*Article* **Plate Fixation for Irreducible Proximal Humeral Fractures in Children and Adolescents—A Single-Center Case Series of Six Patients**

**Florian Freislederer 1,\*, Susanne Bensler <sup>2</sup> , Thomas Specht <sup>1</sup> , Olaf Magerkurth <sup>3</sup> and Karim Eid 1,\***


**Abstract:** Background: Recommended treatment for severely displaced proximal humeral fractures in children is the closed reduction and percutaneous fixation by K-wires or intramedullary nailing. Methods: From January 2016 to January 2017 6, 21 children/adolescents (range 8 to 16 years) with proximal humeral fractures were treated surgically for severe displacement. In these six patients, several attempts of closed reduction were unsuccessful, and an open reduction was performed. The humeral head was fixed with a 3.5 mm T-plate without affecting the growth plate. Plate removal was performed at a mean interval of 132 days after initial surgery. Two years after initial surgery, the clinical outcome was assessed by the Constant–Murley score and QuickDASH score (including sport/music and work) and the shoulder joint was evaluated with a standardized sonographic examination for the rotator cuff and the conjoint tendon. Results: In all six patients, dorsal displacement of the fracture was irreducible due to the interposition of tendinous or osseous structures. Intraoperatively, the interposed structures were the long biceps tendon in two, periosteal tissue in two, a bony fragment in one, and the long biceps tendon together with the conjoint tendon in one case. At mean follow-up of 26 months (range 22 months to 29 months), patients showed very good clinical results with an excellent mean Constant–Murley score of 97.5 (range 91 to 100) and mean QuickDASH score (including sport/music and work) of 5.5 (range 0–20.8). An X-ray follow-up 6 weeks after surgery demonstrated early consolidation and correct alignment in all patients. A sonographic evaluation at 2 years post injury showed that the biceps and the conjoined tendon were intact in all patients. Conclusions: If a proximal humeral fracture is not reducible by closed means, a tissue entrapment (most likely biceps tendon) should be considered. Treatment with an open reduction and plate fixation yields very good clinical and radiological results and preserves interposed structures as the biceps and conjoint tendon.

**Keywords:** humerus fracture; proximal humeral fracture; children; plate fixation; ORIF; tissue entrapment; biceps

### **1. Introduction**

Proximal humeral fractures in children and adolescents are rare injuries, representing less than 5% of all pediatric fractures with a peek incidence between the age of 11 and 15 years [1–3]. These fractures can be physeal or metaphyseal. Metaphyseal fractures account for about 70% of the cases [4]. The muscle attachments of the rotator cuff proximally, and of the deltoid, as well as of the pectoralis major distally, are responsible for the specific fracture pattern.

Muscular tension displaces the proximal fragment in varus and posteromedially, whereas the distal fragment moves anteriorly and in adduction (Figure 1A,B).

**Citation:** Freislederer, F.; Bensler, S.; Specht, T.; Magerkurth, O.; Eid, K. Plate Fixation for Irreducible Proximal Humeral Fractures in Children and Adolescents—A Single-Center Case Series of Six Patients. *Children* **2021**, *8*, 635. https://doi.org/10.3390/children 8080635

Academic Editor: Vito Pavone

Received: 14 June 2021 Accepted: 23 July 2021 Published: 26 July 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

**Figure 1.** Radiographic shoulder views (Pat. N° 4) (ap/Neer). (**A**,**B**): Dorsally displaced humeral head. (**C**,**D**): 10 weeks after open reduction and plate fixation. **Figure 1.** Radiographic shoulder views (Pat. N◦ 4) (ap/Neer). (**A**,**B**): Dorsally displaced humeral head. (**C**,**D**): 10 weeks after open reduction and plate fixation.

Another explanation for the anterior displacement of the distal fragment might be the thinner and weaker anterior periosteum [5]. Another explanation for the anterior displacement of the distal fragment might be the thinner and weaker anterior periosteum [5].

Muscular tension displaces the proximal fragment in varus and posteromedially,

whereas the distal fragment moves anteriorly and in adduction (Figure 1A,B).

Neer and Horwitz classified proximal humeral fractures in children in four grades according to the severity of displacement (I: < 5 mm; II: < 1/3 shaft width; III: < 2/3 shaft width; IV: > 2/3 shaft width) [2]. Neer and Horwitz classified proximal humeral fractures in children in four grades according to the severity of displacement (I: <5 mm; II: <1/3 shaft width; III: <2/3 shaft width; IV: >2/3 shaft width) [2].

Most fractures of the proximal humerus in the skeletally immature are not or only minimally displaced and can be treated conservatively [4,6,7]. Most fractures of the proximal humerus in the skeletally immature are not or only minimally displaced and can be treated conservatively [4,6,7].

The management of severely displaced proximal humeral fractures in children is still controversial [3]. Most of the displaced proximal humeral fractures (Neer and Horwitz ≥ III) are treated by closed reduction and intramedullary nailing or by percutaneous fixation with K-wires [2,8,9]. Due to the high remodeling potential, moderate malalignment after closed reduction is acceptable. In addition, since around 80% of the longitudinal growth of the humerus results from the proximal physis, any mechanical interference with the growth plate by implant materials is usually avoided [10,11]. The management of severely displaced proximal humeral fractures in children is still controversial [3]. Most of the displaced proximal humeral fractures (Neer and Horwitz ≥ III) are treated by closed reduction and intramedullary nailing or by percutaneous fixation with K-wires [2,8,9]. Due to the high remodeling potential, moderate malalignment after closed reduction is acceptable. In addition, since around 80% of the longitudinal growth of the humerus results from the proximal physis, any mechanical interference with the growth plate by implant materials is usually avoided [10,11].

However, insufficient alignment after closed reduction may require open reduction and stable fixation which has already been described 40 years ago by Weber with an excellent clinical outcome [4,12–14]. However, insufficient alignment after closed reduction may require open reduction and stable fixation which has already been described 40 years ago by Weber with an excellent clinical outcome [4,12–14].

In our institution, closed reduction and pin fixation are routinely performed to address displaced pediatric proximal humerus fractures. If secondary displacement is observed or primary reduction is not satisfactory, presumably due to tissue entrapment, open reduction and internal fixation are performed. In our institution, closed reduction and pin fixation are routinely performed to address displaced pediatric proximal humerus fractures. If secondary displacement is observed or primary reduction is not satisfactory, presumably due to tissue entrapment, open reduction and internal fixation are performed.

The aim of the study is to assess clinical and radiological results after an open reduction, plate fixation and plate removal of irreducible displaced proximal humeral fractures in older children and to evaluate the functional integrity of the tissue—mainly the long head of the biceps tendon—interposed in the irreducible fracture. The aim of the study is to assess clinical and radiological results after an open reduction, plate fixation and plate removal of irreducible displaced proximal humeral fractures in older children and to evaluate the functional integrity of the tissue—mainly the long head of the biceps tendon—interposed in the irreducible fracture.

#### **2. Material and Methods**

**2. Material and Methods**  Between January 2016 and January 2017, 21 skeletally immature patients with proximal humeral fractures were treated at the authors institute (level 1 trauma hospital). Six of these patients (29%) were treated by open reduction and internal fixation. All of these Between January 2016 and January 2017, 21 skeletally immature patients with proximal humeral fractures were treated at the authors institute (level 1 trauma hospital). Six of these patients (29%) were treated by open reduction and internal fixation. All of these six patients were included in the study.

six patients were included in the study. The inclusion criteria were: available standard x-rays in anteroposterior (AP) and Neer view preoperatively and 6 weeks postoperatively, severely displaced humeral fracture (Neer–Horwitz III/IV), not reducible by closed means, open physis and hardware removal performed. After 2 years, patients were prospectively assessed with a clinical and

sonographic examination. Informed consent was obtained from the patient's or her/his legal representative.

