Elevated, straight-arm traction for supracondylar fractures of the humerus in children

Transcription

1 Children s orthopaedics Elevated, straight-arm traction for supracondylar fractures of the humerus in children A. Gadgil, C. Hayhurst, N. Maffulli, J. S. M. Dwyer From University Hospital of North Staffordshire and Keele University, Stoke on Trent, England Between January 1995 and December 2000, 112 children with a closed displaced supracondylar fracture of the humerus without vascular deficit, were managed by elevated, straight-arm traction for a mean of 22 days. The final outcome was assessed using clinical (flexion-extension arc, carrying angle and residual rotational deformity) and radiographic (metaphyseal-diaphyseal angle and humerocapitellar angle) criteria. Excellent results were achieved in 71 (63%) patients, 33 (29%) had good results, 5 (4.4%) fair, and 3 (2.6%) poor. All patients with fair or poor outcomes were older than ten years of age. Elevated, straight-arm traction is safe and effective in children younger than ten years. It can be effectively used in an environment that can provide ordinary paediatric medical care and general orthopaedic expertise. The outcomes compare with supracondylar fractures treated surgically in specialist centres. A. Gadgil, FRCS, Specialist Registrar Orthopaedics, Department of Orthopaedics, University Hospital of Wales, Heath Park, Cardiff, UK. C. Hayhurst, MRCS, SHO Orthopaedics N. Maffulli, FRCS(Orth), Professor of Orthopaedics J. S. M. Dwyer, FRCS(Orth), Consultant Paediatric Orthopaedics University Hospital of North Staffordshire and Keele University, Stoke on Trent ST4 7QB, UK. Correspondence should be sent to Mr J. S. M. Dwyer British Editorial Society of Bone and Joint Surgery doi: / x.87b $2.00 J Bone Joint Surg [Br] 2005;87-B:82-7. Received 30 April 2003; Accepted after revision 10 November 2003 No single method of management is suitable for all supracondylar fractures of the humerus in children. 1 A patient with a supracondylar fracture and a cold, pale, pulseless hand calls for urgent surgical attention to re-establish circulation and stabilise the fracture. Closed, displaced supracondylar humeral fractures in children without associated neurovascular injury can be successfully managed either surgically or conservatively. Surgical options include closed or open reduction and stabilisation with Kirschner wires (K-wires). Complications including redisplacement of the fracture, 2,3 cubitus varus, 2,4,5 iatrogenic nerve injuries, 6-8 and pin-track infection 8,9 have been reported following surgery. Strict adherence to protocols and surgical expertise are necessary for a successful outcome with K-wires. 4,10 Traction has been used in the management of supracondylar fractures since 1939 when Dunlop 11 described straight-arm traction with the elbow extended. Other authors applied the traction with the elbow slightly flexed. 12 Skeletal traction with a K-wire 13 or a screw anchor 14 have also been reported. Management of the fracture by straight-arm traction seems to have become less popular, even to have fallen into disrepute. Although Prietto 15 reported a high incidence of cubitus varus (33%) with Dunlop traction, other studies using straight-arm traction have reported excellent results. 1,16,17 We have managed children with closed, displaced, isolated supracondylar humeral fractures, excluding those presenting with vascular compromise, using elevated, straight-arm traction with the elbow in extension since We present our method and results with this type of management. Patients and Methods We treated 196 supracondylar fractures of the distal humerus in 195 children between January 1995 and December Of 179 patients treated as inpatients, 141 were boys and 38 girls with a mean age of 6 years 2 months (2 to 11 years). One of the patients had a re-fracture one year after his original injury, which had been treated by straight-arm traction. On the second occasion the fracture was undisplaced and was treated in a collar and cuff. The following groups were excluded from this study: 16 patients with undisplaced fractures who were treated on an outpatient basis with a collar cuff and sling only; 39 patients with Wilkins grade IIa injuries, who were managed by closed reduction under general anaesthesia and immobilisation with a plaster back slab, taping or collar and cuff for three weeks; and 15 patients who underwent accurate closed or open reduction using K-wires. Variously or in combination, the indications for closed or open reduction using K-wires were: an open fracture (11 patients), a supracondylar 82 THE JOURNAL OF BONE AND JOINT SURGERY

