3.1 Surgical debridement and reconstruction

Debridement, combined with concurrent microbiology sampling, of the infected field remains the most important strategy in managing FRI. The debridement should include the excision of non-viable bone and soft tissue, and the removal of all non-essential foreign material, including implant or suture material.¹¹ Bone excision traditionally is performed until punctate bleeding is encountered with the reasoning that infected but bleeding bone is viable and can heal with local or systemic antibiotic therapy.¹² However, this concept has recently been challenged with debridement shifting from tumour like radical to adequate debridement.¹³

Tissue sampling allows for both the confirmation of FRI and to plan appropriate antibiotic therapy. A combination of sampling for both microbiological and histopathological analysis has been shown to improve both the accuracy of diagnosis and allow for more targeted antibiotic therapy¹⁴ Samples should be obtained early in the surgical procedure, before contamination occurs.¹⁵ At least five deep tissue samples should be taken from sites around the fracture and adjacent to any implants.¹⁶ Separate instruments should be used for each sample to avoid cross-contamination.¹⁷ In centres where it is available, sonification of implants and subsequent analysis of the sonicate fluid may provide an additional adjunct in diagnosing infection¹⁸ There is currently little evidence to show that sonification is superior to traditional sampling techniques.¹⁹

An adequate debridement may leave the patient with significant soft tissue loss. Repeat attempts at primary closure under tension in these situations should be avoided as they are unlikely to be successful. Prior discussion and planning with a plastic surgeon should be undertaken to allow for an adequate debridement without creating an unexpected wound closure problem. This may necessitate a transfer of patient care to a tertiary centre experienced in FRI management with a joint ortho-plastics MDT.

Local muscle flaps are useful in the proximal and middle thirds of the tibia, while the lower third of the tibia will often require free tissue transfer.¹⁵ While data from open fracture studies suggest that a muscle flap had a lower non-union rate than a fasciocutaneous flap,²⁰ there is little evidence to recommend one specific flap type over another in FRI cases.²¹

While robust soft tissue cover over infected bone is desirable as soon as practical, there is limited evidence to support exact timings. In unwell patients or those with inflamed or grossly infected tissues, a two-stage procedure of initial debridement followed by later soft tissue coverage may be necessary. Where possible, a single stage procedure has the obvious benefit of fewer trips to the operating theatre for the patient. The exact choice of soft tissue reconstruction will depend on multiple factors, including local anatomy, centre experience and host status. Although single stage management is the standard of care, however, If coverage cannot be obtained as a single stage procedure then negative pressure dressings or an antibiotic cement bead pouch can be used to temporarily cover the underlying tissue. The few comparative studies available would suggest that both are largely equivalent as at temporary defect management, with negative pressure dressings being significantly more expensive and at risk of bacterial colonisation of the dressing sponge.²² WOLLF trial has shown no benefit of negative pressure dressings as compared to standard dressing in severe open fractures of lower limb.

3.2 Irrigation

Irrigation of the surgical field after open fracture decreases both bacterial load and removes loose debris^{15 23}. Low pressure irrigation using normal saline is recommended to avoid driving bacteria further into the surrounding soft tissue and bone.²⁴ There is little evidence to support the addition of antibiotics or antiseptics to the irrigation fluid.²³ The use of additives to the irrigation fluid is therefore currently not advised as they may add to cell toxicity.²⁵ Commercially available antibiofilm wound lavage, such as Bactisure Wound Lavage (Zimmer Biomet, Warsaw, IN, USA) have been shown to reduce biofilm on infected tissues in pre-clinical studies,²⁶ however robust prospective research to support efficacy is lacking.

Irrigation for FRI after a tibial shaft fracture treated initially with an intramedullary nail should ideally include the whole medullary canal. Fluid can be infiltrated using the nail entry site, and can be supplemented by over-reaming of the canal. The use of a Reamer–irrigator–aspirator (RIA) allows for concurrent reaming and irrigation.²⁷ The relatively large reamer heads may limit the use of RIA in patients with smaller tibial medullary canals, while infection within the wider metaphyseal regions may be missed by the reamer head.

