Abstract
Background: Limb-salvage surgery is the preferred treatment for primary bone tumors, yet postoperative imaging and functional outcomes remain variable. This study aims to identify factors affecting postoperative imaging and functional recovery outcomes in patients undergoing limb-salvage surgery for primary bone tumors. Methods: A retrospective case-control study was conducted on 231 adult patients with primary bone tumors who underwent limb-salvage surgery at a single institution between January 2016 and January 2024. Patients were categorized into favorable and adverse outcome groups based on postoperative imaging, local recurrence, and Musculoskeletal Tumor Society scores. Data were collected on demographics, tumor characteristics, surgical methods, nutritional status, adjuvant treatments, supportive care, and complications. Univariate and multivariate logistic regression analyses were performed to identify factors independently associated with favorable outcomes. Results: Significant differences were found between outcome groups in age, lymphocyte count, preoperative chemotherapy, tumor margin clarity, tumor size, resection method, adjuvant treatment, rehabilitation, and infection rates. Patients with favorable outcomes were younger, had higher lymphocyte counts and albumin levels, received more preoperative chemotherapy and adjuvant treatments, displayed clearer tumor margins, underwent wide resection more often, had smaller tumors, and participated more in rehabilitation. The adverse group had higher rates of infection and marginal resections. No significant differences were observed in tumor type, location, reconstruction method, prealbumin and hemoglobin levels, or rates of pain management, psychological support, delayed wound healing, joint stiffness, or muscle atrophy. Multivariate analysis identified younger age, wide resection, and adjuvant treatment as independent predictors of favorable imaging and functional recovery. Conclusion: Younger age, wide surgical resection, and receipt of adjuvant treatment independently predict improved postoperative imaging and functional outcomes following limb-salvage surgery for primary bone tumors. Optimizing perioperative management and surgical strategies may enhance patient recovery and long-term outcomes.
Keywords: Limb-salvage surgery, bone tumors, functional outcome, imaging outcome, prognostic factors, retrospective study
Introduction
Bone tumors, though relatively rare, pose significant diagnostic and therapeutic challenges due to their heterogeneous biological behavior and potential for substantial morbidity [1]. In children, adolescents, and young adults, primary malignant bone tumors like osteosarcoma, chondrosarcoma, and Ewing’s sarcoma are most common [2]. These cancers threaten both bone structure and limb function, necessitating a delicate balance between tumor control and quality of life preservation [3].
Advances in musculoskeletal oncology and reconstructive surgery have made limb-salvage procedures the preferred treatment for many patients with primary bone tumors [4]. Traditional amputation, once the standard approach to ensure tumor-free margins, is now mainly reserved for cases with extensive neurovascular invasion or when tumor removal would severely compromise limb function [5]. Limb-salvage surgery, offering local tumor control similar to amputation in suitable cases, preserves limb structure, leading to better functional outcomes and psychosocial adjustment [6]. Indications for limb-salvage surgery include tumors classified as Enneking I, II, or III, and no extensive involvement of neurovascular structures [7]. Achieving adequate surgical margins, whether wide or marginal, is also crucial. Additionally, sufficient soft tissue coverage is essential for supporting reconstruction and wound healing. The patient must be in good health, able to tolerate lengthy surgeries, and committed to postoperative rehabilitation. Absolute contraindications typically include extensive neurovascular involvement, widespread contamination from pathological fractures, and conditions that hinder healing, such as severe infections. The procedure typically involves wide or marginal excision of the tumor, followed by reconstruction using endoprostheses, allografts, or autografts [8]. While limb-salvage surgery offers significant benefits, it remains technically challenging, with potential failures due to local recurrence, infection, implant issues, or suboptimal functional recovery [9].
Postoperative outcomes following limb-salvage surgery vary significantly, and understanding the determinants that distinguish favorable from adverse imaging and functional results remains a critical area of research. Imaging follow-up, principally using radiography and MRI, is essential for detecting complications, recurrence, and assessment of reconstruction integrity [10]. Functional recovery is typically quantified using standardized systems such as the Musculoskeletal Tumor Society (MSTS) score [11], which assesses limb function and quality of life across six domains: pain, functional activity, emotional acceptance, brace use, walking ability, and gait. The MSTS score has demonstrated good validity and reliability, distinguishing functional differences among surgical methods and rehabilitation levels [12]. It has become an essential tool in evaluating limb-salvage surgery outcomes. However, a considerable proportion of patients fail to achieve satisfactory imaging or functional recovery even when histologically negative margins are achieved [13]. The variability in outcomes is attributed to a complex interplay of patient-related, tumor-specific, surgical, and perioperative factors, but definitive prognostic indicators remain elusive [14].
Previous studies [15,16] have investigated factors influencing limb-salvage outcomes, including tumor characteristics, patient demographics, surgical technique, and perioperative management. However, many studies have limitations, such as heterogeneous patient populations, variable follow-up durations, or a narrow focus on specific outcomes like recurrence or prosthetic complications, rather than integrated imaging and functional results. The evolving landscape of adjuvant therapies, advancements in imaging modalities, and improvements in prosthetic and graft technologies highlight the need for updated evidence to guide clinical decision-making and personalize patient management [17].
To address these gaps, the present retrospective case-control study systematically analyzed factors influencing both imaging and functional recovery outcomes in patients undergoing limb-salvage surgery for bone tumors. By leveraging a comprehensive dataset - including detailed patient demographics, clinical history, laboratory values, tumor features, surgical and reconstructive methods, nutritional indices, postoperative care strategies, and one-year follow-up data - this investigation aims to identify both well-established and novel predictors for successful limb preservation.
Methods
Research design
In this retrospective case-control study, a total of 231 patients who underwent limb-salvage surgery at our hospital between January 2016 and January 2024 were included, with follow-up until January 2025 or until death. The patients were divided into two groups based on their imaging and functional recovery outcomes one year after surgery.
Patients in the adverse outcome group met one or more of the following criteria: postoperative imaging findings indicated residual tumor or unclear margins, local recurrence was observed during follow-up, or the functional score (MSTS Score) was below 21. The MSTS score assesses functional recovery across six domains: pain, functional activity, emotional acceptance, external support (use of braces), walking ability, and gait [18]. Each domain is scored from 0 to 5, with a total score of 30 points. Scoring was conducted independently by a senior orthopedic surgeon. A higher total score indicates better functional recovery. In total, 110 patients were categorized into the poor outcome group. Conversely, patients in the good outcome group (n=121) demonstrated complete tumor resection with clear margins, no residual tumor on postoperative imaging, no local recurrence during follow-up, and an MSTS score of 21 or higher.
