Comparison of Crossed Versus Lateral Pinning in Fixation of Extension Type of Supracondylar Humerus Fracture in Pediatric Age Group- A Randomized Control Trial

Abstract

Background

Various techniques for fixation of pediatric supracondylar humerus fracture have been in practice, but which is a better option is still controversial.

Purpose of study

There is a varied consensus on the outcome of fracture fixation based on different pinning methods. Hence, this study aims to compare the functional and radiological outcomes of pediatric supracondylar humerus fracture fixed by percutaneous cross-pinning or lateral pinning.

Methodology

We conducted a prospective, single-blinded, randomized control trial with 60 patients divided into two groups of 30 each, ages ranging from less than one year to 13 years, with Gartland type 2 or 3 supracondylar humerus fractures. The outcome of fracture fixation treated with either crossed or lateral pinning was assessed with a six-month follow-up by comparing the pre-operative and post-operative Baumann angle, post-operative carrying angle and the Flynn criteria.

Result

There was no significant difference in the radiological outcome of the supracondylar humerus fracture fixed with percutaneous cross-pinning or lateral pinning. No patient had any iatrogenic ulnar nerve injury. All 60 patients had excellent Flynn criteria at the end of the six-month follow-up.

Conclusion

Both the techniques of percutaneous pinning showed comparable bony union. However, lateral pinning had better functional outcomes due to minimal stiffness and a good range of movements at four weeks. The iatrogenic ulnar nerve injury seen commonly in cross-pinning could be avoided with a 1 cm skin incision over the medial condyle to visualize the ulnar nerve and the medial kirschner wire entry made anteriorly.

Clinical trial registry-india registration number

CTRI202404066233

Graphic abstract

Introduction

Supracondylar humerus fracture is more common in paediatric age due to the thin supracondylar metaphyseal ridge, which converts the axial force to bending force in this area during a fall on an outstretched hand [1, 2]. Various surgical methods exist to fix Gartland type 2 and 3 fractures. Closed reduction percutaneous pinning is preferred with different techniques of Kirschner-wire insertion, crossed with medial–lateral pin entry and only lateral pin entry. Even though crossed-pinning method is found to be more biomechanically stable when compared to lateral-pinning method, the risk of iatrogenic ulnar nerve injury cannot be negated and newer studies have proven that lateral-pinning provided comparable torsional stability compared to the crossed-pinning method [3]. Hence, this study aims to compare the functional and radiological outcomes of pediatric supracondylar humerus fractures treated with crossed-pinning or lateral-pinning following closed reduction of the fracture. This study was done focussing only on closed supracondylar humerus fractures, in particular, extension type (most commonly noticed type) without any pre-operative neurovascular insult, as not many studies have been conducted under these specifications.

Materials and Methods

We conducted a prospective randomised control study in the Department of Orthopaedics, in our institution, for one year, after obtaining institutional ethical committee approval and registering the trial in the national clinical trial registry. Before participating in this study, written informed assent was taken from the parents/legal guardians of the participants.

The sample size was calculated using the formula: $n\ge \frac{{{[Z}_{1-\frac{\alpha }{2}}+Z_{1-\beta }\text{]}}^2\text{[}{\sigma }_1^2+{\sigma }_2^2\text{]}}{{{(\mu }_1-{\mu }_2)}^2}$

Where $Z_{1-\frac{\alpha }{2}}$ = 1.96 corresponded to a 95% confidence interval; $Z_{1-\beta }$ = 0.84 corresponded to 80% power; ${\sigma }_1$= 1.6 and ${\sigma }_2$= 3 were taken as the standard deviation for both the groups respectively; ${\mu }_1$= 28.3; ${\mu }_2$= 30 were taken as the means of the two groups respectively and hence the required sample size was calculated to be 30 participants in each group and the sample size was calculated based on the study conducted by Xing et al. [4].

Sixty four patients were recruited for the study (Fig. 1). Children of ages ranging from less than one year to 13 years [5] with Gartland type 2 or 3 supracondylar humerus fractures who presented on the day of the injury were included in the study. The children who presented with a compound fracture, vascular injury, flexion type of fracture pattern, aged above 13 years and whose parents were unwilling to assent to the study were excluded. Hence, two patients who presented with a flexion type of fracture on the day of injury and two patients who presented on the second day of injury were excluded from the study. However, all four patients received the necessary treatment for the fracture. The study proceeded with 60 patients. None of the patients dropped out of the study or were lost to follow-up.

