Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI

Mohammad Haris, Rakesh Kumar Gupta, Anup Singh, Nuzhat Husain, Mazhar Husain, Chandra Mohan Pandey, Chhitij Srivastava, Sanjay Behari, Ram Kishore Singh Rathore

Research output: Contribution to journalArticle

55 Citations (Scopus)

Abstract

Introduction: It is not always possible to differentiate infective from neoplastic brain lesions with conventional MR imaging. In this study, we assessed the utility of various perfusion indices in the differentiation of infective from neoplastic brain lesions. Methods: A total of 103 patients with infective brain lesions (group I, n=26) and neoplastic brain lesions (high-grade glioma, HGG, group II, n=52; low-grade glioma, LGG, group III, n=25) underwent dynamic contrast-enhanced MR imaging. The perfusion indices, including relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), transfer coefficient (ktrans) and leakage (ve), were calculated and their degree of correlation with immunohistologically obtained microvessel density (MVD) and vascular endothelial growth factor (VEGF) determined. The rCBV was corrected for the leakage effect. Discriminant analysis for rCBV, rCBF, ktrans and ve was performed to predict the group membership of each case and post hoc analysis was performed to look for group differences. Results: The rCBV, rCBF, ktrans, ve, MVD and VEGF were significantly different (P<0.001) between the three groups. Discriminant analysis showed that rCBV predicted 73.1% of the infective lesions, 84.6% of the HGG and 72.0% of the LGG. The rCBF classified 86.5% of the HGG, 80.0% of the LGG and 65.4% of the infective lesions. The ktrans discriminated 98.1% of the HGG, 76.0% of the LGG and 88.5% of the infective lesions correctly. The ve classified 98.1% of the HGG, 76.0% of the LGG and 84.6% the infective lesions. The rCBV was correlated significantly with MVD and VEGF, while the correlation between ktrans and MVD was not significant. Conclusion: Physiological perfusion indices such as k trans and ve appear to be useful in differentiating infective from neoplastic brain lesions. Adding these indices to the current imaging protocol is likely to improve tissue characterization of these focal brain mass lesions.

Original languageEnglish
Pages (from-to)531-540
Number of pages10
JournalNeuroradiology
Volume50
Issue number6
DOIs
Publication statusPublished - Jun 2008
Externally publishedYes

Fingerprint

Cerebrovascular Circulation
Microvessels
Brain
Vascular Endothelial Growth Factor A
Perfusion
Discriminant Analysis
Glioma
Cerebral Blood Volume

Keywords

  • Blood-brain permeability
  • Brain abscess
  • Brain tuberculoma
  • Brain tumor
  • DCE-MRI

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Clinical Neurology
  • Radiological and Ultrasound Technology

Cite this

Haris, M., Gupta, R. K., Singh, A., Husain, N., Husain, M., Pandey, C. M., ... Rathore, R. K. S. (2008). Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI. Neuroradiology, 50(6), 531-540. https://doi.org/10.1007/s00234-008-0378-6

Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI. / Haris, Mohammad; Gupta, Rakesh Kumar; Singh, Anup; Husain, Nuzhat; Husain, Mazhar; Pandey, Chandra Mohan; Srivastava, Chhitij; Behari, Sanjay; Rathore, Ram Kishore Singh.

In: Neuroradiology, Vol. 50, No. 6, 06.2008, p. 531-540.

Research output: Contribution to journalArticle

Haris, M, Gupta, RK, Singh, A, Husain, N, Husain, M, Pandey, CM, Srivastava, C, Behari, S & Rathore, RKS 2008, 'Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI', Neuroradiology, vol. 50, no. 6, pp. 531-540. https://doi.org/10.1007/s00234-008-0378-6
Haris, Mohammad ; Gupta, Rakesh Kumar ; Singh, Anup ; Husain, Nuzhat ; Husain, Mazhar ; Pandey, Chandra Mohan ; Srivastava, Chhitij ; Behari, Sanjay ; Rathore, Ram Kishore Singh. / Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI. In: Neuroradiology. 2008 ; Vol. 50, No. 6. pp. 531-540.
@article{1d932d1605c94da7b375eb5f47f8d11f,
title = "Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI",
abstract = "Introduction: It is not always possible to differentiate infective from neoplastic brain lesions with conventional MR imaging. In this study, we assessed the utility of various perfusion indices in the differentiation of infective from neoplastic brain lesions. Methods: A total of 103 patients with infective brain lesions (group I, n=26) and neoplastic brain lesions (high-grade glioma, HGG, group II, n=52; low-grade glioma, LGG, group III, n=25) underwent dynamic contrast-enhanced MR imaging. The perfusion indices, including relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), transfer coefficient (ktrans) and leakage (ve), were calculated and their degree of correlation with immunohistologically obtained microvessel density (MVD) and vascular endothelial growth factor (VEGF) determined. The rCBV was corrected for the leakage effect. Discriminant analysis for rCBV, rCBF, ktrans and ve was performed to predict the group membership of each case and post hoc analysis was performed to look for group differences. Results: The rCBV, rCBF, ktrans, ve, MVD and VEGF were significantly different (P<0.001) between the three groups. Discriminant analysis showed that rCBV predicted 73.1{\%} of the infective lesions, 84.6{\%} of the HGG and 72.0{\%} of the LGG. The rCBF classified 86.5{\%} of the HGG, 80.0{\%} of the LGG and 65.4{\%} of the infective lesions. The ktrans discriminated 98.1{\%} of the HGG, 76.0{\%} of the LGG and 88.5{\%} of the infective lesions correctly. The ve classified 98.1{\%} of the HGG, 76.0{\%} of the LGG and 84.6{\%} the infective lesions. The rCBV was correlated significantly with MVD and VEGF, while the correlation between ktrans and MVD was not significant. Conclusion: Physiological perfusion indices such as k trans and ve appear to be useful in differentiating infective from neoplastic brain lesions. Adding these indices to the current imaging protocol is likely to improve tissue characterization of these focal brain mass lesions.",
keywords = "Blood-brain permeability, Brain abscess, Brain tuberculoma, Brain tumor, DCE-MRI",
author = "Mohammad Haris and Gupta, {Rakesh Kumar} and Anup Singh and Nuzhat Husain and Mazhar Husain and Pandey, {Chandra Mohan} and Chhitij Srivastava and Sanjay Behari and Rathore, {Ram Kishore Singh}",
year = "2008",
month = "6",
doi = "10.1007/s00234-008-0378-6",
language = "English",
volume = "50",
pages = "531--540",
journal = "Neuroradiology",
issn = "0028-3940",
publisher = "Springer Verlag",
number = "6",

