The deleterious relationship of Pseudomonas aeruginosa and Induced Membranes in infected non-union treatment

The Masquelet technique was originally described for the treatment of an infected non-union with an extensive bone defect where a staged protocol was needed to (cid:977)irst eliminate an infection then secondarily bone graft a defect. Though this is a versatile technique, certain limitations/ complications must be recognized. The study was done between 2012 to 2019 at SRIHER uni-versity. 19 patients in whom the Masquelet technique has failed is taken into study. 17 male and two females, with a mean age of 31 years (range of 13 yrs. – 51 yrs.) with a mean follow up of 12 months. The 19 patients who presented with Pseudomonas aeruginosa infected non-union of the tibia and femur with bone defects underwent the Masquelet technique. All patients failed to form adequate induced membrane at the non-union site. Infected non-union with a bone defect is dif(cid:977)icult to treat. Bone defects of 2cms can be treated by cancellous bone grafting. Defects more than 4-5cms will require specialized reconstructive procedures to prevent amputation. The two common techniques used are Ilizarov technique with bone transport and bone graft into an induced membrane as described by Masquelet. This study shows a high failure rate of the Masquelet technique with Pseudomonas infection. The most dif(cid:977)icult issue faced by the surgeon in treating P. aeruginosa is its ability to develop resistance to multiple classes of antibiotics during the course of treating the patient. Masquelet technique is used extensively for the treatment of infective non-union. Pseudomonas secretes a slime layer that may lead to a weak or de(cid:977)icient formation of the induced membrane. And the elution of antibiotics may not be adequate for intramedullary osteomyelitis with pseudomonas growth. This limits the Masquelet technique in the management of infected non-union with pseudomonas infection.

Pseudomonas, Induced Membrane, Masquelet, Bio ilm ABSTRACT The Masquelet technique was originally described for the treatment of an infected non-union with an extensive bone defect where a staged protocol was needed to irst eliminate an infection then secondarily bone graft a defect. Though this is a versatile technique, certain limitations/ complications must be recognized. The study was done between 2012 to 2019 at SRIHER university. 19 patients in whom the Masquelet technique has failed is taken into study. 17 male and two females, with a mean age of 31 years (range of 13 yrs. -51 yrs.) with a mean follow up of 12 months. The 19 patients who presented with Pseudomonas aeruginosa infected non-union of the tibia and femur with bone defects underwent the Masquelet technique. All patients failed to form adequate induced membrane at the non-union site. Infected non-union with a bone defect is dif icult to treat. Bone defects of 2cms can be treated by cancellous bone grafting. Defects more than 4-5cms will require specialized reconstructive procedures to prevent amputation. The two common techniques used are Ilizarov technique with bone transport and bone graft into an induced membrane as described by Masquelet. This study shows a high failure rate of the Masquelet technique with Pseudomonas infection. The most dif icult issue faced by the surgeon in treating P. aeruginosa is its ability to develop resistance to multiple classes of antibiotics during the course of treating the patient. Masquelet technique is used extensively for the treatment of infective non-union. Pseudomonas secretes a slime layer that may lead to a weak or de icient formation of the induced membrane. And the elution of antibiotics may not be adequate for intramedullary osteomyelitis with pseudomonas growth. This limits the Masquelet technique in the management of infected non-union with pseudomonas infection.

INTRODUCTION
The Masquelet technique was originally described for the treatment of an infected non-union with an extensive bone defect where a staged protocol was needed to irst eliminate an infection then secondarily bone graft a defect (Pelissier et al., 2002). Posttraumatic segmental bone defects remain a challenge for the orthopaedic surgeon. These complex lesions are dif icult to treat and lead to signi icant morbidity. This technique has the theoretical advantage of providing an infected site with a scaffold, that is the induced membrane, to protect the bone graft from resorption. The induced membrane not only contains the bone graft and prevents its resorption at the early stages; it plays a vital role in angiogenesis and bone formation throughout the regeneration process (Pelissier et al., 2002(Pelissier et al., , 2004. This is a two-staged procedure, and soft tissue reconstruction is essential for the formation of healthy membrane (Pelissier et al., 2004;Flick et al., 1987). Though this is a versatile technique, certain limitations and complications must be recognized. Overall, the union rate was achieved in 88% of the cases and infection cured in 93% (Morelli et al., 2018). This paper describes the experiences we have had with 19 patients with bone defects regarding the failure to form or maintain the induced membrane due to a P. aeruginosa infection t the infected non-union site. Gram-negative bacteria are intrinsically resistant to many antibiotics. Species that have acquired multidrug resistance and cause infections that are effectively untreatable present a serious threat to public health (Zgurskaya et al., 2015).

