Advertisement

Comparison of internal and external fixation after debridement in the Masquelet technique for Cierny-Mader type IV tibial post-traumatic osteomyelitis

Open AccessPublished:November 11, 2022DOI:https://doi.org/10.1016/j.injury.2022.11.030

      Highlights

      • Masquelet technique become one of the most utilized procedures for treating chronic osteomyelitis.
      • Both methods were effective for C-M type Ⅳ chronic post-traumatic osteomyelitis of the tibia.
      • Internal fixation did not increase the rate of infection and not impact on bone healing.
      • Masquelet technique has a lower complication rate, infection recurrence is the most serious one.

      Abstract

      Objective

      To compare the effect of internal fixation vs. external fixation after debridement in stage I of the Masquelet technique for Cierny-Mader (C-M) type Ⅳ chronic post-traumatic tibial osteomyelitis.

      Methods

      This retrospective observational study included patients with tibial osteomyelitis who underwent staged treatment with the Masquelet technique between January 2016 and June 2020 at the 920 Hospital of Joint Logistic Support Force of the PLA. The patients were grouped according to the fixation they received after stage I. Infection recurrence, time to radiological bone healing and full weight-bearing, self-rating anxiety scale (SAS) score, Hospital for Special Surgery (HSS) Knee Score, and American Orthopedic Foot and Ankle Society (AOFAS) Ankle-Hindfoot Score were compared.

      Results

      Sixty-three patients were included (50 males and 13 females). There were 40 and 23 patients with internal and external fixation, respectively. There were no significant differences between the two groups regarding the preoperative and intraoperative data (all P>0.05). After stage I operation, the infection control rates were 85.0% and 82.6% in the internal and external fixation groups (P=0.803), and these rates were 92.5% and 95.7% after stage II (P=0.621). There were no differences in the SAS scores (P=0.278), time to radiological union (P=0.795), time to full weight-bearing (P=0.725), AOFAS scores (P=0.302), HSS scores (P=0.085), and complication rates (P=0.593). There were 27 times complications in 19 patients, with an incidence of 42.9%, without significant differences between groups.

      Conclusion

      There were no differences between the two fixation methods after debridement in stage I of the Masquelet technique for C-M type Ⅳ chronic post-traumatic tibia osteomyelitis.

      Keywords

      Introduction

      Chronic post-traumatic osteomyelitis is the most difficult scenario for most orthopedic and trauma surgeons [
      • Conterno L.O.
      • Turchi M.D.
      Antibiotics for treating chronic osteomyelitis in adults.
      ,
      • Panteli M.
      • Giannoudis P.V.
      Chronic osteomyelitis: what the surgeon needs to know.
      ]. Because of the anatomical characteristics of the tibia (i.e., poor soft tissue coverage of the anterior tibia and poor blood supply), it is frequently prone to serious soft tissue damage, bone exposure, and contact with exogenous pathogens after trauma, making the tibia the most common site of post-traumatic osteomyelitis [
      • Yang J.
      • Yao J.L.
      • Wu Z.Q.
      • Zeng D.L.
      • Zheng L.Y.
      • Chen D.
      • et al.
      Current opinions on the mechanism, classification, imaging diagnosis and treatment of post-traumatic osteomyelitis.
      ].
      The Cierny-Mader (C-M) classification is based on the anatomy of the bone infection and the host's physiology [
      • Cierny 3rd, G.
      • Mader J.T.
      • Penninck J.J.
      A clinical staging system for adult osteomyelitis.
      ]. It can be used to provide complete treatment guidelines and predict prognosis [
      • Cierny 3rd, G.
      • Mader J.T.
      • Penninck J.J.
      A clinical staging system for adult osteomyelitis.
      ]. C-M type-IV chronic osteomyelitis is the most severe type of post-traumatic osteomyelitis, with extensive bone and soft tissue invasion [
      • Cierny 3rd, G.
      • Mader J.T.
      • Penninck J.J.
      A clinical staging system for adult osteomyelitis.
      ]. It is usually difficult to eliminate the infection non-surgically, and it is widely recognized that the surgical strategy of staged therapy is preferable to conservative treatments for C-M IV chronic osteomyelitis [
      • Cierny 3rd, G.
      Surgical treatment of osteomyelitis.
      ]. A wide range of surgical procedures has been documented. The most often reported procedures include bone transfer, free vascularized fibular graft, and open-air cancellous Papineau grafting [
      • Dendrinos G.K.
      • Kontos S.
      • Lyritsis E.
      Use of the Ilizarov technique for treatment of non-union of the tibia associated with infection.
      ,
      • Gao Y.S.
      • Ai Z.S.
      • Yu X.W.
      • Sheng J.G.
      • Jin D.X.
      • Chen S.B.
      • et al.
      Free vascularised fibular grafting combined with a locking plate for massive bone defects in the lower limbs: a retrospective analysis of fibular hypertrophy in 18 cases.
      ,
      • Beals R.K.
      • Bryant R.E.
      The treatment of chronic open osteomyelitis of the tibia in adults.
      ].
      In recent years, the Masquelet technique has increasingly become one of the most often used procedures for treating chronic osteomyelitis [
      • Siboni R.
      • Joseph E.
      • Blasco L.
      • Barbe C.
      • Bajolet O.
      • Diallo S.
      • et al.
      Management of septic non-union of the tibia by the induced membrane technique. What factors could improve results?.
      ,
      • Sasaki G.
      • Watanabe Y.
      • Miyamoto W.
      • Yasui Y.
      • Morimoto S.
      • Kawano H.
      Induced membrane technique using beta-tricalcium phosphate for reconstruction of femoral and tibial segmental bone loss due to infection: technical tips and preliminary clinical results.
      ,
      • Raven T.F.
      • Moghaddam A.
      • Ermisch C.
      • Westhauser F.
      • Heller R.
      • Bruckner T.
      • et al.
      Use of Masquelet technique in treatment of septic and atrophic fracture nonunion.
      ,
      • Wu H.
      • Shen J.
      • Yu X.
      • Fu J.
      • Yu S.
      • Sun D.
      • et al.
      Two stage management of Cierny-Mader type IV chronic osteomyelitis of the long bones.
      ]. It is divided into two stages. In the first stage, after thorough debridement, the bone defect is filled with antibiotic bone cement to control infection and to stimulate the surrounding soft tissue to form a reactive membrane by 6-8 weeks later. After controlling the infection, the bone cement is removed, and bone reconstruction is performed by bone grafting in the induced membrane.
      Type IV chronic osteomyelitis is intrinsically unstable or rendered unstable because of debridement and segmental bone resection, which need additional fixation to stabilize the proximal and distal bone segments [
      • Cierny 3rd, G.
      Surgical treatment of osteomyelitis.
      ]. The commonly used stabilization methods include external and internal fixation [
      • Giannoudis P.V.
      • Harwood P.J.
      • Tosounidis T.
      • Kanakaris N.K.
      Restoration of long bone defects treated with the induced membrane technique: protocol and outcomes.
      ,
      • Masquelet A.C.
      • Kishi T.
      • Benko P.E.
      Very long-term results of post-traumatic bone defect reconstruction by the induced membrane technique.
      ,
      • Alemdar C.
      • Azboy I.
      • Atic R.
      • Ozkul E.
      • Gem M.
      • Kapukaya A.
      Management of infectious fractures with "Non-Contact Plate" (NCP) method.
      ,
      • Yilihamu Y.
      • Keremu A.
      • Abulaiti A.
      • Maimaiti X.
      • Ren P.
      • Yusufu A.
      Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator.
      ]. Still, the optimal type of fixation has been subject to debate, and practice varies greatly from one surgeon to another. For example, internal fixation is often deemed more stable, but there is a risk of biofilm formation, and such biofilm is difficult to manage using antibiotics [
      • Giannoudis P.V.
      • Harwood P.J.
      • Tosounidis T.
      • Kanakaris N.K.
      Restoration of long bone defects treated with the induced membrane technique: protocol and outcomes.
      ,
      • Masquelet A.C.
      • Kishi T.
      • Benko P.E.
      Very long-term results of post-traumatic bone defect reconstruction by the induced membrane technique.
      ,
      • Alemdar C.
      • Azboy I.
      • Atic R.
      • Ozkul E.
      • Gem M.
      • Kapukaya A.
      Management of infectious fractures with "Non-Contact Plate" (NCP) method.
      ,
      • Yilihamu Y.
      • Keremu A.
      • Abulaiti A.
      • Maimaiti X.
      • Ren P.
      • Yusufu A.
      Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator.
      ].
      It could be hypothesized that internal fixation is a reliable and stable method after debridement, will not increase the rate of infection recurrence, and can achieve a high rate of bone healing. Therefore, this study compared the effectiveness of internal vs. external fixation for C-M type IV tibial chronic traumatic osteomyelitis after debridement in the first stage of the Masquelet technique.

