Novel piston technique versus Ilizarov technique for the repair of bone defect after lower limb infection

Background We aimed to compare the effectiveness and complications of a novel piston technique versus the Ilizarov technique for the repair of bone defects after lower limb infection. Patients and methods We retrospectively reviewed 41 patients who had been treated at our department for lower extremity bone defects following osteomyelitis. There were 38 men and three women with a mean age of 43.41 (range, 12–69 years). The infected bone defects involved 36 tibias and five femurs. The piston technique (PT, group A) was used in 12 patients and the Ilizarov technique (IT, group B) in 29 patients. The mean follow-up period was 28.50 months (PT) and 29.90 months (IT). The modified Application of Methods of Illizarov (ASAMI) criteria was used to evaluate bone healing and functional recovery. Results Complete eradication of the infection and union of docking sites were accomplished in both groups. The mean external fixator index (EFI) was 42.32 days/cm in group A versus 58.85 days/cm in group B (p < 0.001). The bone outcomes were similar between groups A and B (p = 0.558) (excellent [9 vs. 19], good [3 vs.10]); group A showed better functional outcomes than group B (p < 0.05) (excellent [7 vs. 6], good [4 vs. 12], fair [0 vs. 10] and poor [1 vs. 1]). Pain was the most common complaint during follow-up, and group A had fewer cases of pin tract infection (1 vs. 6), adjacent joint stiffness (3 vs. 8), and delayed healing of the joint (0 vs. 3). Conclusions Satisfactory bone healing can be achieved by using both PT and IT, although PT demonstrated better functional results, lower EFI, and allowed early removal of the external fixation. We found that this novel piston technique can improve the comfort of patients, reduce the incidence of complications, and provide rapid and convenient rehabilitation. Supplementary Information The online version contains supplementary material available at 10.1186/s13018-021-02844-1.


Introduction
Infectious bone defects of the lower extremity pose a huge challenge for doctors in clinical practice due to their complicated treatment and unfavorable prognosis. Common causes include acute bone loss, surgical removal of dead or sclerotic bone after infection and ischemic atrophy of nonunion sites [1,2]. There are numerous options for reconstruction, including Ilizarov distraction osteogenesis [3][4][5][6], Masquelet technique [7], allogeneic bone grafts [8] and vascularized autogenous bone grafts (such as ribs, ilium, and fibula) [9]. New tissue engineering techniques [10] are also emerging in the field of bone defect treatment.
The Ilizarov bone transport technique is widely regarded as the standard surgical method for the treatment of bone defects [3] and is used to treat nonunion,  16:704 osteomyelitis, deformity, traumatic bone defects, and unequal length of the lower extremities. Although the Ilizarov technique can achieve satisfactory results in most cases, one of the most common problems associated with this method is that the external fixator needs to be fixed for a long time until the new bone is completely ossified, which involves many complications, including pin trajectory infection, loosening of the retractor, persistent pain, joint stiffness, angled tilt, refracture and delayed union [11,12]. In recent years, a high value has been placed on the Masquelet technique [13,14], and it is widely used. It is, a two-stage surgical technique, which involves debridement and filling of the bone defect with cement spacer in the first stage to control postoperative infection, followed by bone remodeling by filling with cancellous bone after removing the cement inside the socalled induced membrane in the second stage. Due to the risk of residual infection after the first-stage operation for infectious bone defects, a mixture of bone cement and antibiotics has been widely used [15][16][17]. There are several clinical series reports [3-7, 13, 18] on the reconstruction of bone defects using the Masquelet technique. However, in some cases, several bone grafts are needed, and donor site injury can also occur. To overcome these technical limitations, our department used the Masquelet technique combined with the Ilizarov bone transport technique, which is called the piston technique, for the first time to treat bone defects after lower extremity infection. The piston technique involves a similar procedure in the first stage as the Masquelet technique, and we removed the bone cement without disturbing the induced membrane. External fixation was then used to perform distraction osteogenesis in the second stage. This is expected to reduce the time of external fixation and avoid bone grafting.

Open Access
To the best of our knowledge, there are relatively few reports describing the outcomes of the piston technique for the treatment of infected bone defects. Therefore, here we have described this piston technique, and compared it with the Ilizarov bone transport technique for the treatment of bone defects after lower extremity infection. We aimed to determine whether this technique could reduce the fixator duration time in patients and the risk of complications.