Fig. 1.

CONSORT flow diagram (original figure)

They were divided into two groups by the investigator using sealed envelopes and were designated as groups A and B based on computer-generated randomization (random number generator was used to allot patients into either group A, numbered as 1 or group B, numbered as 2) performed by the investigator on the day of presentation to the hospital. This randomized number was only known to the investigator and operating surgeon and not to the patient or the patient’s parents/legal guardians throughout the study period (single blinding of the patient and the patient’s parents/legal guardians). Patient’s parents/legal guardians were counselled on the surgical procedure while obtaining assent for surgery. They were informed about the surgical technique of fracture fixation that the patient was to be undergone as well as the possible surgical and post-operative complications. Baumann’s angle was measured pre-operatively in the fractured limb as well as the uninjured limb by calculating the angle between long axis of shaft of humerus and physeal line of lateral condyle [6]. The carrying angle was measured pre-operatively only on the uninjured limb.

Group A patients underwent closed reduction with percutaneous cross-pinning with a single 2 mm kirschner wire inserted through the medial condyle and a single or two 2 mm kirschner wires inserted through the lateral condyle (Fig. 2). Group B patients underwent closed reduction with percutaneous lateral pinning with either two or three 2 mm kirschner wires inserted through the lateral condyle (Fig. 3). Both groups were immobilised with an above elbow plaster for four weeks after which the kirschner wires were removed and the elbow range of movements was initiated.

Fig. 2.

x-ray of left elbow joint anteroposterior and lateral view of Group A patient; 2a- pre-operative x-ray showing Gartland type 2 supracondylar humerus fracture; 2b- intraoperative image of Kirschner-wires fixed by cross-pinning method; 2c- pre-operative Baumann’s angle calculation, 2d- post-operative x-ray with kirschner-wires in-situ (original figure)

Fig. 3.

x-ray of left elbow joint anteroposterior and lateral view of Group B patient; 3a- pre-operative x-ray showing Gartland type 2 supracondylar humerus fracture; 3b- intraoperative image of kirschner-wires fixed by cross-pinning method; 3c- pre-operative Baumann’s angle calculation, 3d- post-operative x-ray with kirschner-wires in-situ (original figure)

Following the kirschner wire removal at one-month follow-up, the Baumann angle was measured on the operated elbow. The children were started on active elbow range of movements. At three-month and six-month follow-ups, Baumann angle and carrying angle were measured on the operated elbow and Flynn criteria (which considers carrying angle and elbow range of movements to assess the quality of surgical outcome [2]) were assessed.

The results were statistically analysed using SPSS (statistical package for social sciences) version 21.0 and a p-value of < 0.05 was considered statistically significant with a 95% confidence interval by using the unpaired t-test and ANOVA test for measuring the radiological outcome of the Baumann angle and carrying angle and using the unpaired t-test and Fisher’s Exact test for measuring the functional outcome of elbow range of movements and Flynn criteria respectively.

Results

We found that among the 60 patients included in the study, most of the patients were males and the left elbow was most commonly injured compared to the right in both the groups. The age range was 3–9 years in both groups. The mean age was compared using an unpaired t-test for both groups and was found to have no statistical significance (p-value 0.969). The mechanism of injury among a majority of the patients in both groups was slip and fall and had a fall on an outstretched hand. Gartland type 3 supracondylar humerus fracture was most commonly noted in group A (19 patients) and type 2 fracture was noted more commonly in group B (16 patients). There was no statistical significance in the duration of surgery in both groups using an unpaired t-test (p-value 0.237) (Table 1).

Table 1.

