Age-related marrow conversion in the proximal metaphysis of the femur: Evaluation with T1-weighted MR imaging

Kyung Hoi Koo, Robert Dussault, Phoebe Kaplan, Rokho Kim, In Oak Ahn, John Christopher, Hae Ryong Song, Gwo Jaw Wang

Research output: Contribution to journalArticlepeer-review

46 Citations (Scopus)


PURPOSE: To determine the normal chronologic changes in fatty marrow conversion of the proximal femoral metaphysics at magnetic resonance (MR) imaging. MATERIALS AND METHODS: The 'marrow conversion index,' (signal intensity of the proximal femoral metaphysis + signal intensity of the greater trochanter) x 100, was measured in 182 hips of 91 patients who had no evidence of osteonecrosis or diseases involving bone marrow, no history of steroid treatment, and no other risk factors for osteonecrosis. The results were evaluated according to patient age and sex. RESULTS: The index showed a progressive linear increase with age regardless of sex. The index was 65.4% ± 5.6 (standard deviation) in patients aged 15-19 years and increased monotonously with age to 87.5% ± 7.0 in patients aged 60-79 years. Simple linear regression of the index according to age group showed that the marrow conversion index is roughly equal to 63 + 43 x Z (p < .001), where Z is 1.5 for patients aged 15-19 years, 2.0 for patients aged 20-29 years, 3.0 for patients aged 30-39 years, 4.0 for patients aged 40-49 years, 5.0 for patients aged 50-59 years, and 6.0 for patients aged 60-79 years. CONCLUSION: There is a linear association of marrow conversion index and age. The estimating equation may provide baseline information for the marrow conversion index.

Original languageEnglish
Pages (from-to)745-748
Number of pages4
Issue number3
Publication statusPublished - 1998 Mar
Externally publishedYes


  • Bone marrow
  • Bone marrow, MR
  • Hip, MR
  • Hip, necrosis

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Age-related marrow conversion in the proximal metaphysis of the femur: Evaluation with T1-weighted MR imaging'. Together they form a unique fingerprint.

Cite this