To validate the findings and assess the generalizability of the predictive model, we included an additional 109 eligible patients from the same period as an external validation group. These patients were also categorized into adverse outcome (n=51) and favorable outcome (n=51) groups based on the same criteria. This step ensured that our findings were not specific to a particular sample, thereby enhancing the reliability and robustness of our conclusions.
The primary objective of this study is to identify factors influencing imaging and functional outcomes using comprehensive statistical methods. As this study utilized anonymized patient case data retrospectively and did not affect treatment, it was approved by Affiliated Hospital of Hebei University of Engineering ethics committee with a waiver of informed consent.
Inclusion and exclusion criteria
Inclusion Criteria: patients were included if they met the following criteria: (1) a diagnosis with primary bone tumors confirmed by pathology [19]; (2) receipt of limb-salvage surgery; (3) tumor in a site amenable to safe resection without involving critical neurovascular structures; (4) aged 18 years or older; (5) complete medical records, including preoperative assessment, surgical records, postoperative imaging, and functional recovery evaluations; (6) sufficient follow-up data, particularly at least six months of functional recovery and imaging assessment post-surgery; (7) records indicating whether they received adjuvant treatments such as chemotherapy or radiotherapy and the specific treatment protocols.
Exclusion criteria: (1) diagnosed with another malignancy within the past five years and had received treatment; (2) suffered from severe heart disease, lung disease, liver or kidney dysfunction, or other comorbidities that significantly impaired quality of life and functional recovery; (3) underwent amputation surgery instead of limb-salvage surgery; (4) had severe cognitive impairment that prevented cooperation with postoperative assessments and rehabilitation training; (5) had hyperglycemia that could potentially impact recovery and overall outcomes; (6) missing critical data (e.g., preoperative assessment, surgical records, postoperative imaging examinations, functional recovery evaluations).
Figure 1 illustrates the inclusion and exclusion criteria and the flow of patient inclusion in this study.
Figure 1.
The patient inclusion process.
Data collection
Baseline data
Baseline data were collected from the medical record system and included patients’ age, gender, BMI, underlying diseases, and medical history. Blood test results were sourced from comprehensive examinations conducted at the time of admission. Regular follow-up visits included MRI scans (Signa HDxt 1.5T MRI system, General Electric Healthcare, USA) to detect prosthetic loosening, evaluate tumor margins clarity, and monitor for tumor recurrence. These methods ensured a thorough assessment of patient status both at baseline and during subsequent follow-ups, providing critical information for evaluating treatment outcomes and guiding further interventions. The MSTS scoring system was utilized to evaluate functional outcomes and quality of life one year after surgery [20].
Tumor characteristics
Tumor type was confirmed through histopathological examination. The location and size of the tumor were determined via MRI. The Enneking staging system was used to stage bone tumors, evaluating histological grade, anatomic extent, and metastasis. In histological grading, GX represented an unassessable grade, G1 represented low-grade malignancy, and G2 indicated high-grade malignancy. The anatomic extent was classified as T0 (intra-capsular), T1 (intra-compartmental), and T2 (extra-compartmental). Metastasis status was categorized as M0 (no metastasis) and M1 (with metastasis). Malignant tumors were staged as IA (G1T1M0), IB (G1T2M0), IIA (G2T1M0), IIB (G2T2M0), and III (any G, any T but M1). Staging was based on imaging, biopsy results, and a comprehensive assessment by a specialist regarding tumor aggressiveness and extent of spread [21].
Surgical approaches
Surgical approaches include both resection and reconstruction, which are often documented in the patient’s medical records. The choice of surgical approach was based on the specific location, size, and its relationship to surrounding anatomical structures. Preoperative imaging was conducted to assess the tumor’s extent and its relationship with surrounding tissues. Detailed surgical plans, including the extent of resection and reconstruction methods, were formulated based on imaging results. All limb-salvage surgeries were performed by experienced senior orthopedic oncologists. Wide excision aimed to remove the tumor en bloc with a surrounding layer of normal tissue, while marginal excision involved removing the tumor along its pseudocapsule or reactive zone. Intraoperative frozen section analysis was used to confirm adequate resection margins. After tumor resection, reconstruction was tailored to the specific needs of the patient: prosthetic replacement was used for large defects around joints; composite reconstruction combined prostheses with allografts or vascularized autografts (such as fibular grafts); for smaller defects, autologous bone grafting was employed. Following skeletal reconstruction, soft tissue repair was performed as needed, including tendons, ligaments, and skin reconstruction, to restore limb function and appearance.
Nutritional status
On the first day of hospital admission, fasting venous blood samples were collected in the morning. After processing, the samples were frozen and stored at -80°C. Albumin and prealbumin levels were measured using an automated biochemical analyzer (ADVIA Chemistry Systems, Siemens Healthineers, Germany), and hemoglobin levels were measured using an automated hematology analyzer (Sysmex XN-Series, Sysmex Corporation, Japan). The Nutritional Risk Index (NRI) was calculated using the following formula [22]:
Postoperative management
To retrieve nursing and treatment records from the medical record system, we documented whether patients received adjuvant treatments such as radiotherapy or chemotherapy within one year after surgery (before evaluation). Radiotherapy protocols were tailored based on tumor type, location, and surgical resection status, while chemotherapy regimens were selected according to tumor pathology, stage, and patient characteristics. All treatment decisions were made by a multidisciplinary oncology committee, considering the patient’s specific conditions, pathology results, and established clinical guidelines. To support patient recovery, we also recorded whether each patient participated in an individualized rehabilitation program, which included physical therapy, occupational therapy, and exercise therapy aimed at promoting limb function recovery and preventing complications. Rehabilitation plans were tailored to the type of surgery, functional status, and personal needs of the patient. Pain management plays a crucial role in postoperative recovery. We used a multimodal analgesic strategy involving nonsteroidal anti-inflammatory drugs, opioids, and local anesthetics. For patients with chronic pain, antidepressants or anticonvulsants were considered to alleviate neuropathic pain. Psychological support services, such as counseling, cognitive-behavioral therapy, and family support, were also provided to help patients manage anxiety, depression, and fear associated with their illness, enhancing self-efficacy and quality of life.
Postoperative complications
The medical record system was reviewed to identify postoperative complications, including infection, delayed wound healing (both deep and superficial), joint stiffness, and muscle atrophy.