Approval by the Ethical committee Nordwestschweiz Nr. 2018-01405 was obtained.

#### *2.1. Surgical Technique*

All of the operative procedures were carried out under general anesthesia. The patients were positioned in beach chair position. An initial attempt of closed reduction was conducted in all patients. This was performed by gentle longitudinal traction with abduction and external rotation of the arm. An image intensifier was used to monitor reduction. If closed reduction failed, the surgeon proceeded with open reduction by means of an anterior deltopectoral approach. Entrapped tissue or periosteum were gently freed. Once reduction was achieved, a 3.5 mm T-plate was contoured on the anterolateral part of the proximal humerus and temporary K-wires were inserted through the plate holes to hold the reduction. Under image intensifier, attention was drawn not to injure the physeal plate by the drill or the screws. After confirming anatomical reduction, the K-wires were then replaced with conventional cortical screws.

Surgical wounds were closed with absorbable VICRYL (Ethicon, Raritan, NJ, USA; Johnson & Johnson, New Brunswick, NJ, USA) sutures. Skin was closed using absorbable MONOCRYL (Ethicon, Raritan, NJ, USA; Johnson & Johnson, New Brunswick, NJ, USA) sutures.

Postoperatively, a brace (type Gilchrist) was applied for 2 weeks. For the first 6 weeks, range of motion was limited to 90◦ abduction and flexion. For the first 2 weeks, only passive mobilization out of the brace was allowed, followed by actively assisted movements.

#### *2.2. Clinical Assessment*

Patient outcome was assessed by clinical and sonographic evaluation. The clinical examination was carried out by a single independent observer and sonography by a single radiologist specifically trained in musculoskeletal imaging. The Constant–Murley score and QuickDASH (Disability of the Arm, Shoulder and Hand) score (including sport and music/work modules) were used for objective assessment [15,16]. The Constant–Murley score is a 100-point shoulder score, which assesses the range of motion of the treated shoulder joint. Forward flexion, extension, abduction, and low (arm in adduction) and high (arm in 90◦ of abduction) external and internal rotation were measured using the standardized neutral-zero method in degrees. A goniometer was used for the measurement. The abduction strength was measured using a spring balance (Macro-Line 80020, Fa. Pesola), which was attached distally on the forearm adjacent to the wrist with the method described and validated by Bankes et al. [17]. Strength was measured with the arm in 90 degrees abduction, full extension of the elbow, and the palm of the hand in pronation. The patient was asked to maintain this position for five seconds. This procedure was repeated three times, with at least a one-minute time interval. The average in kilograms (kg) was noted. The same procedure was performed with the contralateral arm. The measurement should be pain free. If pain was present or the patient was unable to abduct above 90◦ , the score equaled zero. The strength score was calculated from the highest score of the three attempts. The score corresponds to the force in kilogram.

We also performed this procedure with the contralateral arm to obtain the individual Constant score as described by Fialka et al. [15]. For interpretation, the results of the Constant–Murley score were divided into four subscales: excellent 90–100; satisfactory 80–89; unsatisfactory 70–79; failure > 70. The QuickDASH outcome measure was a 100-point shoulder score. It measured a 30-item (+4 sports/music, +4 work) questionnaire of physical and social function with symptoms in any or all joints in the upper extremity. The lower the DASH score, the better the outcome.

In addition, medical records were reviewed. All patients had a normal shoulder function and no previous operations on the affected side. Radiological evaluation included standard anteroposterior and Neer view of the shoulder. Follow-up radiographs were carried out six weeks and between two and three months postoperatively. Subsequently, hardware removal was performed.

#### *2.3. Ultrasound Imaging*

All patients were scanned in a sitting position with a relaxed arm hanging freely on the side. For the examination, a GE LOGIQ E9 ultrasound system (GE Healthcare; Chicago, IL, USA) with a linear transducer with a bandwidth of 6–15 MHz was used.

The tendons of the subscapularis, supraspinatus and infraspinatus tendon were examined along their long and short axis.

The subscapularis tendon was examined with the arm externally rotated and the elbow fixed at the iliac crest. For the evaluation of the supraspinatus tendon, the patient's arm was placed posteriorly, with the palmar side of the hand on the superior aspect of the iliac wing with the elbow flexed and directed posteriorly. To examine the infraspinatus tendon, the arm was placed anteriorly with the hand on the opposite shoulder.

The long head of the biceps tendon was examined along the long and short axis with the arm placed in slight internal rotation. The integrity of the conjoint tendon was examined also in both planes with the arm placed in external rotation.

#### **3. Results**

There were five boys and one girl with a mean age of 14 years (8–16 years). At the time of injury, mostly accidents during physical activities which caused an isolated injury to the proximal humerus, all fractures were proximal metaphyseal fractures. Five patients had a Neer–Horwitz Grade III fracture and one had a Grade IV (patient N◦ 4, see Table 1) completely displaced fracture of the proximal humerus (Figure 1A,B). In these six patients, a closed reduction was attempted; in five patients, an immediate conversion to open surgery with open reduction and internal plate fixation was necessary.

**Table 1.** Outcome with 2-year follow-up after open reduction, internal plate fixation and early (mean 4 months postoperative) plate removal.


1 secondary open reduction due to dislocation after closed reduction and percutaneous pinning (see also Figure 2).

> The first patient (N◦ 1, Table 1, Figure 2) from this series was initially treated with a closed reduction and percutaneous pinning; due to secondary displacement, an open reduction and internal fixation was necessary.

> For this patient, it was apparent that a closed reduction was impossible due to the interposition of soft tissue at the fracture site. Tissue entrapment was intraoperatively observed in all of the six cases (the long biceps tendon in two cases (Pat. N◦ 1 and 5), periosteal tissue in two cases (Pat. N◦ 3 and 6), a bony fragment in one case (Pat. N◦ 4), and both the long biceps tendon as well as the conjoint tendon in one case (Pat. N◦ 4)) (Figure 3).

*Children* **2021**, *8*, x FOR PEER REVIEW 5 of 10

*Children* **2021**, *8*, x FOR PEER REVIEW 5 of 10

**Figure 2.** Chronologic radiographic images of Pat. N° 1. (**A**) Initial images after trauma with a posteriorly displaced proximal humeral fracture. (**B**) After closed reduction and transepiphyseal fixation. (**C**) X-ray 1 day after surgery shows posterior displacement in the Neer view (right side). (**D**) Open reduction and T-plate fixation. (**E**) X-ray follow-up 10 weeks after open reduction and plate-fixation with anatomical reduction and fracture healing. **Figure 2.** Chronologic radiographic images of Pat. N◦ 1. (**A**) Initial images after trauma with a posteriorly displaced proximal humeral fracture. (**B**) After closed reduction and transepiphyseal fixation. (**C**) X-ray 1 day after surgery shows posterior displacement in the Neer view (right side). (**D**) Open reduction and T-plate fixation. (**E**) X-ray follow-up 10 weeks after open reduction and plate-fixation with anatomical reduction and fracture healing. For this patient, it was apparent that a closed reduction was impossible due to the interposition of soft tissue at the fracture site. Tissue entrapment was intraoperatively observed in all of the six cases (the long biceps tendon in two cases (Pat. N° 1 and 5), periosteal tissue in two cases (Pat. N° 3 and 6), a bony fragment in one case (Pat. N° 4), and both the long biceps tendon as well as the conjoint tendon in one case (Pat. N° 4)) (Figure 3).