3.3 Implant retention

Retention of the original implant during the management of FRI may be attempted provided an adequate debridement can still be performed and soft tissues around the wound remain robust. While this approach of debridement, antibiotics and implant retention (DAIR) has become mainstream within arthroplasty, there is relatively little high quality literature to support DAIR for FRI. In a systematic review by Morgenstern et al., success of DAIR for FRI may depend on the time between onset of infection and surgical debridement.²⁸ Infections less than three weeks managed by DAIR had a reported success rates of 86–100 %. Falling to 82–89 % between weeks 3-to-10, and to 67 % after 10 weeks. Six studies containing a total of 276 patients were included. Follow-up varied between studies, from a minimum of 6 months to a minimum of 2 years and one included double blinded RCT only indirectly compared the influence of time on success rates.²⁹ Given these studies do not focus on FRI after an open tibia fracture, the findings of this review must be interpreted with caution. While some authors have reported satisfactory outcomes in implant retention and infection suppression in early FRI in closed or low grade open fractures, removal of any internal fixation and temporising external fixation remains the standard of care for the majority of infected open tibial fractures due to the risk of infected or exposed metalwork.³⁰

3.4 Antibiotic therapy

Unless the patient displays features of systemic sepsis, current best practice guidelines recommend withholding systemic antibiotics until deep tissue samples have been obtained for microbiology sampling. The exact antimicrobial regime and duration depends on both the organism and local and patient factors; traditionally a course of intravenous and then oral antibiotics given for six weeks have been employed for deep infection. Current standard of practice is 1 week IV and 5 weeks of oral antibiotics.³¹ Prolonged systemic antibiotics have their own risk of complication, including DRESS syndrome, renal or liver impairment, C Diff diarrhoea or allergy. There is increasing evidence that managing complex infections in a multidisciplinary team format reduces both treatment costs and complications compared to antibiotic therapy commenced by the orthopaedic surgeon alone ^9,32.

Placing local antibiotics to the fracture site and surrounding tissues during debridement surgery has become increasingly common in modern surgical practice. Local antibiotics can achieve higher local tissue concentrations with a lower risk of systemic complications compared to intravenously administered antibiotics.³³ These antibiotics may be administered in powder or liquid form or if sustained release is desired, using a carrier. Available carriers include Polymethylmethacrylate (PMMA) cement, ceramics, such as calcium sulphate, or hydrogels. Difference in cost, the heat sensitivity of the antimicrobial, need for removal or time for resorption may influence which carrier is selected to deliver the local antibiotics. Ceramics have been associated with increased wound discharge compared to PMMA spacers; robust wound coverage including the deep fascia is imported to minimise this discharge. Pre-clinical research would suggest that the concentration of local antibiotics is unaffected by negative pressure dressings should one be needed to manage the wound discharged caused by the ceramic carrier.³⁴

PMMA is falling out of favour of calcium sulphate carriers, due to shorter (appx 2 weeks) MIC levels. Calcium sulphate can provide up to 6 weeks (Aikin et al.) of local MIC level's and with an additional advantage of flexibility of antibiotics. Although due to cost and availability PMMA continues to be used is LMIC's.

There is prospective evidence that topical antibiotics reduce the risk of infection in tibial fractures.³⁵ While the available evidence suggests that local antibiotics are safe with a low-risk of adverse effects, the majority of studies have focused on the efficacy of local antibiotics as an adjunct to systemic therapy. Avoiding long courses of systemic antibiotic therapy may be beneficial, with up to 15 % of patients with a bone or joint infection experiencing a complication secondary to their antibiotic therapy.³⁶ The SOLARIO trial is currently comparing short versus long systemic antimicrobial treatment for bone infection, in patients who have had local antibiotics implanted.³⁷ Shorter courses of systemic antibiotics may lower patient burden and have a beneficial impact of healthcare costs after FRI.

3.5 Antibiotic nail

The use of an antibiotic impregnated cement coated IM nailing for infected non-union of tibia and femur fractures has been well-documented in the literature.³⁸ Such nails allow for the management of dead space and the deep delivery of antibiotics where there is no critical bone defect. Unlike beads or cement block spacers, the cement nail provides osseous stability across the fracture site. Particular problems encountered with this technique include broken metalwork and need for subsequent removal, which can prove challenging.³⁸

There is a variety of commercially available antibiotic nails now available; alternatively, the surgeon may create their own custom-made nail. Methods described include selecting an appropriately sized nail and coating it in antibiotic impregnated bone cement, or utilising an Illizarov threaded rod again coated with antibiotic cement. A female post attached to the end of the threated rod allows for easer extraction. Studies comparing commercial to custom nails are few in number and retrospective in nature and show no significant difference in either infection eradication or rate of non-union.³⁹ Custom made nails may also prolong surgical time and some authors have reported de-bonding of the cement from the nail at the time of extraction.³⁹ To date, no prospective RCT has compared a commercial to a surgeon-created antibiotic nail.