Patient demographics

Patient demographics	Group A (n = 30)	Group B (n = 30)	T value	P value	95% Confidence interval
Mean age (in years)	6.27 ± 2.92	6.24 ± 3.19	0.038	0.969	− 1.5505 to 1.6105
Side of injury
Left	17	15
Right	13	15
Gender
Male	20	21
Female	10	9
Mechanism of injury
Slip and fall on outstretched hand	11	12
Slip and fall direct injury to elbow	6	4
Fall from height on outstretched hand	8	11
Fall from height direct injury to elbow	5	3
Gartland
Type 2	11	16
Type 3	19	14
Mean duration of surgery (in minutes)	56.23 ± 9.50	53.60 ± 7.45	1.1932	0.237	− 1.7821 to 7.0421

Though there was no statistical significance found in comparing the Baumann angle of both the groups at each time interval of measurement of the angle by using the unpaired t-test, there was a statistical significance in comparing the Baumann angle preoperatively and at each postoperative follow-up period in group A and group B individually using ANOVA (p-value < 0.00001 for both the groups; F ratio- 77.973 for group A and F ratio- 140.2321 for group B).

Similarly, though there was no statistical significance found in comparing the carrying angle of both the groups at each time interval of measurement of the angle by using the unpaired t-test (Fig. 4), there was a statistical significance in comparing the carrying angle preoperatively and at each postoperative follow-up period in group A and group B individually by using ANOVA (p-value 0.00018 for both the groups; F ratio- 10.04263 for group A, F ratio- 9.98946 for group B) (Table 2).

Fig. 4.

carrying angle at 6 months of post-operative follow-up; 4a- group A patient; 4b- group B patient (original figure)

Table 2.

Outcome measurements

Outcomes	Group A (n = 30)	Group B (n = 30)	T value	P value	95% Confidence interval
Baumann angle
Pre operative	83.51 ± 7.17	85.36 ± 5.95	1.0875	0.281	− 5.2551 to 1.5551
post operative-
1 month	74.28 ± 4.28	74.04 ± 3.84	0.2286	0.820	− 1.8615 to 2.3415
3 months	71.57 ± 2.98	70.84 ± 2.66	1.0010	0.321	− 0.7298 to 2.1898
6 months	68.76 ± 2.74	68.57 ± 2.57	0.2770	0.782	− 1.1829 to 1.5629
Carrying angle
Pre operative	6.1 ± 2.98	6.03 ± 2.76	0.0944	0.92	− 1.4144 to 1.5544
Post operative-
3 months	6.83 ± 2.69	6.97 ± 2.7	0.2012	0.84	− 1.5329 to 1.2529
6 months	6.13 ± 3.0	6.03 ± 2.76	0.1344	0.89	− 1.3898 to 1.5898
Fixed flexion deformity at 1 month	32.83 ± 7.03	21.83 ± 6.76	6.1776	< 0.0001	7.4357 to 14.5643
Elbow range of movements
3 months			2.9360	0.0048	− 5.05 to − 0.95
6 months			0.3359	0.7381	− 0.83 to 1.16
Flynn criteria
3 months
Fair	20	11		0.02
Excellent	10	19		0.02
6 months
Fair	0	0		< 0.00001
Excellent	30	30		0.0003

Though fixed flexion deformity was noted in the operated elbow at the one-month follow-up after surgery in patients of both the study groups (Fig. 5), the range of movements was found to improve throughout the follow-up period and no loss of range of movements was noted by the end of six-month follow-up in both the groups. The fixed flexion deformity was found to be higher in the group fixed with cross-pinning compared to the lateral pinning group using the unpaired t-test (p-value 0.0001).

Fig. 5.

post-operative elbow range of movement at 1-month follow-up; 5a- elbow extension in group A patient; 5b- elbow flexion in group A patient; 5c- elbow extension in group B patient 5d- elbow flexion in group B patient (original figure)

In group A, at three-month follow-up, the elbow range of movements were 0–120° in 20 patients, 0–125° in seven patients and 0–130° in three patients. At six-month follow-ups, the elbow range of movements was 0–130° in 29 patients while one patient had an elbow range of movements of 0–120°. On comparing the elbow range of movements at three-month and six-month follow-ups, it was found to be statistically significant by using an unpaired t-test (p-value < 0.0001, with 95% confidence interval − 8.85973 to − 5.96855).