Survival analysis
Patients were followed until the end of the study or until a death event occurred, with outcomes recorded. Overall survival (OS) and progression-free survival (PFS) were documented.
Data analysis
Statistical analysis were performed using SPSS version 29.0 (IBM Corporation, Armonk, NY, USA). To ensure data validity and accuracy, categorical variables were summarized using frequencies and percentages [n (%)]. The chi-square test (χ2) was used for categorical data comparisons. Continuous data that followed a normal distribution were reported as means ± standard deviations (M±SD), and group comparisons were performed using independent samples t-tests. The significance level was set at α = 0.05.
To identify the factors influencing postoperative imaging and functional recovery outcomes in patients undergoing limb-salvage surgery for bone tumors, both univariate and multivariate logistic regression analyses were conducted. In the univariate logistic regression analysis, the impact of individual factors on functional recovery outcomes was assessed. These factors included clinical characteristics, laboratory indicators, and treatment-related variables. Variables showing statistical significance (P < 0.05) in the univariate analysis were subsequently included in the multivariate logistic regression model. The specific formula is as follows:
Where P is the probability of experiencing an adverse outcome; X represents the independent variables; β0 is the intercept; βn are the corresponding regression coefficients.
This approach allowed us to adjust for potential confounders and identification of independent predictors of functional recovery outcomes. Results from the logistic regression analyses were presented as odds ratios (ORs) with their respective 95% confidence intervals (CIs) and p-values. Additionally, the performance of the multivariate model was evaluated using the area under the receiver operating characteristic curve (AUC), which measures the model’s discrimination ability.
For external validation, ROC curve analysis was performed on an independent dataset. The AUC was calculated to assess the model’s discriminatory ability. Sensitivity and specificity were determined at various cut-off points, with the optimal cut-off identified using the Youden Index.
Result
Baseline data
Among the 231 patients included in the study, 110 were classified into the adverse outcome group and 121 into the favorable outcome group based on imaging and functional recovery (Table 1). Patients in the adverse outcomes group were older and had lower mean lymphocyte counts compared to those in favorable outcome group. The adverse outcome group also showed significantly higher neutrophil counts and alkaline phosphatase (ALP) levels. There were no significant differences in alanine aminotransferase (ALT) levels between the groups. Fewer patients in the adverse outcome group received preoperative chemotherapy, and tumor margins were less clear on preoperative imaging. Tumor recurrence occurred in 6.36% of patients in the adverse outcome group, while none in the favorable outcome group. The postoperative MSTS score was significantly lower in the adverse outcome group. No significant differences were observed between groups in gender distribution, smoking or alcohol history, BMI, muscle mass, osteoporosis, pathological fracture, hyperlipidemia, hypertension, or prosthetic loosening.
Table 1.
Baseline data
| Adverse Outcome Group (n=110) | Favorable Outcome Group (n=121) | t/χ2 | P | |
|---|---|---|---|---|
| Age (years) | 48.09 ± 8.24 | 45.22 ± 8.37 | 2.618 | 0.009 |
| Gender | 0.174 | 0.677 | ||
| Female | 47 (42.73%) | 55 (45.45%) | ||
| Male | 63 (57.27%) | 66 (54.55%) | ||
| Smoking History (yes) | 33 (30.00%) | 41 (33.88%) | 0.399 | 0.527 |
| Alcohol Consumption History (yes) | 50 (45.45%) | 56 (46.28%) | 0.016 | 0.900 |
| Education Level | 0.241 | 0.887 | ||
| Elementary School or Below | 31 (28.18%) | 32 (26.45%) | ||
| Junior High School/High School | 52 (47.27%) | 56 (46.28%) | ||
| University and Above | 27 (24.55%) | 33 (27.27%) | ||
| BMI (kg/m2) | 21.26 ± 1.28 | 20.97 ± 1.39 | 1.636 | 0.103 |
| Muscle Mass (%) | 28.71 ± 2.47 | 29.34 ± 2.55 | 1.878 | 0.062 |
| Osteoporosis (yes) | 42 (38.18%) | 43 (35.54%) | 0.173 | 0.677 |
| Pathological Fracture (yes) | 5 (4.55%) | 6 (4.96%) | 0.022 | 0.883 |
| Hyperlipidemia (yes) | 12 (10.91%) | 13 (10.74%) | 0.002 | 0.968 |
| Hypertension (yes) | 15 (13.64%) | 17 (14.05%) | 0.008 | 0.928 |
| Lymphocyte Count (×109/L) | 2.70 ± 0.51 | 2.84 ± 0.55 | 2.049 | 0.042 |
| Neutrophil count (×109/L) | 5.72 ± 1.63 | 5.21 ± 1.47 | 2.488 | 0.014 |
| ALT (U/L) | 46.13 ± 13.47 | 43.52 ± 12.63 | 1.519 | 0.130 |
| ALP (U/L) | 213.53 ± 55.27 | 198.24 ± 51.24 | 2.182 | 0.030 |
| Preoperative Chemotherapy (yes) | 91 (82.73%) | 111 (91.74%) | 4.259 | 0.039 |
| Prosthetic Loosening (yes) | 11 (10.00%) | 5 (4.13%) | 3.077 | 0.079 |
| Preoperative Tumor Margin Clarity (yes) | 89 (80.91%) | 109 (90.08%) | 3.960 | 0.047 |
| Tumor Recurrence (yes) | 7 (6.36%) | 0 (0%) | 5.923 | 0.015 |
| Postoperative MSTS Score (points) | 19.47 ± 1.36 | 24.19 ± 1.48 | 25.192 | < 0.001 |
BMI: body mass index; MSTS: Musculoskeletal Tumor Society Scoring System; ALT: Alanine aminotransferase; ALP: Alkaline phosphatase.
Tumor characteristics
Analysis of tumor characteristics revealed no significant differences between the two groups in terms of tumor type or tumor location (Figure 2). The distribution of osteosarcoma, chondrosarcoma, Ewing’s sarcoma, and giant cell tumor of bone was similar between groups, as was the distribution of tumors at the proximal femur, distal femur, and tibia. However, the Enneking stage differed significantly between groups; a higher proportion of patients in the favorable outcome group had stage IIA tumors, whereas stage IIB tumors were more frequent in the adverse outcome group. Additionally, the adverse outcome group had larger tumors, with a greater mean maximum tumor diameter compared with the favorable outcome group.
Figure 2.
Tumor Characteristics. A: Tumor Type; B: Tumor Location; C: Enneking Stage; D: Tumor Size (Maximum Diameter).
