Format

Send to

Choose Destination
Radiology. 2010 Sep;256(3):906-15. doi: 10.1148/radiol.10091461. Epub 2010 Jul 15.

Posttreatment high-grade glioma: usefulness of peak height position with semiquantitative MR perfusion histogram analysis in an entire contrast-enhanced lesion for predicting volume fraction of recurrence.

Author information

1
Department of Diagnostic Radiology, Ajou University School of Medicine, Yeongtong-gu, Suwon, Gyeonggi-do 442-749, Korea. J978005@lycos.co.kr

Abstract

PURPOSE:

To determine whether semiquantitative histogram analysis of the normalized cerebral blood volume (CBV) for an entire contrast material-enhanced lesion could be used to predict the volume fraction of posttreatment high-grade glioma recurrence compared with posttreatment change.

MATERIALS AND METHODS:

The institutional review board approved this retrospective study. Informed consent was obtained. Thirty-nine patients with pathologically proved predominant tumor recurrence (tumor recurrence group, tumor fraction > or =50% [n = 14]), mixed tumor and posttreatment change (mixed group, tumor fraction > or =20% and <50% [n = 10]), and predominant posttreatment change (treatment change group, tumor fraction <20% [n = 15]) were evaluated. Histogram parameters of normalized CBV-histogram width, peak height position (PHP), and maximum value (MV)-were measured in entire contrast-enhanced lesions and used as discriminative indexes. Ordered logistic regression was used to determine independent factors for predicting the diseases of posttreatment contrast-enhanced lesions. Leave-one-out cross-validation was used to determine diagnostic accuracy.

RESULTS:

PHP was an independent predictive factor (P = .003) for differentiating contrast-enhanced lesions in patients with posttreatment gliomas. According to receiver operating characteristic curve analyses, PHP provided sensitivity of 90.2% and specificity of 91.1% for differentiating tumor recurrence from mixed and treatment change groups at an optimum threshold of 1.7 by using leave-one-out cross-validation. MV helped distinguish treatment change group from tumor recurrence and mixed groups at an optimum threshold of 2.6 (sensitivity, 96.5%; specificity, 93.1%).

CONCLUSION:

PHP can be used to predict the volume fraction of posttreatment high-grade glioma recurrence.

PMID:
20634429
DOI:
10.1148/radiol.10091461
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center