Imaging cellularity in benign and malignant peripheral nerve sheath tumors: Utility of the “target sign” by diffusion weighted imaging

Research output: Contribution to journalArticle

Abstract

Objective: To determine the utility of “target sign” on diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) mapping for peripheral nerve sheath tumor (PNST) characterization. Materials and methods: This IRB–approved, HIPAA–compliant study retrospectively reviewed the MR imaging (comprised of T2- FS, DWI (b-values 50, 400, 800 s/mm2and ADC mapping), and static contrast-enhanced (CE) T1-W imaging) of 42 patients (mean age: 40 years (range 8–68 years), 48% (20/42) females) with 15 malignant PNSTs (MPNSTs) and 33 benign PNSTs (BPNSTs). MPNSTs were histologically confirmed while BPNSTs were histologically-proven or with stable clinical and imaging appearance for at least 12 months. Two radiologists assessed imaging characteristics (size, signal intensity, heterogeneity, perilesional edema or enhancement) and the presence or absence of “target sign,” on each sequence. A “target sign” was defined as a biphasic pattern of peripheral hyperintensity and homogeneous central hypointensity. Descriptive statistics are reported. Cohen's κ statistic or interclass correlation coefficient (ICC) were used to evaluate interobserver agreement between two observers. Univariate and multiple logistic regression analysis were performed to identify MRI features with predictive values. Results: MPNSTs were larger than BPNSTs (6.3 ± 2.5 cm vs 3.5 ± −2.1 cm, p = 0.0002), had perilesional edema (87%(13/15) vs 18%(6/33), p < 0.0001), heterogeneous enhancement (71%(10/14) vs 13%(4/31), p = 0.0001) and perilesional enhancement (79%(11/14) vs 18%(6/31), p = 0.0001), respectively. The “target sign” was present in: 24%(8/33) BPNSTs vs 0/15 MPNST on T2-FS (p = 0.26); 39%(13/33) BPNSTs vs 20%(3/15) MPNST on DWI using b-value = 50 s/mm2 (p = 0.5); 55%(18/33) BPNSTs vs 6%(1/15) MPNST on DWI using b-value = 400 s/mm2 (p = 0.002); 48%(16/33) BPNSTs vs 6%(1/15) MPNST on DWI using b-value = 800 s/mm2 (p = 0.005) and 64%(21/33) benign vs 0/15 MPNST on ADC mapping(p < 0.0001). By CE-T1 imaging, 32%(10/31) BPNSTs and 7%(1/14) MPNST had a target sign(p = 0.07). The odds of an MPNST in cases with minimum ADC ≤ 1.0 × 10(−3) mm(2)/s are 150 times higher than in cases with ADC > 1.0 × 10(−3). Conclusion: In this explorative study, a “target sign” suggests a benign PNST and is more often visible on DWI using high b-values and ADC maps compared with anatomic sequences.

Original languageEnglish (US)
Pages (from-to)195-201
Number of pages7
JournalEuropean Journal of Radiology
Volume102
DOIs
StatePublished - May 1 2018

Fingerprint

Neurilemmoma
Nerve Sheath Neoplasms
Edema
Logistic Models
Regression Analysis

Keywords

  • ADC mapping
  • Diffusion weighted MRI
  • Neurofibromatosis
  • Peripheral nerve sheath tumor
  • Soft tissue mass
  • Target sign

