ORIGINAL ARTICLE

Supracondylar humerus fractures in children treated with closedreduction and percutaneous pinning

Michelangelo Scaglione • Daniele Giovannelli •

Luca Fabbri • Dario Dell’Omo • Andrea Goffi •

Giulio Guido

Received: 11 November 2011 / Accepted: 11 May 2012 / Published online: 22 July 2012

� Springer-Verlag 2012

Abstract Supracondylar fractures of the humerus in

children are important for frequency and type of associated

serious complications. The management of this kind of

fractures is still controversial (Skaggs et al. in J Bone Joint

Surg Am 86:702–707, 2004; Kalllio et al. in J Pediatr

Orthop 12:11–15, 1992). We are going to present our

experience in the treatment of supracondylar humeral

fracture in children. In the Orthopedic Department of Pisa,

we treated 150 cases from 1989 to 2006. We are used to

perform, emergency or within 12 h, reduction and two

lateral-entry percutaneous pins fixation. The mean age was

7.5 years. We checked 125 cases, because we excluded all

the cases with follow up less then 5 years. The mean follow

up was 8.2 years. We used Gartland classification modified

by Wilkins. We evaluated 125 cases by using the Flynn

classification: 100 % of patients did not have impairment

of the elbow joint mobility. We had seven valgus deviation,

one of which was more then 10�. We also had 17 varus

deviations, 11 of which were not over 8� and only 2 of

them were 15�. The average value of the joint Baumann

angle was calculated as great as 16�. The obtained results

were classified as very good 80 %, good 11 %, sufficiently

good 6 %, and bad 3 %. In our experience, all the fractures

type II and III by Gartland have to be treated within 12 h,

with closed reduction and stabilization with lateral-entry

K-wire technique. The conservative treatment by cast is

indicated only in type I fracture. The trans olecranic

treatment is not realizable, for the stiffness which can

occur, for the risk of iatrogenic ulnar nerve lesion, and for

long-time hospitalization. The open reduction remains the

first choice treatment for exposed or nonreducible frac-

tures, and in cases of vascular injury.

Keywords Supracondylar fractures � Humerus �Children � Percutaneous fixation

Introduction

The importance of supracondylar humeral fractures in

children is mostly in their complications, which may vary

in frequency and type. The management of this kind of

fractures is still controversial [1, 2].

These sort of trauma represent 75 % of all elbow frac-

tures in children. They are usually due to forces that cause

hyperextension of the upper limb (97 %) and rarely they

can be caused by strong impact on the posterior surface of

the humerus while the elbow is in a flexion position (3 %).

The majority of these fractures are caused by a fall from

a height of about two meters (i.e., from Monkey Barr, or

from slides and swings) [3], but a small percentage of

theme can occur after a fall on the ground level (i.e.,

bicycle or skateboard) [4].

These fractures are very dangerous, because the anterior

displacement of the proximal fragment can damage vas-

cular and nervous structures. Early complications are

humeral artery injury (3.7–7 %), nerve lesions (6–12 %)

(radial, median and anterior interosseus, and ulnar nerve [5,

6]), and Wolkmann syndrome [7], which is the most

serious among them. Delayed complications are varus or

valgus deviations of the axes (6–16 %), and rarely elbow

stiffness.

Therefore, the treatment of these type of fractures

should be immediate, accurate, and less invasive as

M. Scaglione (&) � D. Giovannelli � L. Fabbri � D. Dell’Omo �A. Goffi � G. Guido

Department of Orthopedics, University of Pisa,

Via Paradisa 2, Ed 3, 56100 Pisa, Italy

e-mail: xario@hotmail.it; m.scaglione@ao-pisa.toscana.it

123

Musculoskelet Surg (2012) 96:111–116

DOI 10.1007/s12306-012-0204-5

possible. In this article, we present our experience at the

Orthopedic department of Pisa in treating these fractures

with emergency reduction and two lateral-entry percuta-

neous pins fixation.

Materials and methods

At the Orthopedic Department of Pisa University, we dealt

with 150 cases; 82 of these were male patients, 68 were

females, with a mean age of 7.5 years (range 20 months–

12 years). In this article though, we considered just 125

cases, 79 males and 46 females excluding all those with a

shorter than 5-year follow up. The average follow up was

8.2 years, in a range between 5 and 15.

According to the Gartland system, modified by Wilkins

[8], one patient had Gartland type I injury, 48 patients had

type II injuries, 76 had type III injuries, 66 of which were

III A and 10 III B (Table 1).