Surgical resection and reconstruction methods
A significant difference in resection methods was observed between groups, with wide resection was more frequently performed in the favorable outcome group and marginal resection more common in the adverse outcome group (Figure 3). No significant difference was observed in reconstruction methods between groups, with similar distributions for prosthetic replacement, composite reconstruction, and autograft bone grafting or other techniques.
Figure 3.
Surgical resection and reconstruction methods. A: Resection Method; B: Reconstruction Method.
Nutritional status
Analysis of nutritional status revealed that patients in the favorable outcome group had significantly higher mean albumin levels and higher Nutritional Risk Index scores compared with the adverse outcome group (Table 2). There were no significant differences in prealbumin or hemoglobin levels between groups.
Table 2.
Nutritional status
| Adverse Outcome Group (n=110) | Favorable Outcome Group (n=121) | t | P | |
|---|---|---|---|---|
| Albumin (g/dL) | 4.21 ± 0.48 | 4.35 ± 0.51 | 2.089 | 0.038 |
| Prealbumin (mg/L) | 195.72 ± 15.33 | 199.31 ± 14.82 | 1.808 | 0.072 |
| Hemoglobin (g/dL) | 14.68 ± 1.05 | 14.92 ± 1.01 | 1.749 | 0.082 |
| NRI (score) | 82.25 ± 2.76 | 83.34 ± 2.68 | 3.041 | 0.003 |
NRI: Nutritional Risk Index.
Adjuvant treatments and supportive care
Analysis of adjuvant treatments and supportive care demonstrated that a significantly higher proportion of patients in the favorable outcome group received adjuvant treatment and rehabilitation training compared with the adverse outcome group (Figure 4). Pain management and psychological support rates did not differ significantly between groups.
Figure 4.
Adjuvant treatments and supportive care.
Postoperative complications
Analysis of postoperative complications revealed that infection occurred significantly more frequently in the adverse outcome group compared with the favorable outcome group (Table 3). There were no significant differences between groups in the rates of delayed wound healing, joint stiffness, or muscle atrophy.
Table 3.
Postoperative complications
| Adverse Outcome Group (n=110) | Favorable Outcome Group (n=121) | χ2 | P | |
|---|---|---|---|---|
| Infection (yes) | 13 (11.82%) | 5 (4.13%) | 4.737 | 0.030 |
| Delayed Wound Healing (yes) | 4 (3.64%) | 4 (3.31%) | 0 | 1.000 |
| Joint Stiffness (yes) | 15 (13.64%) | 13 (10.74%) | 0.453 | 0.501 |
| Muscle Atrophy (yes) | 27 (24.55%) | 25 (20.66%) | 0.498 | 0.480 |
Survival analysis
Patients in the favorable outcome group exhibited significantly longer OS and PFS compared to those in the adverse outcome group (Figure 5). No other indicators showed significant differences between the two groups. The results highlight the clear distinction in survival outcomes, underscoring the importance of identifying factors contributing to better prognosis.
Figure 5.
Survival curves. A. OS (months); B. PFS (months). OS: overall survival; PFS: progression-free survival.
Correlation analysis of various factors with postoperative imaging and functional recovery outcomes
Correlation analysis showed that younger age, higher lymphocyte count, preoperative chemotherapy, clearer preoperative tumor margins, smaller tumor size, wide resection, adjuvant treatment, rehabilitation training, and absence of infection were significantly associated with better postoperative imaging and functional recovery outcomes, whereas Enneking stage and albumin level did not reach statistical significance (Table 4). Additionally, higher neutrophil count and elevated ALP levels were also significantly associated with poorer outcomes. Infection showed perfect sensitivity but lacked specificity.
Table 4.
Correlation analysis of various factors with postoperative imaging and functional recovery outcomes
| rho | P | |
|---|---|---|
| Age (years) | -0.185 | 0.005 |
| Lymphocyte count (<2.5×109/L vs. ≥2.5×109/L) | 0.138 | 0.036 |
| Neutrophil count (≥6.0×109/L vs. <6.0×109/L) | -0.402 | 0.014 |
| ALP (≥250 U/L vs. <250 U/L) | -0.421 | 0.030 |
| Preoperative Chemotherapy (yes) | 0.136 | 0.039 |
| Preoperative Tumor Margin Clarity (yes) | 0.131 | 0.047 |
| Enneking stage | -0.094 | 0.153 |
| Tumor Size (Maximum Diameter <8 cm vs. ≥8 cm) | -0.130 | 0.049 |
| Resection Method (Wide Resection/Marginal Resection) | 0.139 | 0.035 |
| Albumin (<4.0 g/dL vs. ≥4.0 g/dL) | 0.127 | 0.054 |
| Adjuvant Treatment (yes) | 0.156 | 0.017 |
| Rehabilitation Training (yes) | 0.136 | 0.039 |
| Infection (yes) | -0.143 | 0.030 |
In the univariate regression analysis (Table 5), younger age, higher lymphocyte count, preoperative chemotherapy, clearer tumor margins, smaller tumor size, wide resection, higher albumin, adjuvant treatment, rehabilitation training, and absence of infection were significantly associated with favorable outcomes, while Enneking stage was not. Elevated neutrophil count and higher ALP levels were significantly associated with poorer outcomes. Multivariate regression (Table 6) identified age, wide resection, and adjuvant treatment as independent predictors of better imaging and functional recovery. In contrast, associations for lymphocyte count, neutrophil count, preoperative chemotherapy, tumor margin clarity, tumor size, albumin, ALP, rehabilitation training, Enneking stage, and infection were not statistically significant in multivariate analysis.
Table 5.