2 ELEVATED, STRAIGHT-ARM TRACTION FOR SUPRACONDYLAR FRACTURES OF THE HUMERUS IN CHILDREN 83 Table I. Relationship between neurovascular deficit and type of displacement Deficit Posterior displacement Posterolateral displacement Posteromedial displacement Absent radial pulse Anterior interosseous nerve palsy Median nerve palsy Radial nerve palsy Ulnar nerve palsy fracture associated with an ipsilateral fracture of the radius and ulna (one patient), a vascular injury (four patients) and a supracondylar fracture in a patient with multiple skeletal injuries. An additional 14 patients who were managed by elevated, straight-arm traction were excluded from the study either because they were lost to follow-up or because the notes and radiographs did not allow accurate determination of the outcome. In all, 112 fractures in 112 patients were included in the study. Of these, 107 were extension fractures (Wilkins modification of Gartland s classification 18 ) types IIb & III, and five were flexion fractures. Eight patients were older than ten years of age. In five patients, the radial pulse was not palpable on admission but the hand was still warm and pink with a capillary return of two seconds. They were successfully managed conservatively in straight-arm traction. At presentation, two patients had a radial nerve palsy, six had an anterior interosseous nerve palsy, seven had a median nerve palsy and two had an ulnar nerve palsy. The five flexion-type fractures all had anterolateral displacement of the distal fragment, but none was associated with neurovascular compromise. Of the 107 extension-type fractures, the distal fragment was displaced posteriorly in 30, posterolaterally in 30, and posteromedially in 47 patients. Posterolateral displacement of the distal fragment was most frequently associated with marked swelling and neurovascular deficit (Table I). Of the 112 patients treated by elevated straight-arm traction, 48 had a manipulation under general anaesthesia but, as a perfect reduction was not achieved due to the swelling or comminution, they were treated by straight-arm traction. As the study progressed, it became apparent that manipulation under general anaesthesia was not necessary except in patients with extreme distress, deformity, or at risk of skin necrosis over the anterior aspect of the proximal fragment. The remaining 64 patients were managed by straight-arm traction from the outset without being taken to the operating theatre. The mean duration of traction was 22 days (14 to 28) and all children tolerated it well. Since 1997, our standard management has been undertaken in the casualty department and the ward. After providing adequate analgesia and making a complete assessment of the patient, including the neurovascular status of the injured limb, the child is admitted to the ward, and below-elbow, straight-arm skin traction is applied (Fig. 1). Application of traction. Skin traction is applied with the elbow extended and forearm in supination in the straight Fig. 1 A patient in straight-arm traction. Note that the arm is suspended from the bed surface by traction. The bed is elevated on the ipsilateral side in order to provide counter traction. lateral position. Adhesive tapes are applied to the forearm and held with an elasticated bandage. The shoulder is kept at 90 abduction. The cord from the skin traction passes over a universal joint pulley and enough weight is attached to the end of the cord (0.45 to 1.8 kg) so that the arm is just lifted from the surface of the bed. The traction side of the bed is elevated with blocks under the legs of the bed to supply counter-traction. A single dose of weight-adjusted, orally administered, opiate provides the necessary analgesia for the application of traction. Once instituted, the traction itself provides the arm with sufficient support so that further opiate administration is rarely required. The neurovascular status is monitored for the first four days with frequent inspection by the ward staff and doctors to ensure that the arm remains elevated from the bed. This is because the elbow will flex if the arm comes to rest on the bed. Abduction of the shoulder needs to be adjusted by sliding the pulley so that the carrying angle of the injured limb is the same as the uninjured arm. When setting up the traction it is helpful, but not crucial, to consider the likely pattern of displacement of the distal fragment. For fractures with posteromedial displacement, when the distal fragment is often rotated internally or in varus angulation, traction with the forearm supinated and shoulder in wide abduction gradually reduces the fracture. The position is easily maintained by placing the pulley on VOL. 87-B, No. 1, JANUARY 2005

