Infected non-union of the tibia remains to be a great challenge to orthopaedic surgeons. Due to its subcutaneous location, the tibia is vulnerable to be deprived of its soft tissue coverage with subsequent desiccation and infection of exposed bone . Deep infection would ruin any reconstructive attempt unless properly debrided. Management of infected non-union of the tibia necessitates adequate debridement and reconstruction of the resultant bone and soft tissue defect . Such complex reconstruction is preferably performed in a staged manner to avoid recurrence of infection . After debridement, the resultant dead space invites collection of hematoma and infected transudate. Elimination of this dead space, either by acute limb shortening or insertion of a spacer would prevent such collection [4, 5].
Several articles have described the results of debridement of bone infection and the use of antibiotic-cement spacer. However, these studies included heterogeneous anatomical and etiological group of patients.
The aim of this prospective study was to evaluate the results of an integrated treatment protocol in a uniform group of patients with severely infected nonunion of the tibia. This protocol is based on staged debridement and elimination of the dead space in preparation for reconstruction of composite bone and soft tissue defect by distraction histogenesis using Ilizarov external fixator. With emphasis on the role of negative pressure wound therapy in the post-debridement phase.
Twenty three adult patients aged 19 to 52 years (average 24 years) were treated in this prospective study in the period between January 2014 and June 2017. There were 4 female and 19 male patients. The study included patients with actively infected nonunion of the tibia whether following operative fixation or open fractures in adults. All patients had previous debridement procedures ranged from two to four times. Ten patients (43.5%) had retained hardware, six cases (26.1%) infected open fracture and seven (30.4)% presented with infected nonunion. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study. General clinical examination of the patient was done to exclude the presence of general debilitating or systemic illness. The limb was examined for the neurovascular status, skin condition, shortening, range of motion of adjacent joints as well as for clinical and biochemical evidence of active infection (draining sinus, local inflammation, pain, tenderness, swelling and elevated ESR, CRP and total leuckocytic count). Radiographs of the limb were obtained and assessed for the level and size of bone defect, bone quality and length discrepancy as well as radiological evidence of infection (cavities, sequestrated and sclerotic bone).
Multiple pus, soft tissue and bony specimens were obtained intra-operatively for culture and antibiotic sensitivity testing, and for pathology as well. Intravenous antibiotics, according to culture results, were essential in all patients for three weeks postoperatively. This was followed by oral antibiotics for further three weeks.
All patients were treated according to the protocol put forward by our surgical team [Tanta University Integrated Protocol (TUIP) for management of post-traumatic osteomyelitis of the tibia] (Figure 1).
All cases were operated under general or epidural anesthesia in the supine position. A tourniquet was applied on the upper thigh and inflated until completion of debridement then was deflated for haemostasis.
Skin incision: Previous skin incisions were utilized with excision of the main sinuses and unhealthy skin. The dissection was performed by sharp knife elevating the skin with the deep fascia and periosteum as one layer.
Bone debridement (Phase I): The fracture-site was debrided to excise all infected and devitalized tissue with a safety margin of at least 5 mm. (oncological debridement). All retained hardware was removed. Transverse “Square” osteotomy of the bone ends was performed, to increase the bone contact surface area as the bone ends come in close proximity after completion of bone transport at the docking site. The osteotomy was performed so that the bone ends were cut shorter than the skin edge which is hanging, like a curtain, to cover the bone ends (soft tissue curtain). The medullary cavity was reamed on each side with removal of all the dead bone and infected granulation tissues.
Management of dead space (Phase II): After completion of debridement the tourniquet was deflated for haemostasis and to ensure resection of all non-viable tissues to bleeding edges. The resultant tissue defect was copiously irrigated with normal saline and packed with wet gauze. The surgical drapes and gloves were exchanged by new ones and the procedure was continued without tourniquet.