}

TY - JOUR

T1 - Differentiation of infective from neoplastic brain lesions by dynamic contrast-enhanced MRI

AU - Haris, Mohammad

AU - Gupta, Rakesh Kumar

AU - Singh, Anup

AU - Husain, Nuzhat

AU - Husain, Mazhar

AU - Pandey, Chandra Mohan

AU - Srivastava, Chhitij

AU - Behari, Sanjay

AU - Rathore, Ram Kishore Singh

PY - 2008/6

Y1 - 2008/6

N2 - Introduction: It is not always possible to differentiate infective from neoplastic brain lesions with conventional MR imaging. In this study, we assessed the utility of various perfusion indices in the differentiation of infective from neoplastic brain lesions. Methods: A total of 103 patients with infective brain lesions (group I, n=26) and neoplastic brain lesions (high-grade glioma, HGG, group II, n=52; low-grade glioma, LGG, group III, n=25) underwent dynamic contrast-enhanced MR imaging. The perfusion indices, including relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), transfer coefficient (ktrans) and leakage (ve), were calculated and their degree of correlation with immunohistologically obtained microvessel density (MVD) and vascular endothelial growth factor (VEGF) determined. The rCBV was corrected for the leakage effect. Discriminant analysis for rCBV, rCBF, ktrans and ve was performed to predict the group membership of each case and post hoc analysis was performed to look for group differences. Results: The rCBV, rCBF, ktrans, ve, MVD and VEGF were significantly different (P<0.001) between the three groups. Discriminant analysis showed that rCBV predicted 73.1% of the infective lesions, 84.6% of the HGG and 72.0% of the LGG. The rCBF classified 86.5% of the HGG, 80.0% of the LGG and 65.4% of the infective lesions. The ktrans discriminated 98.1% of the HGG, 76.0% of the LGG and 88.5% of the infective lesions correctly. The ve classified 98.1% of the HGG, 76.0% of the LGG and 84.6% the infective lesions. The rCBV was correlated significantly with MVD and VEGF, while the correlation between ktrans and MVD was not significant. Conclusion: Physiological perfusion indices such as k trans and ve appear to be useful in differentiating infective from neoplastic brain lesions. Adding these indices to the current imaging protocol is likely to improve tissue characterization of these focal brain mass lesions.

AB - Introduction: It is not always possible to differentiate infective from neoplastic brain lesions with conventional MR imaging. In this study, we assessed the utility of various perfusion indices in the differentiation of infective from neoplastic brain lesions. Methods: A total of 103 patients with infective brain lesions (group I, n=26) and neoplastic brain lesions (high-grade glioma, HGG, group II, n=52; low-grade glioma, LGG, group III, n=25) underwent dynamic contrast-enhanced MR imaging. The perfusion indices, including relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), transfer coefficient (ktrans) and leakage (ve), were calculated and their degree of correlation with immunohistologically obtained microvessel density (MVD) and vascular endothelial growth factor (VEGF) determined. The rCBV was corrected for the leakage effect. Discriminant analysis for rCBV, rCBF, ktrans and ve was performed to predict the group membership of each case and post hoc analysis was performed to look for group differences. Results: The rCBV, rCBF, ktrans, ve, MVD and VEGF were significantly different (P<0.001) between the three groups. Discriminant analysis showed that rCBV predicted 73.1% of the infective lesions, 84.6% of the HGG and 72.0% of the LGG. The rCBF classified 86.5% of the HGG, 80.0% of the LGG and 65.4% of the infective lesions. The ktrans discriminated 98.1% of the HGG, 76.0% of the LGG and 88.5% of the infective lesions correctly. The ve classified 98.1% of the HGG, 76.0% of the LGG and 84.6% the infective lesions. The rCBV was correlated significantly with MVD and VEGF, while the correlation between ktrans and MVD was not significant. Conclusion: Physiological perfusion indices such as k trans and ve appear to be useful in differentiating infective from neoplastic brain lesions. Adding these indices to the current imaging protocol is likely to improve tissue characterization of these focal brain mass lesions.

KW - Blood-brain permeability

KW - Brain abscess

KW - Brain tuberculoma

KW - Brain tumor

KW - DCE-MRI

UR - http://www.scopus.com/inward/record.url?scp=44449131471&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=44449131471&partnerID=8YFLogxK

U2 - 10.1007/s00234-008-0378-6

DO - 10.1007/s00234-008-0378-6

M3 - Article

VL - 50

SP - 531

EP - 540

JO - Neuroradiology

JF - Neuroradiology

SN - 0028-3940

IS - 6

ER -