Patients and methods
The study was done between 2012 to 2019 at SRI-HER university.19 patients in whom the Masquelet technique has failed is taken into study. 17 male and two females, with a mean age of 31 years (range of 13 yrs. -51 yrs.) with a mean follow up of 12 months. The anatomical locations included the Tibia (14 patients) and the femur (5 patients). The diagnoses were infected non-unions (18 patients) and one pseudoarthrosis of the tibia. The initial presentation of the patients was open fractures, which proceeded to infected non-unions from different ixation methods. 14 patients included in our study were initially treated by intramedullary nail, of which 8 patients were with a femur fracture, and six patients were tibial fracture.
Five patients with tibial open fracture were treated by Ilizarov ring ixator and later removed and converted to a functional cast bracing for the tibia. Pseudomonas aeruginosa was grown by culture from a deep swab for all the patients. The mean size of the defects after the radical debridement was 6.3cm on the AP and 6.5cm on the lateral view. Surgical technique -In the irst stage as per the technique used by Masquelet et al. Pelissier et al. (2002), a thorough debridement was done initially after removal of the implant or Ilizarov. The area of fracture non-union site was carefully debrided and irrigated.
Non-viable tissue and sclerotic bone at the nonunion site were removed. Method of ixation was decided based on the level of the fracture and location. After achieving alignment of the limb, an intramedullary nail or external ixator was applied to stabilise the fracture. Antibiotic coated PMMA spacer was placed in the bone defect. This was placed in a such a way as to overlap the bone ends by 2cm and as per culture sensitivity, Meropenem (3 vials -500mg/vial) were mixed into a single pack of palacos cement.
Wound closure by primary suturing was possible in all the patients. The patients who underwent application of external ixator were mobilized with strict non-weight bearing while the patients who underwent intramedullary nailing were mobilized with partial weight-bearing.
All patients were treated with a sensitive or broadspectrum antibiotic for 2 weeks and then received oral antibiotics for 2-3 weeks. Erythrocyte sedimentation rate and C-reactive protein were done on a weekly basis and showed no signi icant elevation (ESR stayed below 16mm/hr and CRP was <0.6mg/dl). X-ray was done prior to the second stage procedure of bone grafting.
The second procedure of autologous bone grafting was planned for all the patients at 4-8 weeks after the irst surgery (average 5.2 weeks). On opening the fracture non-union site with PMMA, we observed a de iciency or total absence of induced membrane. Hence, the second procedure of autologous Iliac crest bone grafting was abandoned. A deep culture was again sent, which revealed the presence of Pseudomonas aeruginosa.

RESULTS
The 19 patients who presented with Pseudomonas aeruginosa infected non-union of the tibia and femur with bone defects underwent the Masquelet technique. The average age of the patients was 31 years (range 13-51). The site of bone defect: 14 cases of middle third tibia and 5 cases of femur middle third infected non-union. One month after the irst stage, wound dehiscence and super icial wound infection were seen in 2 patients. All the other patients had good wound healing after the primary procedure of cement implantation. All patients failed to form adequate induced membrane at the non-union site.
A: X-ray appearance showing the non-union with a bony defect. B: Intra-op picture showing the defect. After IMIL nail removal. C: 8 weeks post-op after application of the PMMA cement showing de icient formation of induced membrane (Figure 1).
All the patients were then treated with an Ilizarov ring-ixator and bone transport. 17 patients went in for bone healing with distraction osteogenesis using the Ilizarov method. Two patients subsequently Figure 1: 50 years old male with an infected non-union of the tibia with IMIL nailin-situ and a pus culture, which grew P. aeruginosa.

