Open tibial fractures in children under 13 years of age 10 years experience

Injury, Int. J. Care Injured 34 (2003) 776 780 Open tibial fractures in children under 13 years of age 10 years experience B.G. Jones, R.D.D. Duncan The Royal Hospital for Sick Children, Yorkhill, Glasgow G3 8SJ, Scotland, UK Accepted 17 January 2003 Abstract Objective: To evaluate the results of treatment in children with open tibial fractures. Design: Retrospective review. Subjects: Eighty three children under 13 years of age treated for an open tibial fracture between 1989 and 1999. Main outcome measures: Patient demographics, mechanism of injury, fracture classification, treatment method, clinical outcome and complications. Results: Eighty one percent of children had an open tibial fracture as their only injury. According to the level of contamination, soft tissue injury and size of wound 46% were Gustilo grade I injuries, 30% grade II, and 22% (18) grade III (6 IIIa, 5 IIIb, 3 IIIc and 4 not otherwise specified). Sixteen fractures (19%) were treated using an external fixator and 65 (78%) using a cast. The average time to union was 15.5 weeks (range 9 31 weeks) for those treated with a frame and 10.4 weeks (range 5 40 weeks) for those treated with a cast. No deep infections, one delayed union and one non-union were recorded. Conclusions: Ninety four percent of these injuries were a result of a motor vehicle accident and involved a significant trauma-related energy transfer. Despite this the associated morbidity was low lending support to the literature, which suggests that open fractures of the tibia in younger children heal more predictably and with less complications than those occurring in adolescents or adults. 2003 Elsevier Science Ltd. All rights reserved. 1. Introduction In adults, open tibial fractures are associated with significant morbidity [6,9,11,12]. However, the literature relating to these injuries in children is contradictory. Some reports suggest that they behave in a similar manner to adult injuries and others that they behave in a more benign fashion, with more favourable outcomes in younger children [15,18,20]. The aim of this study was to evaluate our results of treatment in children below the age of 13 years with open tibial fractures. 2. Patients and methods Our unit is a large children s teaching hospital. It provides secondary care to a population of 900,000 and tertiary care to three million individuals. All surgical disciplines with the exception of paediatric neurosurgery are provided on-site. Corresponding author. Tel.: +44-141-201-0276; fax: +44-141-201-0275. Children above 13 years of age are treated in adult hospitals within the city. We identified 1276 children from the departmental database who had sustained a tibial fracture in the 10 year period 1989 1999. Eighty three of these were open fractures (6.5%). A case record review of these 83 children was performed. Physeal injuries were excluded. Demographic information, the mechanism of injury and any associated injuries were recorded. All injuries were treated operatively with wound debridement and irrigation under intravenous antibiotic cover. Where necessary the advice of a plastic surgeon was sought at the earliest opportunity. The fractures were classified according to the wound size, periosteal stripping and contamination as described by Gustilo et al. [11,12]. The grade assigned by the surgeon who performed the debridement was used for comparative purposes. The fracture management was documented and complications noted. Time to union was judged on clinical and radiological grounds. All patients were followed up until complete resolution of all complications had occurred or until definitive uncomplicated union of the fracture had taken place. The average follow up was 26.5 weeks (range 5 140 weeks). 0020-1383/$ see front matter 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/s0020-1383(03)00031-7

Table 1 Details of graded open tibial fractures Gustilo grade Age (mean/median, year) Side Mechanism of injury Associated injuries Mean time external fixation (week) Mean time long leg cast (week) I(n = 38) 7.4/7.5 19R/18L/1? 2 fall, 1 bicycle vs. car, 35 road traffic accidents II (n = 24) 7.1/6.5 16R/7L/1? 1 fall, 1 crush, 22 road traffic accidents 2 humeral #s, 1 mandible #, 1 femur, # 1 metatarsal #, 1 skull #, 7 minor head injuries 1 posterior cruciate ligament rupture, 4 metatarsal #s, 4 minor head injuries, 1 skull # 0 6.5 5.8 5.6 IIIa (n = 6) 6.2/6.0 3R/3L 6 road traffic accidents D12, L1#s 11.5 8.6 IIIb (n = 5) 7.6/8.0 1R/3L/1? 5 road traffic accidents 1 humerus #, 1 head injury and 14.3 7.5 seizure, 1 disruption Tibialis Anterior +Extensor Hallucis Longus IIIc (n = 3) 7.0/8.0 2R/1L 1 crush, 1 road traffic 1 disruption Tibialis Anterior, 1 11.25 8 accidents, 1 bicycle vs. car humerus #, 1 depressed skull #, 1 disruption anterior + lateral compartments III (no grade, n = 4) 6.5/5.0 2R/1L/1? 