Antibiotic-loaded cement spacer insertion: Antibiotic (vancomycin powder) was added to the dry component of the cement and mixed manually (in a ratio of 3gm for each 40 gm of cement). The spacer was inserted as a cylindrical block enforced by two Kirschner wires passing into each side of the medullary cavity. This was done during the later stages of polymerization to allow for proper handling and shaping of the cement. The ends of the spacer were molded around the bone edges. It is to be noted that over-sizing of the spacer would hinder soft tissue closure.
Skin closure: After setting of the cement spacer, the wound was closed in layers over a wide-bore suction drain for 72 hours (closed wound technique) (Figure 2). If a residual soft tissue defect did not allow wound closure, then the skin edges were only approximated leaving part of the spacer exposed and a negative pressure wound therapy device (NPWT) was applied(open wound technique) (Figures 3 and 4).
Immobilization: The leg was immobilized on a rigid removable splint to be replaced by a walking cast after wound healing. No external nor internal fixation was attempted in any case at this point.
Postoperative management: Postoperative leg elevation was necessary to avoid postoperative edema. Weight-bearing as tolerated was allowed throughout the treatment.
Reconstruction (Phase III): Three weeks later, the patient was readmitted for reconstruction phase in cases of open-wound technique. While in closed wound technique, it was performed six weeks after debridement phase. The cement spacer was exposed and removed en bloc or in a piecemeal fashion. An osteotome can be used to longitudinally split the cement spacer into smaller pieces before removal. After spacer removal, bone and soft tissue reconstruction were performed as one procedure using bone transport by Ilizarov external fixator for advancement of an osteomyocutaneous composite flap to bridge the defect after performing metaphyseal osteotomy. (Figures 3 and 4).
Follow up: The transport was started 7 days post-operative at a rate of 1 mm per day. Patients were taught to do distraction-compression and dressing of the wound before discharge from the hospital. Patients were seen in the clinic every two weeks during the distraction phase (to check for soft tissue problems during transport) and then every month during the consolidation phase. Radiographs were obtained in each visit to check the progress of bone transport and healing. The external fixator was removed after bone consolidation. Criteria for bone consolidation and safe removal of the external fixator have been: 1) radiological: at least three intact bone cortices and radiological density of the regenerate bone comparable to adjacent normal bone. 2) clinical: the patient should be able to stand unsupported on the treated limb for 30 seconds (the single – leg stance test).
After fixator removal, a well-fitted plaster cast was applied for one month with weight-bearing. Patients were then followed up every three months and then yearly by making radiographs and laboratory infection markers.
Evaluation of treatment outcome: The results were assessed based on both objective (clinical and radiographic evaluation) and subjective criteria (limb function and patient’s satisfaction) using our system of results’ evaluation (Table 1) . The final results were considered to be satisfactory or unsatisfactory based on these findings.
|Bony union||United||Not united|
|Residual leg length discrepancy||<2.5 cm||>2.5 cm|
|Recurrent infection||No more infection||Bone and/or soft tissue infection|
|Soft tissue healing||No exposed bone||Soft-tissue defect remaining|
|Permanent joint contracture||<5°||>5°|
|Persistent pain||No or mild pain||Moderate or Incapacitating pain|
|Return to previous work||Yes||Has to changejob|
|Patient’s satisfaction||Satisfied||Not satisfied|
Table 2 summarizes the results of the study cases. The mean size of bone defect after debridement was 6 cm (ranged from 4 to 10 cm). All patients completed the follow up which ranged from 16 to 36 months (average 28 months). Successful reconstruction with bone union and no recurrence of infection was achieved in all cases (100%) without the need of a skin graft or a muscle flap. Spontaneous consolidation of the induced biological membrane was reported during transport in 5 cases (21.7%) before reaching the docking site (Figure 5). Revision of the docking site was needed in 15 cases (65.2%) at the end of transport to clear out the invaginated skin between the bone ends and improve contact of bone ends. There was no need to insert bone graft in any case. Superficial pin tract infection occurred in all cases and was treated by daily pin site care and oral antibiotics. External fixator index, which is the number of days the patients wore the frame per cm of lengthening, ranged from 35 to 60 days/cm (average 45 days/cm). Residual limb length discrepancy (LLD) occurred in 4 cases (17.4%). The main cause for residual LLD was patient’s intolerance to the procedure. The functional results were satisfactory in 20 cases (86.9%) and unsatisfactory in 3 cases (13.1%) due to residual leg length discrepancy, joint stiffness, and persistent pain.
