Fractures of the Lateral Condyle of the Humerus M ilch described two types of lat- eral condyle fractures (Figure 1). In Milch type I, the fracture ex- tends through the ossification center of the capitellum and enters the joint lateral to the trochlear groove. In Milch type II, the fracture extends medially into the trochlear groove. The most widely used system (not identified by name) identifies three fracture patterns (Figure 2). In a type I fracture, the articular surface is in- tact and the fracture is nondisplaced and stable. In types II and III, the frac- ture enters the joint. Type II fractures are minimally displaced (2 to 3 mm); type III fractures are displaced >4 mm and may be rotated. (For additional discussion of these systems, see Milch, 1 Jakob et al, 2 Ogden, 3 Herring, 4 Wilkins et al, 5 and McIntyre. 6 ) Indications Type I fractures and type II fractures displaced <2 mm may be treated by closed means. 7 Closed reduction and percutaneous pinning should be at- tempted in type II fractures displaced 2to3mm; 8 however, if anatomic re- duction is not obtained, open reduc- tion and internal fixation is required. Type II fractures displaced >2 to 3 mm and all type III fractures are unstable. Displaced fractures have an increased propensity to nonunion. Properly managed fractures treat- ed by closed manipulation and per- cutaneous pinning or by open reduc- tion and internal fixation have a 95% union rate, making these the preferred methods of treatment. Contraindications There are very few contraindications to performing percutaneous or open reduction and internal fixation in the properly selected patient. The nondisplaced, stable fracture does not require surgical treatment; cast immobilization is sufficient. When an underlying medical condition prevents surgery or an anesthetic risk, then either nonsurgical treat- ment is required or the medical con- dition must be managed before un- dertaking a surgical procedure. Surgical Technique The patient is positioned supine on the operating table and a general an- esthetic is induced. A small child should be positioned with the arm and forear m lying on the operating table but close enough to the edge that the operative limb can be brought over the edge of the table for use with either standard fluoroscopy or the mini C-arm fluoroscopy unit. Some surgeons use the receiving unit of the fluoroscopy unit as an op- erating surface. The arm is prepared and draped in a sterile manner, then is exsan- guinated and the tourniquet inflated. When closed reduction and percuta- neous pinning is considered, we per- form this technique with fluoro- scopic imaging ( video). If this fails or if open treatment is the method of choice, a curvilinear later- al incision is made centered over the lateral condyle. Minimal dissection is preferred to avoid periosteal strip- ping of the blood supply. The surgi- cal approach involves directly enter- ing the fracture hematoma and visualizing the fragment. I recom- mend the surgical interval between the brachioradialis and the triceps. Once the fragment is identified, the fracture site is ir rigated thoroughly J. Andy Sullivan, MD The video that accom- panies this article is ″Supracondylar Fractures of the Humerus in Children,″ available on the Orthopaedic Knowledge On- line Website, at http://www5.aaos.org/ oko/jaaos/surgical.cfm Dr. Sullivan is Don H. O’Donoghue Professor and Chief Medical Officer, Department of Orthopedic Surgery & Rehabilitation, University of Oklahoma Health Sciences Center, Children’s Hospital, Oklahoma City, OK. Neither Dr. Sullivan nor the department with which he is affiliated has received anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article. Reprint requests: Dr. Sullivan, Children’s Hospital, Room 2MR2000, 940 NE 13th Street, Oklahoma City, OK 73104. J Am Acad Orthop Surg 2006;14: 58-62 Copyright 2006 by the American Academy of Orthopaedic Surgeons. Surgical Techniques 58 Journal of the American Academy of Orthopaedic Surgeons to remove all hematoma and to im- prove visualization. Dissection is kept to a minimum. Usually no dissection is necessary on the distal fragment. Distal and poste- rior dissection should be avoided in order to avert damage to the circula- tion of the fragment, which can cause osteonecrosis. The periosteum of the proximal fragment, which overhangs the fracture site, may have to be stripped back slightly to remove it from the fracture site. It is important to adequately visu- alize the joint articular surface. In some cases, the fragment is rotated 180°; in these situations, I have found it easier to visualize the fragment by applying a varus movement to the el- bow, which reproduces the mecha- nism of injury and opens the fracture site so that it may be easily seen. Once the hematoma has been evac- uated, the distal fragment is manip- ulated into position onto the end of the proximal fragment (distal hu- merus). To accomplish this reduction, the distal fragment is grasped with a bone-holding forceps or a towel clip and rotated back into proper align- ment. This maneuver is facilitated by flexing the elbow in order t o take ten- sion off the distal fragment. My preferred technique is to place a large towel clip or a small-bone point-to-point forceps with one tong in the lateral condyle, then hook the other tong into the periosteum of the proximal fragment (Figure 3). The el- bow is flexed and the clamp gently closed. It is important at this point to place retraction in the wound to directly visualize the joint articular surface. The key component of this procedure is to ensure anatomic re- duction of the joint articular surface. At times, the joint surface cannot be entirely visualized. By palpation an- teriorly and posteriorly, one