- Email: [email protected]
Treatment of severely displaced proximal humeral fractures in children with retrograde elastic stable intramedullary nailing
Injury, Int. J. Care Injured (2008) 39, 1453—1459
www.elsevier.com/locate/injury
Treatment of severely displaced proximal humeral fractures in children with retrograde elastic stable intramedullary nailing ¨rfer, T. Wirth F.F. Fernandez *, O. Eberhardt, M. Langendo Department of Orthopaedics, Klinikum Stuttgart, Olgahospital, Bismarckstr. 8, D-70176 Stuttgart, Germany Accepted 3 April 2008
KEYWORDS Proximal humeral fractures; Children; Elastic stable intramedullary nailing (ESIN); Nancy nail
Summary We report our experience of elastic stable intramedullary nailing (ESIN) of severely displaced proximal humeral fracture Type III and IV according to Neer in children. Thirty-five children (mean age 12.7 years) with 19 Salter—Harris Type II epiphyseal fractures and 16 children with metaphyseal fractures of the proximal humerus were treated with ESIN by one-nail or two-nail-technique. There were the following complications: two perforations of the nail at the head of the humerus with loss of position, one loss of position without nail perforation, one misplacement of a nail, one revision due to haematoma and two difficult removals of metal. For followup examinations after 26 months the children were classified by use of the Constant— Murley score. On average they gained 99 points. All children were able to resume their sporting activities as before. For the treatment of displaced proximal humeral fractures in children more than 10 years old the retrograde ESIN represents a safe minimal invasive surgical procedure. # 2008 Elsevier Ltd. All rights reserved.
Introduction Proximal humeral fractures in children are rare and represent about 2% of all paediatric fractures. Eighty-five percent of these fractures are either undisplaced or only slightly displaced fractures and only 15% demonstrate severe displacement of * Corresponding author. Tel.: +49 711 992 3017; fax: +49 711 992 3820. E-mail address: [email protected] (F.F. Fernandez).
metaphyseal or growth plate involving fractures.1,7,10,14 Purely epiphyseal fractures and apophyseal lesions are only discussed in case reports8,11,12 and are fractures of the adolescents and adults. There is consensus for treating slightly displaced proximal humeral fractures by non-operative means. The treatment of severely displaced fractures of the upper end of the humerus however is still under debate. These fractures most commonly occur before the age of 3 years and in children older than 12 years.4,14
0020–1383/$ — see front matter # 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2008.04.001
1454 Non-operative treatment is still regarded as gold standard for treating subcapital humeral fractures in childhood. It is reported that independent of the therapeutic approach all fractures unite and function will be fully restored. Generally displaced or slightly displaced proximal humeral fractures in childhood and adolescence are perfectly well treated by a short immobilisation period of 2—4 weeks using a Gilchrist type bandage. In contrast treatment of severely displaced Grades III and IV fractures of children over 10 years is controversial. Remodelling potential of a Neer III and Neer IV displacement in children older than 10—11 years is very limited. Dameron reported that children aged 11 years and older can only partially remodel deformities of over 208.4 The fact that immobilisation in Dessault bandage, thorax abduction plaster or even traction through the olecranon do not meet the requirements of modern fracture treatment with early functional postoperative care have led to alternative treatment methods. Most proximal humeral fractures are extraarticular lesions. During childhood avulsions of the greater or lesser tuberosity are extremely rare and usually only occur in adolescence. The growth plate of the proximal humerus accounts for 80% of longitudinal growth of the upper arm and hence has an enormous potential for correction of residual axial deformities.14 Up to the age of 10 years an axial malalignment of as much as 608 of deformity in varus-, anti- or retroversion can be corrected.14 Beyond the age of 10 the potential of correction is much less with the possibility of correcting axial deformities of up to 20—308.7,14 The corrective potential is least for valgus positions beyond 108. For surgical treatment a variety of osteosynthetic material is used such as K-wires and screws.5,13 The most preferred method is by using percutaneous K-wires. This method however exhibits some important disadvantages such as unstable osteosynthesis and the need for postoperative immobilisation. Further most of the protruding K-wires irritate the muscles and soft tissues. There is a considerable risk of perforation of the head of the humerus. The introduction of the retrograde elastic stable intramedullary nailing (ESIN) for fixing proximal humeral fractures adds a surgical technique which respects the principles of minimal invasive ortheosynthesis which does not compromise the soft tissues or the joint and most favourably allows early mobilisation of the injured arm.15 The goal of this study is to evaluate the clinical results following retrograde elastic intramedullary stabilisation of proximal humeral fractures during childhood and adolescence.
