Epiphysis Salvage Reconstruction and Associated Complications Following Tumor Resections in Skeletally Immature Patients

Abstract

The purpose of this study was to discuss the epiphysis salvage reconstruction procedure outcomes and complications in skeletally immature patients. The study included 12 patients with pediatric malignant tumors (osteosarcoma, Ewing’s sarcoma) located close to the epiphysis, who underwent epiphyseal-preservation surgery with vascularized fibula in the plastic surgery and orthopedics and traumatology clinic between January 2008 and November 2018. The study was designed retrospectively. Neoadjuvant and adjuvant chemotherapy was administered to all patients. None of the patients received radiotherapy. Functional evaluation was assessed using the Musculoskeletal Tumor Society (MSTS) scoring system. The patients were followed up for a mean period of 51 (21‑115) months. The mean age of patients was 10.4 (range, 5 to 15) years. Union times in the target region (epiphyseal region) and diaphyseal region were reported separately. Accordingly, the mean total union time in the target region of all patients was 7.9 months (5‑11), and the mean total union time in the diaphyseal region was 6.5 months (5‑9). According to the MSTS grading system, the mean score of the patients was 27.50 (range, 21‑30). The mean operative time was 480 (390‑540) min, the mean intraoperative blood loss was 790 (580‑1100) cc, and the mean length of hospital stay was 6 (4‑17) days. Six patients had shortness, 5 patients had delayed union, 3 patients had angular deformity, and 2 patients had infection. Epiphyseal-preservation surgery may be an alternative to other techniques used for malignant tumors located close to the growth line in skeletally immature patients. Despite the fact that this technique has plenty of complications, many of them can be successfully treated with secondary operations without leaving sequelae.

Introduction

The vast majority of malignant bone tumors are located close to the epiphyseal region of long tubular bones. Limb-salvage operations are currently preferred in 90% of these patients rather than amputation [1]. The vast majority of malignant tumors located in the metaphyseal region are resected over a large margin, including the articular surfaces, and the majority of resulting intercalar defects are repaired with endoprosthetic reconstruction. Although such reconstructions offer great advantages in the early period, there can be serious complications in the late period [2–4].

Resection of the epiphysis in growing children presents unique challenges in terms of limb length discrepancy throughout the child’s growth. In these patients, expendable endoprostheses are used to eliminate limb length discrepancy. However, 25‑80% of these prostheses are also revised within the first 5 years, and the risk of developing at least one complication in 10 years is approximately 80% [5–9].

Current multidisciplinary approaches include imaging techniques, multi-agent chemotherapies, surgical techniques, and careful postoperative care, which have led to improvements in overall survival rates in malignant tumors. These advances have changed the direction of surgical treatment in these tumors. In particular, there is now an increased tendency from non-biological techniques to biological techniques in growing children. Many techniques are used in biological limb-salvage surgery, with reconstruction options including vascularized fibular graft, the interposition of an allograft, a combination of vascularized fibular flap and allograft, insertion of an extracorporeally irradiated autograft, segmental transportation, and modified fibular flap technique [10–13]. However, with the use of these options, complications such as non-union of the allograft-host junction, infection, and fracture are common, especially in allografts and devitalized autografts used alone [14, 15].

In tumors close to the growth plate in pediatric patients, epiphyseal-preservation surgeries have come to the fore in recent years [16–18]. Advances in imaging techniques, multi-agent chemotherapies, and surgical techniques have been important developments bringing “epiphyseal-preservation surgery” to prominence. The most important advantage of this technique is that the biological joint is preserved, providing excellent limb function.

“Epiphyseal-preservation surgery” has been claimed to have very important advantages in growing children compared to non-biological techniques. The aim of this study was to discuss the outcomes and complications of this technique performed on 12 children in which the intercalar defect formed was repaired with a combined free vascularized fibular flap.