4 road traffic accidents 1 Extensor Hallucis Longus disruption, 1 talus #, 1 cyanosed foot 5 7 Mean time below knee cast (week) Time to union (mean/median, week) Flap/graft coverage Complications Comments 5.4 10.5/10 1 delayed split skin graft Delayed union femur #, 1 delayed tibial union, 2 wound infections, 4 angular deformities, 1 rotational deformity, 1 Extensor Hallucis Longus tethering behavioural problem 4.9 10.9/10 1 latissimus dorsi flap to foot, 1 associated split skin graft 1 pin track infection, 1 angular deformity, 1 Extensor Hallucis Longus dysfunction 3.7 12.8/12 1 delayed split skin graft 1 non-union, 1 angular deformity 8.3 17.4/16 3 latissimus dorsi flaps, 1 osteocutaneous flap, 1 fasciocutaneous flap, 1 Extensor Digitorum Brevis flap, 4 associated split skin grafts 1 bulky flap, 1 pintrack infection, 1 odd gait 1 subcutaneous fasciotomy, head injury, delayed surgery 3 fasciotomies (1 foot), 1 external fixator dynamised, 1 intensive care stay Bone grafting required, 1 external fixator dynamised, 1 sub-cutaneous fasciotomy, 1 primary wound closure 1 external fixator dynamised, 1 intensive care stay, head injury, delayed surgery 0 10.2/9 2 split skin grafts 1 foot drop 1 external fixator dynamised, serial casting for equinus 5 11.8/10 1 split skin graft, 1 gastrocnemius flap, 1 associated split skin graft 1 angular deformity, 1 Extensor Hallucis Longus dysfunction 1 fasciotomy, 1 primary wound closure B.G. Jones, R.D.D. Duncan / Injury, Int. J. Care Injured 34 (2003) 776 780 777

778 B.G. Jones, R.D.D. Duncan / Injury, Int. J. Care Injured 34 (2003) 776 780 3. Results The results are presented in Table 1. The 83 children comprised 62 boys and 21 girls. Their ages ranged from 2 to 12 years, mean 7.2 years. There were no bilateral open tibial fractures. Seventy eight fractures (94%) were sustained in a road traffic accident, 76 involving a pedestrian versus a motor vehicle and two a cyclist versus a motor vehicle. The peak time of injury was late afternoon (range 0815 2130 h). The remaining fractures were caused by either a fall (three children), or a crush injury (two children). Sixty seven (81%) children had the open tibial fracture as their only injury. Associated injuries included an associated long bone fracture in five children (including one ipsilateral femoral fracture) and D12 and L1 wedge fractures in one child, but there were no significant chest or abdominal injuries. Twelve children had head injuries, with three including skull fractures. Only three children had delayed primary surgery because of their head injury. None of the series required neurosurgical intervention. There were no deaths. According to the level of contamination, soft tissue injury and size of wound 38 (46%) fractures had been graded as Gustilo grade I injuries, 24 (30%) grade II, and 18 (22%) grade III (six IIIa, five IIIb, three IIIc and four not otherwise specified). Three children had sustained open fractures which had not been assigned a Gustilo grade by the performing surgeon, no attempt has been made to grade these retrospectively but they have been included in the total figures. There were no gunshot or farmyard injuries and no amputations were performed. Most fractures were situated in the tibial diaphysis (middle/lower, 1/3). The ipsilateral fibula was fractured in 66 cases. The mean time between injury and surgery was 5.9 h (S.D. ± 3.7 h). Intravenous antibiotics, most commonly a cephalosporin, were given for a mean of 2.9 days (range one dose to 15 days) after injury. No formal antibiotic protocol was in use during the study period. Twenty-one wounds were closed at the initial debridement including two injuries which had been classified as grade III injuries. Wound cover was achieved using four free latissimus dorsi flaps, two local myocutaneous flaps, one fasciocutaneous flap and 10 split skin grafts. An osteocutaneous flap was performed in one of the two children with significant bone loss. Skin cover was achieved within 24 h in all but two children. Prophylactic fasciotomies were performed in seven children (one foot and six calf), however, no compartment syndromes were seen. Thirty seven children had at least one (range 1 6, median one) further wound inspection or debridement under general anaesthetic, but 31 (37%) children had their soft tissue injury managed using a plaster window. The mean hospital stay was 10 days (range 2 46 days). The choice of immobilisation technique was dictated by the operating surgical team. Sixteen (19%) were treated using a monolateral external fixator and 65 (78%) using a plaster cast. In two additional cases plaster cast immobilisation was augmented by the use of Kirschner wires. External fixation was used for the duration of treatment in two children. It was discontinued and a