In group B, at three-month follow-up, the elbow range of movements was 0–120° in 11 patients, 0–125° in seven patients and 0–130° in 12 patients. At six-month follow-ups, the elbow range of movements were 0–130° in 28 patients while one patient had an elbow range of movements of 0–120° and one had an elbow range of movements of 0–125°. On comparing the elbow range of movements at three-month and six-month follow-ups, it was found to be statistically significant by using an unpaired t-test (p-value < 0.0001, with 95% confidence interval − 8.91 to − 6.09).

Though there was statistical significance in comparing the elbow range of movements between both the groups during the three-month follow-up period (p-value 0.0048), there was found to be no statistical significance at the six-month follow-up period (p-value 0.7381).

There was statistical significance between both the groups at three-month and six-month follow-ups on comparing the Flynn criteria for fair and excellent by using the Fisher exact test (Table 2).

There was no iatrogenic nerve injury in any patient. There was no pin tract infection or pin loosening. One patient sustained loss of reduction on the sixth post-operative day in the cross-pinning group and underwent a secondary procedure where it was converted to lateral pinning. Following the second surgery, the patient did not develop any complications and had excellent Flynn criteria at the end of the study.

Discussion

Supracondylar humerus fractures are more commonly seen in children before the age of ten years and the most common type is the extension type of fracture. The mechanism attributed to this type of fracture is a fall on the outstretched hand leading to hyperextension of the elbow and the olecranon is driven into the olecranon fossa [6]. The number of pins used and its correlation with fracture stability have been controversial [7, 8].

The primary objective of the treatment is to achieve anatomical reduction and ensure stable internal fixation. Closed reduction with kirschner wire fixation has been the gold standard for managing these injuries. While open reduction allows for direct visualization of the fracture site, facilitating stable fixation, it is generally not recommended as the first-line treatment due to the risk of postoperative stiffness, which can result from inadequate handling of surrounding soft tissues, leading to the formation of scar tissue [9]. Hence, in our study, open reduction was not performed for any patient. Kirschner wires are preferred because they are easy to use, cost-effective, and can shorten hospital stays [10].

Kropelnicki et al. [6] found that the fracture is more common on the non-dominant hand, i.e., the left side, which was also noticed in our study. Similar to the study conducted by Pavone et al. [11], we also observed that more boys sustained supracondylar humerus fractures.

Most studies have considered only Gartland type 3 supracondylar humerus fracture for fixation and outcome analysis due to its rotational instability but our study has included both type 2 and type 3 fractures for fixation as the instability and cosmetic complications seen in type 2 fractures are not negligible [12]. The residual deformity seen at skeletal maturity in patients treated conservatively with plaster for type 2 fracture has led to its inclusion for fixation in our study. The duration of surgery for both the groups showed no statistical significance as it was noted that the time consumption in group A was during the mini-incision, while in group B, it was for achieving the divergence between the lateral k-wires.

Bilgili et al., suggested that fixing the lateral pins in a divergent manner to fix through the medial and lateral columns of the supracondylar humerus provides more stability in the axial and sagittal plane [3]. Hence some fractures in group B, which were found to be more rotationally unstable were fixed with divergent pins, though they were not analysed statistically to the pins fixed in a parallel manner. Similarly, Feng et al., concluded that the divergent pins had better stability in all directions compared to the medial–lateral crossed pins except for valgus stability and also noted that three divergent pins were superior to two divergent pins in fixing onto the medial and lateral columns, though not statistically significant [13]. In fixing the medial and lateral columns, the distal lateral k wire was passed via the trans-olecranon technique to achieve four cortical purchases, as it has been proved to have superior stability compared to two cortical purchases, as suggested by Kasirajan et al. [14]. And this method can be used in cases with medial comminution too.

Studies assessing the size and distance between the Kirschner wires suggested that 2 mm pins show better stability and an entry distance of 5 mm or more between the pins at the insertion on the lateral condyle reduces the chance of loss of reduction compared to a distance of less than 5 mm [15, 16] which was adopted in our study.

