Restoring the Anatomy of Long Bones with Large Septic Non-Union Defects with the Masquelet Technique

Abstract

Objectives: Septic non-union in long-bone fractures represents a challenging clinical entity. Management of lower extremity segmental bone defects, aiming to restore functional anatomy, remains extremely difficult and controversial. Masquelet technique is a reconstruction method for large diaphyseal bone defects, based on the notion of the induced membrane. The principle of the induced membrane is to create a foreign body reaction by placing cement spacer in the bone defect. The purpose of this study was to assess the success rate of induced membrane technique (IMT) in treating lower extremity large bone defects due to septic non-union.

Methods:This is a retrospective observational study performed in a single referral center in France, Europe, which is specialized in complex bone and joint infections. All patients operated for septic non-union were identified from a prospectively maintained database. Patients treated with the IMT for septic femoral or tibial non-union between 2013 and 2017 were enrolled in this study. Exclusion criteria were infection of a continuous bone, aseptic non-union, or patients with less than one year of follow-up after antibiotic treatment ending.

Results:Twenty-three cases (19 patients) with an average age of 41.3 years were included in the present study. There were 19 tibial and four femoral fractures. The mean bone defect was 65.3 mm. The mean time interval from initial trauma to the first surgical phase was 17 months, while that between the two surgical phases was 77.7 days. After the first surgical phase, samples were positive in 13 cases (68.5%), isolating Staphylococcus (26%) and more than one pathogen in 22% of cases. Bone union was successful in 16 of 23 cases (69.6%, 14 patients). There were seven failures: five amputations due to mechanical and/or infection-related failure and two failed unions.

Conclusion:This study found that 69% of cases with septic non-union of tibial or femoral fracture treated with the two-step surgical protocol achieved bone union and infection eradication within about 13.2 months after the second stage of the procedure. The study revealed promising results in patients suffering large-size bone defect; hence, the IMT may prove beneficial in the management of such cases.

Keywords:septic non-union, surgical anatomy, osseous infection, bone defects, referral center for bone and joint complex infection.

INTRODUCTION

The Masquelet technique or induced membrane technique (IMT) was described in 1986 and represents a twostep procedure for treating non-unions, especially in long-bones, and the concomitant osseous defects (1). It is a relatively new approach for reconstructing segmental bone defects, caused by resection of infectious, tumorous and/ or traumatic injured bone tissue (2).

The general steps of IMT have been documented and are well established. Nevertheless, the details of the procedure are highly variable and dictated more by surgeon preference and patient conditions, rather than evidence (1-3). During the first stage of the procedure, the affected bone is resected and polymethyl methacrylate (PMMA) cement is placed to fill the defect. Internal or external fixation could be performed, according to surgeon preference and the patient’s condition (3). Following the first step of the technique, a period of at least 4-8 weeks is given, so that the autologous foreign-body “induced” membrane around the cement spacer may be formed. The second stage of the procedure includes the cement spacer removal and bone graft placement in the cavity (1, 2).

Septic non-union represents a major complication for long bone fractures. Large osseous defects may be present, complicating bone healing and the patient’s quality of life (4). Therapeutic interventions in these cases require medical and surgical treatment and multidisciplinary approach is of utmost importance.

There are scarce data regarding the IMT, especially in treating septic non-union and the concomitant bone defects in long bones. Many issues, including the time-interval between the two stages, the factors that affect infection’s eradication and bone union, remain vague in the literature so far. There is only a few small series describing the outcomes of this technique in septic union of long bones. Furthermore, many variants are depended on surgeon’s preference, such as the use and type of fixation, the impregnation of cement with antibiotics and type of bone graft.

The purpose of the present study was to evaluate the efficacy of the IMT in treating femoral and/or tibial large bone defects and restoring the functional anatomy due to septic non-union as well as in eradicating the infection.