is usual- ly able to get a good idea of the suit- ability of the reduction. After anatomic reduction is achieved, the arm is brought out over the fluoroscopy machine, and anteroposterior and lateral radio- graphic views are obtained. In my experience with displaced fractures, when anatomic reduction has not been obtained, the most common position is lateral displacement. Usually reduction can be improved by loosening the clamp, placing pres- sure on the fragment to push it more medially, then closing the clamp. Once the anatomic reduction is obtained, internal fixation should be secured with Kirschner wires (K- wires) (Figure 4). Other fixation tech- niques include compression screws and absorbable pins. I have found that K-wires are simple, efficient, in- expensive, and effective. In most of these patients, a 0.62-in K-wire is sufficient. In a very small child, I recommend 0.45-in K-wires. Some surgeons advocate placing all fixation pins or screws in the metaphyseal fragment, thus avoiding the ossific nucleus and physis of the lateral condyle. With a very large fragment, this is usually possible. However, in many of these condylar fractures, the reduction and articular surface are difficult to visualize. A sufficiently large fragment is needed to confirm that the pin has adequate purchase. However, when the frac- ture fragment is small, I do not hes- itate to place the pins through the condyle and across the physis into the medial aspect of the distal hu- Figure 1 A, Milch type I fracture. The fracture line is through the ossific nucleus of the capitellum. B, Milch type II fracture. The fracture line is lateral to the ossific nucleus. Figure 2 Fracture types I through III. A, In a type I fracture, the fracture line does not violate the articular surface and therefore is stable. B, A type II fracture is through the articular surface but minimally displaced. C, A type III fracture is displaced and often rotated. J. Andy Sullivan, MD Volume 14, Number 1, January 2006 59 merus. The anatomy of the distal hu- merus is such that most of the growth occurs in the lateral condyle and the trochlea. Occasionally, pa- tients end up with a fishtail appear- ance to the distal humerus, which does not interfere with function. In my experience, I have never seen a true arrest of the entire lateral condyle. A variety of pin configurations may be used. Some advocate parallel pins, as in the video ( video); how- ever, I prefer convergent pins placed through the lateral condyle and up into the shaft of the humerus. A sec- ond pin is placed transversely across the fracture line through the meta- physeal fragment. This provides good stability and divergence of the pins. Parallel pins or diverging pins are more stable than converging pins. What is to be avoided is having the pins converge at the fracture site be- cause this is a less stable construct. An important intraoperative tech- nique is to pick up the skin of the pos- terior aspect of the incision gently with forceps and start the K-wires away from the incision in order to avoid having the pins come out through the incision. In most in- stances, the pins can be cut subcuta- neously to bring them retrograde through the skin (by pressing the skin over them). Pin placement and adequacy of re- duction should be confirmed and doc- umented radiographically. This pro- vides a baseline for postoperative care. The fracture site should be inspected visually and by palpation to ensure continuity of the joint articular sur- face. The wound is irrigated, and layer closure is accomplished with absorb- able sutures, including an absorbable subcuticular suture. When the frac- ture treatment has been delayed and there is excessive swelling or concern about skin tension, then mattress or tension-releasing sutures should be used in the skin. Either absorbable or nonabsorbable may be used, depend- ing on surgeon preference. The K-wires previously inserted are cut off outside the skin and bent over ( video). Nonadherent gauze is placed around the pins, and a felt pad is cut and placed over the pin. The arm is wrapped with cotton cast padding, and a posterior splint is ap- plied. For elevation to prevent depen- dent edema, the arm is placed in a sling after the procedure. Follow-up Care The patient is seen at 1 week after pri- mary treatment; the cast is removed, the pin sites are examined, and radio- graphs are obtained. If the pins are in satisfactory position and the reduc- tion maintained, a fiberglass long arm cast is applied. In children younger than age 6 or 7 years, an additional 2 weeks of immobilization is recom- mended, although 3 weeks o r more of immobilization is preferred after a cast application, giving a total of 4 Figure 3 A, A type III fracture with displacement and rot ation. B, The fracture has been reduced and is being held with a clamp. Figure 4 Anteroposterior (A) and lateral (B) views of fixation secured with K-wires. Fractures of the Lateral Condyle of the Humerus 60 Journal of the American Academy of Orthopaedic Surgeons weeks of immobilization. (I have not had any problems with the children regaining elbow motion after lateral condylar fractures and find that I sleep a little easier with the extra week of immobilization.) At 4 weeks, anteroposterior and lateral radiographs are obtained. If there is new bone formation indicat- ing the early stage of healing, the K-wires are removed. If there is no ev- idence of bone formation (which is extremely uncommon, in my experi- ence), pins still can be removed and additional cast immobilization per- formed. Reexamination at 6 weeks is necessary. Failure to demonstrate union of the fracture at that point would require an additional 6 weeks of cast or splint immobilization. Be- yond 12 weeks, I would consider the fracture to be nonunited and would proceed with bone grafting. Although I have not done a formal review, I do not recall having had a nonunion in a lateral condylar frac- ture that was treated acutely, nor do I recall one that was not sufficiently united at 4 weeks in which I could not remove the pins and begin mobi- lization. Delayed Presentation or Nonunion Nonunion of the lateral condyle can result in cubitus valgus deformity and in a tardy ulnar nerve palsy. 