plaster cast applied after a median of 9 weeks (range 3 24) in the remaining 14 children. Fixator dynamisation was not routine. Thirteen casts required wedging, without anaesthetic and seven remanipulations under anaesthetic were performed. The plaster treated group were immobilised for 10.4 weeks (S.D. ± 4.8 weeks), with long leg casts used for an average of 5.9 weeks before conversion to short-leg walking casts. One child took 40 weeks for the fracture to unite in a cast (grade I injury), and another developed a non-union for which grafting was performed at 6 months (grade IIIA fracture, cast immobilisation group). The average time to union was 15.5 weeks (range 9 31 weeks) for those treated with a frame and 10.4 weeks (range 5 40 weeks) for those treated with a cast. Seven fractures healed with a residual angulation of more than 5 ; two in varus (7 8 ), four in valgus (10 20 ) and one procurvatum (12 15 ). One child had both sagittal and coronal plane angulation. No corrective osteotomies have been performed and no child has a significant leg length discrepancy. No child developed a deep infection. Two superficial wound infections developed, both grew Pseudomonas aeruginosa and both resolved with oral antibiotics. There were two pin track infections in the fixator group where the fixators had been used for 8 and 11 weeks, respectively, both infections were treated effectively with antibiotics. At the time of writing, only three children continue under review. One with an equinus deformity of the ankle subsequent to anterior compartment muscle loss, one with an unexplained gait anomaly without neuromuscular deficit, and one with no activity in Extensor Hallucis Longus. 4. Discussion National guidelines have been published on the management of open tibial fractures in adults [1] and are based on initial wound debridement and irrigation, fracture stabilisation and delayed wound closure [10,13]. Several authors have demonstrated that open fractures in children appear to behave in a similar manner and have advocated a similar approach, to reduce problems such as deep infection, delayed and non-union [3,4,7,15,18]. It may be that these injuries in adolescents and older children are associated with the same problems seen in adults, but our results support the observations made by Song [20] that open fractures in younger children (under 13 years) appear to heal more predictably and with less complications than their adult counterparts. The average time to union was 12.9 weeks in this group which is much shorter than previously described, although longer than the 8.5 weeks taken for closed fractures to heal in a similar age group [14]. Song also observed a shorter

B.G. Jones, R.D.D. Duncan / Injury, Int. J. Care Injured 34 (2003) 776 780 779 healing time in children under 11 years [20]. The presence of infection delays healing [4,7] the fact that we had no deep infections may be relevant. The longer time to union in the fixator group has been previously reported [3,15] and probably reflects more fracture instability and soft tissue disruption. However, no child in our series treated with an external fixator required bone grafting to achieve union. External fixation has been associated with a clinically detectable leg length discrepancy [3,4] but this was not seen in our children. Infection rates in children s open tibial fractures range from 7 to 25% [3,7], which compares with 9.5 24% in adult grade III injuries [12,10,16]. No deep infections occurred in our series. All children were managed initially with intravenous antibiotics, but no formal protocol was in place during the 10 years period. Factors known to increase the rate of infection are the degree of contamination, a delay in wound management and a delay in achieving soft tissue cover [5,9,18]. Seventy one percent of our children had their debridement within 6 h of injury and definitive skin coverage was achieved in under 24 h in all but two children. Pin track sepsis occurred in two children (12.5%). Only two of the 16 children treated with external fixation did not have a cast applied after fixator removal. Some authors have advocated early removal of a fixator to reduce the incidence of pin-related sepsis seen in adult and paediatric series [6,7,20]. The median time of external fixator use in our series was 9 weeks (range 3 24 weeks). Whilst we recognise that interobserver agreement with the Gustilo and Anderson classification is variable [2] and so may affect direct comparison between series, our findings suggest young children with open tibial fractures behave differently from open tibial fractures in adolescents and adults. Our results may be partly explained by the fact that our study group contained fewer children with associated injuries than some series [3,4,7], and that there were no gunshot or farmyard injuries, amputations or