3 84 A. GADGIL, C. HAYHURST, N. MAFFULLI, J. S. M. DWYER Fig. 2 Serial anteroposterior (AP) and lateral radiographs of a Gartland IIIb fracture treated in straight-arm traction showing progressive remodelling and an excellent final result: a) and b) AP and lateral radiographs on the day of injury; c) and d) after one week in traction; e) and f) after four weeks in traction; g) and h) at three months follow-up; i) and j) at nine months follow-up. the injured side, towards the head of the bed. Correction for fractures with posterolateral displacement, when the distal fragment is often externally rotated or in valgus, is achieved with traction positioned so that the shoulder is in abduction of 90 or less. The forearm can be allowed to rest in the midprone position although in the supinated position the carrying angle is more readily assessed. The pulley for these fractures is positioned midway between the foot and the head ends of the bed. Rather than adopting rigid rules, however, it is better to adjust the degree of abduction of the arm in traction early in treatment as dictated by the direction of displacement of the distal fragment in the coronal plane according to the apparent carrying angle. Anteroposterior (AP) and lateral radiographs are obtained in bed each week with the arm in traction. The traction is discontinued when the patient is able to actively flex the elbow in traction. Follow-up. The children were reviewed at one week, three months, nine months and at two years after discharge from hospital, with AP and lateral radiographs of the injured elbow at each visit (Fig. 2). If the remodelling was not complete or if a full range of movement was not achieved by 24 months, a further follow-up was arranged at yearly intervals. Patients were discharged sooner if and when a full range of movement, function and remodelling were achieved. The mean duration of follow-up was 24 months (9 to 48). Outcome assessment. We used three clinical and two radiographic criteria to assess the outcome (Table II). Each was graded as excellent, good, fair or poor and the outcome chosen was the lowest grade that the child achieved out of all of the above criteria. Clinical assessment. We used Flynn, Matthews and Benoit s 19 criteria to assess the function (flexion-extension) and the carrying angle at the last follow-up appointment. In addition, any residual rotational deformity was measured and recorded. The senior author (JD) has used the 90 to 90 method (Fig. 3) to measure the residual rotational deformity in the injured elbow with both glenohumeral and elbow joints flexed to 90. It is assumed that after treatment both shoulders have a similar range of movement and that before injury the upper limbs were symmetrical in skeletal terms. The angles subtended by the forearm axis with the horizontal in maximum internal and external rotation of both the sides, are compared. If the supracondylar fracture has malunited with internal rotation of the distal fragment, there is an apparent and measurable increase in the internal rotation of the shoulder on the affected side. For malunion in external rotation, the opposite applies. Radiographic assessment. The metaphyseal-diaphyseal (MD) angle 20 was measured on the AP radiograph and the humerocapitellar (HC) angle 7,21-23 was measured on the lateral radiograph (Fig. 4). THE JOURNAL OF BONE AND JOINT SURGERY