|Total number of study cases||23 cases||100%|
|Age||19–52 years||Average 24 years|
|Previous surgery||23 cases||100%|
|(2–4 operations)||Average 3.1/case|
|Active infection||23 cases||100%|
|10 With retained hardware||43.5%|
|6 infected open fracture||26.1%|
|7 infected nonunion||30.4%|
|Revision of docking site||15 cases||65.2%|
|Residual LLD||4 cases||17.4%|
|Fixator index||35–60||Average 45 days/cm|
|Follow up||16–36 months||Average 28 months|
|Satisfactory outcome||20 cases||86.9%|
|Unsatisfactory outcome||3 cases||13.1%|
The management of infected nonunion depends on several factors including the host’s physiological status, the size of the defect, level of the defect, quality of the surrounding soft tissue, the presence of deformity and limb length discrepancy [1, 2, 3, 4, 5, 6, 7]. Three keys for successful reconstruction of infected nonunion are: 1) Wide resection of the infected tissues with a safety margin to ensure eradication of infection. 2) square osteotomy of bone ends, for better bone contact and healing, and 3) soft tissue curtain, to avoid desiccation of exposed bone ends . Adequate debridement of bone infection usually results in bone and soft tissue defects. It may be possible to achieve primary closure of the wound by direct suturing or plastic procedures at the same sitting of the debridement . But this is usually non-applicable because of the bad local condition and the size of the bone defect after debridement. Various procedures were described in the literatures to overcome the problem of considerable bone defect after debridement with or without soft tissue defect. Vascularized bone autograft is a popular method in reconstruction of a segmental bone defect allowing concurrent soft-tissue coverage. However, many complications were reported including infection and stress fracture. Also, the method is technically demanding and requires expert microsurgical specialist [10, 11, 12]. An alternative approach to the problem is the induced membrane (Masquelet) technique. This technique precludes filling the defect by antibiotic impregnated cement spacer for a period followed by removal of the spacer and cancellous bone grafting with definitive internal fixation. Encouraging union rates have been reported with reported complications including re-infection, refracture and risk of graft donor site morbidity [13, 14, 15, 16]. Another well-known method in the management of infected tibial nonunion is the bone transport technique which entails debridement of all the infected tissues followed by gradual transport from far osteotomy site to close the bone and soft tissue defect. Gradual bone transport allows free drainage of infected nonunion after thorough debridement by keeping the wound open and gradually closed by means of compression – distraction. Several advantages are expected with this technique including high rate of infection clearance, bone and soft tissue reconstruction, deformity correction and limb length equalization with less need to free or local tissue grafts [6, 17, 18, 19].
A common obstacle is the presence of medullary infection during debridement. This medullary infection is usually the cause of failure of any of the above-mentioned procedures with considerable recurrence rate of infection. In the present study we present the results of an integrated management protocol of severely infected non-union of the tibia with medullary contamination by staged debridement and elimination of the dead space using antibiotic- impregnated cement spacer followed by bone transport using Ilizarov external fixator. The included cases were patients with severely infected tibial nonunion after failed previous debridement procedures. The protocol included three phases 1) oncological bone and soft tissue debridement. 2) filling the resultant space by antibiotic impregnated cement spacer with or without soft tissue closure. 3) tissue reconstruction by distraction histogenesis using the Ilizarov external fixator.