METHODS

This is an observational retrospective study of a prospectively collected database. From January 2013 to December 2017, all femoral and tibial septic non-union cases treated with two-stage IMT were enrolled in research.

Data of patients’ age, comorbidities and history of trauma were recorded. Risk factors were evaluated using the Non-Union Scoring System (NUSS) on 100 points, including bone (54 points), soft tissue (12 points) and patient-specific factors (34 points). The NUSS score has been useful when assessing patients with fracture non-union. It takes into consideration the bone quality, bone involvement, soft tissue management and patients’ risk factors, while higher scores imply a higher difficulty to procure union. This score categorizes patients into four groups, including group 1 (0-25), group 2 (26-50), group 3 (51-75) and group 4 (76-100), with group 4 having the highest severity.

Meticulous assessment was performed prior to treatment initiation at the end of antimicrobial treatment and then at least 12 months after the final surgical intervention. Bone consolidation was assessed with orthogonal X-rays, defined by bringing callus of two cortices on the two X-ray views (antero-posterior and lateral). Functional results and bone quality were evaluated with the Paley-Maar score. Table 1 summarizes the values of this score, which was initially described for the segmental transport used by the Ilizarov technique and fixator.

All patients underwent the same rigorous surgical and medical protocol. Their medical history was reviewed during a multidisciplinary meeting, involving orthopaedic and plastic surgeons, infectious disease specialists as well as microbiologists.

The procedure was performed in two steps: initially, resection of the infectious site and PMMA cement spacer loaded with Gentamicin (PALACOS R+G®, Heraeus, Germany) was placed. During this step, at least five separate samples were sent for microbiological and histological examination. Stabilization with external fixator could be used, according to surgeon’s preference, while skin closure was performed with direct suture when it was permitted by soft tissue quality. If not, a pedicle flap or a free flap was performed by plastic surgeon. Then, patients received intravenous (iv) antimicrobial treatment for 15 days, including, initially, empirical treatment with vancomycin and piperacillin/tazobactam, while treatment was adapted according to microbiological and histological results. Following hospital discharge, patients were commenced on per os appropriate antimicrobial treatment for at least 30 days. Infection eradication was assessed according to patient’s signs and symptoms, laboratory examination and imaging results.

Following infection eradication, the reconstruction phase was initiated. The second stage of the surgical intervention was performed, including careful incision of the induced membrane, removal of the cement spacer and placement of autologous cancellous bone combined with bonegraft substitute [Cerament® (bioresorbable hydroxyapatite/ calcium sulphate cement)]. Finally, definitive osteosynthesis with locking plate was performed (Figure 1).

The time interval between the two surgical phases was at least six weeks, while it related to the microbiological diagnosis, the chronicity of the infection and the healing of soft tissues.

The study has been approved from the scientific committee of Hospices Civils de Lyon. This study has also been included in the registry of clinical trials as NCT02817711.

RESULTS

A total of 19 patients with 23 septic non-union tibial and/or femoral cases were included in this cohort.

The studied population’s main characteristics are highlighted in Table 2. Most patients were males (13; 68.4%). Patients’ mean age was 42.7 years [standard deviation (SD) 13.6], while their mean body mass index (BMI) was 25.4 kg/m² (SD 5.1). The mean follow-up, after the second stage of the operation, was 44.6 months (SD 21.01). Fourteen patients (73.6%) were heavy smokers.

Most cases referred to tibial (19; 82,6%), followed by femoral fractures (4; 17.4%), while the majority of the cases (17; 74%) were related to open fractures.

The mean number of surgical interventions prior to IMT was 2.8 procedures (SD 1.7), while the mean bone defect was 65.3 mm (SD 15.2). At the time of inclusion to the protocol, the mean NUSS score was 54.6 (SD 13.4). According to NUSS groups, one patient (4.3%) was in group 1, seven subjects (30.4%) in group 2, 15 patients (65.2%) in group 3 and none in group 4.