4 ELEVATED, STRAIGHT-ARM TRACTION FOR SUPRACONDYLAR FRACTURES OF THE HUMERUS IN CHILDREN 85 Table II. Outcome assessment in supracondylar fractures managed by straight-arm traction Result (number of patients) Criteria Excellent ( ) Good ( ) Fair ( ) Poor ( ) Clinical Loss of flexion-extension compared with normal elbow 19 0 to 5 (92) 6 to 10 (12) 11 to 15 (5) > 15 (3) Change in carrying angle compared with normal side 19 0 to 5 (90) 6 to 10 (21) 11 to 15 (1) > 15 (0) Clinically measured rotational deformity 0 to 5 (107) 6 to 10 (3) 11 to 15 (2) > 15 (0) Radiographical* Change in MD angle 20 (normal = 90 ) 0 (77) 1 to 5 (33) 6 to 10 (2) > 10 (0) Change in lateral HC angle 22 (normal = 20 ) 0 to 5 (71) 6 to 10 (33) 11 to 15 (5) > 15 (3) * MD, metaphyseal-diaphyseal; HC, humerocapitellar Fig. 3 Figure 3 The 90 to 90 method of measuring rotational deformity at the elbow. Figure 3a Shoulder is flexed at 90 ; elbow is flexed at 90. In neutral rotation the axis of the vertical forearm makes a 90 angle with the horizontal. Figure 3b External rotation, measured as 12 in this normal arm. If the supracondylar fracture has united in external rotation, an apparent increase in external rotation can be measured by comparison with the normal arm. Figure 3c Internal rotation is measured to be 51 in this normal arm. If the supracondylar fracture has united in internal rotation, an increase will be apparent. Results Of the 112 patients studied, 71 (63%) had an excellent, 33 (29%) a good, five (4.4%) a fair, and three (2.6%) a poor outcome. A total of 92 patients regained full flexion and extension at a mean of 12 months (9 to 24) after the injury; 20 patients had a mean restriction of flexion of 10 (6 to 20), and 11 developed hyperextension of the elbow compared with the normal side of 5 (3 to 9). The lateral HC angle was less than 10 in eight patients all of whom had more than 10 of restriction of elbow flexion (normal, 20 to 25 ); 90 children had a carrying angle which was comparable with the normal side at the end of the period in traction. An increase in cubitus valgus was noticed in 22 children, in comparison with the normal side of 7 (6 to 10). No patient developed a cubitus varus deformity. The MD angle was normal (90 ) in 77 children but in 34 it was reduced to a mean of 86 (80 to 87). A rotational deformity was observed in five children of whom four had a mild external rotational (mean 6 ) and one an internal rotational deformity of 7. All were cosmetically and functionally acceptable to the children and their parents. There were no neurovascular or infective complications as a consequence of treatment in straight-arm traction. Children younger than six years of age (59) required fewer days in traction (mean of 18 days (15 to 21)). They achieved active flexion in traction earlier than the older patients (53) with a mean of 26 days (22 to 29). Discussion The goals of management of a displaced supracondylar fracture are to recover full function in a cosmetically normal elbow. The application of longitudinal traction relieves muscle spasm and realigns the fracture. Within two to four weeks, sufficient callus is formed so that the length of the limb will be maintained without traction. Straight-arm traction corrects most cases of lateral or medial translocation, external or internal rotation and posterior angulation of the distal fragment, but is probably less effective at correcting posterior translocation. Purely posterior angulation 24 and posterior translocation 25 remodel most satisfac- VOL. 87-B, No. 1, JANUARY 2005

5 86 A. GADGIL, C. HAYHURST, N. MAFFULLI, J. S. M. DWYER Age (yrs) Number of patients Excellent Good Fair & poor Fig. 5 Scatter plot showing the various outcomes and the age of each child. All patients with fair or poor results were older than ten years. Most children under the age of six years had an excellent result. Fig. 4 Figures 4a and b AP and lateral radiographs showing a type IIb supracondylar fracture on admission. Figure 4c AP radiograph at final followup after straight arm traction. Technique to measure the metaphyseal-diaphyseal angle: 20 on the AP radiograph, a transverse line is drawn through the widest point of the metaphysis, a longitudinal line is drawn through the axis of the diaphysis, and the angle subtended by the lateral portion of metaphyseal line and the proximal portion of the diaphyseal line is measured. The normal value of the metaphyseal-diaphyseal angle is 90. An angle of more than 90 indicates varus. Figure 4d Lateral radiograph at final follow-up. Technique of measuring the humerocapitellar angle: the angle subtended between a line drawn along the anterior cortex of the humeral shaft and the axis of the epiphysis of the capitellum is measured. 22 torily (Figs 2 and 4). Fractures with posteromedial displacement are at a risk of malunion, with the distal fragment rotated internally or with a varus angulation. 26 This is prevented by straight-arm traction with the forearm supinated and the shoulder in wide abduction. Similarly, fractures with posterolateral displacement are at a risk of malunion with the distal fragment externally rotated and at an excessive carrying angle. 