Methods

The study included 12 patients with pediatric malignant tumors located close to the epiphysis, who underwent epiphyseal-preservation surgery with vascularized fibula in the plastic surgery and orthopedics and traumatology clinic between January 2008 and November 2018, whose late follow-ups could be done, and who were not lost to follow-up. The mean age of the patients, five of whom were female and seven of whom were male, was 10.4 (5‑15) years. Seven were diagnosed with osteosarcoma, and five were diagnosed with Ewing sarcoma. Six of the tumors were located in the proximal tibia, two in the distal tibia region, and four in the distal femoral region. The study was designed retrospectively. Patients with metastasis and who did not sufficiently respond to neoadjuvant chemotherapy were excluded from the study. Because they could not be followed up for a long time, three of the patients treated with this technique were excluded because of their short life span. Endoprosthetic replacement therapy was applied to 2 patients over 10 years of age with short life expectancy excluded from the study.

Prior to the operation, plain radiography, CT, MRI, and PET-CT images were obtained from all patients. Neoadjuvant (3‑4 cyclus) and adjuvant (4‑6 cyclus) chemotherapy was administered to all patients. Euromac protocol including methotrexate, cisplatin, and doxorubicin was applied to patients with osteosarcoma. The Ewing 2008 protocol was applied to patients with Ewing sarcoma. None of the patients received radiotherapy. Functional evaluation was made using the Musculoskeletal Tumor Society (MSTS) scoring system, which includes pain, function, patient acceptance, need for external supports, walking ability, and gait [19]. In addition, the oncological outcomes, operation details, time to union, and complications of all the patients were reported.

Time to union was evaluated according to both ends of the transferred fibula. In other words, union times in the target region (epiphyseal region) and diaphyseal region were reported separately. If there was no finding of union at both ends after the third month, these cases were evaluated as delayed union. If there was no finding of union at both ends after the sixth month, these cases were also evaluated as non-union.

Surgical Technique

All the patients included in the study underwent surgery after receiving neoadjuvant chemotherapy, and when their general condition improved. First, tumor tissue was resected with a 1-cm margin of healthy tissue in line with the preoperative planning. The resection of the epiphyseal region was performed through the transphyseal line in 5 patients and the transepiphyseal line in the remaining 7 patients (Fig. 1). The combined free vascularized fibular flap (FVFF) technique was used in all patients during the operation, and stabilization was performed with plate screw. In all patients, the vascularized fibula was harvested from the contralateral leg. The harvested vascularized fibula was combined with autoclaved autograft in 4 patients and with massive allograft bone in the remaining 8 patients (Fig. 2). The specimens were heated in an autoclave machine at 130° for 8 min. FVFF was placed into a cortical strut allograft or autograft, with one end touching the epiphysis and the other touching the diaphysis. After this combination was placed in the defect region and stabilized, the autogenous spongious graft was placed in the target area (epiphyseal region) of 7 patients (Fig. 3). None of the other 5 patients received autogenous spongious grafts. In addition, no graft was applied to the diaphyseal junction of any patient. Cortical support allografts of a size suitable for the preserved epiphyseal form were added to the last 3 patients who were treated with massive allograft combination (Fig. 3). K wire or cannulated screw and/or plate screw fixation were applied for stabilization. All patients received four to six cycles of intravenous chemotherapy following surgery, depending on the response to neoadjuvant chemotherapy.

Fig. 1.

Transepiphyseal and transphyseal resection margins

Fig. 2.

a A 12-year-old female patient. Preoperative plain radiographs and MR images of the patient diagnosed with osteosarcoma located in the proximal metaphyseal region of the tibia. b Early postoperative X-ray of the same patient. She was treated with free vascularized fibular flap and a massive allograft combination. c Late follow-up radiographs of the same patient. This patient developed osteoarthrosis and delayed union complications in addition to valgus and shortness deformities

Fig. 3.

a Preoperative plain radiographs and MR images of a 5-year-old patient diagnosed with osteosarcoma. b Intraoperative, postoperative, and late radiographs of the same patient: (1) supporting the preserved epiphyseal part with specially shaped cortical allografts and gaining normal metaphyseal width; (2) femoral distal transepiphysial resection and placement of allograft + FVFF complex in the center in all three planes of the epiphyseal region; (3) supporting the fixation with K wires and minimal osteosynthesis materials; (4) adequacy of compression screws, one in the proximal and one in the distal region of the complex; (5) target region and addition of primary autogenous spongious graft; and (6) diaphysial junction region

Results

The patients comprised 5 females and 7 males with a mean age of 10.4 years (range, 5‑15 years). Osteosarcoma was diagnosed in 7 cases, and Ewing sarcoma in 5. The mean size of the intercalar defect formed was 12.5 (7‑20) cm. Surgery was performed within 1 week after neoadjuvant chemotherapy. Adjuvant chemotherapy was applied 10‑15 days after surgical treatment. Six of the tumors were located in the proximal tibia, two in the distal tibia region, and four in the distal femoral region.

