The value of lumbar spine MRI in the assessment of the pars interarticularis

Clin Radiol. 1997 Sep;52(9):666-71. doi: 10.1016/s0009-9260(97)80029-3.


The aim of this study is to assess the ability of conventional lumbar spine magnetic resonance imaging (MRI) to demonstrate the pars interarticularis. The right and left L4 and L5 pars in 50 patients undergoing lumbar spine MRI for low back pain, with or without radiculopathy, were assessed and classified into four types. Based on their appearance, 59 (29.5%) were classified as normal (Type 1), 59 (29.5%) as hypointense (Type 2), 53 (26.5%) could not be demonstrated (Type 3) and 29 (14.5%) were thought to have defects (Type 4). The gold standard was set by the consensus opinion of two consultant musculoskeletal radiologists on the available plain radiographs of the lumbar spine. MRI was found to have a sensitivity of 57% to 86%, specificity of 81% to 82%, positive predictive value of 14% to 18% and negative predictive value of 97% to 99% for the diagnosis of a pars defect.

Publication types

  • Clinical Trial

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Female
  • Humans
  • Low Back Pain / etiology*
  • Lumbar Vertebrae / pathology*
  • Magnetic Resonance Imaging*
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Sensitivity and Specificity
  • Spondylolysis / complications
  • Spondylolysis / diagnosis*