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

@article{0304597bad6f46afa7a1255136a134fa,
title = "Imaging cellularity in benign and malignant peripheral nerve sheath tumors: Utility of the “target sign” by diffusion weighted imaging",
abstract = "Objective: To determine the utility of “target sign” on diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) mapping for peripheral nerve sheath tumor (PNST) characterization. Materials and methods: This IRB–approved, HIPAA–compliant study retrospectively reviewed the MR imaging (comprised of T2- FS, DWI (b-values 50, 400, 800 s/mm2and ADC mapping), and static contrast-enhanced (CE) T1-W imaging) of 42 patients (mean age: 40 years (range 8–68 years), 48{\%} (20/42) females) with 15 malignant PNSTs (MPNSTs) and 33 benign PNSTs (BPNSTs). MPNSTs were histologically confirmed while BPNSTs were histologically-proven or with stable clinical and imaging appearance for at least 12 months. Two radiologists assessed imaging characteristics (size, signal intensity, heterogeneity, perilesional edema or enhancement) and the presence or absence of “target sign,” on each sequence. A “target sign” was defined as a biphasic pattern of peripheral hyperintensity and homogeneous central hypointensity. Descriptive statistics are reported. Cohen's κ statistic or interclass correlation coefficient (ICC) were used to evaluate interobserver agreement between two observers. Univariate and multiple logistic regression analysis were performed to identify MRI features with predictive values. Results: MPNSTs were larger than BPNSTs (6.3 ± 2.5 cm vs 3.5 ± −2.1 cm, p = 0.0002), had perilesional edema (87{\%}(13/15) vs 18{\%}(6/33), p < 0.0001), heterogeneous enhancement (71{\%}(10/14) vs 13{\%}(4/31), p = 0.0001) and perilesional enhancement (79{\%}(11/14) vs 18{\%}(6/31), p = 0.0001), respectively. The “target sign” was present in: 24{\%}(8/33) BPNSTs vs 0/15 MPNST on T2-FS (p = 0.26); 39{\%}(13/33) BPNSTs vs 20{\%}(3/15) MPNST on DWI using b-value = 50 s/mm2 (p = 0.5); 55{\%}(18/33) BPNSTs vs 6{\%}(1/15) MPNST on DWI using b-value = 400 s/mm2 (p = 0.002); 48{\%}(16/33) BPNSTs vs 6{\%}(1/15) MPNST on DWI using b-value = 800 s/mm2 (p = 0.005) and 64{\%}(21/33) benign vs 0/15 MPNST on ADC mapping(p < 0.0001). By CE-T1 imaging, 32{\%}(10/31) BPNSTs and 7{\%}(1/14) MPNST had a target sign(p = 0.07). The odds of an MPNST in cases with minimum ADC ≤ 1.0 × 10(−3) mm(2)/s are 150 times higher than in cases with ADC > 1.0 × 10(−3). Conclusion: In this explorative study, a “target sign” suggests a benign PNST and is more often visible on DWI using high b-values and ADC maps compared with anatomic sequences.",
keywords = "ADC mapping, Diffusion weighted MRI, Neurofibromatosis, Peripheral nerve sheath tumor, Soft tissue mass, Target sign",
author = "Shivani Ahlawat and Fayad, {Laura M}",
year = "2018",
month = "5",
day = "1",
doi = "10.1016/j.ejrad.2018.03.018",
language = "English (US)",
volume = "102",
pages = "195--201",
journal = "European Journal of Radiology",
issn = "0720-048X",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - Imaging cellularity in benign and malignant peripheral nerve sheath tumors

T2 - Utility of the “target sign” by diffusion weighted imaging

AU - Ahlawat, Shivani

AU - Fayad, Laura M

PY - 2018/5/1

Y1 - 2018/5/1

N2 - Objective: To determine the utility of “target sign” on diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) mapping for peripheral nerve sheath tumor (PNST) characterization. Materials and methods: This IRB–approved, HIPAA–compliant study retrospectively reviewed the MR imaging (comprised of T2- FS, DWI (b-values 50, 400, 800 s/mm2and ADC mapping), and static contrast-enhanced (CE) T1-W imaging) of 42 patients (mean age: 40 years (range 8–68 years), 48% (20/42) females) with 15 malignant PNSTs (MPNSTs) and 33 benign PNSTs (BPNSTs). MPNSTs were histologically confirmed while BPNSTs were histologically-proven or with stable clinical and imaging appearance for at least 12 months. Two radiologists assessed imaging characteristics (size, signal intensity, heterogeneity, perilesional edema or enhancement) and the presence or absence of “target sign,” on each sequence. A “target sign” was defined as a biphasic pattern of peripheral hyperintensity and homogeneous central hypointensity. Descriptive statistics are reported. Cohen's κ statistic or interclass correlation coefficient (ICC) were used to evaluate interobserver agreement between two observers. Univariate and multiple logistic regression analysis were performed to identify MRI features with predictive values. Results: MPNSTs were larger than BPNSTs (6.3 ± 2.5 cm vs 3.5 ± −2.1 cm, p = 0.0002), had perilesional edema (87%(13/15) vs 18%(6/33), p < 0.0001), heterogeneous enhancement (71%(10/14) vs 13%(4/31), p = 0.0001) and perilesional enhancement (79%(11/14) vs 18%(6/31), p = 0.0001), respectively. The “target sign” was present in: 24%(8/33) BPNSTs vs 0/15 MPNST on T2-FS (p = 0.26); 39%(13/33) BPNSTs vs 20%(3/15) MPNST on DWI using b-value = 50 s/mm2 (p = 0.5); 55%(18/33) BPNSTs vs 6%(1/15) MPNST on DWI using b-value = 400 s/mm2 (p = 0.002); 48%(16/33) BPNSTs vs 6%(1/15) MPNST on DWI using b-value = 800 s/mm2 (p = 0.005) and 64%(21/33) benign vs 0/15 MPNST on ADC mapping(p < 0.0001). By CE-T1 imaging, 32%(10/31) BPNSTs and 7%(1/14) MPNST had a target sign(p = 0.07). The odds of an MPNST in cases with minimum ADC ≤ 1.0 × 10(−3) mm(2)/s are 150 times higher than in cases with ADC > 1.0 × 10(−3). Conclusion: In this explorative study, a “target sign” suggests a benign PNST and is more often visible on DWI using high b-values and ADC maps compared with anatomic sequences.