The first surgical phase was performed after 17 months (SD 14.1) from the initial trauma. During the first stage, external fixation was performed in 16 (69.6%) of cases. In five cases (21.7%) during the first step, a musculocutaneous flap was also required for wound coverage. After the first surgical phase samples were positive in 13 cases (68.5%), isolating ` in six cases (31.6%), followed by methicillin resistant Staphylococcus aureus (MRSA) and Proprionibacterium acnes in three cases each (15.8%). In five cases (26.3%), more than one organism was isolated.

Dual or triple appropriate antimicrobial therapy was administrated awaiting the second phase, according to multidisciplinary team discussion considering the microbiological and histological results. After the second surgical phase, samples were positive in four cases, isolating the same organism as in first phase in one case (4.3%).

In the second phase, bone autograft was harvested from the anterior iliac crest in two patients (10.5%) and from both anterior and/or posterior iliac crests in 17 subjects (89.5%), while allograft (bone bank) was additionally used in 18 patients (94.7%). Bone stabilization was achieved by internal fixation with locking plates. The mean period between the two surgical steps was 77.7 days (SD 46.2).

Bone union was achieved in 16 of 23 cases (14 patients, 69.6%) at a mean of 13.2 months (SD 7.77) after the second stage of the procedure. According to the Paley-Maar Bone Score, five patients (21.7%) had an excellent score, eight subjects (34.8%) a good score and three patients (13%) a poor score. There were two recurrences of sepsis, which were treated with below-knee amputation. In terms of mechanical-related failure (loosening of the hardware or plate breakage) there were five cases leading to three amputations and two non-unions.

DISCUSSION

The treatment of large septic femoral or tibial non-union is challenging and frequently requires multidisciplinary approach. In France, specialized team treatment of these infections is the norm with the creation of referral centers for bone and joint complex infections, eg, “Centres de réféférence des infections ostéo-articulaires complexes” (CRIOAc) in 2008. These centers permit a multidisciplinary team treatment associating orthopedic surgeons, infectious disease specialists, microbiologists and plastic surgeons.

The main finding of the present study is that the IMT achieved 69.5% union rate at a mean follow-up of 13.2 months in patients suffering large bone defects due to long-bone septic non-unions. Siboni et al evaluated 19 cases of septic non-union of the tibia, with a mean size of bone defect at 52.4 mm after the first stage, having 89% successful bone union in a period of 16 months (7). Pesciallo et al (8) studied 21 cases of infected femoral and tibial bone defects with a median bone defect size at 45 mm and found union rate at 95.23% over a median time of 7 (range 6-12) months. In the present study, 23 cases with mean bone defect of 65.3 mm were evaluated.

The present cohort showed that the majority of patients (14; 73.6%) with large septic tibial or femoral non-union were heavy smokers. Furthermore, there were two recurrences of sepsis, which were treated with below-knee amputation, while in terms of mechanical-related failure there were five cases leading to three amputations and two non-unions. It should be noted that these “failed” cases were all among the heavy smokers group. This has been already documented. Siboni et al reported that only one of the eight patients achieving union after the second phase without any supplementary procedure was an active smoker (7). In addition, Bauer et al showed that most patients with infected non-union of the tibia or the femur were active smokers (9). Nevertheless, in a retrospective study of 36 patients evaluating the vascularity of the induced membrane, Niikura et al found that smoking did not affect the IM vascularity and suggested a relationship between vascularity and osteogenesis inside the IM (10).

In the present study, after the first stage, the most commonly isolated organism was Staphylococcus epidermitis (six cases; 31.6%), followed by methicillin-resistant Staphylococcus aureus (MRSA) (3; 15.8%). Similarly, Scholz et al (11) identified the types of organisms in a series of 13 cases of diaphyseal bone defects treated with the IMT. They also isolated Staphyloccus (69%) as the primary pathogen and 23% of those was MRSA, followed by Enterococci (30.8%). S. epidermitis has also been associated with orthopaedic implant related infections. In the present study, only one case with multi-bacterial infection which led to amputation was recorded. However, due to the small number of cases, a correlation between infection with multiple organisms and failure of the technique could not be made.