F.F. Fernandez et al.
Surgical technique The patient is placed in a supine position and the whole arm including the shoulder joint is entirely prepped and draped. For reduction it is mandatory that the arm can be moved in all directions at the level of the shoulder joint. The upper arm including the shoulder joint must be accessible for intraoperative fluorography. We favour the unilateral technique with the radial approach to the distal humerus. The skin incision of 2 cm in length is just above the lateral epicondyle of the humerus. Posteriorly to the crista supraepicondylaris the intramedullary cavity is opened with an awl. The entrance to the bone is opened up enough by turning the awl obliquely in a 458 position without perforating the opposite cortex. The Nancy nail is prebent 108 in the proximal 1/5. The first nail is inserted and brought up to the fracture site. The closed reduction is performed by traction, abduction of the arm to 908 and external rotation. The nail is then passed over the fracture line into the proximal fragment. The tip of the nail must perforate the growth plate. By rotation of the nail the proximal fragment can be moved and reduction can be improved. Subsequently the second nail is inserted into the humerus using the same entrance point. The nails should diverge in the head of the humerus. If decision is made to use the ‘‘one-nailtechnique’’ it is important that this nail is bigger in size. So it is recommended to have the nail over the lateral cortex by 1.5—2 cm for easier hardware removal.
Patients and methods Over a time period of the last 12 years all children and adolescents with proximal humeral fracture treated by retrograde intramedullary nailing were included in the study. When intramedullary nailing was introduced in our institution initially, we began with the one-nail-technique. With emerging experience, the two-nail-technique was increasingly performed. The charts and the X-rays of all children were evaluated with regard to the cause of the accident, additional injuries, extent of dislocation according to Neer and Horwitz,11 hospital stay, complications, after treatment, surgical dates at the primary operation and metal removal. Further the postoperative treatment as well as the cosmetic and functional long-term results were looked at. All patients could be followed and evaluated by use of the Constant—Murley score.3
Proximal humeral fractures in children Table 1 Neer proximal humeral epiphyseal fracture displacement grading Grade Grade Grade Grade
I II III IV
<5 mm shaft diameter 1/3 shaft diameter 2/3 shaft diameter >2/3 shaft diameter
Results There were 35 patients (15 boys and 20 girls) with mean age of 12.7 [8—17] years. The age distribution was: 4 patients <10 years, 19 patients between 10 and 13 years and 12 patients >13 years. In 18 cases the left arm, in 17 cases the right arm was injured. All 35 patients demonstrated a fracture dislocation degree III and IV after Neer and Horwitz11 (Table 1). The accident causes were trivial falls in 17 cases, falls from a greater height 6 times, ski and snowboard injuries 5 times, fall from the bicycle in 4 cases and fall from the horse in 3 cases. There were 19 Salter—Harris Type II fractures and 16 metaphyseal fractures of the proximal humerus. All fractures were closed injuries without damage to vascular and neural structures. In 33 children the fracture was a single lesion with one child having an additional glenoid fracture and one with an additional fracture of the olecranon. Twenty-seven children were primarily treated in our institution. Seven children were
1455 Table 2
Complications (n = 7)
Implant perforations with loss of position, n Misplacement of a nail, n Revision due to haematoma, n Loss of correction, n Difficult removals of metals, n
2 1 1 1 2
treated non-operatively before in another hospital and operated on secondarily after referral to our institution. One child was treated by percutaneous K-wires elsewhere and we had to change the stabilisation to intramedullary nailing due to irritation of the soft tissue by the introduced K-wires. The mean primary hospital stay was 4.1 days.2— 8 For metal removal the mean hospital stay was 2.8 days. This quite long time can be explained with the fact that at the beginning, a drain was inserted for about 1 or 2 days. Nowadays, removal of metalwork is done as a day case procedure. The mean time of surgery was 54 min (24—85). All children were treated under general anaesthetics. In 15 children stabilisation was achieved with one nail (4 Neer III and 11 Neer IV) and in 20 children by the use of two elastic nails (11 Neer III and 9 Neer IV). The diameter of the nails used in the one-nailtechnique varied from 3.0 to 4.0 mm and in the double-nail-technique from 2.5 to 3.5 mm. One patient had to have an open reduction because of
Figure 1 (A) 13-year-old girl after traffic accident. Metaphyseal fracture with type Neer IV dislocation. (B) Intramedullary stabilisation in the one-nail-technique. (C) Twelve months later, radiological healing and complete remodelling.