Univariate regression analysis of various factors associated with postoperative imaging and functional recovery outcomes
| Indicator | coefficient | OR | CI lower | CI upper | P |
|---|---|---|---|---|---|
| Age (years) | -0.042 | 0.959 | 0.928 | 0.990 | 0.011 |
| Lymphocyte count (<2.5×109/L vs. ≥2.5×109/L) | 0.511 | 1.667 | 1.022 | 2.764 | 0.043 |
| Neutrophil count (≥6.0×109/L vs. <6.0×109/L) | 0.382 | 0.224 | 0.658 | 0.049 | |
| ALP (≥250 U/L vs. <250 U/L) | 0.425 | 0.254 | 0.709 | 0.047 | |
| Preoperative Chemotherapy (yes) | 0.841 | 2.318 | 1.046 | 5.423 | 0.043 |
| Preoperative Tumor Margin Clarity (yes) | 0.762 | 2.143 | 1.014 | 4.719 | 0.050 |
| Enneking stage (per lower stage) | -0.218 | 0.804 | 0.563 | 1.142 | 0.225 |
| Tumor Size (Maximum Diameter <8 cm vs. ≥8 cm) | -0.231 | 0.794 | 0.636 | 0.984 | 0.038 |
| Resection Method (Wide Resection/Marginal Resection) | 0.657 | 1.929 | 1.047 | 3.608 | 0.037 |
| Albumin (<4.0 g/dL vs. ≥4.0 g/dL) | 0.555 | 1.742 | 1.034 | 2.984 | 0.039 |
| Adjuvant Treatment (yes) | 1.219 | 3.383 | 1.245 | 10.775 | 0.024 |
| Rehabilitation Training (yes) | 0.841 | 2.318 | 1.046 | 5.423 | 0.043 |
| Infection (yes) | -1.134 | 0.322 | 0.100 | 0.886 | 0.037 |
Table 6.
Multivariate regression analysis of various factors associated with postoperative imaging and functional recovery outcomes
| Indicator | Coefficient | OR | OR CI Lower | OR CI Upper | P |
|---|---|---|---|---|---|
| Age (years) | -0.044 | 0.957 | 0.924 | 0.991 | 0.014 |
| Lymphocyte count (<2.5×109/L vs. ≥2.5×109/L) | 0.478 | 1.613 | 0.943 | 2.758 | 0.081 |
| Neutrophil count (≥6.0×109/L vs. <6.0×109/L) | -0.329 | 0.724 | 0.404 | 0.132 | 0.274 |
| ALP (≥250 U/L vs. <250 U/L) | -0.371 | 0.854 | 0.473 | 1.542 | 0.589 |
| Preoperative Chemotherapy (yes) | 0.861 | 2.366 | 0.955 | 5.862 | 0.063 |
| Preoperative Tumor Margin Clarity (yes) | 0.667 | 1.948 | 0.854 | 4.442 | 0.113 |
| Enneking stage (per lower stage) | -0.229 | 0.796 | 0.537 | 1.179 | 0.255 |
| Tumor Size (Maximum Diameter <8 cm vs. ≥8 cm) | -0.244 | 0.783 | 0.609 | 1.006 | 0.056 |
| Resection Method (Wide Resection/Marginal Resection) | 0.794 | 2.213 | 1.109 | 4.416 | 0.024 |
| Albumin (<4.0 g/dL vs. ≥4.0 g/dL) | 0.404 | 1.498 | 0.841 | 2.669 | 0.170 |
| Adjuvant Treatment (yes) | 1.161 | 3.194 | 1.030 | 9.912 | 0.044 |
| Rehabilitation Training (yes) | 0.812 | 2.253 | 0.918 | 5.527 | 0.076 |
| Infection (yes) | -0.812 | 0.444 | 0.137 | 1.435 | 0.175 |
External validation of the predictive model
In the external validation set, significant differences were observed between the adverse outcome group and the favorable outcome group for several parameters (Table 7). The adverse outcome group had significantly higher age, lower lymphocyte count, and less frequent preoperative chemotherapy. Tumor margin clarity was also lower, and tumor recurrence occurred exclusively in the adverse outcome group. Postoperative MSTS scores were significantly lower in the adverse outcome group. Other significant differences included Enneking stage distribution, tumor size, and resection method, all favoring the favorable outcome group. Serum albumin levels were lower, and adjuvant treatment was less frequent in the adverse outcome group. Rehabilitation training was also less common, while infection rates were higher. No significant differences were found in gender, smoking history, alcohol consumption, education level, BMI, muscle mass, osteoporosis, pathological fracture, hyperlipidemia, hypertension, or prosthetic loosening.
Table 7.
Comparison of parameters between adverse and favorable outcome groups in the external validation set
| Indicator | Adverse Outcome Group (n=51) | Favorable Outcome Group (n=58) | t/χ2 | P |
|---|---|---|---|---|
| Age (years) | 47.62 ± 7.61 | 44.03 ± 7.33 | 2.511 | 0.014 |
| Gender | 0.005 | 0.942 | ||
| Female | 19 (37.25%) | 22 (37.93%) | ||
| Male | 32 (62.75%) | 36 (62.07%) | ||
| Smoking History (yes) | 17 (33.33%) | 18 (31.03%) | 0.066 | 0.798 |
| Alcohol Consumption History (yes) | 23 (45.10%) | 25 (43.10%) | 0.044 | 0.834 |
| Education Level | 0.622 | 0.733 | ||
| Elementary School or Below | 11 (21.57%) | 16 (27.59%) | ||
| Junior High School/High School | 27 (52.94%) | 27 (46.55%) | ||
| University and Above | 13 (25.49%) | 15 (25.86%) | ||
| BMI (kg/m2) | 20.82 ± 1.34 | 21.07 ± 1.25 | 1.018 | 0.311 |
| Muscle Mass (%) | 29.14 ± 2.32 | 29.21 ± 2.35 | 0.168 | 0.867 |
| Osteoporosis (yes) | 18 (35.29%) | 21 (36.21%) | 0.010 | 0.921 |
| Pathological Fracture (yes) | 3 (5.88%) | 3 (5.17%) | 0 | 1.000 |
| Hyperlipidemia (yes) | 5 (9.80%) | 6 (10.34%) | 0.009 | 0.925 |
| Hypertension (yes) | 6 (11.76%) | 8 (13.79%) | 0.100 | 0.752 |
| Lymphocyte Count (×109/L) | 2.72 ± 0.48 | 2.91 ± 0.43 | 2.175 | 0.032 |
| Preoperative Chemotherapy (yes) | 36 (70.59%) | 50 (86.21%) | 3.976 | 0.046 |
| Prosthetic Loosening (yes) | 6 (11.76%) | 5 (8.62%) | 0.296 | 0.587 |
| Preoperative Tumor Margin Clarity (yes) | 33 (64.71%) | 48 (82.76%) | 4.633 | 0.031 |
| Tumor Recurrence (yes) | 5 (9.80%) | 0 (0%) | 3.930 | 0.047 |
| Postoperative MSTS Score (score) | 19.68 ± 1.31 | 23.83 ± 1.39 | 15.950 | < 0.001 |
| Enneking Stage | 6.084 | 0.048 | ||
| I | 6 (11.76%) | 14 (24.14%) | ||
| IIA | 20 (39.22%) | 28 (48.28%) | ||
| IIB | 25 (49.02%) | 16 (27.59%) | ||
| Tumor Size (Maximum Diameter, mm) | 9.19 ± 1.06 | 8.73 ± 1.12 | 2.162 | 0.033 |
| Resection Method | 4.429 | 0.035 | ||
| Wide Resection | 29 (56.86%) | 44 (75.86%) | ||
| Marginal Resection | 22 (43.14%) | 14 (24.14%) | ||
| Albumin (g/dL) | 4.14 ± 0.45 | 4.33 ± 0.47 | 2.155 | 0.033 |
| Adjuvant Treatment (yes) | 42 (82.35%) | 55 (94.83%) | 4.311 | 0.038 |
| Rehabilitation Training (yes) | 39 (76.47%) | 54 (93.10%) | 5.995 | 0.014 |
| Infection (yes) | 7 (13.73%) | 1 (1.72%) | 4.118 | 0.042 |
BMI: body mass index; MSTS: Musculoskeletal Tumor Society Scoring System.
