Measuring diagnostic accuracy of imaging parameters in pelvic lipomatosis

Eur J Radiol. 2012 Nov;81(11):3107-14. doi: 10.1016/j.ejrad.2012.05.031. Epub 2012 Jul 1.

Abstract

Objectives: To study whether the individual radiological findings can help predict diagnosis of pelvic lipomatosis (PL) or, specifically appreciate its progression.

Methods: Data from 32 clinically proven cases of PL and 25 controls were collected. Two reviewers were recruited for a blinded evaluation, image features were recorded in terms of: (1) bladder shape; (2) bladder-rectosigmoid morphological indexes including ratio of superior-inferior to anterior-posterior length of bladder (SI/AP), angle between anterior and posterior wall (AAP), relative length of posterior urethra (rLPU), angle between bladder and seminal vesicle (ABS) and rectosigmoid morphological index (RMI); (3) secondary complications. Results were evaluated by an unpaired t test and ROC analysis.

Results: The sensitivity and specificity were 40.6% and 100% for pear and banana-shaped bladder, 62.5% and 100% for SI/AP, 40.6% and 100% for AAP, 62.5% and 100% for ABS, 78.1% and 72% for rLPU, 59.4% and 96% for RMI, respectively. These radiological findings partially correlated with the severity of disease weighted by hydronephrosis and treatment grade. Image analysis demonstrated high prevalence of glandular cystitis (100%) and hydronephrosis (73.4%).

Conclusion: We conclude that PL is a progressive disease involving multiple pelvic organs with high prevalence of intractable cystitis and hydronephrosis. The imaging characteristics can help predict diagnosis and, specifically appreciate progression.

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Algorithms*
  • Female
  • Humans
  • Image Enhancement / methods
  • Image Interpretation, Computer-Assisted / methods*
  • Lipomatosis / diagnosis*
  • Magnetic Resonance Imaging / methods*
  • Male
  • Middle Aged
  • Observer Variation
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Single-Blind Method
  • Tomography, X-Ray Computed / methods*
  • Urinary Bladder Diseases / diagnosis*

Supplementary concepts

  • Pelvic lipomatosis