All fractures were treated in emergency, from minimum

1 h to max 5 h after trauma (minimum time required for

preoperative exams and a safe anesthesia).

Type-1 fractures have been treated with a cast in only

one case. After 7 days, X-rays showed a loss of reduction,

and a treatment with two K-wires was performed.

At the moment of their admission, the radial pulse was

absent in five patients, and returned after reduction. 11

Patients had nerve involvement (median nerve in four

patients, interosseus anterior nerve in two patients, and

radial nerve in five patients), 10 of these were resolved,

whereas a median nerve neurolysis was necessary in one

case, after a month.

We have used the surgical technique, conceived by

Arino [9], (lateral cross-wiring) with two parallel Kirschner

wires introduced through the lateral condyle across the

fracture and into the medial cortex.

The cosmetic and functional results were evaluated with

X-rays and Flynn’s Criteria [10] (Table 2). Moreover, we

investigated Baumann’s angle (9�–26�) for all patients.

Surgical technique

Under general anesthesia, the patient is placed supine on

the operating table and the fracture is analyzed with an

image intensifier.

The surgeon applies a gradual longitudinal traction with

one hand, placing the forearm in supinated position and

flexing the elbow at 30�; his other hand holds the fracture’s

proximal fragment and the reduction of the fracture is

performed with small movements in varus valgus, with the

help of a fluoroscope.

At this point, the surgeon pronates the forearm and

flexes the elbow, while maintaining the traction on the

forearm. In this way, he uses the forearm as a joystick, so

that he can control and reduce both rotation and angulation

of the distal fracture fragment.

The gap of the anterior bone cortex is then reduced by

flexing the elbow.

Fluoroscopy checks are then performed in the antero-

posterior (AP), lateral (LL), and Jones views, rolling the

C-arm, without the risk of losing the obtained reduction 2.

Once a satisfactory reduction is obtained, percutaneous

pinning is performed with two parallel or slightly divergent

K-wires (1.5–2 mm), from the base of the lateral condyle

to the medial cortex of the fracture’s proximal fragment.

The K-wires must be placed as distant from each other as

possible and the first wire must be positioned very close to

the olecranon bone.

The reduction is then X-rays checked in AP and LL

projection with the elbow in extended position, and the

stability of the synthesis is evaluated performing passive

flexion–extension movements.

Afterward the upper limb is immobilized in a cast for a

period of 28 days. The position of the immobilized arm can

be decisive for the outcome of the reduction. Arnold and

Nasca [11], using electromyographic studies on corpses,

have highlighted the role of biomechanical factors in pre-

venting deformities. These Authors have demonstrated

how the position of the forearm can affect the outcome,

altering muscles’ tension: fractures with deformities in

varus should be immobilized with the forearm in pronated

position, while the ones in valgus with the forearm in

supinated position. When the forearm is pronated, the

brachioradialis muscle and wrist’s extensor muscles move

the distal fragment of the fracture into a valgus position,

whereas the flexor muscles and the pronator teres move the

distal fragment of the fracture to the varus position.

In order to reduce the risk of local vascular disorders,

the elbow should be placed within the cast at an angle of

Table 1 Gartland system modified by Wilkins

Type I Nondisplaced 1

Type II Displaced intact posterior cortex 48

Type III Displaced no cortical contact 76

A Posteromedial 66

B Posterolateral 10

Table 2 Flynn’s criteria

Rating Cosmetic factor:

carrying angle loss (�)

Functional factor:

motion loss (�)

Excellent 0–5 0–5

Good 6–10 6–10

Fair 11–15 11–15

Poor [15 [15

112 Musculoskelet Surg (2012) 96:111–116

123

40�, avoiding extreme pronated position. After 28 days, we

reviewed all the children, removed the cast, made an X-ray

check, removed the two K-wires, and started a functional

rehabilitation. The results were evaluated after 2 months, 6

months, and 1 year (Figs. 1, 2).

Results

We evaluated 125 cases which we classified according to

the Wilkins modification of the Gartland classification: one

patient had a Gartland type I injury, 48 patients had type II

injuries, 76 had type III injuries, 66 of which were III A

and 10 III B.

We then used the Flynn classification in order to survey

the overall outcome: 101 patients (81 %) had no valgus nor

varus deviations. The remaining 19 %, which belonged to

type III of the Gartland classification, presented a valgus

deviation in 7 cases (6 %), nine of which were less then

10�, and a varus deviation in 17 cases (13 %), 11 of which

were not over 8� and 2 of them were 15�.