1456
F.F. Fernandez et al.
Figure 2 (A) 11-year-old girl falls while playing. Epiphyseal fracture type Salter II with type Neer III dislocation. (B) Intramedullary stabilisation in the two-nail-technique. Five months later radiological healing and complete remodelling.
the incarceration of the biceps tendon into the fracture. Postreduction position was anatomical in 25 children and improvement to dislocation Grade I after Neer and Horwitz11 in 10 patients. There were the following complications (Table 2): there was a perforation of the nail through the humeral head in two patients. The tip of the nail left the humeral head laterally without need for revision. There were two patients who experienced a loss of position, one 13-year-old girl with a Salter II fracture treated in the one-nailtechnique and an 11-year-old boy with a metaphyseal fracture treated in the two-nail-technique. In a 14-year-old boy with a subcapital humeral fracture the nail was initially misplaced and needed to be changed. One child needed to be revised for release of postoperative haematoma at the side of pin insertion. In two children the metal removal was extremely difficult. There was one more patient treated in the one-nail-technique with a loss of position without perforation of the nail. This fracture healed in a varus position of 108. There were no iatrogenic lesions to the radial nerve, pseudarthrosis, infections or wound breakdowns (Figs. 1—3). The mean time for bony consolidation was 7.1 weeks. Removal of metal was usually performed 5.2 months after the fracture. The radiographs before metal removal that were made at an average of 4.5 months after the original injury demonstrated that
there was almost completely remodelling of all postoperatively detected malpositions.
Follow-up examination All 35 children were seen for follow-up. The mean follow-up time was 26 months (between 6 and 58 months) post-injury. Median follow-up time was 35 months. The children were evaluated with the Constant— Murley score3 which rates pain (maximum 15 points), daily activities (maximum 20 points), power (maximum 25 points) and range of motion (maximum 40 points). The maximum is 100 points with 91—100 points representing excellent results. All children achieved excellent results independent from age, sex, fracture type and extent of displacement. In the Constant—Murley score patients reached 99 points (Table 3). Almost all children could return to their sporting activities again. Two children however who had a fall from a horse did not return to horse riding. There were no arm length discrepancies. Two children complained of some aching in the area of the injured shoulder joint. From the cosmetic aspect all patients were satisfied with the result with scars of a mean length of 2.8 cm. In 31 cases parents and children were very happy with the treatment results, in 3 cases they were satisfied with the
Proximal humeral fractures in children
1457
Figure 3 (A) 13-year-old girl with epiphyseal fracture type Salter II with type Neer IV dislocation. (B) Intramedullary stabilisation in the one-nail-technique. (C) Four months later radiological healing and implant perforation with loss of correct position.
1458
F.F. Fernandez et al.
Table 3 Constant—Murley shoulder score (n = 35) Pain (max. 15 points) Daily activities (max. 20 points) Range of motion (max. 40 points) Power (max. 25 points)
results and only one child was not happy with the outcome.