External validation ROC
The ROC curve analysis yielded an AUC of 0.811, indicating good discrimination between positive and negative cases (Figure 6). At a threshold of 0.663, the test achieved a sensitivity of 0.961 and a specificity of 0.466 (with 1-specificity being 0.534). This highlights the model’s capability to effectively identify true positives while maintaining reasonable specificity, demonstrating strong predictive power beyond random chance.
Figure 6.
External validation ROC curve.
Discussion
Limb-salvage surgery has become the preferred strategy for managing primary bone tumors, offering patients both tumor eradication and preservation of limb function [23]. Despite significant advancements, significant heterogeneity exists in postoperative imaging and functional outcomes [24]. This retrospective study, conducted at a single tertiary institution, sought to identify the clinical and perioperative factors that play pivotal roles in achieving favorable imaging resolution and functional status after limb-salvage surgery. The major findings highlight several complex and interrelated mechanisms that influence the recovery process in this patient population.
One of the primary insights from our analysis is the influence of patient age on postoperative recovery. Younger patients demonstrated superior imaging and functional results, a finding consistent with the greater regenerative capacity of younger tissues and superior physiological reserves that facilitate healing [25]. Younger patients are also more likely to mount robust immune responses, better tolerate intensive multimodal therapies, and adhere more effectively to rehabilitation protocols [26]. Age influences not only musculoskeletal healing and neurovascular adaptation but also the capacity for bone remodeling and integration of prosthetic or graft constructs [27]. These biological advantages likely enable younger patients to recover more effectively from surgical trauma, and achieve optimal restoration of limb integrity and function, whereas advanced age is associated with delayed wound healing, diminished musculoskeletal plasticity, and higher susceptibility to postoperative complications [27].
Surgical strategy emerged as another critical determinant of postoperative outcome, with the resection method being particularly influential. Patients who underwent wide resection, characterized by en bloc removal of the tumor with a clear margin of healthy tissue, had significantly better outcomes than those who received marginal resection. Several mechanisms may explain this. Wide resection maximizes the probability of complete tumor eradication, thereby reducing the risk of residual disease and local recurrence, both of which can compromise imaging results and functional recovery [28]. From an oncological perspective, inadequate removal of tumor-infiltrated tissue can promote micrometastasis and recurrence, negating the long-term benefits of limb preservation [29]. Moreover, wide resection typically results in more predictable reconstructive scenarios, as the excised volume and surrounding anatomy are more conducive to stable prosthetic fixation or biological reconstruction, which supports better limb functionality after surgery [30].
The importance of perioperative adjuvant therapy was reinforced by our findings, as receipt of adjuvant treatment - particularly chemotherapy - was independently associated with improved postoperative outcomes. The mechanistic basis of this benefit is rooted in the cytotoxic effect of chemotherapy on microscopic residual disease, both at the primary tumor site and systemically [31]. This adjuvant effect reduces the risk of locoregional recurrence and metastatic dissemination, thereby alleviating the secondary burden on healing tissues and facilitating improved imaging findings. Furthermore, well-timed chemotherapy can shrink tumor mass preoperatively, facilitating the feasibility of wide resection while sparing crucial neurovascular and musculoskeletal structures, which leads to better functional preservation.
Our study also underscores the significant role of rehabilitation training and comprehensive supportive care, although their independent predictive power diminished upon multivariate adjustment. Rehabilitation plays a crucial role in maximizing musculoskeletal recovery, reducing postoperative stiffness, and retraining muscle groups weakened by surgical exposure or immobilization [32]. The lack of consistent independent significance in multivariable analysis may reflect collinearity with other strong predictors, such as the extent of resection or age. It is also possible that the greatest benefits from rehabilitation accrue in synergy with optimal surgical and oncological management.
Nutritional and immunological status, as indicated by albumin and lymphocyte count, were associated with favorable outcomes in univariate but not multivariate analyses. These markers reflect a patient’s general physiological resilience and capacity for tissue repair. Hypoalbuminemia and lymphopenia can impede collagen synthesis, angiogenesis, and cellular immune responses required for wound healing and defense against infection [33]. However, their effects may be overshadowed by more direct determinants such as surgical margin status and tumor burden in the multivariate context. Given the established relationship between perioperative nutrition and surgical recovery, further research exploring targeted nutritional supplementation as part of the perioperative protocol could be warranted.
Tumor-specific factors, such as size and Enneking stage, demonstrated associations with postoperative outcome in the univariate setting, also lost statistical significance in the multivariate analysis. Tumor size largely dictates the technical feasibility of achieving clear surgical margins and influences the complexity of limb reconstruction [34]. Larger tumors often extend into soft tissue planes and may abut neurovascular structures, compromising function or precluding curative resection if these structures are affected. The Enneking staging system incorporates tumor grade, anatomical compartment, and metastatic status, all of which are critical in evaluating recurrence risk and long-term prognosis. The lack of independent predictive value in this cohort may reflect the overriding influence of surgical and systemic management decisions, which are closely tied to tumor staging and biology.
Moreover, postoperative infection emerged as a significant complication associated with adverse outcomes in univariate analysis but failed to maintain independent significance in multivariate modeling. The detrimental effects of infection are twofold: direct tissue damage impedes wound healing, while the systemic inflammatory response may necessitate reoperation or delay adjuvant therapy, both of which hinder recovery [12,35]. The decreased statistical impact in multivariate analysis suggests that infection may act as a mediator or consequence of other clinical variables, such as advanced age or compromised nutritional status, rather than an independent determinant.