None of the subjects showed impairment of the elbow

joint R.O.M. at the end of rehabilitation.

We also considered the Baumann joint angle: the average

value was 16� (range 10�–27�), supporting the reference

normal values of 9�–26� range. The obtained results were

classified as very good 80 %, good 11 %, sufficiently good

6 %, and bad 3 %, according to Flynn’s criteria (Table 3).

Patients did not show significant clinic symptoms, nor did

they complain about pain. In the 3 % that was classified as

‘‘bad’’, the only problem was related to esthetics. An inter-

esting case was represented by a child who had a type 4

fracture, which subsequently developed into a chondral injury

of the humeral condyle, as shown by the X-ray after a year

from the trauma. This complication was probably due to the

severity of the initial trauma, the diameter of the K-wires used,

or fixation of the K-wires to the cast. After 2 years, X-rays

showed that the nucleus of the bone epiphysis was still frag-

mented in the correct position, with an undoubtful functional

improvement. After 4 years, X-rays showed a good healing,

with complete functional recovery of the joint R.O.M.

We did not have infections, pins loosening, or loss of

reduction during the follow up, nor the need of second

surgery as completion of the procedures. We did not

observe any involvement of the ulnar nerve, or Wolkmann

syndrome cases.

Fig. 1 MR, 7-year-old white male: a, b type III-B supracondylar humerus fracture according to Gartland system modified by Wilkins; c,

d radiograph 30 days later to operation; e–h excellent radiological and functional results 13 years later

Musculoskelet Surg (2012) 96:111–116 113

123

Discussion

Supracondylar humeral fractures in children represent a deli-

cate trauma to treat because of the risk of severe and some-

times invalidating complications. Up till now, uniformity in

literature about the best techniques or treatments is missing.

The commonly used techniques are open reduction, per-

cutaneous K-wires, transolecranon traction, or conservative

Fig. 2 SA, 7-year-old white female: a, b type III-A supracondylar

humerus fracture according to Gartland system modified by Wilkins;

c, d radiograph 30 days later to operation; e, f radiograph 90 days

later, with adequate restoration of all radiologic indices; g–j excellent

radiological and functional results 11 years later

Table 3 Results obtained in our cases according to Flynn’s criteria

Rating Carrying angle

loss (�)

Motion

loss (�)

Results

(%)

Excellent 108 125 80

Good 11 11

Fair 4 6

Poor 2 3

114 Musculoskelet Surg (2012) 96:111–116

123

treatment for not displaced fractures. All these methods

usually give good functional results, but occasionally can be

followed by an elbow deformity. In particular among all

deformities, varus elbow is the most common [11–13].

Complications like infections, pins mobilization, iatro-

genic nerve lesions, Wolkmann syndrome, and stiffness

have greatly reduced their incidence thanks to the modern

surgical techniques.

The conservative treatment (cast) is indicated only in

type I fracture 2.3. In all the other types of fractures, the

results of conservative treatment seem to be worse than

those deriving from a surgical treatment. Furthermore, the

need to apply the cast with the elbow in a flexed position

(120�) would determine a major risk of vascular involve-

ment, and difficulty in X-rays performing [7, 11, 12, 14].

Transolecranon traction is not advisable, not only for the

stiffness which can occur and for the risk of iatrogenic

ulnar nerve lesion, but also for the need of a long hospi-

talization (2–3 weeks) [2, 11].

Most Authors prefer percutaneous K-wires reduction

surgery and cast immobilization [1–3, 11, 14, 15]. There

are still many diverging opinions about the use of crossing

or parallel K-wire techniques. Ulnar nerve lesion can be a

consequence of the crossing wire technique [1, 2, 13, 16],

mostly because of the edema and constriction of the ulnar

nerve in the cubital tunnel, and rarely to the K-wire

penetrating the nerve by a medial-entry approach [1,

16–19]. In most cases, this situation resolves spontaneously

in 4–6 months [16]. In order to avoid nerve involvement, it

is suggested to perform a small incision and a slight

extension of the elbow during the insertion of the medial-

entry K-wire [1, 13, 16, 20].

Ayadi et al. [21] describes 1880 cases and reports 55

ulnar lesions (23 of which were postoperative). During our

experience, we have never had a case of iatrogenic nerve

lesion.

Ulnar nerve involvement can also be avoided using the

two lateral-entry K-wires technique.

