's angle, κ = 0.317 (95 % CI, 0.145–0.488) for a disrupted hourglass configuration, κ = 0.271 (95 % CI, 0.105–0.437) for a breach of posterior cortex and κ = 0.131 (95 % CI, −0.127 – 0.389) for the presence of medial comminution.

We found a low inter-rater reliability for the modified Gartland classification. The breach of anterior and posterior cortex, Baumann's angle and hourglass disruption have higher inter-rater reliability.

1 Introduction

Supracondylar humerus fractures (SCHFs) are common in children, accounting for up to 60 % of elbow fractures and 18 % of all pediatric fractures.^1–4 Left untreated, children with SCHF can go on to develop neurovascular complications including compartment syndrome, malunion, cubitus varus or valgus deformities.⁵^,⁶ Such fractures also incur significant direct costs to the families as well as loss of productivity due to absence from school and work.⁷

SCHFs can be classified into either flexion or extension-type fractures, with the latter being more common and most often caused by a fall on an outstretched hand on a hyperextended elbow.⁸^,⁹ Clinicians have traditionally divided the management of SCHF into conservative versus operative treatment, guided by the Gartland classification.²^,¹⁰ The American Academy of Orthopedic Surgeons (AAOS) Clinical Practice guidelines advocates for conservative management for Gartland Type I fractures and closed reduction with surgical fixation for Gartland Type II and III fractures.³^,¹¹^,¹² Similarly, the British Orthopaedic Association Standard for Trauma (BOAST) also advocates for early surgical fixation of displaced (Gartland Type II and III) fractures.¹³ With the Wilkin's modification, others have suggested that Gartland Type IIa fractures should be managed non-operatively while Gartland Type IIb fractures should undergo operative management.¹⁴^,¹⁵

Although there is concurrence that Gartland Type I fractures should be managed conservatively and Gartland Type III with surgical intervention, the treatment of Gartland Type II fractures remains controversial.¹⁶ Some advocate the use of operative management for Gartland Type II fractures to obtain better reduction, reduce the risk of developing compartment syndrome in the short-term and malunion in the long-term.^17–19 Others prefer not to operate on these fractures, citing good clinical outcomes without the added surgical risks.²⁰

Given the need for a reliable tool to classify SCHFs to determine subsequent management, there is a need to appraise how reliable the Gartland classification is. Several authors have reported low rates of both intra-rater and inter-rater agreement in distinguishing between Gartland Type IIa and Gartland Type IIb fractures.¹⁶^,²¹

2 Aims

We aim to assess the inter-rater reliability of the Gartland classification among children less than 12 years old with a diagnosis of SCHF. In addition, we also intend to investigate the corresponding inter-rater reliability of other commonly used fracture classification criteria such as Baumann's angle, breach of anterior cortex, breach of posterior cortex, presence of medial comminution, and a disrupted hourglass configuration.

3 Methods

3.1

3.1 Study design

This is a retrospective observational cohort study between January 2017 and October 2017.

Our ED is sited in a tertiary pediatric institution with a specialist orthopaedic department that provides 24/7 on-site support to the ED, 7 days a week. Trauma cases account for about 25,000 attendances a year, a large proportion of which are minor trauma cases.

3.2

3.2 Study population

We included children from birth to the age of 12 years old with a clinical diagnosis of SCHF as outlined by the International Classification of Diseases (ICD-10). Children with polytrauma, presence of a concomitant ipsilateral upper limb fracture, repeat injury to the same elbow and pathological fractures were excluded. Children with flexion type fractures were excluded as well due to the different mechanism of injury resulting in such fractures when compared to extension type fractures.²² Furthermore, flexion type fractures tend to have a higher morbidity compared to extension type fractures and often result in different complications such as an increased risk of ulnar nerve injury rather than median nerve injury seen in extension type fractures.²³^,²⁴ We chose to focus on SCHFs that constituted a diagnostic dilemma.¹⁶ Therefore, we excluded occult SCHFs (fat pad present only) and un-displaced fractures. We also excluded obviously displaced fractures that would clearly necessitate surgical management.

3.3

3.3 Data variables

The following clinical variables were obtained: patient demographics (age, gender, weight, co-morbidities), triage priority, clinical history and examination (mechanism of injury, laterality of fracture, neurovascular compromise, displacement, compound or simple fracture), clinical outcomes and management (manipulation and reduction or surgical fixation), and time spent in the ED.