All patients received three cycles of neoadjuvant chemotherapy. The patients were followed up for a mean period of 51 (21‑115) months. Time to union was evaluated according to both ends of the transferred fibula, so union times in the target region and diaphyseal region were reported separately. Accordingly, the mean total union time in the target region of all patients was 7.9 (5‑11) months, and the mean total union time in the diaphyseal region was 6.5 (5‑9) months.

The mean operative time was 480 (390‑540) min, the mean intraoperative blood loss was 790 (580‑1100) cc, and the mean length of hospital stay was 6 (4‑17) days.

Complications were evaluated in three categories as infection, union problems, and deformities (Table 1). In terms of infection, two patients developed a superficial infection. One patient developed superficial skin necrosis. In terms of union problems, five patients had delayed union, in the epiphyseal region in 4 patients and in the diaphyseal region in one patient. One patient developed non-union in the target area. In terms of deformity, six of the twelve patients developed shortness > 2 cm. The mean shortness was 3.6 (3‑5) cm. One patient developed genu varus deformity, two patients developed genu valgus deformity, and one patient developed epiphyseal collapse. Ligamentous laxity was not detected in any patient. Normal joint flexion extension was achieved in 11 patients.

Table 1.

Demographic characteristics and treatments of complications in the cases

Patient no.	Sex/age (first application)	Treatment	Complication	Complication treatment	Follow-up (month)	Result
1	12/F	FVFF + allograft	1-LLD 2-OA 3-Valgus deformity 4- Delayed union	1-Follow-up 2-Corrective osteotomy 3-Autogenous grafting	115	CDF
2	7/F	FVFF + allograft	1-LLD 2-Delayed union	1-Follow-up 2-Autogenous grafting	32	CDF
3	9/M	FVFF + allograft	1-LLD 2-Delayed union 3-LR 4-Süperficial infection	1-Follow-up 2-Autogenous grafting 3-Wide resection 4-Dressing + AB	49	CDF
4	10/M	FVFF + allograft	1-Delayed union 2-LLD 3-Skin necrosis	1-Autogenous grafting 2-Lengthening 3-Free flap + AB	44	CDF
5	8/F	FVFF + allograft	1-Non-union 2-LR 3-Varus deformity	1-İmplant replacement + autogenous grafting 2-Wide resection	21	CDF
6	5/F	FVFF + auto-clavised bone	1-AVN 2-Süperficial infection	1-Corrective osteotomy 2-Dressing + AB	51	CDF
7	6/M	FVFF + auto-clavised bone	1-LLD	Lengthening	96	CDF
8	12/M	FVFF + allograft	1-Valgus deformity 2-LLD 3-OA 4-Epiphysial collapse	1-Corrective osteotomy 2-Follow-up	34	CDF
9	14/F	FVFF + allograft			44	CDF
10	15/M	FVFF + auto-clavised bone			48	CDF
11	13/M	FVFF + auto-clavised bone		1. Follow-up	43	CDF
12	14/M	FVFF + allograft			40	CDF

FVFF free vascularized fibula flap, CDF continuous disease free, LR local recurrence, LLD limb length discrepancy, L lengthening, OA osteoarthritis, AVN avascular necrosis, AB anti-biotherapy

Oncological Outcome

Three of the patients treated with this technique died; two due to chemotherapy complications, and one due to lung metastasis. As these patients died before a sufficient follow-up duration, they were not included in the study evaluations. Local recurrence in the soft tissue developed in 2 patients. Both were treated with large resection and were evaluated as disease free in the final follow-up examination.

Functional Outcome

Based on the revised MSTS rating system, the patients had a mean score of 27.50 (range, 21‑30).