      Patients and methods

      Study design and patients

      This retrospective observational study included patients with tibial osteomyelitis who underwent staged treatment with the Masquelet technique between January 2016 and June 2020 at the 920 Hospital of Joint Logistic Support Force of the PLA.
      The inclusion criteria were 1) tibial post-traumatic osteomyelitis in the diaphysis or metaphyseal region, 2) anatomic C-M type IV, and 3) bone defect of no less than 4 cm after segmental bone resection during debridement. The exclusion criteria were 1) local defects after debridement (i.e., during osteomyelitis debridement, the bone was debrided without a segmental “en bloc” resection, but a small part of cortex was preserved), 2) physiological C-M class C, 3) intra-articular infections, and 4) external fixation was used in stage II, 5) incomplete clinical data.

      Surgical procedure

      The routine surgical method used during the study period was similar to that reported in the literature [
      • Yilihamu Y.
      • Keremu A.
      • Abulaiti A.
      • Maimaiti X.
      • Ren P.
      • Yusufu A.
      Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator.
      ]. The operation was divided into two stages. In stage I, internal fixation (locking plate) or external fixation (circular external fixator, monolateral external fixator, or locking compression plate as external fixator) was used to stabilize the bone ends. In stage II, internal fixation (locking plate, intramedullary nail, or locking plate + intramedullary nail) and autologous iliac bone/allogeneic bone graft were used.
      In stage I, thorough debridement was first carried out. According to the preoperative plan, the involved bone was resected segmentally (en bloc) until there was punctate bleeding in the broken end (Paprika sign). The deep tissue was harvested for bacterial culture and pathological examination. In the internal fixation group, the bone was fixed with a locking plate. The reconstruction locking plate or humeral locking compression plate was the most commonly choice because of the relatively small volume. The appropriate plate was chosen so that two to four screws could be placed on each side of the two bone ends. The bone defect was filled with vancomycin-loaded bone cement to wrap the broken bone ends and the full length of the plate. In the external fixation group, the appropriate external fixation device was selected after debridement. The fixation pin was placed away from the bone defect. The axis between the broken ends of the bone defect was adjusted. Then, the bone defect was filled with bone cement. If there was a previous external fixation before debridement, it could be retained and adjusted appropriately to facilitate operation. A flap could be used for those who had soft-tissue defects after debridement, and the skin could not be sutured directly. After the operation, routine broad-spectrum antibiotics were applied intravenously. If the bacterial culture and drug sensitivity tests were positive, sensitive antibiotics were selected for 14 days. Later, oral broad-spectrum antibiotics were administered for 4 weeks. White blood cells (WBCs), C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) were checked every 2 weeks. Only after they were normal for more than two consecutive times could the stage II operation be performed. In case of infection recurrence or flap necrosis, debridement was required again until the infection was controlled and the wound was covered with additional flaps.
      Stage II operation was generally performed 6-8 weeks after infection control of stage I. In the external fixation group, the external fixation was removed 1 week before the second operation and changed to a brace. The bone cement and internal fixation were removed. The locking compression plate (LCP) or/and intramedullary nail were placed to fix the two broken bone ends. The cancellous bone of the iliac bone harvested in advance was cut into 2-mm3 pieces and placed to fill the bone defect. If the autogenous bone was not enough, allogeneic bone could be used, and the ratio of autogenous to allogeneic bone did not exceed 3:1. Antibiotics were continued for another 2 weeks. No weight-bearing was allowed on the affected limb within 3 months after the operation. It is because, absolute stability should be ensured as necessary to promote graft union and revascularization by the membrane in this stage [
      • Masquelet A.
      • Kanakaris N.K.
      • Obert L.
      • Stafford P.
      • Giannoudis P.V.
      Bone repair using the Masquelet technique.
      ,
      • Masquelet A.C.
      Induced Membrane technique: pearls and pitfalls.
      ].