**Figure 3.** Intraoperative findings. (**A**) Entrapment of the long biceps tendon (white arrow) in the fracture gap (Pat. N° 1, see also Figure 2); (**B**) freed long biceps tendon (Pat. N° 1, see also Figure 2); (**C**) freed conjoint tendons (white asterisk) (Pat. N° 4, see also Figure 1). **Figure 3.** Intraoperative findings. (**A**) Entrapment of the long biceps tendon (white arrow) in the fracture gap (Pat. N◦ 1, see also Figure 2); (**B**) freed long biceps tendon (Pat. N◦ 1, see also Figure 2); (**C**) freed conjoint tendons (white asterisk) (Pat. N◦ 4, see also Figure 1).

For this patient, it was apparent that a closed reduction was impossible due to the

**Figure 3.** Intraoperative findings. (**A**) Entrapment of the long biceps tendon (white arrow) in the fracture gap (Pat. N° 1, see also Figure 2); (**B**) freed long biceps tendon (Pat. N° 1, see also Figure 2); (**C**) freed conjoint tendons (white asterisk) (Pat. N° 4, see also Figure 1). The scheduled removal of the hardware was performed in all six patients. The implants were removed under general anesthesia as a day case procedure without difficulty at a mean time of 4.4 months after surgery (range 3–5.3 months). The scheduled removal of the hardware was performed in all six patients. The implants were removed under general anesthesia as a day case procedure without difficulty at a mean time of 4.4 months after surgery (range 3–5.3 months).

The scheduled removal of the hardware was performed in all six patients. The implants were removed under general anesthesia as a day case procedure without difficulty at a mean time of 4.4 months after surgery (range 3–5.3 months). The mean follow-up was 26 months (range: 22 months to 29 months) after fracture fixation. The Constant–Murley Shoulder and QuickDASH (including sport and music/work modules) scores are presented in (Table 1). The Constant–Murley score at the final followup was 97.5 (range 91 to 100) and the mean overall QuickDASH score (including sport and music/work) was 5.5 (range 0–20.8). Analyzing the subtypes of the QuickDASH score, we

found a score of 3 for disability (range 0–10), 3.125 for sport and music (range 0–12.5) and 0 for work. ing sport and music/work) was 5.5 (range 0–20.8). Analyzing the subtypes of the Quick-DASH score, we found a score of 3 for disability (range 0–10), 3.125 for sport and music (range 0–12.5) and 0 for work.

The mean follow-up was 26 months (range: 22 months to 29 months) after fracture fixation. The Constant–Murley Shoulder and QuickDASH (including sport and music/work modules) scores are presented in (Table 1). The Constant–Murley score at the final follow-up was 97.5 (range 91 to 100) and the mean overall QuickDASH score (includ-

*Children* **2021**, *8*, x FOR PEER REVIEW 6 of 10

All fractures showed advanced radiological healing at the 10–12 weeks follow-up (Figure 1C,D). Postoperatively, there was no loss of reduction, residual deformity or screw migration. All fractures showed advanced radiological healing at the 10–12 weeks follow-up (Figure 1C,D). Postoperatively, there was no loss of reduction, residual deformity or screw migration.

A sonographic examination of the shoulder two years postoperatively showed normal rotator cuff, long head of biceps and conjoint tendon (Figure 4). A sonographic examination of the shoulder two years postoperatively showed normal rotator cuff, long head of biceps and conjoint tendon (Figure 4).

**Figure 4.** Sonographic scans 2 years postoperatively of the soft tissue surrounding the shoulder joint. **Figure 4.** Sonographic scans 2 years postoperatively of the soft tissue surrounding the shoulder joint.

No major complication was observed related to primary surgery or plate removal. None of the patients presented with vascular or neurological complications. All patients showed a rather apparent skin scar, known to appear frequently in this location [4]. Three out of six patients reported the scar to be esthetically disturbing. Two patients described a feeling of irritation at the operative scar, which they attributed to the plate irritation. No major complication was observed related to primary surgery or plate removal. None of the patients presented with vascular or neurological complications. All patients showed a rather apparent skin scar, known to appear frequently in this location [4]. Three out of six patients reported the scar to be esthetically disturbing. Two patients described a feeling of irritation at the operative scar, which they attributed to the plate irritation. This resolved completely after plate removal. At the 2-year follow-up, five patients were very satisfied and one was satisfied with the outcome.

#### **4. Discussion**

Recommended treatment for severely displaced proximal humeral fractures in children is closed reduction and percutaneous fixation by K-wires or intramedullary nailing [4,5,18]. We presented a series of six patients treated by an open reduction and

plate fixation, in which either secondary displacement occurred, or closed reduction was unsatisfactory. tion, in which either secondary displacement occurred, or closed reduction was unsatisfactory.

This resolved completely after plate removal. At the 2-year follow-up, five patients were

Recommended treatment for severely displaced proximal humeral fractures in children is closed reduction and percutaneous fixation by K-wires or intramedullary nailing [4,5,18]. We presented a series of six patients treated by an open reduction and plate fixa-

*Children* **2021**, *8*, x FOR PEER REVIEW 7 of 10

very satisfied and one was satisfied with the outcome.

**4. Discussion** 

Two years postoperatively, we were able to demonstrate excellent shoulder function and preserved anatomical integrity of the entrapped tissues, namely, the biceps and conjoint tendon. To the best of our knowledge, this is the first report on the integrity of the interposed structures. It might well be questioned, whether these structures would have been intact, if a closed reduction had been accepted. Two years postoperatively, we were able to demonstrate excellent shoulder function and preserved anatomical integrity of the entrapped tissues, namely, the biceps and conjoint tendon. To the best of our knowledge, this is the first report on the integrity of the interposed structures. It might well be questioned, whether these structures would have been intact, if a closed reduction had been accepted.

The first patient of this series (Figure 2) was initially treated with a closed reduction and percutaneous pinning. Due to secondary displacement, a revision with an open reduction and internal fixation was necessary caused by a biceps entrapment. Subsequently, we treated unreducible fractures by an open reduction and internal plate fixation and found tissue entrapment to be present in all cases. The first patient of this series (Figure 2) was initially treated with a closed reduction and percutaneous pinning. Due to secondary displacement, a revision with an open reduction and internal fixation was necessary caused by a biceps entrapment. Subsequently, we treated unreducible fractures by an open reduction and internal plate fixation and found tissue entrapment to be present in all cases.

All fractures healed completely, and functional scores were excellent at a 2 year followup with symmetrical shoulder movement (Figure 5). The obstacle to reduction was, in most cases, the entrapped biceps tendon. In one case, the entrapped conjoint tendons inhibited reduction and in one other case periosteal tissue. All fractures healed completely, and functional scores were excellent at a 2 year follow-up with symmetrical shoulder movement (Figure 5). The obstacle to reduction was, in most cases, the entrapped biceps tendon. In one case, the entrapped conjoint tendons inhibited reduction and in one other case periosteal tissue.

**Figure 5.** Range of motion after plate fixation of a proximal humerus fracture on the right site (Patient N° 6). **Figure 5.** Range of motion after plate fixation of a proximal humerus fracture on the right site (Patient N◦ 6).