Several studies have found a higher rate of iatrogenic ulnar nerve injury in the cross-pinning group, loss of reduction in the lateral pinning group and pin tract infection and loosening in both groups [4, 17, 18]. However, our study faced one complication of revision surgery due to loss of reduction in the cross-pinning group on the sixth post-operative day, the reason for which was found retrospectively to be due to patient non-compliance to the above-elbow plaster applied which was found to be broken on the sixth day after surgery and not due to the surgical technique used. We avoided the chance of iatrogenic nerve injury by posterior displacement of soft tissues over the medial condyle aided with the mini-open technique suggested by Green et al. [19] and anterior placement of the kirschner wire over the medial condyle.

There was no significant difference in comparing the Flynn criteria, elbow range of movements, Baumann angle and carrying angle between our study groups, similar to the study conducted by Radaideh et al. [20].

With a proper surgical team, careful insertion of k-wires, careful protective measures for the ulnar nerve and sterile postoperative care, complications following a supracondylar humerus fracture fixation can be avoided [21]. Early mobilization of the elbow joint has been shown to improve the range of movements faster. Plaster application following surgery has been shown to prevent post-operative loss of reduction and mal-positioning of the kirschner wires due to elbow movement, which is arrested by the above elbow plaster.

In conclusion, both the techniques of percutaneous pinning, i.e., cross-pinning and lateral-pinning, showed comparable bony union. However, the functional outcomes were better at 3 months follow-up in the lateral-pinning group compared to cross-pinning. Nonetheless, at 6 months after surgery there were no differences in the outcomes in both the groups. Though the duration of surgery was not statistically significant between the groups, cross-pinning group was found have a longer time for fracture fixation. The trans-olecranon technique of fixation can be adopted as it has been shown to provide superior stability with four cortical purchases. However, more research on this technique is warranted.

The limitation of this study was a small sample size. A larger sample size could have given a better insight into the demographics of the study population. Including flexion type of supracondylar humerus fracture and compound fracture would have given a broader insight into the outcome. The study was concluded at 6 months after surgery as there was no statistical significance among both groups regarding the outcome analysis, however, a longer follow-up period till skeletal maturity was needed as there could be delayed development of varus deformity. We did not include the number of k-wires inserted into consideration for statistical analysis, which could have affected the outcome analysis. The fracture pattern and the outcome could have been analysed. The inclusion of neurovascular injury could have given a detailed outcome analysis and insight into post-operative management. Hence, future studies conducted on pediatric supracondylar humerus fractures can include the above limitations for analysis in their study for a better-detailed understanding of this most common upper limb injury in the pediatric age group.

Authors Contributions

Concept and design: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, Dr. Muthukumar Balaji S, Dr. Gowtham Senguttuvan definition o f intellectual content: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, literature search: Dr. Eswaran Mythilisri, Dr. Gowthaman N, Dr. Muthukumar Balaji S, Dr. Gowtham Senguttuvan data acquisition: Dr. Eswaran Mythilisri, Dr. Gowthaman N, Dr. Muthukumar Balaji S, data analysis: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, manuscript preparation: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, Dr. Gowtham Senguttuvan, manuscript editing: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, Dr. Muthukumar Balaji S, Dr. Gowtham Senguttuvan, manuscript review: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, Dr. Muthukumar Balaji S, Dr. Gowtham Senguttuvan approval of final version of the manuscript for publication: Dr. Eswaran Mythilisri, Dr. Rajavel K, Dr. Gowthaman N, Dr. Muthukumar Balaji S, Dr. Gowtham Senguttuvan guarantors: Dr. Eswaran Mythilisri, Dr. Gowthaman N.

Funding

This study received no specific grant from any funding agency in the public, commercial or not-for-profit sectors for conducting this study or for the preparation and submission of this manuscript.

Data availability

The data used in this study are available with the corresponding author and can be accessed upon reasonable request via email.

Declarations

Conflict of Interest

On behalf of all authors, the corresponding author states that there is no conflict of interest. The final version of the manuscript has been read and approved by all the authors. All authors have contributed significantly towards this article.

Ethics Approval

The study was conducted per the principles of the Declaration of Helsinki. Approval was granted by the institutional ethics committee.

Informed Consent

Informed and written consent was obtained from all the parents/legal guardians of all the participants as they were children before enrolling into the study. Informed and written consent was obtained from all the parents/legal guardians of all the participants for publishing their data. A separate informed and written consent was obtained from the parents/legal guardians of the participants whose clinical pictures and identifying details were included in this article.