Bilateral Knee Osteonecrosis in an Adolescent With Leukemia on Prolonged Corticosteroids: Discordant Femoral–Tibial Findings on Bone Scan SPECT–CT

ABSTRACT

Osteonecrosis (avascular necrosis, AVN) is a disabling complication of corticosteroid exposure and is well recognized in adolescents treated for hematologic malignancies, including acute lymphoblastic leukemia (ALL). We report a 17‐year‐old female with ALL and later acute myelogenous leukemia (AML) after hematopoietic stem cell transplantation (HSCT), who received documented post‐HSCT prednisolone therapy and presented with progressive bilateral knee pain. MRI demonstrated multifocal osteonecrosis involving both distal femoral condyles and proximal tibiae, with classic serpiginous infarct morphology and a double‐line sign. A [99mTc]‐MDP three‐phase bone scan with hybrid SPECT/CT demonstrated discordant activity: intense uptake and marked sclerosis in the femoral condyles but minimal uptake in the tibial lesions. This pattern may reflect heterogeneous lesion activity or different reparative stages rather than a validated predictor of collapse. The case highlights the potential value of bone scintigraphy/SPECT/CT as an adjunct to MRI for lesion activity mapping and surveillance planning in selected high‐risk pediatric oncology patients while emphasizing that its prognostic role requires longitudinal validation.

Key Clinical Message

In pediatric leukemia patients receiving prolonged corticosteroid‐containing therapy, knee osteonecrosis may be multifocal. MRI remains the preferred diagnostic modality, while bone scintigraphy/SPECT–CT may provide complementary metabolic‐anatomic information for lesion activity mapping and surveillance planning. Its prognostic role in predicting collapse remains unstandardized.

1. Introduction

Osteonecrosis (avascular necrosis, AVN) arises from compromised bone perfusion leading to osteocyte death and potential subchondral collapse. Corticosteroids are a major nontraumatic cause and a well‐described therapy‐related toxicity in pediatric ALL. Adolescents, particularly patients older than 10 years, and females are at increased risk, and outcomes can include persistent pain and functional limitation [1, 2, 3, 4]. Osteonecrosis in ALL is frequently multifocal and may involve the knee [5]. MRI is preferred for diagnosis, while bone scintigraphy and SPECT–CT can add functional information and anatomical localization that may support lesion activity assessment and follow‐up planning in selected cases [6, 7]. We present an adolescent leukemia patient with multifocal knee osteonecrosis showing discordant metabolic activity between femoral and tibial lesions on SPECT–CT.

2. Case Presentation

2.1. Patient Information

A 17‐year‐old female with a 3‐year history of hematologic malignancy (ALL with later AML) after hematopoietic stem cell transplantation (HSCT) was evaluated for bilateral knee pain. The documented post‐HSCT corticosteroid regimen consisted exclusively of prednisolone 25 mg daily for approximately 1 year, corresponding to an estimated post‐HSCT cumulative prednisolone exposure of 9125 mg. Details of corticosteroid exposure before HSCT were not available; therefore, the total lifetime cumulative corticosteroid dose could not be calculated. There was no history of knee trauma.

2.2. Clinical Findings

She presented with progressively worsening bilateral knee pain for 20 days, associated with difficulty ambulating and reduced weight‐bearing by the time of presentation. The patient denied fever, erythema, or other systemic features suggestive of infection.

2.3. Physical Examination

Examination showed bilateral knee tenderness with mild swelling, most prominent over the medial aspects. There was no increased local temperature or erythema. Range of motion was limited due to pain.

2.4. Timeline

Time point	Clinical event
Before HSCT	Treatment for hematologic malignancy; exact pre‐HSCT corticosteroid exposure was not available from the records reviewed.
After HSCT	Prednisolone 25 mg daily was administered as the documented corticosteroid regimen.
Approximately 1 year after post‐HSCT prednisolone exposure	Documented cumulative post‐HSCT prednisolone exposure was approximately 9125 mg.
20 days before presentation	Progressive bilateral knee pain developed, with increasing difficulty ambulating and reduced weight‐bearing.
Initial diagnostic evaluation	MRI of both knees demonstrated multifocal osteonecrosis involving the bilateral distal femoral condyles and proximal tibiae.
14 days after MRI	[99mTc]‐MDP three‐phase bone scan with hybrid SPECT/CT demonstrated intense uptake in the femoral condyles and minimal uptake in tibial lesions.
After imaging	Orthopedic consultation recommended conservative management with reduced joint loading and avoidance of stair climbing and high‐impact weight‐bearing activities.
5‐month follow‐up	Knee radiography demonstrated sclerosis around the left distal femoral metaphysis. Follow‐up MRI/SPECT–CT was not available; therefore, healing, stability, progression, or collapse risk could not be confirmed.