The elbow radiographs on presentation to the Emergency Department were extracted from the electronic health record. For each patient, both anteroposterior and lateral radiographs of the elbow were deidentified and assessed independently by a fellowship accredited Radiologist and an Orthopaedic Surgeon with more than a decade of experience in pediatric orthopedics. Both individuals were blinded to patient history and examination findings. The radiographs were reviewed for the following characteristics: Gartland classification, Baumann's angle, breach of anterior cortex, breach of posterior cortex, the presence of medial comminution and disrupted hourglass configuration.

For the Gartland classification, we divided them into: Non-displaced (Gartland Type I), displaced with intact posterior cortex without rotational deformity (Gartland Type IIa), displaced with intact posterior cortex with rotational deformity (Gartland Type IIb) and complete displacement (Gartland Type III).¹⁴^,²⁵

We defined a normal Baumann angle as a range of 64° to 81°.¹^,²⁶^,²⁷ The Baumann angle was therefore deemed abnormal if it fell out of this range.

3.4

3.4 Analytical plan

All continuous and categorical variables were reported as mean (standard deviation (SD)) and frequency (percentage) respectively. Reliability of the orthopedic surgeon's and a radiologist's judgement on whether individuals had a Gartland Type I, Type IIa, Type IIb or Type III SCHF, normal or abnormal Baumann angle, breach of anterior cortex, breach of posterior cortex, presence of medial comminution and disrupted hourglass configuration was performed using Cohen's kappa. The kappa values obtained were subsequently interpreted according to the guidelines outlined by Landis and Koch (Table 1).²⁸

Table 1 Interpretation of kappa statistic.

Kappa Statistic	Strength of the Agreement
<0.000	Poor
0.000–0.200	Slight
0.210–0.400	Fair
0.410–0.600	Moderate
0.610–0.800	Substantial
0.810–1.000	Almost perfect

This study was given approval by our local ethics board 2018/2268.

4 Results

Out of the 601 children involved in the study, we analyzed 129 patients with SCHF (Fig. 1). Out of these 129 patients, 101 of these patients were conservatively managed while 28 of these patients were managed operatively.

Among the 129 patients studied with SCHF, 40 (31.0 %) of these patients sustained SCHFs due to a fall on an outstretched hand and 33 (25.6 %) of these patients due to a fall on an elbow. The remaining 59 (43.4 %) patients had SCHFs due to an unwitnessed fall or a fall by which the mechanism was unclear.

Among the 129 patients studied, the mean time spent in the ED was 109 ± 78.0 minutes. While in the ED, 9 (7.0 %) patients required sedation with intramuscular (IM) ketamine and 6 (4.7 %) patients required sedation with intranasal (IN) fentanyl. 94 (72.9 %) of the patients required Orthopedics specialist review in the ED. 107 (82.9 %) patients were eventually discharged from the ED, 21 (16.3 %) patients required hospital admission and 1 (0.8 %) patient was discharged against medical advice.

The demographics and radiological characteristics of the population analyzed is summarized in Table 2.

Table 2 Demographics, fracture characteristics and management outcome of study population (n = 129).

Variable	Conservative (n = 101) N (%)	Operative (n = 28) N (%)	Overall (n = 129) N (%)
Age (Years), mean ± SD	5.5 ± 2.5	5.3 ± 2.3	5.4 ± 2.4
Sex, n(%)
Male	60 (59.4)	17 (60.7)	77 (59.7)
Laterality, n(%)
Left	57 (56.4)	16 (57.1)	73 (56.6)
Right	44 (43.6)	12 (42.9)	56 (43.4)
Radiological Characteristics
Anterior cortex breached, n(%)	72 (71.3)	27 (96.4)	99 (76.7)
Posterior cortex breached, n(%)	34 (33.7)	15 (53.6)	49 (38.0)
Presence of medial comminution, n(%)	1 (1.0)	5 (17.9)	6 (4.7)
Hourglass configuration disrupted, n(%)	54 (53.5)	22 (78.6)	76 (58.9)
Baumann's angleMean (°) ± SD	71.0 ± 5.9	72.8 ± 8.3	71.4 ± 6.5
Abnormal Baumann's angle, n(%)	12 (11.9)	10 (35.7)	22 (17.1)

Among all 129 patients studied, the mean age was 5.4 ± 2.4 years and 77 (59.7 %) were males. 73 (56.6 %) had a left-sided fracture while 56 (43.4 %) had a right-sided fracture. Among 101 conservatively managed patients, 72 (71.3 %), 34 (33.7 %), 1 (1.0 %), 54 (53.5 %) and 12 (11.9 %) of had a breach of the anterior cortex, breach of the posterior cortex, presence of medial comminution, disrupted hourglass configuration and abnormal Baumann's angle respectively.