Dobbs et al. examined a subgroup of older adolescents and mentioned patients with irreducible fractures due to tissue entrapment which needed an open reduction [11]. Dobbs et al. examined a subgroup of older adolescents and mentioned patients with irreducible fractures due to tissue entrapment which needed an open reduction [11].

The entrapment of the long head of the biceps tendon or periosteum was mentioned earlier but has not been identified as a major cause for irreducible fractures [4,5,11]. Lucas The entrapment of the long head of the biceps tendon or periosteum was mentioned earlier but has not been identified as a major cause for irreducible fractures [4,5,11]. Lucas et al. did not find entrapment of the tendon of the long head of the biceps in the fracture site in four patients, which were assessed by magnetic resonance imaging.

In contrast, Bahrs et al. described open reduction in Neer III and IV fractures in 17 of their 31 patients. They found that in nine of these patients, the biceps tendon was entrapped in the fracture site. They concluded that a failed closed reduction should be interpreted as a possible soft tissue entrapment (most likely biceps tendon) and that these cases should be addressed with an open reduction and the removal of the entrapped structures [14].

Performing an open reduction, we could liberate the tissue interposed in the fracture side and achieve an anatomical reduction. To achieve a stable fixation of our reduction without crossing the physeal plate, we decided to use a T-plate fixation.

The use of a plate for an internal fixation after an open reduction has rarely been considered. In the aforementioned study of Bahrs et al., twelve patients were treated with a K-wire or screw fixation, and a plate fixation was used only in five patients. As early as 42 years ago, Weber et al. described treatment of severely displaced or irreducible infratubercular proximal humerus fractures by an open reduction and plate fixation without complications and symmetrical function of the shoulders.

In contrast to the generally preferred use of k-wires, a plate fixation avoids direct injury to the physis if screws are placed with the use of an image intensifier. If planned hardware removal is performed 3–6 months after initial surgery, tethering of the epiphyseal plate is not to be anticipated [14]. The drawback of the plate fixation, however, is the necessity of its removal at 3 months.

A plate fixation provides a stable fixation of the reduced fracture. Anatomical fracture healing is of upmost importance in adolescents (>12 years), where the remodeling capacity is limited [19]. In the present study, the average age of patients was 13 years. There is only limited data on the outcome of adolescent patients with this fractures [11].

In contrast to the plate fixation, any wire or elastic nail in metaphyseal and epiphyseal fractures will pass the epiphyseal plate and damage it to a certain degree. Excellent outcomes without limb shortening or axial deviation of the proximal humerus after K-wire or intramedullary nailing are reported [4,6,8]. Nevertheless, a physeal arrest and progressive deformity can be a potential risk of any crossing stabilization [20]. Peterson et al. reported on physeal injury in physeal fractures at three different sites (proximal humerus, distal humerus, distal femur) and recorded 100% premature closure in the three cases in which a K-wire internal fixation had passed the physis [21]. Intramedullary retrograde stabilization with ESIN has been recommended as the standard fixation method for proximal humeral fractures in children and adolescents, but this technique has some major drawbacks, such as nail penetration into the joint cavity, humeral head perforation, physeal damage due to multiple perforation and the displacement of the proximal fragment by pushing with the ESIN tips [4]. Zivanovic et al. observed complications in 5 of 16 patients: 2 humeral head perforations, 10◦ of residual varus deformation in 2 patients and difficulties in nail extraction in one patient. Similar complications were reported by other authors [8,22].

Our study has its limitations. First, the number of patients treated was small and general treatment recommendations could not be deducted. We did not have a control group that would demonstrate any superiority compared to other treatment strategies.

Second, due to the ethical restrictions, the study lacked a late radiographic follow-up at two years. However, as clinical results were excellent and ultrasound did not show any deformity of the tuberosities or the humeral head, it seems reasonable to state, that the growth plate was not subjected to any injury or tethering.

A drawback of our proposed treatment is the requirement of a second surgery.

We do agree with the generally accepted age and deformity-based decision making [4,5,18,23], but want to emphasize that tissue entrapment, which inhibits closed reduction, is very likely in Neer III and IV fractures, and may be underestimated in the literature so far.

#### **5. Conclusions**

If a proximal humeral fracture is not reducible by closed means, a tissue entrapment (most likely biceps tendon and conjoined tendon) has to be considered as an obstacle to the reduction. An open reduction and plate fixation not only yield excellent clinical results, but allow the functional and anatomical integrity of the entrapped tendons.

**Author Contributions:** Conceptualization, F.F., T.S. and K.E.; methodology, F.F., T.S., S.B., O.M. and K.E.; validation, F.F. and K.E. investigation, F.F. and S.B.; data curation, F.F.; writing—original draft preparation, F.F. and S.B.; writing—review and editing, F.F. and K.E.; visualization, F.F.; supervision, K.E.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board (or Ethics Committee) of Ethical committee Nordwestschweiz Nr. 2018-01405, 05. September 2018.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are stored and openly available in the secretary bureau of the Department of Orthopaedics and Traumatology, Kantonsspital Baden, Im Ergel 1, 5404 Baden, Switzerland.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Tu Ngoc Vu 1,2,3 , Son Hong Duy Phung 1,3,\* , Long Hoang Vo <sup>4</sup> and Uoc Huu Nguyen 1,3,\***

	- Hanoi 100000, Vietnam

**Abstract:** (1) Background: This study aims to describe the clinical and paraclinical characteristics of and the diagnostic approach to brachial artery injuries in pediatric supracondylar humerus fractures, as well as to evaluate intraoperative vascular anatomical lesions and early postoperative results. (2) Methods: A retrospective, hospital-based analysis of medical records at Viet Duc University Hospital (Vietnam), using a sample of children under 16 years who met the diagnostic criteria for supracondylar humerus fractures with brachial artery injuries between January 2016 and December 2020, was performed. A total of 50 patients were included in the analysis. (3) Results: Out of 50 pediatric patients, 36 patients were male (72%) and the mean age was 5.85 years (range, 1.5–14 years). Before treatment, there were 46 patients with severely displaced fractures which were classified as Gartland type III (92%). Following casting, the percentage of those with severely displaced fractures was reduced significantly to 12%, while there were no patients with Gartland type III fractures after percutaneous pinning. Doppler sonography failed to assess vascular lesions at the fracture site before and after casting in most patients. Two-thirds of surgical cases had only vasospasm, without physical damage to the vessel wall or intravascular thrombosis. Preoperative Doppler spectrum analysis was not consistent with the severity of intraoperative brachial artery injury. Out of 24 patients with vasospasm, we performed vascular blockade using papaverin in 11 cases and intraoperative balloon angioplasty of the brachial artery using the Fogarty catheter in 13 cases. Brachial artery graft was performed with 12 patients who had anatomical damage to the vascular wall. A complication of embolism occurred in one patient immediately after surgery, and two patients had superficial infections. One month following surgery, 2 out of 36 patients had a temporary loss of sensation in the area of incision. (4) Conclusions: Most pediatric patients did not present with symptoms of critical limb ischemia similar to those associated with lower extremity vascular injuries. The diagnosis and treatment of pediatric supracondylar humerus fractures with vascular injury is difficult and time-consuming, especially in cases of transverse fractures.