2.5. Diagnostic Assessment

MRI demonstrated mild joint effusion and multifocal geographic marrow lesions involving the medial and lateral distal femoral condyles bilaterally and the proximal tibial diaphyses. Lesions demonstrated serpiginous low‐signal rims on T1‐weighted images and a double‐line sign on fluid‐sensitive sequences, consistent with osteonecrosis. There were no aggressive imaging features such as cortical destruction, periosteal reaction, diffuse marrow replacement, or soft‐tissue mass. (Figure 1).

FIGURE 1.

MRI of the right knee (A1 coronal T1‐weighted) shows geographic lesions with a hypointense serpiginous rim and relative central fat signal in the femoral condyles and proximal tibial diaphysis. (A2 sagittal STIR) shows edema and the double‐line sign without aggressive features. MRI of the left knee (B1 coronal T1‐weighted, B2 sagittal STIR) shows multiple geographic marrow lesions with a double‐line sign consistent with osteonecrosis.

Although MRI established the diagnosis of osteonecrosis, [99mTc]‐MDP bone scintigraphy with SPECT–CT was obtained to assess whole‐skeleton involvement, evaluate relative metabolic activity of the lesions, and anatomically correlate tracer uptake with CT‐defined sclerosis. Plain radiographic findings were not available in the records reviewed.

A [99mTc]‐MDP three‐phase bone scan performed 14 days after MRI showed pronounced hyperemia on flow and blood‐pool phases in the distal femora bilaterally. Delayed images demonstrated persistent focal increased tracer uptake corresponding to sclerotic changes in the femoral condyles. Tibial lesions identified on MRI were isoactive on scintigraphy. Hybrid SPECT–CT localized intense uptake to the femoral condyles with corresponding marked sclerosis, whereas tibial lesions showed only thin sclerotic rims and no hypermetabolism. These findings demonstrated discordant femoral‐tibial metabolic activity but were not interpreted as a validated predictor of lesion progression or collapse. (Figures 2, 3).

FIGURE 2.

[99mTc]‐MDP three‐phase bone scan. (A) Flow images show hyperperfusion above the knees, more prominent on the right. (B) Blood‐pool images of the knees, anterior (B1) and posterior (B2), show hyperemia of the bilateral femoral condyles. (C) Delayed whole‐body images obtained 2 h after injection of Tc‐99 m methylene diphosphonate, anterior (C1) and posterior (C2), show intense radiotracer uptake in the bilateral femoral condyles, more prominent in the right medial condyle.

FIGURE 3.

Hybrid SPECT/CT. 3D and multiplanar views demonstrate intense uptake in bilateral femoral condyles corresponding to pronounced sclerosis on CT. Tibial lesions show subtle sclerosis with minimal tracer uptake.

Laboratory evaluation did not support infectious, inflammatory, autoimmune, or infiltrative etiologies. Parathyroid hormone and vitamin D levels were within normal limits. Inflammatory and autoimmune markers, including CRP, ESR, RF, ANA, and FANA, were normal or negative. Wright, Coombs Wright, and 2‐mercaptoethanol (2ME) tests were negative. CBC, WBC count, and LDH were within normal limits. Bone mineral density assessment showed values below the expected range for age, with Z‐scores of −2.6 at the lumbar spine, −2.1 at the femoral neck, and −2.3 at the forearm.

2.6. Diagnostic Challenges and Differential Diagnosis

In leukemia patients, knee pain may reflect osteonecrosis, infection, leukemic marrow infiltration, secondary malignant bone lesions, therapy‐related arthralgia, inflammatory arthritis, metabolic bone disease, or trauma/stress injury. In this patient, the diagnosis of osteonecrosis was favored because MRI showed characteristic serpiginous infarct morphology and a double‐line sign, while clinical, laboratory, and imaging findings did not suggest aggressive infection or malignant infiltration.

Differential diagnosis	Findings arguing against the diagnosis in this patient
Osteomyelitis or septic arthritis	No fever, erythema, or local warmth; CRP, ESR, WBC count, and CBC were within normal limits; MRI lacked abscess, destructive cortical change, or aggressive infectious features.
Leukemic marrow infiltration or secondary malignant lesion	CBC, WBC count, and LDH were within normal limits; MRI demonstrated classic serpiginous osteonecrosis morphology without diffuse marrow replacement, cortical destruction, periosteal reaction, or soft‐tissue mass.
Autoimmune or inflammatory arthritis	RF, ANA, FANA, ESR, and CRP were normal or negative.
Brucellosis‐related osteoarticular disease	Wright, Coombs Wright, and 2ME tests were negative.
Metabolic bone disease due to PTH or vitamin D abnormality	PTH and vitamin D were within normal limits.
Trauma or stress injury	There was no history of knee trauma, and imaging showed multifocal bilateral osteonecrosis rather than an isolated traumatic or stress‐related pattern.
Low bone mineral density	Bone mineral density was below the expected range for age and may have contributed to skeletal vulnerability, but it did not explain the characteristic MRI pattern as an alternative diagnosis.

