Evaluation of giant cell tumors by diffusion weighted imaging–fractional ADC analysis

Research output: Contribution to journalArticlepeer-review

9 Scopus citations


Background: A single ADC value is used in clinical practice on multi b-value acquisitions. Low b-value acquisitions are affected by intravoxel incoherent motion, which is dependent on perfusion. Giant cell tumors (GCTs) are known to exhibit early arterial enhancement and low ADC values. Mean, minimum and fractional ADC characteristics of osseous and tenosynovial GCTs are systematically evaluated. Methods: Tenosynovial and osseous GCTs were included. Each lesion was evaluated on conventional MRI and DWI by two musculoskeletal radiologists. ADC was measured by placing an ROI on the most confluent enhancing portion of the lesion. Fractional and best fit ADC calculations were performed using MATLAB software. Results: No statistically significant difference was found between tenosynovial and osseous lesions’ ADC values. Mean ADC for all lesions was 1.0 × 10 −3  mm 2 /s (SD = 0.2 × 10 −3  mm 2 /s) and minimum ADC was 0.5 × 10 −3  mm 2 /s (SD = 0.3 × 10 −3  mm 2 /s). Average mean ADC value obtained from B50–B400 slope was 1.1 × 10 −3  mm 2 /s (SD = 0.2 × 10 −3  mm 2 /s), and the average mean ADC value obtained from B400–B800 slope was 0.8 × 10 −3  mm 2 /s (SD = 0.1 × 10 −3  mm 2 /s) [p-value <0.01]. Conclusion: Tenosynovial and osseous GCTs demonstrate similar and low ADC values, which become even lower when using high b-value pairs. Our study also supports the theory of intravoxel incoherent motion that becomes apparent at low b values as related to giant cell tumors, which are known to be hyperperfused.

Original languageEnglish (US)
JournalSkeletal Radiology
StatePublished - Jan 1 2019


  • DWI
  • Giant cell tumor
  • MRI
  • Osseous
  • Tenosynovial

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Evaluation of giant cell tumors by diffusion weighted imaging–fractional ADC analysis'. Together they form a unique fingerprint.

Cite this