      Data collection and definition

      X-ray films of the affected limbs and WBCs, ESR, and CRP were checked at 1, 2, 3, 6, 9, and 12 months after the operation and then every 3-6 months. X-ray films were taken to evaluate bone healing.
      Data were collected, including recurrence of infection, Self-rated Anxiety Scale (SAS) score [
      • Yilihamu Y.
      • Keremu A.
      • Abulaiti A.
      • Maimaiti X.
      • Ren P.
      • Yusufu A.
      Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator.
      ], the recurrence rate after stages I and II, radiographic bone healing time and full weight-bearing time, Hospital for Special Surgery (HSS) Knee Score [
      • Zhang Z.
      • Chai W.
      • Zhao G.
      • Zhang Q.
      • Chen Z.
      • Wang X.
      • et al.
      Association of HSS score and mechanical alignment after primary TKA of patients suffering from constitutional varus knee that caused by combined deformities: a retrospective study.
      ], American Orthopedic Foot and Ankle Society (AOFAS) Ankle-Hind Foot Score [
      • Ceccarelli F.
      • Calderazzi F.
      • Pedrazzi G.
      Is there a relation between AOFAS ankle-hindfoot score and SF-36 in evaluation of Achilles ruptures treated by percutaneous technique?.
      ], and complications, including infection recurrence, donor site pain, delayed wound healing of donor site, length discrepancy, refracture, and ankylosis. The criteria for the recurrence of infection were local redness and swelling, purulent secretion or sinus tract, or increased WBCs, CRP, and ESR. Radiographic bone healing was defined as three sides of cortical bridging on the anteroposterior (AP) and lateral X-ray films [
      • Simpson A.
      • Robiati L.
      • Jalal M.M.K.
      • Tsang S.T.J.
      Non-union: Indications for external fixation.
      ].

      Statistical analysis

      SPSS 25.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis. The Shapiro-Wilk test was used to test the normality of the data. The continuous data conforming to the normal distribution were expressed as means ± standard deviation and analyzed using Student's t-test for comparison between groups. The categorical data were expressed as n (%) and analyzed using the chi-square test or Fisher's exact test. Two-sided P-values <0.05 were considered statistically significant.

      Results

      Twelve patients (ten with internal fixation and two with external fixation) with missing outcome data, 7 with external fixation in the stage I and II, and 4 with local defects after debridement were excluded. Sixty-three patients were included (50 males and 13 females). After admission, 6 patients were first treated with initial debridement and external fixation, but a few days later, the external fixation was changed to internal fixation in the stage I, and internal fixation was also used in the stage II. Finally, 40 patients were included in the internal fixation group and 23 patients in external fixation group. They were 37.2±7.3 years old (range, 17-55 years) and had an average bone defect length of 7.5±1.4 cm (range, 4.0-12.0 cm). In the external fixation group, the patients were 36.6±7.0 years old (range, 26-53 years) and had an average bone defect length of 6.9±1.3 cm (range, 5.0-10.0 cm). There were no significant differences in preoperative and intraoperative data between the two groups (Table 1). The duration of the spacer was 9.2±4.0 vs. 8.4±1.6 weeks (P=0.384).
      Table 1Characteristics of the patients
      ItemsInternal (n=40)External (n=23)P
      Sex0.870
       Male32 (80.0%)18 (78.3%)
       Female8 (20.0%)5 (21.7%)
      Age (years)37.2±7.336.6±7.00.754
      Type of injury0.865
       Open injury27 (67.5%)16 (69.6%)
       Closed injury13 (32.5%)7 (30.4%)
      Initial injury mechanism0.356
       Traffic accident19 (47.5%)9 (39.1%)
       Falling14 (35.0%)12 (52.2%)
       Direct violence7 (17.5%)2 (8.7%)
      Number of previous operations5.0±1.85.3±1.60.609
      Sinus tract33 (82.5%)20 (87.0%)0.641
      Previous fixation0.148
       No6 (15%)2 (8.7%)
       Plate fixation15 (37.5%)9 (39.1%)
       Intramedullary nail fixation6 (15%)0 (0%)
       External fixation13(32.5%)12 (52.2%)
      Location12/17/115/9/90.597
       Proximal12 (30.0%)5 (21.7%)
       Medial17 (42.5%)9 (39.1%)
       Distal11 (37.5%)9 (39.1%)
       Multiple injuries13 (42.5%)7 (30.4%)0.865
      Soft tissue defect21 (52.5)14 (60.9%)0.520
      Bone defect size (cm)7.5±1.46.9±1.30.145
      Flap coverage15 (37.5%)11 (47.8%)0.423
      Duration of spacer (weeks)9.2±4.08.4±1.60.384
      In 35 patients (21 in the internal fixation group and 14 in the external fixation group), there was a soft tissue defect after debridement, with an area of 2.0 cm × 2.5 cm to 18.0 cm × 6.0 cm. The length of the bone defect was 4-12 cm, with an average of 7.3 cm. Flap coverage was performed in 26 patients, including 22 with a pedicled fascial flap and four with a free flap.
      The average follow-up was 32.9 (range, 22-73) months. After stage I operation, there were 10 cases of infection recurrence, with six cases in the internal fixation group and four in the external fixation group. The infection control rates were 85.0% and 82.6% in the internal and external fixation groups (P=0.803). After debridement, infection was controlled in five cases. The others underwent multiple debridement, and the infection was ultimately controlled. In the external fixation group, there were no pin tract infections. The duration of cement spacer between stage I and stage II were 9.2±4.0 and 8.4±1.6 weeks, respectively (P=0.803). The range was 6 to 31 weeks. Seventeen patients could not undergo secondary bone reconstruction surgery within 6-8 weeks for comorbidities or financial or other reasons, and the longest interval was 31 weeks. After stage II, there were three cases of infection recurrence in the internal fixation group and one case in the external fixation group. The infection control rates were 92.5% and 95.7% after stage II (P=0.621). There were no significant differences in the SAS scores (P=0.278), time to radiological bone healing (P=0.795), time to full weight-bearing (P=0.725), AOFAS scores (P=0.302), HSS scores (P=0.085), and complication rates (P=0.593) between the two groups (Table 2).
      Table 2Postoperative data after stage II
      ItemsInternalExternalP
      Follow-up period (months)33.1±9.832.7±8.50.863
      Infection control casesStage Ⅰ34 (85.0%)19 (82.6%)0.803
      Stage Ⅱ37 (92.5%)22 (95.7%)0.621
      Self-rated Anxiety Scale (SAS) (no/mild/moderate/sever)31/7/2/014/5/3/10.278
      Time to radiological healing (months)6.4±1.86.6±2.40.795
      Time to full weight-bearing (months)7.6±2.47.9±2.90.725
      AOFAS score89.4±7.587.2±9.10.302
      HSS score86.9±5.889.4±4.70.085
      Complications [cases (times)]13 (16)6 (11)0.593
      Among the 63 patients, there were 27 times complications in 19 patients, with an incidence of 42.9%. Complications included infection recurrence (n=4), donor site pain (n=5), delayed wound healing of donor site (n=4),length discrepancy (n=3), re-fracture (n=1), and ankylosis of the joints (knee and ankle; n=10) (Table 3).
      Table 3Complications after stage Ⅱ (times)
      ItemsInternalExternalTotal
      Infection recurrence314
      Pain of donor site235
      Delayed wound healing of donor site224
      Length discrepancy123
      Refracture101
      Ankylosis of the joints7310
      Total161127