Insertion of a cylinder spacer in post-debridement defects, in infected nonunion, allows for temporization until eradication of infection. Moreover, it obliterates the dead space, permits local delivery of antibiotics, creates a tunnel for bone transport, and provides mechanical support for the surrounding soft tissue which permits application of negative pressure wound therapy (NPWT) postoperatively.
Although NPWT was originally used for staged management of open fractures with associated soft-tissue defects, The use of closed incision negative-pressure wound therapy to manage delayed wound healing and the benefits of this treatment modality include wound contraction with diminished tensile forces, stabilization of the wound environment, decreased edema and improved removal of exudates, and increased blood and lymphatic flow. In orthopaedic surgery, NPWT has been shown to be clinically effective for incisions at high risk for perioperative complications. Proven effects include an increase in blood flow, stimulation of angiogenesis, and a decrease in wound surface area. In this study, the NPWT was applied in cases of open wound technique over the spacer. We observed its benefit in suction of any residual exudates and promoting healthy granulation tissue formation. This was comparable to other studies which focused on the importance of vacuum assisted dressing [20, 21].
Bone transport overcomes the problem of bone graft shortage and decreases donor site morbidity. No bone graft was needed in our series. Furthermore, considering bone transport an advancement of osteomyocutaneous flap, it tapers off the need for soft tissue flaps. All soft tissue defects were spontaneously closed at the end of the bone transport. All soft tissue defects were spontaneously closed at the end of the bone transport. This supports the hypothesis of the osteomyocutaneous transport which entails the simultaneous transport of the soft tissue with underlying bone. Functionally, results were satisfactory in 20 cases (86.9%) and unsatisfactory in 3 cases (13.1%).
A limitation of this study is the lack of control group and relatively small number of cases.
Our results are encouraging and comparable to other studies which used the same technique of transport over induced membrane in management of infected nonunion of the tibia [21, 22]. In contrary to other studies, our protocol highlights the role of using NPWT following debridement of infected nonunion and the use of structural cement-spacer support in the latency period before embarking on tissue reconstruction.
The proposed staged protocol in the present study gave the benefit of elimination of the infection followed by the gradual tissue-transport, by Ilizarov frame, in a healthy non infected media. This allowed sound union with no need for bone grafting of the docking site, no plastic procedures and no recurrence of infection. Further longer-term study with larger sample size is planned to support and validate our protocol in management of such cases.
No financial support was received for this study.
There was no funding received for this paper.
The authors have no competing interests to declare.
Prof. Mahmoud El-Rosasy, MD is the guarantor.
This is a non-commissioned paper that has undergone external peer review according to journal policy.
Dinh, P, Hutchinson, BK, Zalavaras, C and Stevanovic, MV. Reconstruction of osteomyelitis defects. Semin Plast Surg. 2009; 23: 108–118. DOI: https://doi.org/10.1055/s-0029-1214163
Boriani, F, UlHaq, A, Baldini, T, Urso, R, Granchi, D, Baldini, N, Tigani, D, Tarar, M and Khan, U. Orthoplastic surgical collaboration is required to optimize the treatment of severe limb injuries: A multi-centre, prospective cohort study. J Plast Reconstr Aesthet Surg. 2017 Jun; 70(6): 715–722. DOI: https://doi.org/10.1016/j.bjps.2017.02.017
Malkawi, H, Shannak, A and Sunna, P. Active treatment of segmental defects of long bones with established infection: A prospective study. Clin Orthop. 1984; 184: 241–8. DOI: https://doi.org/10.1097/00003086-198404000-00040
El-Rosasy, MA. Appraisal of the role of external skeletal fixation in the management of sequelae of open tibial fractures. Indian J Orthop. 2008 Oct; 42(4): 420–425. DOI: https://doi.org/10.4103/0019-5413.43388
Marais, LC, Ferreira, N, Aldous, C and Le Roux, TLB. The management of chronic osteomyelitis: Part II – Principles of post-infective reconstruction and antibiotic therapy. SA orthop. j. 2014; 13: 32–39.