Discussion In accordance to the literature2,6,12,15 intramedullary nailing was characterised by a low complication rate in our hands. The most prominent complication was the perforation of the nail through the humeral head which was reported by other authors as well.2,6 The nails in our cases did penetrate laterally without need for revision. The two revisions in our series were due to a release of a haematoma and inadequate stabilisation of the fracture in one case using the one-nail-technique. Three children demonstrated a loss of position resulting in a varus position of the humeral head of 108. In two cases there was a perforation of the nail through the humeral head without functional deficits. Similar results were reported by Chee et al.2 who also describes two children with a loss of position after bony consolidation in varus by being stabilised in the one-nail-technique. In spite of these results he prefers the one-nail-technique because of the danger of iatrogenic growth plate damage, longer surgery time and higher costs. However he recommends a collar and cuff for 2 weeks for protection of the osteosynthesis. In this series there were also two losses of position in the one-nailtechnique and one similar result in the two-nailtechnique. Patients treated in the two-nail-technique were immobilised in a gilchrist bandage for 2—4 days for pain relief. The children treated by the one-nailtechnique were treated according to Chee’s regimen, which means sling for 2 weeks. In 34 cases the reduction was achieved by closed means. In one case the long biceps tendon needed to be removed from the fracture site. This is a rare complication which can only happen with a complete anteriorly dislocation of the humeral shaft.9 Closed reduction and improvement of the reduction by using the tip of the nail resulted in an anatomic reduction in 25 children. In 10 cases a postoperative malposition of a Neer I situation could be accepted because the corrective potential of the growth plate in that age group is big enough to lead to complete remodelling of the
33 35 30 31
patients: patients: patients: patients:
15 20 40 25
points; 2 patients: 10 points points points; 5 patients: 38 points points; 4 patients: 21 points
proximal humerus. According to the literature we like to emphasise that stable osteosynthesis can only be achieved with anchoring nails in the epiphysis. We did not see any premature closures of the growth plate and the arm length was equal in all children at follow-up. This is in contrast to Chee who reported a shortening of the humerus of 1.5 cm after perforation of the growth plate.2 All children could keep up their daily and sporting activities at a mean of 26 months from the injury. No child demonstrated any functional deficits. We could not detect a difference in the Constant— Murley score for the children treated with either one-nail or two-nail-technique. To summarise the unilateral radial retrograde intramedullary stabilisation of proximal humeral fractures with proximal epiphyseal anchoring of the nail has the following advantages: Early functional rehabilitation in the two-nailtechnique. Excellent functional and cosmetic results. Minimal invasive technique without release of the fracture haematoma and iatrogenic compromise to the soft tissue. Safe and reliable surgical technique. Low complications rate. Quick learning curve. The treatment of severely displaced subcapital humeral fractures in children older than 10 years should reflect the advantages and disadvantages of non-operative treatment and surgical treatment by intramedullary stabilisation. Because of many obvious advantages of the surgical approach we prefer this method.
Conflict of interest statement None.
References 1. Beringer DC, Weiner DS, Noble JS, Bell RH. Severely displaced proximal humeral epiphyseal fractures: a follow-up study. J Pediatr Orthop A 1998;18:31—7. 2. Chee Y, Agorastides I, Garg N, et al. Treatment of severely displaced proximal humeral fractures in children with elastic
Proximal humeral fractures in children
3.
4.
5.
6. 7. 8. 9.
stable intramedullary nailing. J Pediatr Orthop B 2006;15: 45—50. Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clin Orthop 1987;214(January):160—4. Dameron TB, Reibel DB. Fractures involving the proximal humeral epiphyseal plate. J Bone Joint Surg Am 1969;51A:289—97. Hilton M, Yngve DA, Carmichael KD. Proximal humerus fractures sustained durino the use of restraints in adolescents. J Pediatr Orthop 2006;26(1):50—2. Hohl JC. Fractures of the humerus in children. Othop Clin North Am 1976;7:557—71. Knorr P, Joeris A, Lieber J, et al. The use of ESIN in humerus fractures. Eur J Trauma 2005;1:12—8. Larsen FC, Kiaer T, Lindequist S. Fractures of the proximal humerus in children. Acta Orthop Scand 1990;61(3):255—7. Levine B, Pereira D, Rosen J. Avulsion fractures of the lesser tuberosity of the humerus in adolescents: review of the literature and case report. J Orthop Trauma 2005;19(5): 349—352.
1459 10. Lucas JC, Mehlmann CT, Laor T. The location of the biceps tendon in completely displaced proximal humerus fractures in children: a report of four cases with magnetic resonance imaging and cadaveric correlation. J Pediatr Orthop 2004;24:249—53. 11. Neer CS, Horwitz BS. Fractures of the proximal humeral epiphyseal plate. Orthopedics 1965;41:24—31. 12. Ross GJ, Love MB. Isolated avulsion fracture of the lesser tuberosity of the humerus: report of two cases. Radiology 1989;172:833—4. 13. Sarwark JF, King EC, Luhmann SJ. Proximal humerus, scapula, and clavicle. In: Rockwood and Wilkins’ fractures in children6th ed., Lippincott Williams and Wilkins; 2006 . 14. Schmittenbecher PP, Blum J, David ST, et al. Die Behandlung von Humerusschaftfrakturen und subcapitalen Humerusfrakturen im Kindesalter. Unfallchirurg 2004;107: 8—14. 15. Sessa S, Lascombes P, Prevot J, et al. Centromedullary nailing in fractures of the upper end of the humerus in children and adolescents. Chir Pediatr 1990;31:43—6.