      Typical cases

      Case 1

      A 35-year-old male had a chief complaint of recurrent pain and pus after operations for trauma in the left leg for more than 3 months. The left leg was fixed with an external fixation frame, and a sinus tract was formed in the lower segment with purulent discharge. The diagnosis was post-traumatic osteomyelitis of the left lower tibia (C-M type IV). The Masquelet technique was used to debride and fill the bone defect with antibiotic-loaded bone cement and fixed with a locking plate in stage I. After 8 weeks, the bone cement and locking plate were removed in stage II, and the bone defect was fixed using an intramedullary nail and filled with the bone graft. Radiological bone union was observed 6 months after stage II and the limb achieved full weight-bearing at 11 months after surgery. At the last follow-up, there were no infection recurrences (Fig. 1).
      Fig 1
      Fig. 1Thirty-five-year-old male. Recurrent pain and pus after operations for the trauma of the left leg for >3 months. Deep tissue culture revealed Staphylococcus aureus. (A) Preoperative X-ray image. (B-C) After debridement, about 8 cm of infected bone was resected, and a bone defect was formed. (D) After debridement, the bone defect was fixed with a locking plate. (E) X-rays after stage II, 3 days. The bone defect was fixed with a tibial intramedullary nail(Double med. China.). (F) X-rays after stage II, 6 months. (G-H). X-rays after stage II, 32 months. (I-J) X-rays after the second stage II, 48 months. (K) Appearance of the affected limb 48 months after the stage II.

      Case 2

      A 43-year-old female had a chief complaint of redness, swelling, and pus in the left leg for 8 months after trauma. A circular external fixation frame fixation of the lower segment of the left leg was seen, and a sinus tract was formed in the lower segment with purulent discharge. The diagnosis was post-traumatic osteomyelitis of the left tibial (C-M type IV). The debridement and reconstruction were performed with the Masquelet technique, and the external fixation frame was maintained after stage I. The external fixation frame was removed, and replaced by the intramedullary nail in stage II. At 7 months after stage II, the bone was healed, and the limb achieved full weight-bearing. At the last follow-up, there were no infection recurrences, and the internal fixation was removed (Fig. 2).
      Fig 2
      Fig. 2Forty-three-year-old female. Bone infection after open fracture for 8 months. Deep tissue culture revealed Staphylococcus aureus. (A) Preoperative X-ray image. (B-C) After debridement, about 5 cm of infected bone was resected, and a bone defect was formed. (D) After debridement, the defect was fixed with the original external fixation. (E) X-rays after stage II, 8 days. The bone defect was fixed with a retrograde intramedullary nail of tibia (Double med. China.). (F) X-rays after stage II, 7 months. (G-H) X-rays after stage II, 18 months. (I-J) X-rays after stage II, 25 months, the intramedullary nail was removed. (K) Appearance of the affected limb 28 months after the stage II.