**Keywords:** supracondylar humerus fracture; vascular injury; brachial artery injuries

## **1. Introduction**

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Supracondylar fractures of the humerus are the most prevalent kind of fractures, accounting for approximately 60% of all elbow fractures and 3–7% of pediatric fractures [1,2]. They are most common in young children between 5 and 10 years of age. More than 95% of these fractures are extension fractures, which may result in a variety of neurological and vascular complications. About 10% to 20% of displaced supracondylar fractures present alterations in vascular status [3,4]. An absent radial pulse was observed in 7% to 12% of all fractures and up to 19% in displaced fractures. Brachial artery injuries are often a consequence of stretching, entrapping or disrupting the neurovascular structures on the proximal

**Citation:** Vu, T.N.; Phung, S.H.D.; Vo, L.H.; Nguyen, U.H. Diagnosis and Treatment for Pediatric Supracondylar Humerus Fractures with Brachial Artery Injuries. *Children* **2021**, *8*, 933. https://doi.org/ 10.3390/children8100933

Academic Editor: Christiaan J. A. van Bergen

Received: 5 September 2021 Accepted: 11 October 2021 Published: 18 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

fragment, as well as, though less frequently, of reduction maneuvers or immobilization of the elbow in the hyperflexion position [5,6].

The diagnosis and morphology of the anatomical lesion as well as the surgical treatment of this form of injury in children are markedly different from the extremity arterial injuries which are more common in adults. Through treating this pathology at our institution, we identified the following key concerns. Firstly, most brachial artery injuries in children do not have obvious symptoms of ischemia. Therefore, the implementation of intensive diagnostic procedures, as well as emergency surgery which is similar to that performed in cases of adult vascular trauma, can increase the risk of complications and consume unnecessary medical and social resources. Secondly, the application of Doppler ultrasound and multislice computed tomography angiography (CT-A) in children is not easy, especially for those who are not yet of school age. Anesthesia must be used in order to ensure the safety and effectiveness of these methods. As a result, the duration of the evaluation is prolonged and the pediatric patient may have to undergo anesthesia several times during the course of treatment, thereby increasing the risk of complications from anesthesia. Thirdly, once the vessel wall is opened for examination, the potential for endothelium lesion damage leads to actual occlusion of the artery because of the small diameter of the pediatric brachial artery. Once the vessel wall is opened for examination, the potential for endothelium lesion damage leads to actual occlusion of the artery because of the small diameter of the pediatric brachial artery. If the lesion is long, there is no alternative solution with artificial vessels or saphenous veins, as in adults. In addition, intervention on brachial arteries in pediatric patients has potential risks for adverse outcomes, such as vasodilation or vasoconstriction [7,8], large and bad scars [7,9] and osteomyelitis [10].

To better inform practice so as to minimize the duration of treatment and avoid the unnecessary consumption of medical resources, while ensuring that pediatric patients receive effective treatment for their injuries and avoid more complicated vascular complications, is the ultimate goal of this research. Our aims were to describe the clinical and paraclinical characteristics and the diagnostic approach to brachial artery injuries in pediatric supracondylar humerus fractures, and to evaluate intraoperative vascular anatomical lesions and early postoperative results.

#### **2. Methods**

#### *2.1. Patients*

This study was retrospectively performed using medical records at Viet Duc University Hospital, one of the oldest and largest surgical public hospitals in Vietnam, in the period from January 2016 to December 2020. We examined the data of patients who were under 16 years of age, had a diagnosis of traumatic arterial injury in the upper extremity and underwent treatment at our institution. Diagnostic criteria for supracondylar humerus fractures with brachial artery injuries included radiographs showing supracondylar humerus fractures and loss of the ulnar/radial pulse. We excluded those patients who had experienced previous elbow fracture(s) that caused limited movement and deformity.

#### *2.2. Data Collection*

Clinical parameters: age (years), age group (<3 years/3–<13 years/≥13 years), gender (male/female), injured side in the upper extremity (right/left), mechanism of injury (high energy trauma/other), types of bone fractures (closed fracture/open fracture), time interval between the beginning of trauma and arrival to the first health facility (hours), time interval between the beginning of trauma and arrival to the operating room (hours) and clinical symptoms.

Paraclinical parameters:


c. Multislice CT-A for vascular injuries: short lesion, <5 mm/long lesion, ≥5 mm. Multislice CT-A was indicated for those with ischemic symptoms after unsuccesful conservative fracture treatment using casts.

Treatment:


#### *2.3. Data Analysis*

Data were sorted, cleaned, coded and entered into Epidata 3.1. Then, a software program (SPSS, version 23.0; IBM, Armonk, NY, USA) was used for all statistical analyses. Descriptive statistics, such as frequency, percentage, mean, standard deviation and interquartile range, were used to summarize preoperative, intraoperative and postoperative parameters.

#### **3. Results**

Among 50 pediatric patients, 36 patients were male (72%) and the mean age was 5.85 years (range, 1.5–14 years). The mean time interval between the beginning of trauma and arrival to the first health facility was 12 h (range, 1–120 h), and the mean time interval between the beginning of trauma and arrival to the operating room was 52.8 h (range, 4–168 h). Injuries were commonly the result of high energy trauma (*n* = 49, 98%). 31 patients injured their left arm (62.0%), while 19 injured their right arm (38.0%). Most patients were diagnosed with Gartland type III fractures (*n* = 46, 92%), the rest with Gartland type II fractures (*n* = 4, 8%). Pink hand was present in 49 patients with supracondylar fractures (98%), and purple hand in only 1 case (2%) (Table 1).

**Table 1.** Clinical characteristics at admission.


**Table 1.** *Cont*.


*Children* **2021**, *8*, x FOR PEER REVIEW 4 of 11

Right arm 19 (38.0%) Left arm 31 (62.0%)

\* This involved the arm being rammed into by a buffalo.

Injured side, no. (%)

Figure 1 indicates changes of the Gartland classification of supracondylar humerus fractures following casting and percutaneous pinning. Before treatment, there were 46 patients with severely displaced fractures; these were classified as Gartland type III (92%). Following casting, the percentage of those with severely displaced fractures decreased significantly to 12%, while there were no patients with Gartland type III fractures after pinning. fractures following casting and percutaneous pinning. Before treatment, there were 46 patients with severely displaced fractures; these were classified as Gartland type III (92%). Following casting, the percentage of those with severely displaced fractures decreased significantly to 12%, while there were no patients with Gartland type III fractures after pinning.

Figure 1 indicates changes of the Gartland classification of supracondylar humerus

**Figure 1.** Changes of the Gartland classification of supracondylar humerus fractures after casting and after pinning. **Figure 1.** Changes of the Gartland classification of supracondylar humerus fractures after casting and after pinning.

Doppler sonography failed to assess vascular lesions at the fracture site before and after casting in most patients (Table 2). Doppler sonography failed to assess vascular lesions at the fracture site before and after casting in most patients (Table 2).

**Table 2.** Brachial artery injuries and flow velocity around the fracture site with Doppler sonography classification, before and after casting.  **Before Cast (***N* **= 50) After Cast (***N* **= 50)**  Brachial artery injuries, no. (%) Out of 50 patients, all cases were firstly treated by casting, from them, 14 cases were successfully treated with a cast. Other 36 cases were indicated for surgery. As was shown in Table 3, all pediatric patients in the age group of 13 years or over intraoperatively had found vessel injury. Most lesions having length of ≥5 mm were indicated for surgery, while lesions < 5 mm were treated conservatively.