www.elsevier.com/locate/injury
Treatment of severely displaced proximal humeral fractures in children with retrograde elastic stable intramedullary nailing ¨rfer, T. Wirth F.F. Fernandez *, O. Eberhardt, M. Langendo Department of Orthopaedics, Klinikum Stuttgart, Olgahospital, Bismarckstr. 8, D-70176 Stuttgart, Germany Accepted 3 April 2008
KEYWORDS Proximal humeral fractures; Children; Elastic stable intramedullary nailing (ESIN); Nancy nail
Summary We report our experience of elastic stable intramedullary nailing (ESIN) of severely displaced proximal humeral fracture Type III and IV according to Neer in children. Thirty-five children (mean age 12.7 years) with 19 Salter—Harris Type II epiphyseal fractures and 16 children with metaphyseal fractures of the proximal humerus were treated with ESIN by one-nail or two-nail-technique. There were the following complications: two perforations of the nail at the head of the humerus with loss of position, one loss of position without nail perforation, one misplacement of a nail, one revision due to haematoma and two difficult removals of metal. For followup examinations after 26 months the children were classified by use of the Constant— Murley score. On average they gained 99 points. All children were able to resume their sporting activities as before. For the treatment of displaced proximal humeral fractures in children more than 10 years old the retrograde ESIN represents a safe minimal invasive surgical procedure. # 2008 Elsevier Ltd. All rights reserved.
Introduction Proximal humeral fractures in children are rare and represent about 2% of all paediatric fractures. Eighty-five percent of these fractures are either undisplaced or only slightly displaced fractures and only 15% demonstrate severe displacement of * Corresponding author. Tel.: +49 711 992 3017; fax: +49 711 992 3820. E-mail address: [email protected] (F.F. Fernandez).
metaphyseal or growth plate involving fractures.1,7,10,14 Purely epiphyseal fractures and apophyseal lesions are only discussed in case reports8,11,12 and are fractures of the adolescents and adults. There is consensus for treating slightly displaced proximal humeral fractures by non-operative means. The treatment of severely displaced fractures of the upper end of the humerus however is still under debate. These fractures most commonly occur before the age of 3 years and in children older than 12 years.4,14
0020–1383/$ — see front matter # 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2008.04.001
1454 Non-operative treatment is still regarded as gold standard for treating subcapital humeral fractures in childhood. It is reported that independent of the therapeutic approach all fractures unite and function will be fully restored. Generally displaced or slightly displaced proximal humeral fractures in childhood and adolescence are perfectly well treated by a short immobilisation period of 2—4 weeks using a Gilchrist type bandage. In contrast treatment of severely displaced Grades III and IV fractures of children over 10 years is controversial. Remodelling potential of a Neer III and Neer IV displacement in children older than 10—11 years is very limited. Dameron reported that children aged 11 years and older can only partially remodel deformities of over 208.4 The fact that immobilisation in Dessault bandage, thorax abduction plaster or even traction through the olecranon do not meet the requirements of modern fracture treatment with early functional postoperative care have led to alternative treatment methods. Most proximal humeral fractures are extraarticular lesions. During childhood avulsions of the greater or lesser tuberosity are extremely rare and usually only occur in adolescence. The growth plate of the proximal humerus accounts for 80% of longitudinal growth of the upper arm and hence has an enormous potential for correction of residual axial deformities.14 Up to the age of 10 years an axial malalignment of as much as 608 of deformity in varus-, anti- or retroversion can be corrected.14 Beyond the age of 10 the potential of correction is much less with the possibility of correcting axial deformities of up to 20—308.7,14 The corrective potential is least for valgus positions beyond 108. For surgical treatment a variety of osteosynthetic material is used such as K-wires and screws.5,13 The most preferred method is by using percutaneous K-wires. This method however exhibits some important disadvantages such as unstable osteosynthesis and the need for postoperative immobilisation. Further most of the protruding K-wires irritate the muscles and soft tissues. There is a considerable risk of perforation of the head of the humerus. The introduction of the retrograde elastic stable intramedullary nailing (ESIN) for fixing proximal humeral fractures adds a surgical technique which respects the principles of minimal invasive ortheosynthesis which does not compromise the soft tissues or the joint and most favourably allows early mobilisation of the injured arm.15 The goal of this study is to evaluate the clinical results following retrograde elastic intramedullary stabilisation of proximal humeral fractures during childhood and adolescence.