9 If the fracture is not united in a patient who presents between 6 and 12 weeks, the standard surgical approach described is recommended. W ith nonunion, the surgical procedure is more difficult. In the established nonunion, the frac- ture site is separated from the prox- imal fragment with sharp dissection with a scalpel or a small osteotome. Once the two fragments are sepa- rated, curets are used to remove the reactive fibrous tissue, taking care to dissect the distal fragment as mini- mally as possible. The surfaces of the proximal and distal fragments must be cleaned of this fibrous tissue. Once this is achieved, I t ry to obtain as near an anatomic reduction as possible. There are no good anatomic land- marks at this point, and achieving an- atomic reduction is more difficult than in an acute case. Once the anatomic position is de- termined, the fracture fragments are reduced with a towel clip and pins are inserted. I still prefer iliac crest bone as the bone donor site. There are bone graft substitutes available that may be equally effective. With an established nonunion, I prefer internal fixation and bone graft, followed by 6 weeks of cast im- mobilization. Most fractures in my experience have united with this treatment at 6 to 12 weeks. Patients may present with established non- unions that are in an older age group. Roye et al 10 have demonstrated that, even in adolescents, treatment can be successful by means of open re- duction, cleaning the bony surfaces, inserting a cancellous screw, and making use of bone grafting. With surgical intervention, the stability of the elbow is improved, and the risk of cubitus valgus is reduced. 11 References 1. Milch H: Fractures and fracture dislo- cations of the humeral condyles. J Trauma 1964;15:592-607. 2. Jakob R, Fowles JV, Rang M, Kassab MT: Observations concerning frac- tures of the lateral humeral condyle in children. J Bone Joint Surg Br 1975; 57:430-436. 3. Ogden JA: Skeletal Injury in the Child, ed 2. Philadelphia, PA: WB Pearls • Position the patient close enough to the edge of the operating table that the arm can be visualized by fluoroscopy, but with sufficient room that the arm is completely supported during the surgical pro- cedure. • In a larger child, I recommend use of a hand table. During surgery, extend the elbow and apply varus movement to visualize the frac- ture. Flex the arm at the moment of reduction to relax the distal frag- ment. • Warn the family preoperatively of potential complications, includ- ing nonunion, cubitus varus, osteonecrosis, and protuberance of the lateral condyle. Protuberance of the lateral condyle is the most com- mon. The other complications are extremely rare but are serious. Clinical series report that up to half of cases of lateral condyle frac- ture have a lateral protuberance. Although there is no functional dis- ability from the protuberance and no surgical treatment is required, it is disconcerting to parents to see the deformity, so it is best to have warned them preoperatively. Although the exact cause is not known, it is thought that dissection of the periosteum may increase the likelihood of this result. For this reason, dissection should be limited to that necessary to expose the fracture site. Pitfalls • Failure to visualize the intra-articular component of the fracture and obtain anatomic reduction • Inadequate internal fixation or fracture reduction with rotational displacement • Failure to recognize displacement or rotation in fractures treated closed or with percutaneous pinning • Aggressive early return to sports. It is best to keep patients out of sports for 2 months following fracture treatment. J. Andy Sullivan, MD Volume 14, Number 1, January 2006 61 Saunders, 1990. 4. Herring JA: Tachdjian’s Pediatric Or- thopedics, ed 3. Philadelphia, PA: WB Saunders, 2002. 5. Wilkens KE, Beaty JH, Chambers HG,Toniolo RM: Fractures and dislo- cations of the elbow region, in Rock- wood CA Jr, Wilkens KE, Beaty JH (eds): Fractures in Children. Philadel- phia, PA: Lippincott-Raven, 1996, vol 3, pp 653-904. 6. McIntyre W: Lateral condylar frac- tures of thehumerus, in Letts RM(ed): Management of Pediatric Fractures. New York, NY: Churchill Living- stone, 1994, pp 241-258. 7. Foster DE, Sullivan JA, Gross RH: Lat- eral humeral condylar fractures in children. J Pediatr Orthop 1985;5:16- 22. 8. Mintzer CM, Waters PM, Brown DJ, Kasser JR: Percutaneous pinning in the treatment of displaced lateral condyle fractures. J Pediatr Orthop 1994;14:462-465. 9. Masada K, Kawai H, Kawabata H, Masatomi T, Tsuyuguchi Y,Yamamo- to K: Osteosynthesis for old, estab- lished non-union of the lateral condyle of the humerus. J Bone Joint Surg Am 1990;72:32-40. 10. Roye DP, Bini SA, Infosino A: Late surgical treatment of lateral condylar fractures in children. J Pediatr Orthop 1991;11:195-199. 11. Morrissy RT, Wilkins KE: Deformity following distal humeral fracture in childhood. J Bone Joint Surg Am 1984; 66:557-562. Fractures of the Lateral Condyle of the Humerus 62 Journal of the American Academy of Orthopaedic Surgeons . at the moment of reduction to relax the distal frag- ment. • Warn the family preoperatively of potential complications, includ- ing nonunion, cubitus varus, osteonecrosis, and protuberance of