No assessment 43 (86%) 45 (90%)


**Table 2.** Brachial artery injuries and flow velocity around the fracture site with Doppler sonography classification, before and after casting. Contusion 2 (4%) 3 (6%) Vasospasm 4 (8%) 2 (4%)

Thrombosis 1 (2%) 0 (0%)

*Children* **2021**, *8*, x FOR PEER REVIEW 5 of 11

**Table 3.** Frequency of treatment methods with age group and length of lesions on multislice CT-A.  **Surgical Treatment Conservative Treatment** 

**Table 3.** Frequency of treatment methods with age group and length of lesions on multislice CT-A.


Two-thirds of surgical cases had only vasospasm without vessel wall contusion or intravascular thrombosis (*n* = 24, 66.7%) (Figure 2). intravascular thrombosis (*n* = 24, 66.7%) (Figure 2).

Two-thirds of surgical cases had only vasospasm without vessel wall contusion or

**Figure 2.** Intraoperative blood vessel injury in pediatric patients undergoing surgical treatment.

**Figure 2.** Intraoperative blood vessel injury in pediatric patients undergoing surgical treatment.

No patients under 3 years of age had anatomical damage of the vessel wall, while there were two patients over 13 years of age who were diagnosed with vessel wall contusion. Of the cases with long-segment lesions on multislice CT-A (≥5 mm) that were operated on, only one-third of the lesions were actually vascular contusion. Preoperative Doppler spectrum analysis was not consistent with the severity of intraoperative brachial No patients under 3 years of age had anatomical damage of the vessel wall, while there were two patients over 13 years of age who were diagnosed with vessel wall contusion. Of the cases with long-segment lesions on multislice CT-A (≥5 mm) that were operated on, only one-third of the lesions were actually vascular contusion. Preoperative Doppler spectrum analysis was not consistent with the severity of intraoperative brachial artery injury. Out of 24 patients with vasospasm, we performed intraoperative vascular blockade using papaverin in 11 patients and intraoperative balloon angioplasty of the brachial artery

using the Fogarty catheter in 13 patients. A brachial artery graft using great saphenous vein was performed in 12 patients with vascular wall contusion (Table 4).

**Table 4.** Association of intraoperative vascular injury with age group, lesions on multislice CT, preoperative Doppler spectrum analysis and brachial artery graft in pediatric patients undergoing surgical treatment.


As shown in Table 5, we recorded a complication of embolism occurring in one patient immediately after surgery and superficial infections in two patients. One month following surgery, 2 out of 36 patients experienced a temporary loss of sensation around an incision.

**Table 5.** Postoperative results and one-month re-examination (*N* = 36).


#### **4. Discussion**

#### *4.1. Clinical Condition*

Non-dominant hand injuries occur more frequently in pediatric insupracondylar humerus fractures. The incidence of fractures in men and women is almost equal [11,12]. Most fractures occurred in male children. The most common age is around 5–6 years old. This is the age when children are preparing or starting school and their awareness is still immature; it is difficult for them to control their movements and body postures, yet they are eager to learn and explore the world around them. 62% of children had fractures in the left hand, which is mostly non-dominant and less flexible compared to the right hand. Most were closed fractures due to high energy trauma, only a single case admitted to the hospital being an open fracture, caused by a buffalo ramming into the arm. The injury is mainly due to falling against the hand and the grounding distance is not too large, so all patients were in a stable condition when they were admitted to hospital.

Most of the patients had type III fractures based on the Gartland classification (92%). This is consistent with injury to the brachial artery in supracondylar humerus fractures, where the artery is trapped in the fracture. According to Pham Quang Tri [13], six out of eight patients with supracondylar humerus fractures with vascular injury had Gartland type III displacement. Most of the children who came to the hospital did not have obvious symptoms of ischemia. Only a few children showed cold extremities (22%) and cyanosis (2%), while none of them showed signs of ischemia or of severe or irreversible limb bleeding, which is common in adult limb vascular trauma. In our study, time intervals between the

beginning of trauma and arrival at the first health facility, as well as arrival to the operating room, were much longer than the ideal time (6 h after the accident) to restore circulation after acute extremity embolism in adults. The prolonged time of hospital admission and delaying surgery in children could be explain by several reasons. First, the ischemia of the injury hand is not severe and it don't influence on their condition. Second, diagnostic imaging procedures such as Doppler ultrasound, X-ray, and multislide CT-A, and casting procedure take more time among small children compared to among adults. Finally, the treatment algorism for this injury is casting firstly, after that, if the pulse is restored, we will keep stop on conservative treatment. If the pulse is not restored, we send to multislide CT-A and consider to surgery.

#### *4.2. Preoperative Imaging*

In the study of Phan Quang Tri [13], among 102 cases of supracondylar fractures that were treated with born reposition and percutaneous pinning under C-arm control. There were 8 cases of brachial artery injury, accounting for 7.48% in general and 57.2% of cases with complications. Because the clinical signs and symptoms of ischemia are not clear and difficult to find out in small children, especially those who are not yet of school age. Diagnostic imaging plays an important role in decision between surgical and conservative treatments. Most of the children being at school age, with little or no sense of cooperation, vascular imaging diagnostic procedures are very difficult to perform with high quality. The performance of vascular diagnostic tests requires anesthesia, leading to difficulties in deploying human resources and equipment, as well as carrying an increased risk of adverse consequences.

Because it is performed in emergency conditions with a lack of patient's cooperation, cast, edema, and hematoma, only a very small number (<10%) of Doppler ultrasound analyses can directly evaluate the blood vessels at the level of the fracture, including the vessel wall and lumen. That information is quite needed for verify diagnosis of upper extremity vascular injury. In our study, more than half of the pediatric patients undergo a multislice CT-A of the upper extremity, whereas in adult patients these lesions are very rare to require this diagnostic procedure. Multislice CT-A enables an assessment of the perfusion status of the entire vascular system of the upper extremity but does not accurately assess the damage to the vessel wall and lumen in children. We found that all cases with length of lesion of the brachial artery on CT-A less than 5 mm did not necessarily require surgery. In addition, all study patients who had the lesions longer than 5 mm were indicated to surgery, but in only half of them can find out the vascular contusion intraoperatively. Therefore, multislice CT-A may be useful mainly in the decision of conservative therapy, i.e., those with short lesions (less than 5 mm) (Figure 3). *Children* **2021**, *8*, x FOR PEER REVIEW 8 of 11

**Figure 3.** Brachial artery injuries in multislice CT-A. (**A**) Lesions < 5 mm. (**B**) Lesions ≥ 5 mm. **Figure 3.** Brachial artery injuries in multislice CT-A. (**A**) Lesions < 5 mm. (**B**) Lesions ≥ 5 mm.

For conventional closed supracondylar fractures of the humerus, treatment is

blood flow. However, even with perfect born reposition, surgical treatment is still needed to treat the vascular injury in case having really anatomical damage to the vascular wall. Among the techniques for achieving and maintaining born reposition, pinning is the most effective method, but it is also the most invasive and time-consuming, and is associated with risk of complications. By contrast, casts present the simplest and least invasive technique; they may not gave perfect reposition but could decompress vessel and restore blood flow. In this study, when patients were mainly diagnosed with type III displaced supracondylar fractures of the humerus before treatment (92%), this figure decreased sig-

*4.3. Efficacy of Vascular Rehabilitation after Treatment* 

nificantly after cast to 12% and 0% after pinning (Figure 4).

**Figure 4.** X-rays of supracondylar fractures of the humerus. (**A**) Before reposition—Gartland type III. (**B**) After reposition—

*4.4. Vascular Injury, Surgical Management and Related Factors* 

Gartland type II. (**C**) After operative pinning—Gartland type II.