3.6 Management of bone defects

Bone defects may be encountered either following the index surgery or may be created as part of an appropriate debridement of non-viable bone. Over 50 % of patients with a greater than 1 cm bone defect involving more than 50 % of the cortical bone will require a secondary surgery to achieve bony union.⁴⁰ Bone defects greater than 3 cm are considered critical defects and are unlikely to heal spontaneously.⁴¹ A variety of surgical techniques have been described; to date there is little evidence to support one technique for all circumstances.⁴² The most commonly utilised techniques are described below.

3.7 Masquelet

Traditional bone grafting techniques are limited by uncontrollable graft resorption.⁴³ The Masquelet technique of membrane induced osteogenesis is a two stage procedure in which a PMMA cement spacer is placed into the bone defect after debridement.⁴⁴ In the first stage, the bone ends are stabilised and robust soft tissue coverage is then performed. The PMMA induces a foreign body reaction that results in thick vascularised membranes into which graft material is placed following removal of the PMMA spacer. This second stage procedure is recommended to be performed at 6–8 weeks following the first, when the membranes are considered to be biologically active.

There is limited literature focusing on Masquelet for FRI that results in a critical bone defect following open fracture. Successful union and infection eradication was reported by Revan et al. in 72 % of cases of infected tibial non-union.⁴⁵ Higher rates of union have been reported for contaminated segmental G-A grade 3 open fractures requiring both flap and microvascular repair as simultaneous procedures.⁴⁶ The graft can be supported by either internal or external fixation; internal fixation may produce less patient morbidity but does increase the risk of re-implanting metalwork into a potentially infected field. Where internal fixation is selected, intramedullary nailing appears superior to plate fixation when possible to implant.⁴⁷

3.8 Bone transport

Bone transport utilises the principle of distraction osteogenesis to produce new bone formation during lengthening of the affected bone. Bone transport can be performed using either a monorail, circular frame or over a nail. The accepted safe rate of transport is 1 mm a day. While the transport process can be long, high rates of union have been reported even following segmental excision of bone after tibial FRI. Traditional bone transport relies on external fixation essentially dragging the transport segment to the docking site. This results in longitudinal scars during the transport process. Unique complications can arise when this external fixation moves through compromised skin or a flap. In 1998, Weber et al. modified the previous Ilizarov internal bone transport technique by utilising cables and pullies allowing transport to be performed without excursion of external fixation through the skin.⁴⁸ While published case series appear to be equivalent to traditional transport while avoiding additional soft tissue complications, there is currently no robust prospective evidence comparing the two methods.⁴⁹

A variety of other modifications to traditional bone transport are described in the literature. The defect site can be shortened to allow for easier soft tissue coverage before being gradually re-lengthened. Multi-level bone transport may be employed to shorten the overall transport and frame time.⁵⁰ Again, high rates of union have been reported, although rates of complications reach up to 45 %.⁵⁰ Poor quality of regenerate, soft tissue complications from pin sites and docking site non-union can be common complications after bone transport.

3.9 Vascularised bone graft

A less common strategy to bridge a critical bone defect is vascularised bone grafting (VBG). VBG is performed as a microsurgical procedure harvesting both a bony segment and the supplying bold vessels. Skin, subcutaneous tissue and muscles may also be harvested as a composite graft to provide soft tissue coverage in addition to bridging the critical bone defect. The fibula iliac crest and ribs have all been descried as potential donor sites for VBG ^51,52. Grafting of defects up to 18 cm have been described.⁵³ The results from published studies appear largely comparable to the Masquelet technique, however have the additional potential complication of donor site morbidity due to the nature of the grafting process. Trends towards a faster rate of union are tempered with trends to higher rates of recurrent deep infection in the VBG groups within the available literature.⁵⁴ The procedure is also technically demanding and must be performed in a centre with the appropriate microvascular expertise. The role of VBG as a routine management of critical bone defects following FRI remains unclear given the lack of prospective comparative studies within this challenging population.