Among 28 children who received operative management, 27 (96.4 %), 15 (53.6 %), 5 (17.9 %), 22 (78.6 %) and 10 (35.7 %) of these patients had a breach of the anterior cortex, breach of the posterior cortex, presence of medial comminution, disrupted hourglass configuration and abnormal Baumann's angle respectively.

For the Modified Gartland classification, we found a Cohen's k of 0.141 (95 % CI, 0.049 to 0.233) between two physician's judgement. The agreement was greatest for breach of anterior cortex (κ = 0.654, 95 % CI, 0.493 to 0.815), followed by Baumman's angle (κ = 0.357, 95 % CI, 0.135 to 0.578), and disrupted hourglass configuration (κ = 0.317, 95 % CI, 0. to 0.619). We did not find a strong agreement for breach of posterior cortex and presence of medial communication (Table 3)

Table 3 Interobserver agreement for commonly used fracture classification criteria.

Classification variable	Overall κ, (95 % CI)	Interpretation
Modified Gartland Classification	0.141 (0.049–0.233)	Slight
Anterior cortex breached	0.654 (0.493–0.815)	Substantial
Posterior cortex breached	0.271 (0.105–0.437)	Fair
Presence of medial comminution	0.131 (−0.127–0.389)	Slight
Hourglass configuration disrupted	0.317 (0.145–0.488)	Fair
Baumann's angle	0.357 (0.135–0.578)	Fair

5 Discussion

Based on the results, this study demonstrated only slight interobserver agreement in the radiological classification of SCHF using the modified Gartland classification (Cohen's k of 0.141, 95 % CI, 0.049 to 0.233). We found the Gartland classification to be less reliable than other radiographical measures including breach of anterior context, abnormal Baumman's angle, disruption of the hourglass configuration and breach of posterior cortex.

Prior studies that sought to investigate the reliability of the Gartland Classification for extension type SCHF report varying results. Overall, there tends to be a high level of agreement when looking at occult and obviously displaced SCHFs, but a low level of agreement for fractures of all other severities.¹⁶^,²¹^,^29–31 Our findings reinforce these reports that for fractures that were not clearly occult nor obviously displaced, the level of agreement between 2 independent trained staff is very low.

As compared to other fracture classification criteria such as breach of anterior cortex, breach of posterior cortex, disrupted hourglass configuration and abnormal Baumann angle, the modified Gartland classification scored lower in terms of the level of agreement between the two raters. This would suggest that the modified Gartland classification when used alone, is not reliable when used to evaluate extension-type SCHF. Of note, the classification criterion looking at the breach of anterior cortex showed a substantial level of agreement between the two raters. Our findings suggest that the breach of anterior cortex, abnormal Baumann's angle, and disruption of the hourglass configuration can provide additional radiographic value for SCHF fracture classification, given their higher interrater reliability.

Other studies have also looked at the reliability of other fracture classification criteria and radiographic findings. A study by Silva et al., in 2010 demonstrated excellent inter-rater and intra-observer reliability with Baumann's angle and would prove useful in determining outcomes of SCHFs in children.³² Another study by Murphy-Zane et al. demonstrated greater inter-observer and intra-observer reliability when comparing the Anterior Humeral Line Index which looks at the degree of displacement of the anterior cortex with the Gartland classification.³³ These studies further highlight the possibility of incorporating other radiographic findings when deciding on treatment modality for SCHFs.

We recognize the limitations of this study. Firstly, we chose to include only SCHFs that would pose a diagnostic dilemma, and excluded occult fractures and those that were obviously displaced. While we did not report the reliability of the Gartland classification for these, we believe that we were therefore able to focus our investigation on fractures that posed a diagnostic challenge. Secondly, we limited our study to Gartland and other radiographic features in this study, and did not include clinical factors, including neurovascular status, for fracture severity. The latter may have contributed to decision-making on the need for operative management. Finally, being a single-center study, our findings may not be generalizable to other healthcare institutions and settings. Further prospective research is needed to include radiographic interpretation from physicians of different disciplines (emergency medicine and family medicine), at different stages of training.