      Discussion

      This study suggested that both internal and external fixation methods were effective for C-M type Ⅳ chronic post-traumatic osteomyelitis of the tibia. All 63 patients achieved bone healing, and 93.7% had no infection recurrence.
      The advantages of external fixation have been suggested to include stability in treating bone infection [
      • Simpson A.
      • Robiati L.
      • Jalal M.M.K.
      • Tsang S.T.J.
      Non-union: Indications for external fixation.
      ]. In addition, with external fixation, the pins are far from the infected site, and bacterial adhesion and biofilm formation can be avoided [
      • Mauffrey C.
      • Hake M.E.
      • Chadayammuri V.
      • Masquelet A.C.
      Reconstruction of long bone infections using the induced membrane technique: tips and tricks.
      ]. Furthermore, Masquelet et al. [
      • Masquelet A.C.
      • Giannoudis P.V.
      The induced membrane technique for treatment of bone defects: what have I learned?.
      ] believe that external fixation might enhance micro-movement to form a more active inducted membrane. Still, external fixation has disadvantages. Indeed, the incidence of pin infection and loosening is high [
      • Parameswaran A.D.
      • Roberts C.S.
      • Seligson D.
      • Voor M.
      Pin tract infection with contemporary external fixation: how much of a problem?.
      ], and nursing is difficult, causing difficulties with clothing, sexual life, social life, and compliance [
      • Giannoudis P.V.
      Treatment of bone defects: bone transport or the induced membrane technique?.
      ]. Mathieu et al. [
      • Mathieu L.
      • Tossou-Odjo L.
      • de l'Escalopier N.
      • Demoures T.
      • Baus A.
      • Brachet M.
      • et al.
      Induced membrane technique with sequential internal fixation: use of a reinforced spacer for reconstruction of infected bone defects.
      ] suggested that an external fixator in the femur or tibia for >14 days would result in a higher incidence of postoperative infection, which might be the reason for the late infection in the reconstruction of long bone infection by the Masquelet technique. On the other hand, internal fixation can provide more reliable stability than external fixation, which is of great significance for infection control and bone healing [
      • Worlock P.
      • Slack R.
      • Harvey L.
      • Mawhinney R.
      The prevention of infection in open fractures: an experimental study of the effect of fracture stability.
      ,
      • Einhorn T.A.
      • Gerstenfeld L.C.
      Fracture healing: mechanisms and interventions.
      ]. Despite the advantages of internal fixation, it has disadvantages. Indeed, no matter how thoroughly debridement is performed, residual bacteria can still adhere to the surface of the internal fixation and form a bacterial biofilm [
      • Campoccia D.
      • Montanaro L.
      • Arciola C.R.
      The significance of infection related to orthopedic devices and issues of antibiotic resistance.
      ]. Biofilm-forming bacteria exhibit increased protection against the host defense and enhanced resistance to antibiotics [
      • Campoccia D.
      • Montanaro L.
      • Arciola C.R.
      The significance of infection related to orthopedic devices and issues of antibiotic resistance.
      ,
      • Shi J.
      • Mao N.F.
      • Wang L.
      • Zhang H.B.
      • Chen Q.
      • Liu H.
      • et al.
      Efficacy of combined vancomycin and fosfomycin against methicillin-resistant Staphylococcus aureus in biofilms in vivo.
      ]. Therefore, an antibiotic bone cement-coated plate can be used to fix the bone ends after debridement [
      • Yilihamu Y.
      • Keremu A.
      • Abulaiti A.
      • Maimaiti X.
      • Ren P.
      • Yusufu A.
      Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator.
      ,
      • Jia C.
      • Wang X.
      • Yu S.
      • Wu H.
      • Shen J.
      • Huang Q.
      • et al.
      An antibiotic cement-coated locking plate as a temporary fixation for treatment of infected bone defects: a new method of stabilization.
      ], eliminating the biofilm-forming bacteria [
      • Shi J.
      • Mao N.F.
      • Wang L.
      • Zhang H.B.
      • Chen Q.
      • Liu H.
      • et al.
      Efficacy of combined vancomycin and fosfomycin against methicillin-resistant Staphylococcus aureus in biofilms in vivo.
      ]. Jia et al. [
      • Jia C.
      • Wang X.
      • Yu S.
      • Wu H.
      • Shen J.
      • Huang Q.
      • et al.
      An antibiotic cement-coated locking plate as a temporary fixation for treatment of infected bone defects: a new method of stabilization.
      ] suggested that antibiotic cement-coated locking plates have good clinical effects in controlling bone infection, but the premise is that thorough debridement must have been done first. Although Masquelet did not approve of adding antibiotics to bone cement, he thought that it would not affect the bioactivity of the induced membrane [
      • Masquelet A.C.
      The induced membrane technique.
      ]. Meanwhile, a large body of literature, summarized by Hank et al. [
      • Hake M.E.
      • Young H.
      • Hak D.J.
      • Stahel P.F.
      • Hammerberg E.M.
      • Mauffrey C.
      Local antibiotic therapy strategies in orthopaedic trauma: practical tips and tricks and review of the literature.
      ], demonstrated that the local high concentrations of sustained release of antibiotics is beneficial for infection control. Wang et al. [
      • Luo F.
      • Wang X.
      • Wang S.
      • Fu J.
      • Xie Z.
      Induced membrane technique combined with two-stage internal fixation for the treatment of tibial osteomyelitis defects.
      ] retrospectively analyzed 424 patients with osteomyelitis treated with the Masquelet technique and found that internal fixation after debridement increased the infection recurrence rate. In the present study, compared with external fixation, internal fixation did not increase the rate of infection, and there was no significant impact on bone healing.
      Staged treatment of type IV osteomyelitis is helpful in increasing the rate of infection control and bone healing [
      • Cierny 3rd, G.
      Surgical treatment of osteomyelitis.
      ]. Kanakaris et al. [
      • Kanakaris N.K.
      • Tosounidis T.H.
      • Giannoudis P.V.
      Surgical management of infected non-unions: an update.
      ] believed that one-stage surgery is mainly suitable for uncomplicated lesions, while staged surgery is suitable for more complex ones. Struijs et al. [
      • Struijs P.A.
      • Poolman R.W.
      • Bhandari M.
      Infected nonunion of the long bones.
      ] found that the local administration of antibiotics and bone grafting with a staged treatment strategy could reduce the rate of infection recurrence to 0%-18% and increase the rate of bone healing to 93%-100%. In the present series, no recurrence of infection occurred in 59 patients, and the bone union was achieved during follow-up. It is probably owing to the staged treatment and the application of antibiotic bone cement. The staged strategy might allow better control of the infection, which is, in turn, conducive to better healing of the bone and wound. Therefore, the treatment of type IV osteomyelitis should involve staged surgery, and antibiotic bone cement is a good way to achieve infection control, with no apparent compromise or adverse effect of the antibiotic on bone healing.