Biopsy was not performed because the imaging findings were characteristic of osteonecrosis and there were no aggressive clinical, laboratory, or imaging features suggesting infection, leukemic infiltration, or secondary malignant bone lesion. The absence of such features made biopsy unnecessary in the clinical context, although laboratory findings alone cannot definitively exclude focal marrow disease.

2.7. Therapeutic Intervention

The patient was referred for orthopedic evaluation. Conservative management was recommended, including reduction of joint loading, avoidance of stair climbing and other high‐impact weight‐bearing activities, activity modification, and supportive analgesia as needed. No surgical intervention was performed at the time of reporting.

2.8. Follow‐Up and Outcomes

At 5‐month follow‐up, knee radiography demonstrated sclerosis around the left distal femoral metaphysis. This finding was interpreted cautiously as chronic reparative/sclerotic change corresponding to the previously identified femoral osteonecrotic lesion (Figure 4). Follow‐up MRI or SPECT/CT was not available; therefore, complete healing, lesion stability, radiologic progression, or future subchondral collapse risk could not be assessed. The patient had been advised by the orthopedic team to reduce joint weight bearing, including avoidance of stair climbing and other high‐impact activities, and remained under clinical surveillance.

FIGURE 4.

Follow‐up plain radiograph of the left knee obtained 5 months after initial imaging. Anteroposterior radiograph demonstrates patchy sclerosis around the left distal femoral metaphysis/condylar region, corresponding to the previously identified osteonecrotic femoral lesion. The finding is compatible with chronic reparative/sclerotic change; however, sclerosis alone does not confirm complete healing or exclude future structural progression.

2.9. Patient Perspective

The patient and family were distressed by pain‐related functional limitation and preferred nonoperative management with close monitoring.

2.10. Informed Consent

Written informed consent for publication of this case and accompanying images was obtained.

3. Discussion

Osteonecrosis is one of the most consequential musculoskeletal toxicities of pediatric ALL therapy and occurs most commonly in adolescents, with female sex and older age being consistent risk factors across cohorts.1–4 In the Children's Cancer Group analysis (CCG‐1882), symptomatic osteonecrosis was notably more frequent in patients aged 10–20 years treated with intensive chemotherapy and multiple corticosteroid courses.1 Prospective data confirm that a meaningful proportion of affected patients experience persistent symptoms years after therapy, underscoring the long‐term morbidity of this complication.2.

Although the femoral head is classically emphasized, osteonecrosis in ALL is frequently multifocal and may involve the knee, including the femoral condyles and tibiae [5, 8]. In MRI‐based series of knee osteonecrosis in children with leukemia/lymphoma, lesions are often bilateral and may involve both femur and tibia, with progression dependent on the extent and location of subchondral involvement [5, 8].

Corticosteroid‐associated osteonecrosis is thought to reflect a convergence of mechanisms including lipid metabolism disturbance with marrow fat hypertrophy, increased intraosseous pressure, endothelial dysfunction, hypercoagulability, and microvascular occlusion, culminating in ischemia and osteocyte death [9, 10]. The subchondral region is particularly vulnerable due to terminal perfusion and high mechanical load; continued loading of necrotic or weakened subchondral bone can lead to microfracture propagation, collapse, and secondary arthrosis [10, 11, 12].

In the present case, the documented corticosteroid exposure consisted of prednisolone 25 mg daily for approximately 1 year after HSCT, corresponding to an estimated post‐HSCT cumulative prednisolone dose of 9125 mg. Although corticosteroid exposure before HSCT could not be retrieved, the prolonged post‐HSCT prednisolone course represents a substantial risk factor for therapy‐associated osteonecrosis. In addition, this patient had low bone mineral density for age, with Z‐scores below −2.0 at the lumbar spine, femoral neck, and forearm. Reduced bone mineral density may have increased skeletal vulnerability in the setting of prolonged corticosteroid‐containing therapy, although it does not by itself establish the mechanism of osteonecrosis.