Patients and methods
This retrospective study was reviewed and approved by the institutional review board. All procedures were performed in compliance with the Declaration of Helsinki. Informed consent for participation was obtained from all participants.
We retrospectively analyzed the outcomes of the piston technique (PT) or Ilizarov technique (IT) for the treatment of bone defects after lower limb infection between 2013 and 2018. The inclusion criteria for this study were: (1) osteomyelitis diagnosed by bacteriology or histology following intraoperative biopsy based on clinical manifestations and imaging; (2) patients treated with the PT or IT; (3) bone defect > 2.5 cm after debridement; and (4) age < 70 years. Exclusion criteria were as follows: (1) patients with bone tumors, rheumatic diseases and metabolic bone diseases; (2) patients who withdrew from treatment early or did not complete the treatment; or (3) patients who could not be followed up.
Finally, 29 and 12 patients in the IT and PT groups, respectively, were enrolled in the study. There were 38 men and three women and the bone defects included 36 tibias and five femurs. The basic information of the patients is shown in Additional file 1: Tables S1 and S2. The variables included sex, age, site, organism, type of external fixator, previous operations, comorbidities, soft tissue defects and bone defects.

Surgical techniques
The PT and IT techniques were similar in debridement. The patient remained supine on a transmissive operating table. It was ensured that the surgical incision was as consistent as possible with the previous incision, including the sinus. Hardware such as steel plates, screws, or intramedullary nails used in the previous surgery was removed. Then, the infected bone was completely removed until the so-called "paprika sign" (cortical bone bleeding area) appeared ( Fig. 1a, b) [19]. It is of great importance to ensure that the cortex remaining after radical debridement is intact and uninfected.
In group A (PT), the bone defect was measured in the first stage after radical debridement and was filled with a polymethylmethacrylate (PMMA) bone cement spacer (Fig. 1c). We tended to use 4.0 g vancomycin + 4.0 g meropenem per 40 g cement. The types of antibiotics can be adjusted according to the results of bacterial culture and drug sensitivity tests before surgery. The fixation method depends on the type and location of the bone infection defect. Choices include monolateral or circular external fixations (Fig. 1d). The cement spacer was carefully removed without damaging the PMMA-induced membrane 6-8 weeks after the first operation. Next, we used an external fixation extender for comparison of the installation. Under the control of an image intensifier, two or three suitable hydroxyapatite-covered pins were used to fix the femoral or tibial shaft, which were interposed through a pre-drilled path about 2-3 cm above and below the osteotomy site. A subperiosteal transverse osteotomy was performed then. After suturing the periosteum, the incision was closed using a drainage tube. In group B (PT), after debridement and removal of the infected or necrotic bone, we chose external fixation for the tibial and femoral bone defects. The transverse osteotomy method was similar to that in group A.
In both groups, we used an open dressing change or vacuum sealing drainage (VSD) to temporarily seal large and complex soft tissue defects at the site of infection. Whether skin grafts or flaps are needed before bone defect repair depends on the condition of the soft tissue.

Postoperative protocols
Clinical follow-up was conducted every 2 weeks to check the status of the nail path, skeleton stability, range of motion and damage to adjacent joints. In the distraction phase, X-ray examination was performed every 2 weeks, followed by once a month during the consolidation period, to assess the fracture healing and the quality of consolidation (Fig. 2). Radiographic measurements were performed by two surgeons independently. Laboratory indicators, such as erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and blood cell count, were tested at appropriate time points to ensure eradication of the infection. Postoperative complications were also recorded. Dahl's grading was used to assess pin-tract inflammation [20]. The results of bone healing and functional recovery were evaluated according to the modified Application of Methods of Illizarov (ASAMI) criteria (Table 1) [21]. The external fixation time (EFT) represents the number of whole days during which the external fixation was fixed to the bone. The external fixation index (EFI) was defined as EFT (days) divided by the total length (cm).

Statistical analysis
Data were analyzed using SPSS 22.0 software (SPSS Inc, Chicago, Illinois). The Kolmogorov-Smirnov test was used to check for normality. A p value of > 0.05 was defined as a normal distribution. The internal consistency test (kappa value) was used to assess interobserver variability in the measurement of the parameters. Descriptive statistics were used for all the variables. Continuous variables were expressed as means and standard deviations. Dichotomous variables were expressed as percentages and events. The independent samples t-test, chi-square test, or Wilcoxon rank sum test were performed to evaluate the differences between the two groups. Statistical significance was set at p < 0.05.