F.F. Fernandez et al.
Surgical technique The patient is placed in a supine position and the whole arm including the shoulder joint is entirely prepped and draped. For reduction it is mandatory that the arm can be moved in all directions at the level of the shoulder joint. The upper arm including the shoulder joint must be accessible for intraoperative fluorography. We favour the unilateral technique with the radial approach to the distal humerus. The skin incision of 2 cm in length is just above the lateral epicondyle of the humerus. Posteriorly to the crista supraepicondylaris the intramedullary cavity is opened with an awl. The entrance to the bone is opened up enough by turning the awl obliquely in a 458 position without perforating the opposite cortex. The Nancy nail is prebent 108 in the proximal 1/5. The first nail is inserted and brought up to the fracture site. The closed reduction is performed by traction, abduction of the arm to 908 and external rotation. The nail is then passed over the fracture line into the proximal fragment. The tip of the nail must perforate the growth plate. By rotation of the nail the proximal fragment can be moved and reduction can be improved. Subsequently the second nail is inserted into the humerus using the same entrance point. The nails should diverge in the head of the humerus. If decision is made to use the ‘‘one-nailtechnique’’ it is important that this nail is bigger in size. So it is recommended to have the nail over the lateral cortex by 1.5—2 cm for easier hardware removal.
Patients and methods Over a time period of the last 12 years all children and adolescents with proximal humeral fracture treated by retrograde intramedullary nailing were included in the study. When intramedullary nailing was introduced in our institution initially, we began with the one-nail-technique. With emerging experience, the two-nail-technique was increasingly performed. The charts and the X-rays of all children were evaluated with regard to the cause of the accident, additional injuries, extent of dislocation according to Neer and Horwitz,11 hospital stay, complications, after treatment, surgical dates at the primary operation and metal removal. Further the postoperative treatment as well as the cosmetic and functional long-term results were looked at. All patients could be followed and evaluated by use of the Constant—Murley score.3
Proximal humeral fractures in children Table 1 Neer proximal humeral epiphyseal fracture displacement grading Grade Grade Grade Grade
I II III IV
<5 mm shaft diameter 1/3 shaft diameter 2/3 shaft diameter >2/3 shaft diameter
Results There were 35 patients (15 boys and 20 girls) with mean age of 12.7 [8—17] years. The age distribution was: 4 patients <10 years, 19 patients between 10 and 13 years and 12 patients >13 years. In 18 cases the left arm, in 17 cases the right arm was injured. All 35 patients demonstrated a fracture dislocation degree III and IV after Neer and Horwitz11 (Table 1). The accident causes were trivial falls in 17 cases, falls from a greater height 6 times, ski and snowboard injuries 5 times, fall from the bicycle in 4 cases and fall from the horse in 3 cases. There were 19 Salter—Harris Type II fractures and 16 metaphyseal fractures of the proximal humerus. All fractures were closed injuries without damage to vascular and neural structures. In 33 children the fracture was a single lesion with one child having an additional glenoid fracture and one with an additional fracture of the olecranon. Twenty-seven children were primarily treated in our institution. Seven children were
1455 Table 2
Complications (n = 7)
Implant perforations with loss of position, n Misplacement of a nail, n Revision due to haematoma, n Loss of correction, n Difficult removals of metals, n
2 1 1 1 2
treated non-operatively before in another hospital and operated on secondarily after referral to our institution. One child was treated by percutaneous K-wires elsewhere and we had to change the stabilisation to intramedullary nailing due to irritation of the soft tissue by the introduced K-wires. The mean primary hospital stay was 4.1 days.2— 8 For metal removal the mean hospital stay was 2.8 days. This quite long time can be explained with the fact that at the beginning, a drain was inserted for about 1 or 2 days. Nowadays, removal of metalwork is done as a day case procedure. The mean time of surgery was 54 min (24—85). All children were treated under general anaesthetics. In 15 children stabilisation was achieved with one nail (4 Neer III and 11 Neer IV) and in 20 children by the use of two elastic nails (11 Neer III and 9 Neer IV). The diameter of the nails used in the one-nailtechnique varied from 3.0 to 4.0 mm and in the double-nail-technique from 2.5 to 3.5 mm. One patient had to have an open reduction because of
Figure 1 (A) 13-year-old girl after traffic accident. Metaphyseal fracture with type Neer IV dislocation. (B) Intramedullary stabilisation in the one-nail-technique. (C) Twelve months later, radiological healing and complete remodelling.