Although corticosteroid exposure is a major risk factor, osteonecrosis in leukemia patients is multifactorial. Chemotherapeutic agents, disease‐related vascular or marrow changes, metabolic alterations, coagulopathy, immobility, infection, HSCT‐related factors, and supportive medications may contribute. Some antimicrobials or azole antifungals may also alter corticosteroid metabolism and increase systemic exposure. Therefore, osteonecrosis in this case should be interpreted as therapy‐associated and multifactorial rather than solely corticosteroid‐induced.

MRI is the diagnostic reference standard because it sensitively detects early marrow changes and characteristic morphologic features, including a serpiginous rim and double‐line sign [11]. However, MRI primarily depicts morphology and edema and does not directly quantify current osteoblastic activity. Bone scintigraphy offers whole‐skeleton assessment and can help identify metabolically active lesions when symptoms are nonlocalizing. In addition, SPECT/CT can complement MRI by anatomically localizing uptake and correlating metabolic activity with CT‐defined sclerosis or subchondral changes [6, 7].

In this case, femoral condylar lesions demonstrated intense and persistent [99mTc]‐MDP uptake with marked sclerosis, whereas tibial lesions were metabolically inactive or minimally active with only thin sclerotic rims. A cautious interpretation is that the femoral and tibial lesions may have represented heterogeneous stages of multifocal osteonecrosis, differences in lesion burden, biomechanical loading, osteoblastic reparative activity, or technical sensitivity. The tibial isoactivity should not be interpreted definitively as quiescent or burned‐out disease. Similarly, intense uptake in the femoral condyles should not be interpreted as a validated predictor of future collapse in the absence of longitudinal imaging.

The medial femoral condyle is a common load‐bearing site implicated in knee osteonecrosis and subchondral insufficiency‐related collapse [11, 12, 13]. In the osteonecrosis/SIFK literature, subchondral location and mechanical stress are central determinants of progression [11, 12, 13]. While prognostic thresholds for knee SPECT–CT are not standardized, case‐based literature suggests that SPECT–CT may help identify subchondral areas of increased uptake that correlate with clinically significant osteonecrosis on MRI or histology.7 Sakai et al. also reported moderate bone scan sensitivity for femoral condyles but very low sensitivity for metaphyseal/tibial lesions and low likelihood of collapse for noncondylar lesions; therefore, tibial isoactivity in this case may reflect limited sensitivity, smaller lesion burden, different loading conditions, or different lesion stage.

From a management standpoint, early recognition in high‐risk leukemia patients enables timely load modification and orthopedic surveillance. Stage‐ and site‐specific strategies are recommended. Conservative management is commonly used for precollapse stages, while surgical options are considered in cases with collapse or refractory pain [4, 10]. In this patient, SPECT–CT did not replace MRI or establish a different diagnosis, but it provided complementary metabolic‐anatomic information that supported attention to the metabolically active femoral condylar lesions during surveillance planning.

A limitation of this report is the absence of longitudinal follow‐up MRI or SPECT–CT. The 5‐month radiograph showed sclerosis around the left distal femoral metaphysis, which was concordant with the initially increased femoral tracer uptake and may reflect reparative remodeling. However, sclerosis alone should not be equated with complete healing, and the prognostic significance of the initial SPECT/CT uptake pattern remains uncertain without serial MRI/SPECT–CT or longer clinical‐radiographic follow‐up demonstrating absence of progression or collapse.

4. Conclusion

This case report describes bilateral distal femoral condylar and proximal tibial osteonecrosis in an adolescent leukemia patient after HSCT and prolonged post‐HSCT prednisolone therapy. MRI established the diagnosis, while bone scintigraphy and SPECT/CT demonstrated discordant femoral–tibial metabolic activity. Hybrid SPECT/CT may provide complementary metabolic–anatomic information to support lesion activity mapping and surveillance planning in selected high‐risk patients; however, its prognostic role in predicting progression or collapse remains uncertain without longitudinal follow‐up imaging.

Author Contributions

Fatemeh Haghighi: conceptualization, data curation, formal analysis, methodology, project administration, resources, software, validation, writing – original draft. Fatemeh Tahghighi Sharabian: conceptualization, methodology, validation, visualization. Fatemeh Jooshaei: data curation, methodology, software, writing – original draft. Farzaneh Ghaffari Yousefzadeh: data curation, formal analysis, methodology, software, visualization. Marzieh Ebrahimi: software, validation. Zeinab Paymani: conceptualization, methodology, project administration, supervision, writing – review and editing.

Funding

The authors have nothing to report.

Consent

Written informed consent was obtained from the patient and her legal guardian for publication of this report and any accompanying images.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

All data relevant to this report are available upon request from the corresponding author, with patient information de‐identified and in accordance with institutional ethical policies.