6 Conclusion

The results of this study suggest that the level of agreement for the Modified Gartland Classification is low between fellowship trained physicians practicing in Singapore. In addition, other radiological characteristics such as the breach of anterior cortex and Baumann's should be considered when deciding on treatment of Gartland Type II fractures.

Ethics approval

The study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by our Institutional Ethics Review Board. As all data had been anonymised, no consent to participate or to publish were taken. This was also waived by our local ethics board.

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Conflict of interests/funding support

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript. The authors have no relevant financial or non-financial interests to disclose.

CRediT authorship contribution statement

Ethan Choo: Conceptualization, Methodology, Validation, Investigation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Jie Er Janice Soo: Conceptualization, Methodology, Validation, Investigation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Rehena Sultana: Conceptualization, Methodology, Writing – review & editing, Visualization. Kenneth Pak Leung Wong: Conceptualization, Methodology, Writing – review & editing, Visualization. Neeraj Mishra: Conceptualization, Methodology, Writing – review & editing, Visualization. Summaiyya Hanum Ahamed: Conceptualization, Methodology, Writing – review & editing, Visualization. Kar Yee Catrin Kong: Conceptualization, Methodology, Writing – review & editing, Visualization. Arjandas Mahadev: Conceptualization, Methodology, Writing – review & editing, Visualization. Sashikumar Ganapathy: Conceptualization, Methodology, Writing – review & editing, Visualization. Khai Pin Lee: Conceptualization, Methodology, Writing – review & editing, Visualization. Shu-Ling Chong: Conceptualization, Methodology, Validation, Investigation, Writing – original draft, Writing – review & editing, Visualization, Project administration.