#### *4.3. Efficacy of Vascular Rehabilitation after Treatment* achieved mainly by reposition of fractured born and cast or pinning. In cases with vascu-

*4.3. Efficacy of Vascular Rehabilitation after Treatment* 

*Children* **2021**, *8*, x FOR PEER REVIEW 8 of 11

For conventional closed supracondylar fractures of the humerus, treatment is achieved mainly by reposition of fractured born and cast or pinning. In cases with vascular injury, the good born reposition may enable to decompress blood vessels and restore blood flow. However, even with perfect born reposition, surgical treatment is still needed to treat the vascular injury in case having really anatomical damage to the vascular wall. lar injury, the good born reposition may enable to decompress blood vessels and restore blood flow. However, even with perfect born reposition, surgical treatment is still needed to treat the vascular injury in case having really anatomical damage to the vascular wall. Among the techniques for achieving and maintaining born reposition, pinning is the most effective method, but it is also the most invasive and time-consuming, and is associ-

For conventional closed supracondylar fractures of the humerus, treatment is

(**A**) (**B**)

**Figure 3.** Brachial artery injuries in multislice CT-A. (**A**) Lesions < 5 mm. (**B**) Lesions ≥ 5 mm.

Among the techniques for achieving and maintaining born reposition, pinning is the most effective method, but it is also the most invasive and time-consuming, and is associated with risk of complications. By contrast, casts present the simplest and least invasive technique; they may not gave perfect reposition but could decompress vessel and restore blood flow. In this study, when patients were mainly diagnosed with type III displaced supracondylar fractures of the humerus before treatment (92%), this figure decreased significantly after cast to 12% and 0% after pinning (Figure 4). ated with risk of complications. By contrast, casts present the simplest and least invasive technique; they may not gave perfect reposition but could decompress vessel and restore blood flow. In this study, when patients were mainly diagnosed with type III displaced supracondylar fractures of the humerus before treatment (92%), this figure decreased significantly after cast to 12% and 0% after pinning (Figure 4).

**Figure 4.** X-rays of supracondylar fractures of the humerus. (**A**) Before reposition—Gartland type III. (**B**) After reposition— Gartland type II. (**C**) After operative pinning—Gartland type II. **Figure 4.** X-rays of supracondylar fractures of the humerus. (**A**) Before reposition—Gartland type III. (**B**) After reposition— Gartland type II. (**C**) After operative pinning—Gartland type II.

#### *4.4. Vascular Injury, Surgical Management and Related Factors 4.4. Vascular Injury, Surgical Management and Related Factors*

Regarding the cases of conservative treatment when born reposition and cast was applied but the radial pulse was still not found, if Doppler ultrasound analysis showed reduced blood flow and there were long-length of lesions on a multislice CT-A, patients were indicated for surgery with born reposition, pinning and re revascularization.

Intraoperatively found that, two-thirds of surgical cases had vasospasm without anatomical damage of the vessel wall or intravascular thrombosis. In most cases, following born reposition and cast, the blood vessels have been released from the fracture and only a few are still stuck in the fracture. All cases of surgery fractured born was reposited and fixed by pinning with K-wire from the lateral side. In cases with vasospasm. Blood flow will be restored by extra-vessel papaverine blockage or balloon dilatation of the brachial artery using a Fogarty catheter or cutting damaged artery and performing anastomosis end-to-end or replace it by great saphenous graft, while a brachial artery had wall damage (Figure 5).

(Figure 5).

(Figure 5).

*Children* **2021**, *8*, x FOR PEER REVIEW 9 of 11

**Figure 5.** (**A**) Arterial entrapment in the fracture socket. (**B**) Freeing the artery from the fracture socket. (**1**: median nerve; **2**: brachial artery; **3**: brachial vein; **4**: fracture location.) **Figure 5.** (**A**) Arterial entrapment in the fracture socket. (**B**) Freeing the artery from the fracture socket. (**1**: median nerve; **2**: brachial artery; **3**: brachial vein; **4**: fracture location.) With vasospasm, there were several cases of short spasm which is not dilated after

With vasospasm, there were several cases of short spasm which is not dilated after the pinning and blocking using papaverin. This opens the blood flow very weakly but without damage to the vessel wall and thrombosis of the lumen. This explains why on a multislice CT-A there can be seen a loss of the long brachial artery but no damage to the vessel wall during surgery. Therefore, if Doppler ultrasound accurately assesses the status of contusion in the transverse fracture and non-thrombotic lumen, angioplasty will most likely not need to be performed. This leads to a real vascular injury and the risk of a serious With vasospasm, there were several cases of short spasm which is not dilated after the pinning and blocking using papaverin. This opens the blood flow very weakly but without damage to the vessel wall and thrombosis of the lumen. This explains why on a multislice CT-A there can be seen a loss of the long brachial artery but no damage to the vessel wall during surgery. Therefore, if Doppler ultrasound accurately assesses the status of contusion in the transverse fracture and non-thrombotic lumen, angioplasty will most likely not need to be performed. This leads to a real vascular injury and the risk of a serious vascular complication, especially where a specialist in vascular surgery is not available (Figure 6). the pinning and blocking using papaverin. This opens the blood flow very weakly but without damage to the vessel wall and thrombosis of the lumen. This explains why on a multislice CT-A there can be seen a loss of the long brachial artery but no damage to the vessel wall during surgery. Therefore, if Doppler ultrasound accurately assesses the status of contusion in the transverse fracture and non-thrombotic lumen, angioplasty will most likely not need to be performed. This leads to a real vascular injury and the risk of a serious vascular complication, especially where a specialist in vascular surgery is not available (Figure 6).

Regarding the cases of conservative treatment when born reposition and cast was applied but the radial pulse was still not found, if Doppler ultrasound analysis showed reduced blood flow and there were long-length of lesions on a multislice CT-A, patients

Regarding the cases of conservative treatment when born reposition and cast was applied but the radial pulse was still not found, if Doppler ultrasound analysis showed reduced blood flow and there were long-length of lesions on a multislice CT-A, patients

Intraoperatively found that, two-thirds of surgical cases had vasospasm without anatomical damage of the vessel wall or intravascular thrombosis. In most cases, following born reposition and cast, the blood vessels have been released from the fracture and only a few are still stuck in the fracture. All cases of surgery fractured born was reposited and fixed by pinning with K-wire from the lateral side. In cases with vasospasm. Blood flow will be restored by extra-vessel papaverine blockage or balloon dilatation of the brachial artery using a Fogarty catheter or cutting damaged artery and performing anastomosis end-to-end or replace it by great saphenous graft, while a brachial artery had wall damage

Intraoperatively found that, two-thirds of surgical cases had vasospasm without anatomical damage of the vessel wall or intravascular thrombosis. In most cases, following born reposition and cast, the blood vessels have been released from the fracture and only a few are still stuck in the fracture. All cases of surgery fractured born was reposited and fixed by pinning with K-wire from the lateral side. In cases with vasospasm. Blood flow will be restored by extra-vessel papaverine blockage or balloon dilatation of the brachial artery using a Fogarty catheter or cutting damaged artery and performing anastomosis end-to-end or replace it by great saphenous graft, while a brachial artery had wall damage

were indicated for surgery with born reposition, pinning and re revascularization.

were indicated for surgery with born reposition, pinning and re revascularization.