Interestingly, several variables, including gender, BMI, history of metabolic or cardiovascular disease, and reconstruction technique, did not demonstrate significant associations with outcome. These findings suggest that, within the context of surgical and rehabilitation protocols, intrinsic patient and tumor factors, as well as perioperative management strategies, exert a more profound influence on postoperative success than demographic or procedural variables.
Our findings have several clinical implications. First, they reinforce the necessity of individualized surgical planning aimed at achieving clear margins and minimizing tumor burden. Second, they highlight the importance of integrating adjuvant therapies into a multidisciplinary framework to address residual microscopic disease, particularly in high-risk subsets. Third, the results condition clinical equipoise for resource allocation, emphasizing rehabilitation and nutritional support as adjunctive, rather than primary, determinants of success. Finally, awareness of the vulnerable nature of older patients may prompt more aggressive perioperative optimization and follow-up strategies to enhance postoperative outcomes.
In summary, this study elucidates the intricate interplay between patient age, surgical technique, and adjuvant therapy in determining imaging and functional recovery after limb-salvage surgery for bone tumors. The findings underscore the importance of an individualized, multidisciplinary approach that balances oncological safety with optimal support for physiological healing and limb function. Future prospective studies, with detailed analysis of perioperative interventions and long-term quality of life outcomes, will be crucial in refining care protocols and improving the prospects of limb-salvage surgery in this challenging clinical setting.
Acknowledgements
This work was supported by Mandatory Project of Health Commission of Hebei Province (No. 20240140).
Disclosure of conflict of interest
None.
References
- 1.Shehadeh A, Al Qawasmi M, Edilbi A, Sultan I, Ismael T, Yaser S, Al Mousa A. Long-term outcomes of limb-salvage surgery for malignant bone tumors at a single institution in a developing country. Gulf J Oncolog. 2023;1:42–53. [PubMed] [Google Scholar]
- 2.Zhu X, Hu J, Lin J, Song G, Xu H, Lu J, Tang Q, Wang J. 3D-printed modular prostheses for reconstruction of intercalary bone defects after joint-sparing limb salvage surgery for femoral diaphyseal tumours. Bone Jt Open. 2024;5:317–323. doi: 10.1302/2633-1462.54.BJO-2023-0170.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hu X, Chen Y, Cai W, Cheng M, Yan W, Huang W. Computer-aided design and 3D printing of hemipelvic endoprosthesis for personalized limb-salvage reconstruction after periacetabular tumor resection. Bioengineering (Basel) 2022;9:400. doi: 10.3390/bioengineering9080400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ebeid WA, Badr IT, Mesregah MK, Hasan BZ. Outcomes of management of primary benign aggressive or malignant bone tumors around the elbow by limb-salvage surgery. J Exp Orthop. 2023;10:105. doi: 10.1186/s40634-023-00675-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Farooq MZ, Shafiq MB, Ali S, Rafi I. Complications and Outcome of Bone Sarcoma Patients with Limb Salvage using Liquid Nitrogen-treated Bone for Reconstruction. J Cancer Allied Spec. 2024;10:543. doi: 10.37029/jcas.v10i1.543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhu Y, Wu X, Zhang W, Zhang H. Limb-salvage surgery versus extremity amputation for early-stage bone cancer in the extremities: a population-based study. Front Surg. 2023;10:1147372. doi: 10.3389/fsurg.2023.1147372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Xu M, Wang Z, Yu XC, Lin JH, Hu YC. Guideline for Limb-Salvage Treatment of Osteosarcoma. Orthop Surg. 2020;12:1021–1029. doi: 10.1111/os.12702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zan P, Shen J, Liu K, Wang H, Cai Z, Ma X, Sun W. Custom-made semi-joint prosthesis replacement combined ligament advanced reinforcement system (LARS) ligament reconstruction for the limb salvage surgery of malignant tumors in the distal femur in skeletal immature children. Front Pediatr. 2023;11:1168637. doi: 10.3389/fped.2023.1168637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Liu W, Yang Y, Jin T, Sun Y, Li Y, Hao L, Zhang Q, Niu X. What are the results of limb salvage surgery for primary malignant bone tumor in the forearm? Front Oncol. 2022;12:822983. doi: 10.3389/fonc.2022.822983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kumar B, Sharma P, Shantanu K, Kumar S, Agarwal R, Kumar A, Kumar D. Limb salvage strategy amendment for a better future in the era of bone cancer therapy: a cross-sectional study in North India. Cureus. 2023;15:e41768. doi: 10.7759/cureus.41768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hargiss JB, Labott JR, Broida SE, Rose PS, Barlow JD, Houdek MT. Outcome of Scapular Ewing Sarcoma. Anticancer Res. 2022;42:3869–3872. doi: 10.21873/anticanres.15879. [DOI] [PubMed] [Google Scholar]
- 12.Hovav O, Kolonko S, Zahir SF, Velli G, Chouhan P, Wagels M. Limb salvage surgery reconstructive techniques following long-bone lower limb oncological resection: a systematic review and meta-analysis. ANZ J Surg. 2023;93:2609–2620. doi: 10.1111/ans.18335. [DOI] [PubMed] [Google Scholar]
- 13.Houdek MT, Wilke BK, Barlow JD. Management of Scapular Tumors. Orthop Clin North Am. 2023;54:101–108. doi: 10.1016/j.ocl.2022.08.009. [DOI] [PubMed] [Google Scholar]
- 14.Vale SS, Castro R, Andrade A, Faleiro J, Abreu N, Mendes C, Gonçalves JP. Amputation versus limb-salvage surgery as treatments for pediatric bone sarcoma: a comparative study of survival, function, and quality of life. Cureus. 2025;17:e78543. doi: 10.7759/cureus.78543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Liu B, Yang F, Zhang TW, Tan J, Yuan Z. Clinical exploration of the international society of limb salvage classification of endoprosthetic failure using Henderson in the application of 3D-printed pelvic tumor prostheses. Front Oncol. 2023;13:1271077. doi: 10.3389/fonc.2023.1271077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Asano Y, Yamamoto N, Hayashi K, Takeuchi A, Miwa S, Igarashi K, Taniguchi Y, Morinaga S, Tada K, Nojima T, Tsuchiya H. Clinical outcomes of limb-sparing tumor surgery with vascular reconstruction for bone and soft-tissue tumors. Anticancer Res. 2022;42:4619–4626. doi: 10.21873/anticanres.15966. [DOI] [PubMed] [Google Scholar]