References

Williamson D.M., Coates C.J., Miller R.K., Cole W.G., . Normal characteristics of the Baumann (humerocapitellar) angle: an aid in assessment of supracondylar fractures. J Pediatr Orthop. 1992;12:636-639.
[Google Scholar]
Shenoy P., Islam A., Puri R., . Current management of paediatric supracondylar fractures of the humerus. Cureus. 2020;12(5)
[Google Scholar]
Mulpuri K., Wilkins K., . The treatment of displaced supracondylar humerus fractures. J Pediatr Orthop. 2012;32(Supplement 2):S143-S152.
[Google Scholar]
Omid R., Choi P., Skaggs D., . Supracondylar humeral fractures in children. J Bone Jt Surg Am Vol. 2008;90(5):1121-1132.
[Google Scholar]
Oetgen M., Mirick G., Atwater L., Lovejoy J., . Complications and predictors of need for return to the operating room in the treatment of supracondylar humerus fractures in children. Open Orthop J. 2015;9(1):139-142.
[Google Scholar]
Babal J.C., Mehlman C.T., Klein G., . Nerve injuries associated with pediatric supracondylar humeral fractures: a meta-analysis. J Pediatr Orthop. 2010;30:253-263.
[Google Scholar]
Holm A., Lurås H., Randsborg P., . The economic burden of outpatient appointments following paediatric fractures. Injury. 2016;47(7):1410-1413.
[Google Scholar]
Kuoppala E., Parviainen R., Pokka T., Sirviö M., Serlo W., Sinikumpu J., . Low incidence of flexion-type supracondylar humerus fractures but high rate of complications. Acta Orthop. 2016;87(4):406-411.
[Google Scholar]
Kumar V., Singh A., . Fracture supracondylar humerus: a review. J Clin Diagn Res. 2016;10(12):RE01-RE06.
[Google Scholar]
Gartland J.J., . Management of supracondylar fractures of the humerus in children. Surg Gynecol Obstet. 1959 Aug;109(2):145-154.
[Google Scholar]
Mulpuri K., Hosalkar H., Howard A., . AAOS clinical Practice guideline: the treatment of pediatric supracondylar humerus fractures. J Am Acad Orthop Surg. 2012;20(5):328-330.
[Google Scholar]
Howard A., Mulpuri K., Abel M., et al . The treatment of pediatric supracondylar humerus fractures. J Am Acad Orthop Surg. 2012;20(5):320-327.
[Google Scholar]
2020
[Google Scholar]
Alton T., Werner S., Gee A., . Classifications in brief: the Gartland classification of supracondylar humerus fractures. Clin Orthop Relat Res. 2015;473(2):738-741.
[Google Scholar]
Zorrilla S., de Neira J., Prada-Cañizares A., Marti-Ciruelos R., Pretell-Mazzini J., . Supracondylar humeral fractures in children: current concepts for management and prognosis. Int Orthop. 2015;39(11):2287-2296.
[Google Scholar]
Leung S., Paryavi E., Herman M., Sponseller P., Abzug J., . Does the modified Gartland classification clarify decision making? J Pediatr Orthop. 2018;38(1):22-26.
[Google Scholar]
Ariyawatkul T., Eamsobhana P., Kaewpornsawan K., . The necessity of fixation in Gartland type 2 supracondylar fracture of the distal humerus in children (modified Gartland type 2A and 2B) J Pediatr Orthop B. 2016;25(2):159-164.
[Google Scholar]
Kaewpornsawan K., . Comparison between closed reduction with percutaneous pinning and open reduction with pinning in children with closed totally displaced supracondylar humeral fractures: a randomized controlled trial. J Pediatr Orthop B. 2001 Apr;10(2):131-137.
[Google Scholar]
Skaggs D., Sankar W., Albrektson J., Vaishnav S., Choi P., Kay R., . How safe is the operative treatment of Gartland type 2 supracondylar humerus fractures in children? J Pediatr Orthop. 2008;28(2):139-141.
[Google Scholar]
Padman M., Warwick A., Fernandes J., Flowers M., Davies A., Bell M., . Closed reduction and stabilization of supracondylar fractures of the humerus in children: the crucial factor of surgical experience. J Pediatr Orthop B. 2010;19(4):298-303.
[Google Scholar]
Heal J., Bould M., Livingstone J., Blewitt N., Blom A., . Reproducibility of the Gartland classification for supracondylar humeral fractures in children. J Orthop Surg. 2007;15(1):12-14.
[Google Scholar]
Sharma A., . The flexion-type supracondylar humeral fracture in children. JBJS Rev. 2019;7(4)
[Google Scholar]
Babal J.C., Mehlman C.T., Klein G., . Nerve injuries associated with pediatric supracondylar humeral fractures: a meta-analysis. J Pediatr Orthop. 2010;30:253-263.
[Google Scholar]
Steinman S., Bastrom T., Newton P., Mubarak S., . Beware of ulnar nerve entrapment in flexion-type supracondylar humerus fractures. J Children's Orthopaedics. 2007;1(3):177-180.
[Google Scholar]
Wilkins K., . Fractures in children. 1991:509-828.
[Google Scholar]
Smajic N., Smajic J., Sadic S., Jasarevic M., AhmetovicDjug J., Hodzic R., . Correlation between bauman's and carrying angle in children with supracondylar fracture of humerus. Med Arch. 2013;67(3):195.
[Google Scholar]
Camp J., Ishizue K., Gomez M., Gelberman R., Akeson W., . Alteration of Baumannʼs angle by humeral position. J Pediatr Orthop. 1993;13(4):521-525.
[Google Scholar]
Landis J., Koch G., . The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159.
[Google Scholar]
Barton K., Kaminsky C., Green D., Shean C., Kautz S., Skaggs D., . Reliability of a modified Gartland classification of supracondylar humerus fractures. J Pediatr Orthop 2001:27-30.
[Google Scholar]
Rocha I., Faria A., Filho C., Rocha M., . Reproducibility of the AO/ASIF and Gartland classifications for supracondylar fractures of the humerus in children. Revista Brasileira de Ortopedia (English Edition). 2015;50(3):266-269.
[Google Scholar]
Teo T., Schaeffer E., Habib E., et al . Assessing the reliability of the modified Gartland classification system for extension-type supracondylar humerus fractures. J Children's Orthopaedics. 2019;13(6):569-574.
[Google Scholar]
Silva M., Pandarinath R., Farng E., et al . Inter- and intra-observer reliability of the Baumann angle of the humerus in children with supracondylar humeral fractures. Int Orthop. 2009;34(4):553-557.
[Google Scholar]
Murphy-Zane M., Pyle L., . Reliability of the anterior humeral line Index compared with the Gartland classification for posteriorly hinged supracondylar humerus fractures. Orthopedics. 2018;41(4)
[Google Scholar]