**Figure 6.** Vasospasm of brachial artery and its surgical management. (**A**) Vasospasm. (B) Balloon dilatation of the brachial ar-tery using the Fogarty catheter. **Figure 6.** Vasospasm of brachial artery and its surgical management. (**A**) Vasospasm. (**B**) Balloon dilatation of the brachial ar-tery using the Fogarty catheter.

**Figure 6.** Vasospasm of brachial artery and its surgical management. (**A**) Vasospasm. (B) Balloon dilatation of the brachial ar-tery using the Fogarty catheter. Comparing intraoperative finding of vascular lesions with the age group of the patient, we found that patients under 3 years of age only had vasospasm, while those over 13 years of age only had vascular contusion. This can be explained by ossification in the humerus bone. The ossification centers begin to fuse together at age 3 years and the ossification process completes by the age of 13 [14]. Therefore, in children less than 3 years of age, the bone structure is almost entirely cartilage which is not capable of causing damage to the vessel wall, while since the bone structure in children over 13 years old is almost adult, the pathology of vascular trauma is similar to that of an adult with typical vessel contusion and thrombosis. In the study, direct anastomosis end-to-end was performed when the lesions were short (less than 1 cm). In addition, the choice of grafting with saphenous vein is very difficult because of the very small size of vessel in children. However, in older children

with sufficiently large vein size, the artery bypass grafting can be used if the damage to the vessel wall is long, which avoids missing the injury with postoperative embolism as a result. We documented the case of a patient who suffered a thrombosis immediately after direct anastomosis end-to-end surgery and who was then re-operated on using grafting with saphenous vein. The cause of complication may be not completely removement of the damaged segment, short length and tension of vessel, may also be due to the failed technique or the inadequate anticoagulation (Figure 7). used if the damage to the vessel wall is long, which avoids missing the injury with postoperative embolism as a result. We documented the case of a patient who suffered a thrombosis immediately after direct anastomosis end-to-end surgery and who was then re-operated on using grafting with saphenous vein. The cause of complication may be not completely removement of the damaged segment, short length and tension of vessel, may also be due to the failed technique or the inadequate anticoagulation (Figure 7).

Comparing intraoperative finding of vascular lesions with the age group of the patient, we found that patients under 3 years of age only had vasospasm, while those over 13 years of age only had vascular contusion. This can be explained by ossification in the humerus bone. The ossification centers begin to fuse together at age 3 years and the ossification process completes by the age of 13 [14]. Therefore, in children less than 3 years of age, the bone structure is almost entirely cartilage which is not capable of causing damage to the vessel wall, while since the bone structure in children over 13 years old is almost adult, the pathology of vascular trauma is similar to that of an adult with typical vessel contusion and thrombosis. In the study, direct anastomosis end-to-end was performed when the lesions were short (less than 1 cm). In addition, the choice of grafting with saphenous vein is very difficult because of the very small size of vessel in children. However, in older children with sufficiently large vein size, the artery bypass grafting can be

*Children* **2021**, *8*, x FOR PEER REVIEW 10 of 11

**Figure 7.** Physical injury to brachial artery and its surgical management. (**A**) Contusion, thrombosis and brachial artery nearly ruptured. (**B**) Damaged artery excised and directly anastomosed. **Figure 7.** Physical injury to brachial artery and its surgical management. (**A**) Contusion, thrombosis and brachial artery nearly ruptured. (**B**) Damaged artery excised and directly anastomosed.

#### *4.5. Limitations 4.5. Limitations*

Several limitations need to be noted in this article. First of all, in operating on these patients, it was most important to us to restore vascular circulation, the recovery of bone anatomy being secondary. Hence, although there is no anatomical perfection, the bones have been less displaced. Additionally, while there was a follow-up in this study, the period for evaluating patient outcomes was within one month following surgery. There are two main reasons for this which should be acknowledged. Firstly, the main purpose in our study was to focus on treating early-stage vascular damage. Secondly, because COVID-19 lockdowns have been continuously imposed in Vietnam and out-of-province patients are those who have been most affected by this, the re-examination of patients after surgery has been very difficult in our institution. Therefore, longer-term outcomes Several limitations need to be noted in this article. First of all, in operating on these patients, it was most important to us to restore vascular circulation, the recovery of bone anatomy being secondary. Hence, although there is no anatomical perfection, the bones have been less displaced. Additionally, while there was a follow-up in this study, the period for evaluating patient outcomes was within one month following surgery. There are two main reasons for this which should be acknowledged. Firstly, the main purpose in our study was to focus on treating early-stage vascular damage. Secondly, because COVID-19 lockdowns have been continuously imposed in Vietnam and out-of-province patients are those who have been most affected by this, the re-examination of patients after surgery has been very difficult in our institution. Therefore, longer-term outcomes have not been evaluated in this study.

#### have not been evaluated in this study. **5. Conclusions**

**5. Conclusions**  The majority of supracondylar humerus fractures with brachial artery injuries did not present with signs and symptoms of critical limb ischemia similar to peripheral vascular injuries in lower extremity; therefore, emergency management was not required in all cases. Diagnosis and treatment for pediatric supracondylar humerus fractures with The majority of supracondylar humerus fractures with brachial artery injuries did not present with signs and symptoms of critical limb ischemia similar to peripheral vascular injuries in lower extremity; therefore, emergency management was not required in all cases. Diagnosis and treatment for pediatric supracondylar humerus fractures with vascular injury is a difficult and time-consuming procedure, especially in cases of transverse fractures. Two-thirds of patients who underwent surgery had no physical damage to the blood vessel wall and lumen. As a result of this study, we propose to apply a unique protocol in the management of pediatric supracondylar humerus fractures with brachial artery injuries with the aim of shortening treatment duration and minimizing the performance of unnecessary procedures and surgical treatments.

**Author Contributions:** Conceptualization, T.N.V.; methodology, T.N.V., S.H.D.P., L.H.V. and U.H.N.; validation, T.N.V., S.H.D.P. and U.H.N.; formal analysis, T.N.V. and L.H.V.; investigation, T.N.V., S.H.D.P., L.H.V. and U.H.N.; data curation, T.N.V. and L.H.V.; writing—original draft preparation, T.N.V., S.H.D.P. and L.H.V.; writing—review and editing, U.H.N.; visualization, T.N.V., S.H.D.P., L.H.V. and U.H.N.; supervision, U.H.N. All authors have read and agreed to the final version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The studies involving human participants were reviewed and approved by the Ethics Board of the Viet Duc University Hospital (2430/QÐ-VÐ), 12 December 2020.

**Informed Consent Statement:** Informed consent was obtained from all individual participants included in the study. If the participants are under the age of 18, or otherwise legally or medically unable to provide written informed consent, then consent was obtained from their parents or guardians.

**Data Availability Statement:** All the data that support the findings of this study are available from the corresponding author on reasonable request. Requests for access to these data should be made to Tu Ngoc Vu (Email: vungoctu@hmu.edu.vn).

**Acknowledgments:** We sincerely thank the patients, their families, the members working at Department of Cardiovascular and Thoracic Surgery and the Board of Directors of Viet Duc University Hospital for their support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

	- 11. Abzug, J.M.; Kozin, S.H.; Zlotolow, D.A. *The Pediatric Upper Extremity*; Springer: Hoboken, NJ, USA, 2015.
	- 12. Houshian, S.; Mehdi, B.; Larsen, M.S. The epidemiology of elbow fracture in children: Analysis of 355 fractures, with special reference to supracondylar humerus fractures. *J. Orthop. Sci.* **2001**, *6*, 312–315. [CrossRef] [PubMed]