Simpson, AH, Dakin, M and Latham, JM. The effect of the extent of surgical resection on infection-free survival. J Bone Joint Surg Br. 2001; 83: 403–407. DOI: https://doi.org/10.1302/0301-620X.83B3.10727
Rodriguez-Collazo, E, et al. A Systematic Review of Outcomes and Flap Selection Following Lower Extremity Free Tissue Transfer Versus Vascularized Perforator Pedicle Flap Transfer in Lower Limb Reconstruction. International Journal of Orthoplastic Surgery. 2018; 1(2): 55–66. DOI: https://doi.org/10.29337/ijops.15
Yajima, H, Kobata, Y, Shigemetsu, K, et al. Vascularized fibular grafting in the treatment of methicillin-resistant Staphylococcus aureus osteomyelitis and infected nonunion. J Reconstr Microsurg. 2004; 20: 13–20. DOI: https://doi.org/10.1055/s-2004-818044
Doi, K, Kawakami, F, Hiura, Y, Oda, T, Sakai, K and Kawai, S. One-stage treatment of infected bone defects of the tibia with skin loss by free vascularized osteocutaneous grafts. Microsurgery. 1995; 16: 704–712. DOI: https://doi.org/10.1002/micr.1920161009
Masquelet, AC and Begue, T. The concept of induced membrane for reconstruction of long bone defects. OrthopClin N Am. 2010; 41: 27–37. DOI: https://doi.org/10.1016/j.ocl.2009.07.011
Morris, R, Hossain, M, Evans, A and Pallister, I. Induced membrane technique for treating tibial defects gives mixed results. Bone Joint J. 2017 May; 99-B(5): 680–685. DOI: https://doi.org/10.1302/0301-620X.99B5.BJJ-2016-0694.R2
McKee, MD, Li-Band, EA, Wild, LM and Schemitsch, EH. A prospective randomized clinical trial comparing on antibiotic-impregnated bioabsorbable bone substitute with standard antibiotic-impregnated cement beads in the treatment of chronic osteomyelitis and infected nonunion. J Orthop Trauma. 2010; 24: 483–490. DOI: https://doi.org/10.1097/BOT.0b013e3181df91d9
Gitelis, S and Brebach, GT. The treatment of chronic osteomyelitis with a biodegradable antibiotic-impregnated implant. J Orthop Surg (Hong Kong). 2002; 10: 53–60. DOI: https://doi.org/10.1177/230949900201000110
El-Alfy, BS. Unhappy triad in limb reconstruction: Management by Ilizarov method. World J Orthop. 2017 Jan 18; 8(1): 42–48. DOI: https://doi.org/10.5312/wjo.v8.i1.42
Mahaluxmivala, J, Nadarajah, R, Allen, PW and Hill, RA. Ilizarov external fixator: Acute shortening and lengthening versus bone transport in the management of tibial non-unions. Injury. 2005 May; 36(5): 662–8. DOI: https://doi.org/10.1016/j.injury.2004.10.027
Pande, K. Optimal management of chronic osteomyelitis: Current perspectives. Orthopedic Research and Reviews. 2015; 7: 71–81. DOI: https://doi.org/10.2147/ORR.S50753
Marais, LC and Ferreira, N. Bone transport through an induced membrane in the management of tibial bone defects resulting from chronic osteomyelitis. Strategies Trauma Limb Reconstr. 2015; 10: 27–33. DOI: https://doi.org/10.1007/s11751-015-0221-7
Uzel, A-P, Lemonne, F and Casoli, V. Tibial segmental bone defect reconstruction by Ilizarov type bone transport in an induced membrane. Orthop Traumatol Surg Res. 2010; 96: 194–198. DOI: https://doi.org/10.1016/j.otsr.2009.10.017