The type of fixation used for limb stabilization also represents an understudied issue. In this study, external fixation was used in most of cases (69.6%) in the first stage of the procedure. As for the second stage, internal fixation by locking plate was the only stabilization method used by us. Overall, five mechanical failures (21.7%) and 69.5% union rate were recorded. In a retrospective multicenter cohort study, Morwood et al (12) found 85.9% union rate in a series of 64 cases sustaining significant femur and/or tibial bone loss where stabilization was assured by plate. However, in contrast with the present study, the bone defects in those patients were not due only to septic non-union. Therefore, it could be assumed that infection, as cause of osseous defect, could lead to less favorable outcomes. In a systematic review evaluating the results of IMT in the management of critical size bone defects in children, Aurégan et al (13) found no difference in favor of a certain type of fixation used for the second stage. However, it was noted that osteosyntheses deemed “unstable” by the surgeon were more likely to fail. Different fixation methods based on the stage of the procedure have been recommended. Initially, rigid stabilization is favored immediately after the second stage to promote revascularization. Subsequently, a shift towards more flexible stabilization – even with an external fixation – is preferred to enhance corticalization by adjusting the fixation rigidity and introducing dynamization by gradually removing the rods (14).

Reconstruction of long bone defects and restoration of the anatomy and function of of the lower extremity remains a major challenge for an orthopedic surgeon. Commonly, these large bone defects are associated with damage or loss of the surrounding wound tissues. In the literature, in most series the bone defects were classified into four types, according the Karger’s classification (2): type I less than 20 mm, type II between 20-50 mm, type III 50-100 mm and type IV more than 100 mm. Very few studies have provided detailed information regarding the use of IMT for large bone defects. The herein study showed encouraging results of IMT patients with bone defects among class 3 and above.

Generally, the Ilizarov technique and the vascularized fibular autograft represented the most commonly used procedures for the reconstruction of large bone defects (15). These techniques are associated with an increased risk of fracture and are highly demanding in terms of micro-surgical approach. Masquelet technique may provide potential advantages in these challenging cases, creating the favorable biologic cavity to improve bone regeneration and its osteoconductivity regardless of the defect size (16). Bone graft substitute may also be used to complete the insufficient amount of autograft (17, 18). In the present cohort this was necessary for almost all patients (18 subjects; 94.7%).

The present study has some limitations. It is a retrospective single center study which does not comprise a control group. Furthermore, it is a relatively small cohort. Nevertheless, only few such small cohorts exist so far in the literature – hence, it is important to report the outcomes of IMT technique to provide evidence-based data for treatment options of an extremely challenging osseous infection.

CONCLUSION

The present study reported that 69% of patients treated at a referral center for bone and joint complex infections using the IMT achieved bone union and infection eradication within about 13.2 months after the second stage of the procedure. A polymicrobial infection may have an impact on the resolution of the infection and the bacterial profile should be investigated, while smoking seems to lead to non-favorable outcomes. A multidisciplinary approach is of utmost importance for the management of these challenging cases for achieved desired outcomes.

Conflicts of interest: none declared.

Financial support: none declared.

Institutional review board statement: The study has been approved by the scientific committee of Hospices Civils de Lyon, France, and has also been included in the registry of clinical trials as NCT02817711.

TABLE 1.

Paley-Maar bone score

FIGURE 1.

A: anteroposterior (AP) view of the right distal tibia; screw breakdown and sequestrum due to Staphylococcus epidermidis are shown. B: AP view of the first stage of Masquelet technique, plates removal, external fixation and spacer with Gentamicin. C: AP view, second stage, infection eradication, fixation with locking plate. D: AP view, bone union after three months

TABLE 2.

Patients’ main characteristics