Many Authors (Skaggs, Larson) [1, 13, 16, 20, 22]

believe that implants involving two columns (lateral and

medial) are as stable as those with two crossing wires. For

this reason, they recommend to introduce the K-wires from

the lateral cortex of the distal epiphysis of the humerus,

inserting them up to the medial cortex of the proximal

fragment. The K-wires have to be divergent and far enough

from each other (at least 2 mm) [1, 2, 10, 13, 16, 20, 23].

Implant stability can be enhanced by the use of a third

K-wire [1, 13, 16, 20–23], as described by Larson et al. [22].

Furthermore, it is statistically relevant that the ulnar

nerve injury is more frequent in the medial- and lateral-

entry technique than in the two lateral-entry pins technique.

The two techniques have a nerve lesion incidence rate of

3.5 and 1.9 %, respectively.

As Skaggs says ‘‘when correct technique is used, lateral-

entry pins alone provide sufficient fixation stability with

avoidance of the possibility of iatrogenic ulnar nerve

injury’’.

The open reduction, however, remains the first choice

for exposed or nonreducible fractures, and it is also indi-

cated in cases of vascular injuries, with involvement of the

humeral artery, due to the dislocation of the fragments. In

most cases, we assist to a spontaneous resolution of the

vascular problem thanks to the formation of vascular col-

lateral circulation or to a resolution of the vessel occlusion.

Exploration and reduction of the dislocated fragment is

sometimes required. Vascular lesions are rarely present,

and when they occur, the injured tract of the vessel must be

necessarily removed and bypassed [16, 18]. With the open

reduction though, there is a major risk of joint stiffness and

infection [2].

During our experience, we have never had the need to

perform an open reduction, probably because we always

perform closed reduction and K-wires stabilization under

emergency condition, but despite of that there is not for the

time being a univocal opinion about the right time for

surgery.

Many Authors point out that the fractures that are

treated under emergency condition have better results than

those ones treated later [22]. This is due to the fact that the

risk of a Wolkmann syndrome is avoided in early surgery

[23, 24]. On the contrary, other Authors think that the

initial edema can complicate the reduction and recommend

the use of the transolecranon traction for 3–5 days before

surgical treatment [25] is performed.

Gupta [26], in a retrospective study, demonstrated the

absence of a significant difference between the two

methods.

In our experience, all the Gartland type II and III frac-

tures were treated in emergency within the first 12 h, with

closed reduction and K-wires stabilization according to

Arino Technique.

This technique presents many advantages, for instance

an easier reduction and the absence of vascular-nervous

complications, thanks to the fact that an immediate

reduction and synthesis can reduce blood loss, avoiding

vascular lesions and the Wolkmann syndrome.

Moreover, the entry site preserves the joint cartilage and

the fracture site, and it avoids the ulnar nerve lesion.

We think that the loss of correction in the immediate

postoperative period represents a real danger, usually due

to a wrong position of the wires.

We agree with some authors that the axial deviations,

either varus or valgus, are mainly caused by insufficient

reduction or wrong synthesis and are not caused by post-

traumatic anatomic alterations of the humeral epiphysis

(Arino, Aronson, Flynn, Labelle, Sutton) [9, 10, 27].

Musculoskelet Surg (2012) 96:111–116 115

123

According to our experience, deformities that are caused

by a less than 10� degrees rotation of the distal fragment

can improve spontaneously in children, and anyway they

do not reduce the elbow’s R.O.M.

This technique results simple in capable hands and

reduces the hospitalization time. In fact, children only

spend one night in hospital, with little discomfort for the

family.

Taking into account the overall costs (hospitalization

time, used materials, and time of surgery), the method of

transolecranon traction determines a cost around 5,650

euros instead the percutaneous technique about 2,930

euros.

This allows a reduction of sanitary costs, with an

effective saving of about 52 %.

This difference of costs between the two techniques is

even larger in other Authors [27].

Conflict of interest None.

References

1. Skaggs DL, Cluck MW et al (2004) Lateral entry pin fixation in

the management of supracondylar fractures in children. J Bone

Joint Surg Am 86:702–707

2. Kalllio PE, Foster BK et al (1992) Difficult supracondylar elbow

fractures in children: analysis of percutaneous pinning technique.

J Pediatr Orthop 12:11–15

3. Leitch KK, Kay RM et al (2006) Treatment of multidirectionally

unstable supracondylar humeral fractures in children. A modified

gartland type-IV fracture. J Bone Joint Surg Am 88:980–985

4. Farnsworth CL, Silva PD (1998) Etiology of supracondylar

humerus fractures. J Pediatr Orthop 18:38–42

5. Campbell CC, Waters PM et al (1995) Neurovascular injury and

displacement in type III supracondylar humerus fractures.