1456
F.F. Fernandez et al.
Figure 2 (A) 11-year-old girl falls while playing. Epiphyseal fracture type Salter II with type Neer III dislocation. (B) Intramedullary stabilisation in the two-nail-technique. Five months later radiological healing and complete remodelling.
the incarceration of the biceps tendon into the fracture. Postreduction position was anatomical in 25 children and improvement to dislocation Grade I after Neer and Horwitz11 in 10 patients. There were the following complications (Table 2): there was a perforation of the nail through the humeral head in two patients. The tip of the nail left the humeral head laterally without need for revision. There were two patients who experienced a loss of position, one 13-year-old girl with a Salter II fracture treated in the one-nailtechnique and an 11-year-old boy with a metaphyseal fracture treated in the two-nail-technique. In a 14-year-old boy with a subcapital humeral fracture the nail was initially misplaced and needed to be changed. One child needed to be revised for release of postoperative haematoma at the side of pin insertion. In two children the metal removal was extremely difficult. There was one more patient treated in the one-nail-technique with a loss of position without perforation of the nail. This fracture healed in a varus position of 108. There were no iatrogenic lesions to the radial nerve, pseudarthrosis, infections or wound breakdowns (Figs. 1—3). The mean time for bony consolidation was 7.1 weeks. Removal of metal was usually performed 5.2 months after the fracture. The radiographs before metal removal that were made at an average of 4.5 months after the original injury demonstrated that
there was almost completely remodelling of all postoperatively detected malpositions.
Follow-up examination All 35 children were seen for follow-up. The mean follow-up time was 26 months (between 6 and 58 months) post-injury. Median follow-up time was 35 months. The children were evaluated with the Constant— Murley score3 which rates pain (maximum 15 points), daily activities (maximum 20 points), power (maximum 25 points) and range of motion (maximum 40 points). The maximum is 100 points with 91—100 points representing excellent results. All children achieved excellent results independent from age, sex, fracture type and extent of displacement. In the Constant—Murley score patients reached 99 points (Table 3). Almost all children could return to their sporting activities again. Two children however who had a fall from a horse did not return to horse riding. There were no arm length discrepancies. Two children complained of some aching in the area of the injured shoulder joint. From the cosmetic aspect all patients were satisfied with the result with scars of a mean length of 2.8 cm. In 31 cases parents and children were very happy with the treatment results, in 3 cases they were satisfied with the
Proximal humeral fractures in children
1457
Figure 3 (A) 13-year-old girl with epiphyseal fracture type Salter II with type Neer IV dislocation. (B) Intramedullary stabilisation in the one-nail-technique. (C) Four months later radiological healing and implant perforation with loss of correct position.
1458
F.F. Fernandez et al.
Table 3 Constant—Murley shoulder score (n = 35) Pain (max. 15 points) Daily activities (max. 20 points) Range of motion (max. 40 points) Power (max. 25 points)
results and only one child was not happy with the outcome.