26 For these fractures, the shoulder is maintained in traction at less than 90 of abduction. Flynn s grading system 19 is based on the clinical assessment of a lost arc of flexion and extension and a change in the carrying angle from the normal. The clinical data in this study were collected prospectively by the senior author (JD) who personally reviewed every patient. In addition to recording the range of movement and residual deformity, we included Flynn s grading system in our assessment (Table II) of prospectively collected clinical data. We also established radiographic criteria for a retrospective review of films. For this reason, the MD angle rather than Baumann s angle was chosen. Although Baumann s angle is said to correlate with the final carrying angle, its normal values vary between 64 and 81, 27 making it necessary to compare the angle on the affected side with that on the unaffected side. Moreover, the inaccuracy of its measurement increases in young children and adolescents. 28 Slight variation in orientation of the X-ray beam invalidates the measurement. 16 On the lateral radiographs, reduction of the HC angle seems to be associated with loss of flexion of the elbow. However, statistical significance could not be tested as only eight patients showed reduction of flexion at final followup. In the present study, 92 of 112 patients regained full flexion and extension at a mean of 12 months after injury. Many of these patients may well have recovered full flexion earlier than this but we made no formal review between nine and 24 months. In our series, younger patients fared better following conservative management (Fig. 5). Moreover, the younger the patient the fewer the days required in traction. The excellent recovery of flexion following straight-arm traction depends on the remodelling capacity of the bone. All eight patients older than ten years of age were treated with straight-arm traction and had some terminal restriction of flexion (10 to 20 ). This observation suggests that displaced supracondylar fractures in children more than ten years of age might require surgery if the results of such intervention were shown to be better. All the fracture patterns encountered in the present study would have been amenable to accurate closed or open reduction with K- wires. The straight-arm traction method is not without its drawbacks. Some weeks of inpatient management of children can impose a burden on their parents. Some authors THE JOURNAL OF BONE AND JOINT SURGERY

6 ELEVATED, STRAIGHT-ARM TRACTION FOR SUPRACONDYLAR FRACTURES OF THE HUMERUS IN CHILDREN 87 Table III. Comparison of outcomes of this and other studies Author/s Method of treatment Outcome measures used Excellent Davies et al 2 K-wire Flynn s criteria Mazda et al 10 K-wire Flynn s modified criteria Reitmann et al 30 K-wire Flynn s criteria Our study Skin traction Flynn s criteria + rotational deformity radiographic Piggot et al 1 Skin traction Flynn s criteria Jefferiss 17 Skin traction Flynn s criteria (calculated by us from provided data) Dodge 16 Skin traction Mitchell & Adams criteria Good Fair Poor Number of patients argue that traction unduly prolongs hospital stay 29 and increases costs. 15 Closed or open reduction of displaced supracondylar fractures, followed by K-wire fixation, is an excellent method of management in experienced hands with strict adherence to protocols. 4,10 However, O Hara et al 4 have reported a 32% rate of cubitus varus deformity after this form of treatment when guidelines were not strictly followed. Moreover, not all hospitals have access to paediatric anaesthetists. Increasingly, direct involvement of such superspecialists is being recommended in the belief that it will reduce clinical risk. If one was obliged to treat a displaced closed supracondylar fracture without neurovascular deficit surgically for any reason, the surgery can be safely postponed to be undertaken in the daylight hours. 