      The choice of fixation method in stage Ⅰwas based on the following principles. 1) Local soft tissue conditions. After debridement, if the local soft tissue condition was good, the incision could be closed easily with direct suture or a flap, then internal fixation was considered. On the contrary, when the soft tissue condition was poor, external fixation was preferred. Because if there were no cement-coating plates as an internal fixation at the defect region, the volume of the spacer would significantly decrease, which was beneficial for closing the incision. 2) For the patients who were previously treated with external fixation, if the structure of the frame was stable and the pins were far away from the infected zone, the external fixation could be retained. 3) The location of the lesion is very important. After debridement, in case of there were sufficient lengths at both ends of the bone defect for placing the plate, internal fixation was favored. If there was an insufficient length at any end, a trans-articular external fixator or a locking plate external fixation was usually selected. 4) Finally, the will of the patient should be considered. For the above reasons, more patients in this study received an internal fixation instead of an external fixation, resulting in a difference in the sample size between the two groups. It is generally believed that the optimal interval for stage II is 4-8 weeks [
      • Masquelet A.
      • Kanakaris N.K.
      • Obert L.
      • Stafford P.
      • Giannoudis P.V.
      Bone repair using the Masquelet technique.
      ]. Wang et al. [
      • Luo F.
      • Wang X.
      • Wang S.
      • Fu J.
      • Xie Z.
      Induced membrane technique combined with two-stage internal fixation for the treatment of tibial osteomyelitis defects.
      ] found that the structure and level of growth factor release in the induced membrane and periosteum were basically the same after 4-6 weeks, and the thickness and vascular density of the induced membrane decreased gradually with time. Still, it has also been reported that bone reconstruction for a tibial bone defect of >10 cm achieved good results after 8 and a half years of stage Ⅰ primary operation [
      • Assal M.
      • Stern R.
      The Masquelet procedure gone awry.
      ]. It suggests that the interval between the two stages might not affect the bioactivity of the induced membrane. Masquelet et al. [
      • Masquelet A.C.
      Induced Membrane technique: pearls and pitfalls.
      ] believe that the reason is that the recruitment of stem cells begins after bone grafting. In the present study, some patients could not carry out secondary bone reconstruction surgery within this time window due to various reasons (such as comorbidities, economic reasons, etc.), and the longest interval was 31 weeks. Therefore, the only valid criteria to judge whether the second stage operation might be infection control and soft tissue healing [
      • Masquelet A.C.
      Induced Membrane technique: pearls and pitfalls.
      ].
      In this study, the internal fixation of the stage Ⅰ was replaced in the stage Ⅱ for several reasons. 1) To prevent infection recurrence due to the bacteria being dispersed from the biofilm. Despite the application of high concentrations of antibiotics with sustained release in stage I, it is possible that there were small amounts of bacteria in biofilm colonized on the plate surface. 2) In the stage Ⅱ, the reconstruction segment needs a reliable and stable environment to achieve bone healing, the stability of the initial plate alone is not enough. 3) In the process of bone cement removal, the presence of the plate can increase the risk that the induced membrane is scratched by the fragments of the bone cement.
      The type of osteosynthesis to use for the second stage of the Masquelet technique remains under debate. Giannoudis et al. [
      • Giannoudis P.V.
      • Einhorn T.A.
      • Marsh D.
      Fracture healing: the diamond concept.
      ] described the “diamond concept” of bone healing, and mechanical stability is one of the key factors. As an intra-axial load-sharing osteosynthesis device, the intramedullary nail has better mechanical stability than external fixation [
      • Luo F.
      • Wang X.
      • Wang S.
      • Fu J.
      • Xie Z.
      Induced membrane technique combined with two-stage internal fixation for the treatment of tibial osteomyelitis defects.
      ]. Patients were allowed for early functional exercise and showed a reduced volume of the bone graft occupying the central area of the bone defect [
      • Masquelet A.C.
      • Kishi T.
      • Benko P.E.
      Very long-term results of post-traumatic bone defect reconstruction by the induced membrane technique.
      ]. In addition, unlike plate, the intramedullary nailing did not interfere with the contact between the induced membrane and the bone graft. In some cases where horizontal micromotion remains after intramedullary nail fixation [
      • Schuller M.
      • Weninger P.
      • Tschegg E.
      • Jamek M.
      • Redl H.
      • Stanzl-Tschegg S.
      Micromotion at the fracture site after tibial nailing with four unreamed small-diameter nails–a biomechanical study using a distal tibia fracture model.
      ,
      • Ben-Or M.
      • Shavit R.
      • Ben-Tov T.
      • Salai M.
      • Steinberg E.L.
      Control of the micromovements of a composite-material nail design: a finite element analysis.
      ], a narrow, thin reconstructive locking nail plate can be used as an adjunct to increase stability and also reduce stress shielding [
      • Beltran M.J.
      • Collinge C.A.
      • Gardner M.J.
      Stress modulation of fracture fixation implants.
      ,
      • Yang K.H.
      • Won Y.
      • Kim S.B.
      • Oh B.H.
      • Park Y.C.
      • Jeong S.J.
      Plate augmentation and autologous bone grafting after intramedullary nailing for challenging femoral bone defects: a technical note.
      ]. When the bone defect is close to the epiphysis, some new intramedullary nailing applications can be used. In Fig. 2, the bone defect was located in the distal tibia, where a conventional locking intramedullary nail could not be used, but the retrograde intramedullary nail solved this problem. Although a proximally locking screw loosened and withdrew, the bone defect still healed quickly, demonstrating the nail had reliable stability.
      Compared with other reconstruction methods, the Masquelet technique has a lower complication rate [
      • Luo F.
      • Wang X.
      • Wang S.
      • Fu J.
      • Xie Z.
      Induced membrane technique combined with two-stage internal fixation for the treatment of tibial osteomyelitis defects.
      ,
      • Blum A.L.
      • BongioVanni J.C.
      • Morgan S.J.
      • Flierl M.A.
      • dos Reis F.B.
      Complications associated with distraction osteogenesis for infected nonunion of the femoral shaft in the presence of a bone defect: a retrospective series.
      ]. In this study, the incidence of complications was 42.9%, similar to that reported in the literature [
      • Luo F.
      • Wang X.
      • Wang S.
      • Fu J.
      • Xie Z.
      Induced membrane technique combined with two-stage internal fixation for the treatment of tibial osteomyelitis defects.
      ]. The most common complication is joint ankylosis, mainly in the knee joint. These patients have serious joint ankylosis before the operation. The main reasons are long-term infection and fixation of the joint after repeated operations. The joint function cannot be improved by functional exercise because of soft tissue adhesion. The infection recurrence is still the most serious complication, which often means the failure of the operation and has a great psychological impact on patients. Surgical intervention and re-debridement can be carried out in the early stage of infection recurrence. Due to the relatively limited infection, a segmental osteotomy is not required, and debridement and reconstruction are relatively easy. In the 23 patients with external fixation, there were no pin tract infections. It was mainly due to the short interval between the two stages (6-12 weeks, average of 8.4 weeks), during which period antibiotics were applied for 6 weeks, so pin tract infection was reduced.
      This study has some limitations. Firstly, this study was designed as a single center retrospective study with selection and indication bias. Indeed, as indicated above, the indications for selecting one method over the other are based on the patient's condition, leading to bias. Further prospective research is needed to solve the methodological limitations. Second, there were only 23 patients in the external fixation group. Although randomized, multicenter controlled studies are ideal, such studies will encounter ethical difficulties in the proper selection of the patients.