Discussion In accordance to the literature2,6,12,15 intramedullary nailing was characterised by a low complication rate in our hands. The most prominent complication was the perforation of the nail through the humeral head which was reported by other authors as well.2,6 The nails in our cases did penetrate laterally without need for revision. The two revisions in our series were due to a release of a haematoma and inadequate stabilisation of the fracture in one case using the one-nail-technique. Three children demonstrated a loss of position resulting in a varus position of the humeral head of 108. In two cases there was a perforation of the nail through the humeral head without functional deficits. Similar results were reported by Chee et al.2 who also describes two children with a loss of position after bony consolidation in varus by being stabilised in the one-nail-technique. In spite of these results he prefers the one-nail-technique because of the danger of iatrogenic growth plate damage, longer surgery time and higher costs. However he recommends a collar and cuff for 2 weeks for protection of the osteosynthesis. In this series there were also two losses of position in the one-nailtechnique and one similar result in the two-nailtechnique. Patients treated in the two-nail-technique were immobilised in a gilchrist bandage for 2—4 days for pain relief. The children treated by the one-nailtechnique were treated according to Chee’s regimen, which means sling for 2 weeks. In 34 cases the reduction was achieved by closed means. In one case the long biceps tendon needed to be removed from the fracture site. This is a rare complication which can only happen with a complete anteriorly dislocation of the humeral shaft.9 Closed reduction and improvement of the reduction by using the tip of the nail resulted in an anatomic reduction in 25 children. In 10 cases a postoperative malposition of a Neer I situation could be accepted because the corrective potential of the growth plate in that age group is big enough to lead to complete remodelling of the
33 35 30 31
patients: patients: patients: patients:
15 20 40 25
points; 2 patients: 10 points points points; 5 patients: 38 points points; 4 patients: 21 points
proximal humerus. According to the literature we like to emphasise that stable osteosynthesis can only be achieved with anchoring nails in the epiphysis. We did not see any premature closures of the growth plate and the arm length was equal in all children at follow-up. This is in contrast to Chee who reported a shortening of the humerus of 1.5 cm after perforation of the growth plate.2 All children could keep up their daily and sporting activities at a mean of 26 months from the injury. No child demonstrated any functional deficits. We could not detect a difference in the Constant— Murley score for the children treated with either one-nail or two-nail-technique. To summarise the unilateral radial retrograde intramedullary stabilisation of proximal humeral fractures with proximal epiphyseal anchoring of the nail has the following advantages: Early functional rehabilitation in the two-nailtechnique. Excellent functional and cosmetic results. Minimal invasive technique without release of the fracture haematoma and iatrogenic compromise to the soft tissue. Safe and reliable surgical technique. Low complications rate. Quick learning curve. The treatment of severely displaced subcapital humeral fractures in children older than 10 years should reflect the advantages and disadvantages of non-operative treatment and surgical treatment by intramedullary stabilisation. Because of many obvious advantages of the surgical approach we prefer this method.
Conflict of interest statement None.
References 1. Beringer DC, Weiner DS, Noble JS, Bell RH. Severely displaced proximal humeral epiphyseal fractures: a follow-up study. J Pediatr Orthop A 1998;18:31—7. 2. Chee Y, Agorastides I, Garg N, et al. Treatment of severely displaced proximal humeral fractures in children with elastic
Proximal humeral fractures in children
3.
4.
5.
6. 7. 8. 9.
stable intramedullary nailing. J Pediatr Orthop B 2006;15: 45—50. Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clin Orthop 1987;214(January):160—4. Dameron TB, Reibel DB. Fractures involving the proximal humeral epiphyseal plate. J Bone Joint Surg Am 1969;51A:289—97. Hilton M, Yngve DA, Carmichael KD. Proximal humerus fractures sustained durino the use of restraints in adolescents. J Pediatr Orthop 2006;26(1):50—2. Hohl JC. Fractures of the humerus in children. Othop Clin North Am 1976;7:557—71. Knorr P, Joeris A, Lieber J, et al. The use of ESIN in humerus fractures. Eur J Trauma 2005;1:12—8. Larsen FC, Kiaer T, Lindequist S. Fractures of the proximal humerus in children. Acta Orthop Scand 1990;61(3):255—7. Levine B, Pereira D, Rosen J. Avulsion fractures of the lesser tuberosity of the humerus in adolescents: review of the literature and case report. J Orthop Trauma 2005;19(5): 349—352.
1459 10. Lucas JC, Mehlmann CT, Laor T. The location of the biceps tendon in completely displaced proximal humerus fractures in children: a report of four cases with magnetic resonance imaging and cadaveric correlation. J Pediatr Orthop 2004;24:249—53. 11. Neer CS, Horwitz BS. Fractures of the proximal humeral epiphyseal plate. Orthopedics 1965;41:24—31. 12. Ross GJ, Love MB. Isolated avulsion fracture of the lesser tuberosity of the humerus: report of two cases. Radiology 1989;172:833—4. 13. Sarwark JF, King EC, Luhmann SJ. Proximal humerus, scapula, and clavicle. In: Rockwood and Wilkins’ fractures in children6th ed., Lippincott Williams and Wilkins; 2006 . 14. Schmittenbecher PP, Blum J, David ST, et al. Die Behandlung von Humerusschaftfrakturen und subcapitalen Humerusfrakturen im Kindesalter. Unfallchirurg 2004;107: 8—14. 15. Sessa S, Lascombes P, Prevot J, et al. Centromedullary nailing in fractures of the upper end of the humerus in children and adolescents. Chir Pediatr 1990;31:43—6.