Figure 2: 9 years old patient who is a known case of Pseudoarthrosis of tibia, presented with a fracture of both bones of the leg with an implant in-situ.
required amputation after multiple surgical procedures due to persistent infection and diffuse intramedullary osteomyelitis.

DISCUSSION
Infected non-union with a bone defect is dif icult to treat. Bone defects of 2cms can be treated by cancellous bone grafting. Defects more than 4-5cms will require specialized reconstructive procedures to prevent amputation (Pelissier et al., 2002(Pelissier et al., , 2004. The two common techniques used are Ilizarov technique with bone transport and bone graft into an induced membrane as described by Masquelet (Pelissier et al., 2002(Pelissier et al., , 2004Flick et al., 1987). There are few case reports which has shown the failure of the Masquelet technique. This study shows a high failure rate of the Masquelet technique with Pseudomonas infection. Pseudomonas aeruginosa is a ubiquitous organism present in many environmental settings, and it can be isolated from various living sources, including plants, animals, and humans.
Serious infections by Pseudomonas aeruginosa are found to be predominantly hospital-acquired and can be isolated from respiratory therapy equipment, antiseptics, soaps, sinks and medicines. Danielle et al. and Scott C P et al. have shown the ability of Pseudomonas to produce bio ilm and slime layer over bone cement (Scott et al., 1999). They showed that enhanced slime production on antibiotic-loaded bone cement together with the formation of small colonies variants resulted in decreased susceptibility to antibiotics leading to persistent infections (Scott et al., 1999;Rezzouk et al., 2005). Various studies have shown the failure of antibioticloaded bone cement in treating pseudomonas infection, but none of the studies has shown the relationship of pseudomonas and induced membrane. In our study, one patient underwent Masquelet technique for congenital pseudarthrosis of the tibia. Though preliminary results of the induced membrane in CPT showed promising results in many studies (Korompilias et al., 2009), our patient had pseudomonas infection at the non-union site and hence did not respond to the Masquelet technique. Tobramycin, Meropenem and Imipenem have shown to have good ef icacy against pseudomonas when used with bone cement (Rezzouk et al., 2005;Korompilias et al., 2009). In our study meropenem was used in all patients based on culture sensitiv-ity. The most dif icult issue faced by the surgeon in treating P. aeruginosa is its ability to develop resistance to multiple classes of antibiotics during the course of treating the patient. This has been attributed to the way this organism regulates its AmpC, OprD and ef lux pumps (Rezzouk et al., 2005;Gouron, 2011). Though in our study, P. aeruginosa isolated in culture throughout the study appeared to be sensitive to Meropenem the development of resistance experienced in the form of absence of membrane formation was suspected. Through the pioneering work of Masquelet et al., large posttraumatic bone defects have been bridged through this technique. Gerber et al. stated that this membrane prevents connective tissue in iltration and acts as a 'bone-forming chamber' with osteoinductive and osteoconductive properties (Hertel et al., 1994).
Welby et al. stated that this technique could be successfully used in patients who have had the limb irradiated or in the presence of infection (Welby et al., 2004). Even though the mean size of the defect in our study (6.5cm) was much less than the defect bridged by Masquelet et al., membrane formation. The bacterial bio ilm formed are the root of many persistent and chronic bacterial infections (Costerton, 1999). We suspect that the presence of Pseudomonas Aeruginosa with its bio ilm production and slime layer formation in bone cement played a major role in preventing the formation of the induced membrane.

CONCLUSIONS
Masquelet technique is used extensively for the treatment of infective non-union. The development of induced membrane requires adequate soft tissue cover and proper technique. Bone cement induced tissue necrosis may also lead to compromised soft tissue cover at the non-union site-speci ically at the tibial shaft region. Pseudomonas secretes a slime layer that may lead to a weak or de icient formation of the induced membrane. And the elution of antibiotics may not be adequate for intramedullary osteomyelitis with pseudomonas growth. This limits the Masquelet technique in the management of infected non-union with pseudomonas infection. Further studies are needed to establish a relation between pseudomonas and induced membranes.