- 17.Kinoshita H, Kinoshita S, Hagiwara Y, Kamoda H, Ohtori S, Yonemoto T. Postoperative limb function and QOL in elderly patients with malignant bone tumor/soft tissue sarcoma. Anticancer Res. 2023;43:3273–3279. doi: 10.21873/anticanres.16502. [DOI] [PubMed] [Google Scholar]
- 18.Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard DJ. A system for the functional evaluation of reconstructive procedures after surgical treatment of tumors of the musculoskeletal system. Clin Orthop Relat Res. 1993:241–246. [PubMed] [Google Scholar]
- 19.Anderson WJ, Doyle LA. Updates from the 2020 World Health Organization classification of soft tissue and bone tumours. Histopathology. 2021;78:644–657. doi: 10.1111/his.14265. [DOI] [PubMed] [Google Scholar]
- 20.Gomez-Brouchet A, Mascard E, Siegfried A, de Pinieux G, Gaspar N, Bouvier C, Aubert S, Marec-Bérard P, Piperno-Neumann S, Marie B, Larousserie F, Galant C, Fiorenza F, Anract P, Sales de Gauzy J, Gouin F GROUPOS (GSF-GETO RESOS) Assessment of resection margins in bone sarcoma treated by neoadjuvant chemotherapy: Literature review and guidelines of the bone group (GROUPOS) of the French sarcoma group and bone tumor study group (GSF-GETO/RESOS) Orthop Traumatol Surg Res. 2019;105:773–780. doi: 10.1016/j.otsr.2018.12.015. [DOI] [PubMed] [Google Scholar]
- 21.Redondo A, Bagué S, Bernabeu D, Ortiz-Cruz E, Valverde C, Alvarez R, Martinez-Trufero J, Lopez-Martin JA, Correa R, Cruz J, Lopez-Pousa A, Santos A, García Del Muro X, Martin-Broto J. Malignant bone tumors (other than Ewing’s): clinical practice guidelines for diagnosis, treatment and follow-up by Spanish Group for Research on Sarcomas (GEIS) Cancer Chemother Pharmacol. 2017;80:1113–1131. doi: 10.1007/s00280-017-3436-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zeng G, Zhang C, Song Y, Zhang Z, Xu J, Liu Z, Tang X, Wang X, Chen Y, Zhang Y, Zhu P, Guo X, Jiang L, Wang Z, Liu R, Wang Q, Yao Y, Feng Y, Han Y, Yuan J. The potential impact of inflammation on the lipid paradox in patients with acute myocardial infarction: a multicenter study. BMC Med. 2024;22:599. doi: 10.1186/s12916-024-03823-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Durrani MYK, Ali U, Saeed J, Umer M. Thirty days outcomes of limb salvage surgery in pediatric patients treated at a tertiary care hospital. Cureus. 2024;16:e74672. doi: 10.7759/cureus.74672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bartelstein MK, Boland PJ. Fifty years of bone tumors. J Surg Oncol. 2022;126:906–912. doi: 10.1002/jso.27027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rai V, Munazzam SW, Wazir NU, Javaid I. Revolutionizing bone tumor management: cutting-edge breakthroughs in limb-saving treatments. Eur J Orthop Surg Traumatol. 2024;34:1741–1748. doi: 10.1007/s00590-024-03876-z. [DOI] [PubMed] [Google Scholar]
- 26.Rangarajan GK, Krishnakumar R, Raja A, Singh SS, Manivannan N. Role of bone scan and MRI in designing of customized prosthesis for limb salvage surgery of long bone tumors: 4 years of single institution analysis. Indian J Surg Oncol. 2022;13:364–371. doi: 10.1007/s13193-021-01475-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Harshwal RK, Sharma D, Meena DS, Sehrawat M. Limb Salvage Surgery in an Enormously Large Pelvic Malignant Bone Tumor. J Orthop Case Rep. 2023;13:20–24. doi: 10.13107/jocr.2023.v13.i06.3680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Arguello AM, Houdek MT, Barlow JD. Management of Proximal Humeral Oncologic Lesions. Orthop Clin North Am. 2023;54:89–100. doi: 10.1016/j.ocl.2022.08.008. [DOI] [PubMed] [Google Scholar]
- 29.Niculescu SA, Grecu AF, Gheonea C, Grecu DC. Limb salvage surgery in pediatric patients with osteosarcoma. Curr Health Sci J. 2024;50:360–367. doi: 10.12865/CHSJ.50.03.03. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.van Kouswijk HW, van Keeken HG, Ploegmakers JJW, Seeber GH, van den Akker-Scheek I. Therapeutic validity and effectiveness of exercise interventions after lower limb-salvage surgery for sarcoma: a systematic review. BMC Musculoskelet Disord. 2023;24:216. doi: 10.1186/s12891-023-06315-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Li M, Xiao X, Fan J, Lu Y, Chen G, Huang M, Ji C, Wang Z, Li J. Is the capanna technique a reliable method for revision surgery after failure of previous limb-salvage surgery? Ann Surg Oncol. 2022;29:1122–1129. doi: 10.1245/s10434-021-10506-z. [DOI] [PubMed] [Google Scholar]
- 32.Arif F, Mirza A, Yasmeen S, Rahman MF, Shaikh SA. Vascularized free fibula flap for limb salvage after long bone tumor resection in pediatric patients: a single-center seven-year experience from a developing country. Cureus. 2025;17:e80187. doi: 10.7759/cureus.80187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Carolino DKD, Tud AR. Functional outcomes of limb salvage surgery in patients with giant cell tumor of bone of the lower extremities: a cross-sectional comparative study. Acta Med Philipp. 2024;58:34–40. doi: 10.47895/amp.vi0.7795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Eralp L, Rozburch SR, Civan M. Advancing pediatric bone sarcoma care: navigating complications and innovating solutions in limb salvage and reconstruction-why, when, and how to treat limb length inequalities. Acta Orthop Traumatol Turc. 2024;58:142–148. doi: 10.5152/j.aott.2024.24080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lazarides AL, Burke ZDC, Gundavda MK, Clever DC, Griffin AM, Tsoi K, Ferguson PC, Wunder JS. Mapping the course of recovery following limb-salvage surgery for soft-tissue sarcoma of the extremities. J Bone Joint Surg Am. 2024;106:1797–1808. doi: 10.2106/JBJS.23.01007. [DOI] [PubMed] [Google Scholar]