31 The outcomes of management with straight-arm traction in this study were comparable with those of recent large studies reporting the results of closed or open reduction and K- wire fixation (Table III). No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Piggot J, Graham HK, McCoy GF. Supracondylar fractures of the humerus in children: treatment by straight lateral traction. J Bone Joint Surg [Br] 1986;68-B: Davies RT, Gorczyca JT, Pugh K. Supracondylar humerus fractures in children: comparison of operative treatment methods. Clin Orthop 2000;376: Shim JS, Lee YS. Treatment of completely displaced supracondylar fractures of the humerus in children by cross fixation with three K-wires. J Pediatr Orthop 2002;22: O Hara LJ, Barlow JW, Clarke NM. Displaced supracondylar fractures of the humerus in children: audit changes practice. J Bone Joint Surg [Br] 2000;82-B: Reynolds RA, Mirzayam R. A technique to determine proper pin placement of crossed pins in supracondylar fractures of the elbow. J Pediatr Orthop 2000;20: Brown IC, Zinar DM. Traumatic and iatrogenic neurological complications after supracondylar humerus fractures in children. J Pediatr Orthop 1995;15: Skaggs DL, Hale JM, Bassett J, et al. Operative treatment of supracondylar fractures of the humerus in children: the consequences of pin placement. J Bone Joint Surg [Am] 2001;83-A: Srivastava S. The results of open reduction and pin fixation in displaced supracondylar fractures of the humerus in children. Med J Malaysia 2000;55(Suppl): Cheng JC, Lam TP, Shen WY. Closed reduction and percutaneous pinning for type III displaced supracondylar fractures of the humerus in children. J Orthop Trauma 1995;9: Mazda K, Boggione C, Fitoussi F, Pennecot GF. Systematic pinning of displaced extension-type supracondylar fractures of the humerus in children: a prospective study of 116 consecutive patients. J Bone Joint Surg [Br] 2001;83-B: Dunlop J. Transcondylar fractures of the humerus in childhood. J Bone Joint Surg 1939;21: Allen PD, Gramse AE. Transcondylar fractures of the humerus treated by Dunlop traction: report of 21 cases. Am J Surg 1945;67: Smith FM. Kirschner wire traction in elbow and upper arm injuries. Am J Surg 1947; 74: Palmer EE, Niemann KMW, Vesely D, Armstrong JH. Supracondylar fracture of the humerus in children. J Bone Joint Surg [Am] 1978;60-A: Prietto CA. Supracondylar fractures of the humerus: a comparative study of Dunlop s traction versus percutaneous pinning. J Bone Joint Surg [Am] 1979;61-A: Dodge HS. Displaced supracondylar fractures of the humerus in children: treatment by Dunlop s traction. J Bone Joint Surg [Am] 1972;54-A: Jefferiss CD. Straight lateral traction in selected supracondylar fractures of the humerus in children. Injury 1976;8: Wilkins KE. Supracondylar fractures of the humerus. In: Operative management of upper extremity fractures in children. Rosemont, Illinois: American Academy of Orthopaedic Surgeons, Flynn JC, Matthews JG, Benoit RL. Blind pinning of displaced supracondylar fractures of the humerus in children: sixteen years experience with long term follow-up. J Bone Joint Surg [Am] 1974;56-A: Canale ST. Fractures and dislocations in children. In: Crenshaw AH, ed. Campbell s operative orthopaedics. Vol 2, Eighth ed. St Louis etc: Mosby Year Book Inc, 1992: Devito DP. Management of fractures and their complications. In: Morrisy RT, Weinstein SL, eds. Lovell and Winter s pediatric orthopaedics. Fourth ed. Philadelphia: Lippincott-Raven, 1996: Watson-Jones R. Fractures and joint injuries. Vol 2, Sixth ed. Edinburgh: Churchill Livingstone, 1982: Holda ME, Manoli A, Lamont RL. Epiphyseal separation of the distal end of the humerus with medial displacement. J Bone Joint Surg [Am] 1980;62-A: Graham HA. Supracondylar fractures of the elbow in children. Clin Orthop 1967;54: Mann TS. Prognosis in supracondylar fractures. J Bone Joint Surg [Br] 1963;45-B: Arnold JA, Nasca RJ, Nelson CL. Supracondylar fractures of the humerus: the role of dynamic factors in prevention of deformity. J Bone Joint Surg [Am] 1977;59-A: Williamson DM, Coates CJ, Miller RK, Cole WG. Normal characteristics of the Baumann angle: an aid in assessment of supracondylar fractures. J Pediatr Orthop 1992;12: Webb AJ, Sherman FC. Supracondylar fractures of the humerus in children. J Pediatr Orthop 1989;9: Iyengar SR, Hoffinger SA, Townsend DR. Early versus delayed reduction and pinning of type III displaced supracondylar fractures of the humerus in children: a comparative study. J Orthop Trauma 1999;13: Reitmman RD, Waters P, Millis M. Open reduction and internal fixation for supracondylar humerus fractures in children. J Pediatr Orthop 2001;21: Leet AI, Frisancho J, Ebramzadeh E. Delayed treatment of type 3 supracondylar humerus fractures in children. J Pediatr Orthop 2002;22: VOL. 87-B, No. 1, JANUARY 2005