AB - Objective: To determine the utility of “target sign” on diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) mapping for peripheral nerve sheath tumor (PNST) characterization. Materials and methods: This IRB–approved, HIPAA–compliant study retrospectively reviewed the MR imaging (comprised of T2- FS, DWI (b-values 50, 400, 800 s/mm2and ADC mapping), and static contrast-enhanced (CE) T1-W imaging) of 42 patients (mean age: 40 years (range 8–68 years), 48% (20/42) females) with 15 malignant PNSTs (MPNSTs) and 33 benign PNSTs (BPNSTs). MPNSTs were histologically confirmed while BPNSTs were histologically-proven or with stable clinical and imaging appearance for at least 12 months. Two radiologists assessed imaging characteristics (size, signal intensity, heterogeneity, perilesional edema or enhancement) and the presence or absence of “target sign,” on each sequence. A “target sign” was defined as a biphasic pattern of peripheral hyperintensity and homogeneous central hypointensity. Descriptive statistics are reported. Cohen's κ statistic or interclass correlation coefficient (ICC) were used to evaluate interobserver agreement between two observers. Univariate and multiple logistic regression analysis were performed to identify MRI features with predictive values. Results: MPNSTs were larger than BPNSTs (6.3 ± 2.5 cm vs 3.5 ± −2.1 cm, p = 0.0002), had perilesional edema (87%(13/15) vs 18%(6/33), p < 0.0001), heterogeneous enhancement (71%(10/14) vs 13%(4/31), p = 0.0001) and perilesional enhancement (79%(11/14) vs 18%(6/31), p = 0.0001), respectively. The “target sign” was present in: 24%(8/33) BPNSTs vs 0/15 MPNST on T2-FS (p = 0.26); 39%(13/33) BPNSTs vs 20%(3/15) MPNST on DWI using b-value = 50 s/mm2 (p = 0.5); 55%(18/33) BPNSTs vs 6%(1/15) MPNST on DWI using b-value = 400 s/mm2 (p = 0.002); 48%(16/33) BPNSTs vs 6%(1/15) MPNST on DWI using b-value = 800 s/mm2 (p = 0.005) and 64%(21/33) benign vs 0/15 MPNST on ADC mapping(p < 0.0001). By CE-T1 imaging, 32%(10/31) BPNSTs and 7%(1/14) MPNST had a target sign(p = 0.07). The odds of an MPNST in cases with minimum ADC ≤ 1.0 × 10(−3) mm(2)/s are 150 times higher than in cases with ADC > 1.0 × 10(−3). Conclusion: In this explorative study, a “target sign” suggests a benign PNST and is more often visible on DWI using high b-values and ADC maps compared with anatomic sequences.

KW - ADC mapping

KW - Diffusion weighted MRI

KW - Neurofibromatosis

KW - Peripheral nerve sheath tumor

KW - Soft tissue mass

KW - Target sign

UR - http://www.scopus.com/inward/record.url?scp=85044105442&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044105442&partnerID=8YFLogxK

U2 - 10.1016/j.ejrad.2018.03.018

DO - 10.1016/j.ejrad.2018.03.018

M3 - Article

C2 - 29685535

AN - SCOPUS:85044105442

VL - 102

SP - 195

EP - 201

JO - European Journal of Radiology

JF - European Journal of Radiology

SN - 0720-048X

ER -