deaths. Ninety four percent of injuries were, however, sustained in motor vehicle accidents. It is well accepted that trauma-related energy transfer results in a greater force per unit area being applied to children when compared to adults, predominantly because of their small size [17]. Biomechanical data also suggests that children s bones behave differently to adults when a deforming force is applied to them. The structure of immature bone results in a resistance to crack propagation and thus more energy can be absorbed by a child s bone before fracture [8,21,19]. The lack of comminution in children s fractures may be explained by the mechanical properties of the bone tissue and does not necessarily imply a lower level of energy involved in the injury. With these factors taken into consideration and the fact that most were caused in motor vehicle accidents, these fractures could be regarded as high energy injuries and thus automatically graded as Gustilo grade III. In this situation the differences between the behaviour of these injuries in young children and adults are even more apparent. We do not advocate complacency in the management of these injuries and wholeheartedly support early and aggressive wound management, with early involvement of plastic surgeons where indicated. Following debridement and irrigation of the wound it is clear that a cast may be an appropriate method of fracture immobilisation, with a window enabling wound inspection as required. Close attention to detail is of course required, as with all casted fractures. However, it does not appear necessary to treat all open fractures in children with an external fixator, with the risks of pin track sepsis and potential osteitis. Obviously in injuries with little fracture stability and with more extensive soft tissue injuries a fixator would be a more appropriate choice for immobilisation; each case has to be judged on its own merits. References [1] BOA/BAPS working party. The management of open tibial fractures. British Orthopaedic Association; September 1997. [2] Brumback RJ, Jones AL. Interobserver agreement in the classification of open fractures of the tibia. J Bone Joint Surg Am 1994;76- A:1162 5. [3] Buckley SL, Smith G, Sponseller PD, Thompson JD, Griffen PP. Open fractures of the tibia in children. J Bone Joint Surg Am 1990;72-A:1462 9. [4] Buckley SL, Smith G, Sponseller PD, Thompson JD, Griffen PP. Severe (type III) open fractures of the tibia in children. J Paediatr Orthopaed 1996;16:627 34. [5] Caudle RJ, Stern PJ. Severe open fractures of the tibia. J Bone Joint Surg Am 1987;69-A:801 7. [6] Court-Brown CM, Wheelwright EF, Christie J, Mc Queen MM. External fixation for type III open tibial fractures. J Bone Joint Surg Br 1990;72-B:801 4. [7] Cramer KE, Limbird TJ, Green NE. Open fractures of the diaphysis of the lower extremity in children. J Bone Joint Surg Am 1992;74- A:218 32. [8] Currey JD, Butler G. The mechanical properties of bone tissue in children. J Bone Joint Surg Am 1975;57-A:810 4. [9] Fischer MD, Gustilo RB, Varecka TF. The timing of flap coverage, bone-grafting, and intramedullary nailing in patients who have a fracture of the tibial shaft with extensive soft-tissue injury. J Bone Joint Surg Am 1991;73-A:1316 21. [10] Gopal S, Majumder S, Batchelor AG, Knight SL, De Boer P, Smith RM. Fix and flap: the radical orthopaedic and plastic treatment of severe open fractures of the tibia. J Bone Joint Surg Br 2000;82- B:959 66. [11] Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones. J Bone Joint Surg Am 1976;58-A:453 8. [12] Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma 1984;24:742 6. [13] Gustilo RB, Merkow RL, Templeman D. Current concepts review: the management of open fractures. J Bone Joint Surg Am 1990;72- A:299 304. [14] Hansen BA, Greiff J, Bergman F. Fractures of the tibia in children. Acta Orthop Scand 1976;47:448 53. [15] Hope PG, Cole WG. Open fractures of the tibia in children. J Bone Joint Surg Br 1992;74-B:546 53. [16] Keating JF, Blachut PA, O Brien PJ, Court-Brown CM. Reamed nailing of Gustilo grade-iiib tibial fractures. J Bone Joint Surg Br 2000;82-B:1113 6.

780 B.G. Jones, R.D.D. Duncan / Injury, Int. J. Care Injured 34 (2003) 776 780 [17] Krantz BE. ATLS Instructor Course Manual. 6th ed. Chicago: American College of Surgeons; 1997. p. 355 75. [18] Kreder HJ, Armstrong P. A review of open tibia fractures in children. J Paediatr Orthopaed 1995;15:482 8. [19] Nafei A, Danielson CC, Linde F, Hvid I. Properties of growing trabecular ovine bone. Part 1. Mechanical and physical properties. J Bone Joint Surg Br 2000;82-B:910 20. [20] Song KM, Sangeorzan B, Benirschke S, Browne R. Open fractures of the tibia in children. J Paediatr Orthopaed 1996;16: 635 9. [21] Zioupos P, Currey JD. Changes in the stiffness, strength, and toughness of human cotical bone with age. Bone 1998;22: 57 66.