Discussion

Current advances in imaging techniques, multi-agent chemotherapy, and modern surgical techniques are important factors in increasing overall survival rates in malignant bone tumors. In parallel with these advances, there have been new developments in surgical techniques. The increase in overall survival rates in particular has increased the tendency of orthopedic oncologists from non-biological techniques to biological techniques. Although non-biological techniques are still among the first preferred techniques for adults and resections involving the joint, the most important disadvantage of these techniques is that time-dependent complication rates are high. Only 50% of patients survive more than 10 years after this technique, whereas survival rates are higher in biological techniques [3].

In recent years, there has been increasing interest in “epiphyseal-preservation surgery” as a biological technique for pediatric malignant tumors located close to the epiphyseal region [16–18]. In epiphyseal-preservation surgery, the patient retains native joint functions, which has great advantages, especially in the lower limb. Advances in imaging techniques have played an important role in the early recognition of tumors and preoperative surgical planning. Moreover, advances in chemotherapeutic agents have been another important factor in the more widespread use of epiphyseal-preservation surgery. However, there are few studies on the complications of epiphyseal-preservation surgery, which has taken its place among the biological techniques in recent years, and on the biological techniques to be used for the repair of the intercalar defect formed.

Many biological reconstruction techniques are used to repair intercalar defects formed in the skeletal system [10–13], of which the combined VFFF and callus distraction techniques are the most often used [20]. Each technique has its own advantages and disadvantages. The segmental bone transport technique in particular has important advantages as well as serious complications [21].

The biological technique commonly used for the repair of defects secondary to the tumor is free vascularized fibular flaps [22]. However, a long time is required for the fibular graft to functionally thicken. During this period of time, the rate of returning to macro-fracture for micro-fractures is high. Combined techniques such as allograft and recycled autograft have been described to eliminate some of the disadvantages of vascularized fibula used alone. Combined techniques can meet most of the load on the fibula until the fibula is remodeled, thereby minimizing fracture complications in the vascularized fibula [14, 22, 23]. In fact, the basic definition of combined techniques is to combine mechanical activity and biological activity in the same region. In other words, the vascularized bone flap can facilitate the union of inorganic cortical bone (allograft, autoclaved, irradiated bone tissue) with mechanical strength and provide osteointegration of the vascularized fibula as a result of progressive hypertrophy.

In this study, repair of the intercalar defect created was performed by preserving the epiphysis with a combined free vascularized fibular flap (FVFF). Transepiphyseal resection was applied to seven patients and transphyseal resection to the remaining five patients. This technique has two major challenges. The first is the difficulty in determining the intraoperative resection margin due to the status of the growth line. The second is the problem of fixation, especially in patients younger than 10 years old and undergoing transepiphyseal resection. We especially used 3D CT and MR imaging techniques during the preoperative planning to determine the incision margin. Resection was performed under the guidance of fluoroscopy to determine the correct margins intraoperatively, considering the preoperatively determined plan. None of the patients who were operated on according to this plan developed local recurrence in the osseous tissue.

Weitao et al. [24] treated 15 patients with epiphyseal-preservation surgery, and defect repair was performed with structural allograft and autogenous spongious grafts. All of these patients had delayed union and approximately 90% of the union delays occurred in the diaphyseal junction region. In a large-series study by Tinao et al. [17], only three (8.5%) of 35 patients had non-union. However, complications treated with additional surgical procedures were recorded for 19 patients (54%), including three local recurrences, two infections, 11 fractures, and three non-unions. In 10 of these 19 patients, the allograft was removed. Only five of a total of 35 patients (14%) lost the originally preserved epiphysis due to complications. In this study, FVFF was combined with a massive allograft in eight patients and autoclaved autograft in four patients. Structural allograft was preferred in patients where the osseous tissue was completely destroyed. Four of the twelve patients developed no complications, and the mean MSTS score was 27.75 (range, 27‑30). The remaining eight patients developed at least one complication (66.6%). The most common complications seen in these patients were shortness and union problems. Seven of the 12 patients treated with this technique had no union problems and progressed in the normal course. However, four of the remaining five patients developed delayed union (33%), and one patient developed non-union (8%). All of these patients were those combined with allografts. The region where the union delay and non-union occurred was the target region. All of these patients were successfully treated by replacing autogenous spongious graft and fixation materials.