Results
The most common cause of infection was orthopedic implant-related infection, with only a few patients suffering from open fracture and primary hematogenous infection. Both groups were most infected with Staphylococcus aureus. Detailed intraoperative and postoperative data of groups A and B are shown in Tables 2 and 3, respectively. In both groups, complete eradication of the infection and union of docking sites were accomplished. The mean bone defect length and EFT did not differ between groups A and B, while the mean external fixator index (EFI) of group A (42.32 days/cm) was significantly lower than that of group B (58.85 days/cm) (p < 0.001).
According to the bone healing and limb function evaluation criteria recommended by ASAMI, the results of bone healing in group A, in which nine cases were classified as excellent and three cases as good, were similar to   Table 4). The relatively worse functional result was mainly due to prolonged external fixation, leading to significant pain and restricted movement of the adjacent joints. Pain was the most common complaint during the follow-up period. Some patients experienced severe and intolerable pain in the first few days after surgery. In both groups A and B, almost all patients had dull pain during

Discussion
Treatment of lower limb osteomyelitis remains a serious problem in orthopedics. Some measures can be effective in preventing postoperative infection of the lower limbs, such as intramedullary antibiotic-coated nails [22]. However, infection cannot be completely avoided.
Patients who decide to undergo limb salvage treatment usually require multiple surgical treatments, resulting in soft tissue and bone defects. Currently, there are many methods for the treatment of bone defects, including the Masquelet and Ilizarov techniques. In    [23,24]. However, for large segments of bone defects, the classic Masquelet technique cannot obtain a sufficient amount of autologous bone graft and is also prone to complications at the donor site. Furthermore, it may not correct limb length discrepancy (LLD) and limb alignment. The traditional Ilizarov technique has several advantages. Patients can bear weight early after surgery. In addition, IT has the advantages of continuous adjustment of limb alignment with external fixation, and the possibility of a one-stage operation. Unfortunately, the long duration of the external fixator places restrictions on its use and can also result in many complications.
There is no doubt that if the external fixation time can be effectively reduced, this technique will become very attractive. The present study investigated a novel hybrid PT combining induced membrane and bone transport techniques to effectively reduce the duration of external fixation and complications.
Our study demonstrated that the average EFI in the PT group was significantly lower than that in the IT group, and the bone healing results were not negatively affected. Traditional studies on the management of infected nonunion with Ilizarov external fixator reported an average EFI of 54.9 and 54.0 days/cm [25,26]. The characteristics of the induced membrane can be described as a vascularized structure that resembles the periosteum, secreting vascular inducible factors (VEGF) [27,28], and a maturity stage between the 4th and 6th weeks [29]. One of the advantages of the PT is that the blood supply of the induced membrane in the bone defect area and debridement area can promote the migration of the docking sites, thus accelerating the process of distraction osteogenesis.
Delayed union or nonunion of the docking site may occur when the contact area is insufficient or when the docking site does not coincide. Several feasible options have been proposed to prevent this common complication and shorten the duration of the fixator. Some surgeons choose to perform partial autogenous bone grafting, while some prefer the osteotome and curette to polish the bony edges at the docking site [30,31]. Bifocal bone transport has also been reported to be effective [32]. However, our PT provides an exceptional osteogenic microenvironment compared to the IT, where the induced membrane can secrete osteoinductive factors and stimulate bone regeneration. In our study, the PT group had no case of delayed docking site union, while the IT group had three cases requiring further operations for the bone graft.
Removal of the entire bone segment and continuous antibiotic release with bone cement can effectively remove the dead bone and biofilm, thus offering a better chance of eliminating infection compared to the Ilizarov technique. Although no infection recurrence was noted in either group, Roukoz et al. [33] and Shah et al. [34] have reported that antibiotic-loaded bone cement can achieve a better anti-infection effect. A narrative review [35] showed that the rate of infection recurrences in patients with infected or non-infected critical-sized tibial bone defects treated using the Ilizarov method was 4.58%. Although there was no significant difference in the operation times between the two groups, a logistic regression analysis [36] revealed that repeated operations, post-traumatic osteomyelitis, and internal fixation in the first stage were risk factors for recurrence of infection in patients treated with the induced membrane technique. Based on these concerns, we recommend that repeat operations and the surgical design of internal fixation should be carefully considered to allow patients to receive more benefits from PT.
Although the PT requires a two-stage operation, its advantages outweigh the disadvantages. Owing to the potential benefits of the induced membrane, rich blood supply, and satisfactory union of the docking site, the PT can gather remarkably reduced EFI, satisfactory bone healing results, and better functional results. In addition, early removal of the external fixator could significantly reduce the risk of complications. In the PT group, the percentage and severity of pain, joint stiffness and pin tract infection were low, and no complications occurred that required further surgical treatment.

Conclusions
The limitations of our study are the relatively small sample size and short follow-up period. Nevertheless, the PT provides an early opportunity to remove the fixator, allowing for earlier rehabilitation exercises and eliminating patient discomfort. We believe that this new combined technique provides a feasible and improved method for the treatment of patients with bone defects after lower extremity infection.