      Conclusion

      There were no significant differences between the two methods in infection control rates, bone healing, and healing time. Internal fixation and external fixation might all be suitable to fix the bone ends after the Masquelet technique for C-M Ⅳ tibial post-traumatic osteomyelitis. The selection might be based on the patient's soft tissue condition and the original fixation method.

      Declarations

      Ethics approval

      This study was approved by the Institutional Review Board of The 920 Hospital of Joint Logistic Support Force of the PLA. The requirement for individual consent was waived by the committee because of the retrospective nature of the study.

      Consent for publication

      Not applicable.

      Data statement

      The data presented in this study are available in article or supplementary material.

      Funding

      This study was funded by the Clinical Orthopeadic Trauma Medical Center Program of Yunnan Province, China ( ZX20191001 ), the Applied Basic Research Joint Project of Yunnan Science and Technology Department and Kunming Medical University ( 202101AY070001-294 ).

      Declaration of Competing Interests

      The authors declare that they have no competing interests.

      Acknowledgements

      None.

      References

        • Conterno L.O.
        • Turchi M.D.
        Antibiotics for treating chronic osteomyelitis in adults.
        Cochrane Database Syst Rev. 2013; CD004439
        • Panteli M.
        • Giannoudis P.V.
        Chronic osteomyelitis: what the surgeon needs to know.
        EFORT Open Rev. 2016; 1: 128-135
        • Yang J.
        • Yao J.L.
        • Wu Z.Q.
        • Zeng D.L.
        • Zheng L.Y.
        • Chen D.
        • et al.
        Current opinions on the mechanism, classification, imaging diagnosis and treatment of post-traumatic osteomyelitis.
        Chin J Traumatol. 2021; 24: 320-327
        • Cierny 3rd, G.
        • Mader J.T.
        • Penninck J.J.
        A clinical staging system for adult osteomyelitis.
        Clin Orthop Relat Res. 2003; : 7-24
        • Cierny 3rd, G.
        Surgical treatment of osteomyelitis.
        Plast Reconstr Surg. 2011; 127: 190S-204S
        • Dendrinos G.K.
        • Kontos S.
        • Lyritsis E.
        Use of the Ilizarov technique for treatment of non-union of the tibia associated with infection.
        J Bone Joint Surg Am. 1995; 77: 835-846
        • Gao Y.S.
        • Ai Z.S.
        • Yu X.W.
        • Sheng J.G.
        • Jin D.X.
        • Chen S.B.
        • et al.
        Free vascularised fibular grafting combined with a locking plate for massive bone defects in the lower limbs: a retrospective analysis of fibular hypertrophy in 18 cases.
        Injury. 2012; 43: 1090-1095
        • Beals R.K.
        • Bryant R.E.
        The treatment of chronic open osteomyelitis of the tibia in adults.
        Clin Orthop Relat Res. 2005; : 212-217
        • Siboni R.
        • Joseph E.
        • Blasco L.
        • Barbe C.
        • Bajolet O.
        • Diallo S.
        • et al.
        Management of septic non-union of the tibia by the induced membrane technique. What factors could improve results?.
        Orthop Traumatol Surg Res. 2018; 104: 911-915
        • Sasaki G.
        • Watanabe Y.
        • Miyamoto W.
        • Yasui Y.
        • Morimoto S.
        • Kawano H.
        Induced membrane technique using beta-tricalcium phosphate for reconstruction of femoral and tibial segmental bone loss due to infection: technical tips and preliminary clinical results.
        Int Orthop. 2018; 42: 17-24
        • Raven T.F.
        • Moghaddam A.
        • Ermisch C.
        • Westhauser F.
        • Heller R.
        • Bruckner T.
        • et al.
        Use of Masquelet technique in treatment of septic and atrophic fracture nonunion.
        Injury. 2019; 50: 40-54
        • Wu H.
        • Shen J.
        • Yu X.
        • Fu J.
        • Yu S.
        • Sun D.
        • et al.
        Two stage management of Cierny-Mader type IV chronic osteomyelitis of the long bones.
        Injury. 2017; 48: 511-518
        • Giannoudis P.V.
        • Harwood P.J.
        • Tosounidis T.
        • Kanakaris N.K.
        Restoration of long bone defects treated with the induced membrane technique: protocol and outcomes.
        Injury. 2016; 47: S53-S61
        • Masquelet A.C.
        • Kishi T.
        • Benko P.E.
        Very long-term results of post-traumatic bone defect reconstruction by the induced membrane technique.
        Orthop Traumatol Surg Res. 2019; 105: 159-166
        • Alemdar C.
        • Azboy I.
        • Atic R.
        • Ozkul E.
        • Gem M.
        • Kapukaya A.
        Management of infectious fractures with "Non-Contact Plate" (NCP) method.
        Acta Orthop Belg. 2015; 81: 523-529
        • Yilihamu Y.
        • Keremu A.
        • Abulaiti A.
        • Maimaiti X.
        • Ren P.
        • Yusufu A.
        Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator.
        Injury. 2017; 48: 1636-1643
        • Masquelet A.
        • Kanakaris N.K.
        • Obert L.
        • Stafford P.
        • Giannoudis P.V.
        Bone repair using the Masquelet technique.
        J Bone Joint Surg Am. 2019; 101: 1024-1036
        • Masquelet A.C.
        Induced Membrane technique: pearls and pitfalls.
        J Orthop Trauma. 2017; 31: S36-SS8
        • Zhang Z.
        • Chai W.
        • Zhao G.
        • Zhang Q.
        • Chen Z.
        • Wang X.
        • et al.