In this study, contrary to the study by Y. Weitao et al. [24], all of the union problems occurred in the target region. The reasons for this can be considered to be due to insufficient stabilization in this region, inadequate bone stock, and poor blood supply to the epiphyseal region. In addition, the high union in the diaphyseal area was attributed to the vascularized fibula being in closer contact with the periosteum. The addition of autogenous spongious graft to the target region and augmenting the stabilization with cross K wires could minimize this complication. Moreover, the vascularized fibula should be embedded in the epiphyseal level in this region at the maximum level. In addition, fixation should be carried out using a minimum number of materials that will provide maximum stability. Delayed union was observed in only one of the patients who was operated on following these rules.

Limb length discrepancy is one of the most common and inevitable deformities in epiphyseal-preservation surgeries in pediatric patients. In the study by Y. Weitao et al. [24], four of 15 patients developed shortness. In the study by Kiss et al. in their series of 5 patients performed epiphysis preserving surgery, it was stated that an average of 2.6 cm (1.5‑4.5) shortness developed [25]. In this study, 3 out of 5 patients had fractures in the transferred fibula. The fact that no fibula fracture was seen in our study was attributed to its support with cortical allografts. Takeouchi et al. reported a mean shortening of 2.6 cm (0.1‑4.8) in their series of 12 patients, and it was shown that no epiphyseal collapse developed [26]. In the series of 5 diseases of Li et al., 3 patients had shortness problem, but no union problem was observed [27]. In other studies, the rates of limb length discrepancy were not indicated. In the current study, none of the patients aged 13 and over developed shortness > 2 cm, whereas eight patients under the age of 13 developed shortness ranging between 2 and 6 cm. Only two of these patients accepted a lengthening operation. None of the other patients accepted surgical intervention and maintained their functions in this way. Another important complication is angular deformities. In the current study, three of the twelve patients developed frontal plane deformities. All of these patients were those treated with allograft. It was thought that these deformities were caused by not completely supporting the preserved epiphyseal region with cortical allografts, soft tissue imbalance, and resection margin. In order to minimize this deformity, it may be beneficial to avoid this complication by supporting this region with massive metaphyseal allografts suitable for the shape of the preserved epiphysis or with cortical allografts shaped according to the remaining epiphyseal region.

Another complication is the development of osteoarthritic findings in the late follow-up of these patients. This complication can develop for many reasons. Especially in transepiphyseal resections, the impaired blood supply to the preserved epiphyseal part, subsequent shortness, or angular deformities can be important reasons for the development of this complication.

Two of the current study patients developed local recurrence, both of which were in the soft tissue area on the biopsy tract. None of the patients developed recurrence in the resection margins. This complication was successfully treated with large resection.

This study has some important limitations. First of all, this study is a retrospective study and does not include a control group. The low number of cases is another disadvantage. However, reaching a sufficient number of cases involves a long time, considering that these tumors are rare, and this technique is used in selected patients. This is especially difficult in malignant tumors. Another important limitation is that detailed information on the indication limits of this intervention could not be provided due to the insufficient number of cases.

In conclusion, although the small number of cases prevented precise information, epiphyseal-preservation surgery may be an alternative to other techniques used for malignant tumors located close to the growth line in skeletally immature patients. Despite the fact that this technique has many complications, most of these can be successfully treated with secondary operations without leaving sequelae. Complications such as union problems and frontal plane deformities, especially in the target region, can be minimized by measures to be taken.

Authors’ Contributions

Study concepts: Rana Kapukaya, Evren Karaali, Osman Çiloğlu

Study design: Rana Kapukaya, Evren Karaali, Hasan Ulaş Oğur

Data acquisition: Rana Kapukaya, Mehmet Baydar, Hasan Ulaş Oğur

Quality control of data and algorithms: Rana Kapukaya, Hasan Ulaş Oğur, Mehmet Baydar

Data analysis and interpretation: Rana Kapukaya, Evren Karaali, Mehmet Baydar

Statistical analysis: Rana Kapukaya, Osman Çiloğlu, Hasan Ulaş Oğur

Manuscript preparation: Rana Kapukaya, Evren Karaali, Osman Çiloğlu

Manuscript editing: Rana Kapukaya, Evren Karaali, Mehmet Baydar

Manuscript review: Rana Kapukaya, Evren Karaali, Kahraman Öztürk

Data Availability

Not applicable.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Code Availability

Not applicable.