J Pediatric Orthop 15:47–52

6. De Sanctis N, Razzano E et al (1990) La paralisi del nervo int-

erosseo anteriore (N.I.A.) possibile complicanza delle fratture

sovracondiloidee omerali nell’infanzia. Rivista Italiana Ortopedia

Pediatrica 8(2):379–382

7. Mubarak SJ, Carroll NC (1979) Volkmann’s contracture in

children: aetiology and prevention. J Bone Joint Surg (Br) 61:

285–293

8. Gartland JJ (1959) Management of supracondylar fractures of the

humerus in children. Surg Gynecol Obstet August:145–154

9. Arino VL, Lluch EE, Ramirez AM, Ferrer J, Rodriguez L,

Baixauli F (1977) Percutaneous fixation of supracondylar frac-

tures of the humerus in children. J Bone Joint Surg 59-A:914–916

10. Flynn JC, Matthews JG, Benoit RL (1974) Blind pinning of

displaced supracondylar fractures of the humerus in children.

Sixteen years experience with long term follow-up. J Bone Joint

Surg [AM] 56:263–272

11. James AA, Richard JN (1977) Supracondylar fractures of the

humerus. The role of dynamic factors in prevention of deformity.

J Bone Joint Surg 350–355

12. Giannini S, Maffei G, Girolami M, Ceccarelli F (1983) Il trat-

tamento delle fratture sovracondiloidee del bambino con riduzi-

one e fissazione con fili di Kirschner percutanei. GIOT Fasc 2

183–192

13. Boyd DW, Aronson DD (1992) Supracondylar fractures of the

humerus: a prospective study of percutaneous pinning. J Pediatr

Orthop 12:789–794

14. Sankar WN, Hebela NM, Skaggs DL, Flynn JM (2007) Loss of

pin fixation in displaced supracondylar humeral fractures in

children: causes and prevention. J Bone Joint Surg Am 89:

713–717

15. De Pellegrini M, Brivio A, Pescatori E, Tessari L (2008)

Supracondylar humerus fractures in children: closed reduction

and cross pin fixation in prone position. GIOT 34:199–204

16. Pirone AM, Graham HK et al (1979) Management of displaced

extension-type supracondylar fractures of the humerus in chil-

dren. J Bone Joint Surg Am 61:425–428

17. Lyons JP, Ashley E et al (1998) Ulnar nerve palsies after per-

cutaneous cross-pinning of supracondylar fractures in children’s

elbow. J Pediatr Orthop 18:43–45

18. Rasool MN (1998) Ulnar nerve injury after K-wire fixation of

supracondylar humerus fractures in children. J Pediatr Orthop

18(5):686–690

19. Wilkins KE (1991) Fractures and dislocations of the elbow

region. In: Rockwood CA Jr, Wilkins KE, King RE (eds) Frac-

tures in children, vol III. Lippincott, Philadelphia, p 593

20. Kocher MS, Kasser JR (2007) Lateral entry compared with

medial and lateral entry pin fixation for completely displaced

supracondylar humeral fractures in children. A randomizer clin-

ical trial. J Bone Joint Surg Am 89:706–712

21. Ayadi K, Trigui M, Tounsi N, Ellouze Z, Bahloul L, Triki FE

et al (2006) Supracondylar fractures of the humerus in children.

Rev Chir Orthop 92:651–656

22. Larson L, Firoozbakhsh K (2006) Biomechanical analysis of

pinning techniques for pediatric supracondylar humerus fractures.

J Pediatr Orthop 26(5) Sep–Oct

23. Epps CH, Bowen R (1995) Complication in pediatric orthopaedic

surgery. Philadelpjia, JB Lippincott

24. Mastragostino S, Stella G (1990) L’urgenza nelle fratture

sovracondiloidee di omero nei bambini. Rivista Italiana Ortopedia e

Traumatologia Pediatrica 280–284

25. Odgen JA (1990) Skeletal injury in the child. WB Saunders,

Philadelphia

26. Gupta N, Kay RM et al (2004) Effect of surgical delay on peri-

operative complications and need for open reduction in supra-

condylar humerus fractures in children. J Pediatr Orthop 24(3):

245–248

27. Sutton WR, Greene WB, Georgeopoulos G, Dameron TB Jr

(1992) Displaced supracondylar fractures in children. A com-

parison of results and costs in patients treated by skeletal traction

versus percutaneous pinning. Ciln Orthop 278:81–87

116 Musculoskelet Surg (2012) 96:111–116

123