        Association of HSS score and mechanical alignment after primary TKA of patients suffering from constitutional varus knee that caused by combined deformities: a retrospective study.
        Sci Rep. 2021; 11: 3130
        • Ceccarelli F.
        • Calderazzi F.
        • Pedrazzi G.
        Is there a relation between AOFAS ankle-hindfoot score and SF-36 in evaluation of Achilles ruptures treated by percutaneous technique?.
        J Foot Ankle Surg. 2014; 53: 16-21
        • Simpson A.
        • Robiati L.
        • Jalal M.M.K.
        • Tsang S.T.J.
        Non-union: Indications for external fixation.
        Injury. 2019; 50: S73-SS8
        • Mauffrey C.
        • Hake M.E.
        • Chadayammuri V.
        • Masquelet A.C.
        Reconstruction of long bone infections using the induced membrane technique: tips and tricks.
        J Orthop Trauma. 2016; 30: e188-e193
        • Masquelet A.C.
        • Giannoudis P.V.
        The induced membrane technique for treatment of bone defects: what have I learned?.
        Trauma Case Rep. 2021; 36100556
        • Parameswaran A.D.
        • Roberts C.S.
        • Seligson D.
        • Voor M.
        Pin tract infection with contemporary external fixation: how much of a problem?.
        J Orthop Trauma. 2003; 17: 503-507
        • Giannoudis P.V.
        Treatment of bone defects: bone transport or the induced membrane technique?.
        Injury. 2016; 47: 291-292
        • Mathieu L.
        • Tossou-Odjo L.
        • de l'Escalopier N.
        • Demoures T.
        • Baus A.
        • Brachet M.
        • et al.
        Induced membrane technique with sequential internal fixation: use of a reinforced spacer for reconstruction of infected bone defects.
        Int Orthop. 2020; 44: 1647-1653
        • Worlock P.
        • Slack R.
        • Harvey L.
        • Mawhinney R.
        The prevention of infection in open fractures: an experimental study of the effect of fracture stability.
        Injury. 1994; 25: 31-38
        • Einhorn T.A.
        • Gerstenfeld L.C.
        Fracture healing: mechanisms and interventions.
        Nat Rev Rheumatol. 2015; 11: 45-54
        • Campoccia D.
        • Montanaro L.
        • Arciola C.R.
        The significance of infection related to orthopedic devices and issues of antibiotic resistance.
        Biomaterials. 2006; 27: 2331-2339
        • Shi J.
        • Mao N.F.
        • Wang L.
        • Zhang H.B.
        • Chen Q.
        • Liu H.
        • et al.
        Efficacy of combined vancomycin and fosfomycin against methicillin-resistant Staphylococcus aureus in biofilms in vivo.
        PLoS One. 2014; 9e113133
        • Jia C.
        • Wang X.
        • Yu S.
        • Wu H.
        • Shen J.
        • Huang Q.
        • et al.
        An antibiotic cement-coated locking plate as a temporary fixation for treatment of infected bone defects: a new method of stabilization.
        J Orthop Surg Res. 2020; 15: 44
        • Masquelet A.C.
        The induced membrane technique.
        Orthop Traumatol Surg Res. 2020; 106: 785-787
        • Hake M.E.
        • Young H.
        • Hak D.J.
        • Stahel P.F.
        • Hammerberg E.M.
        • Mauffrey C.
        Local antibiotic therapy strategies in orthopaedic trauma: practical tips and tricks and review of the literature.
        Injury. 2015; 46: 1447-1456
        • Luo F.
        • Wang X.
        • Wang S.
        • Fu J.
        • Xie Z.
        Induced membrane technique combined with two-stage internal fixation for the treatment of tibial osteomyelitis defects.
        Injury. 2017; 48: 1623-1627
        • Kanakaris N.K.
        • Tosounidis T.H.
        • Giannoudis P.V.
        Surgical management of infected non-unions: an update.
        Injury. 2015; 46: S25-S32
        • Struijs P.A.
        • Poolman R.W.
        • Bhandari M.
        Infected nonunion of the long bones.
        J Orthop Trauma. 2007; 21: 507-511
        • Assal M.
        • Stern R.
        The Masquelet procedure gone awry.
        Orthopedics. 2014; 37: e1045-e1048
        • Giannoudis P.V.
        • Einhorn T.A.
        • Marsh D.
        Fracture healing: the diamond concept.
        Injury. 2007; 38: S3-S6
        • Schuller M.
        • Weninger P.
        • Tschegg E.
        • Jamek M.
        • Redl H.
        • Stanzl-Tschegg S.
        Micromotion at the fracture site after tibial nailing with four unreamed small-diameter nails–a biomechanical study using a distal tibia fracture model.
        J Trauma. 2009; 66: 1391-1397
        • Ben-Or M.
        • Shavit R.
        • Ben-Tov T.
        • Salai M.
        • Steinberg E.L.
        Control of the micromovements of a composite-material nail design: a finite element analysis.
        J Mech Behav Biomed Mater. 2016; 54: 223-228
        • Beltran M.J.
        • Collinge C.A.
        • Gardner M.J.
        Stress modulation of fracture fixation implants.
        J Am Acad Orthop Surg. 2016; 24: 711-719
        • Yang K.H.
        • Won Y.
        • Kim S.B.
        • Oh B.H.
        • Park Y.C.
        • Jeong S.J.
        Plate augmentation and autologous bone grafting after intramedullary nailing for challenging femoral bone defects: a technical note.
        Arch Orthop Trauma Surg. 2016; 136: 1381-1385
        • Blum A.L.
        • BongioVanni J.C.
        • Morgan S.J.
        • Flierl M.A.
        • dos Reis F.B.
        Complications associated with distraction osteogenesis for infected nonunion of the femoral shaft in the presence of a bone defect: a retrospective series.
        J Bone Joint Surg Br. 2010; 92: 565-570