Arterial Input Functions and Tissue Response Curves in Dynamic Glucose-Enhanced (DGE) Imaging: Comparison Between glucoCEST and Blood Glucose Sampling in Humans

Linda Knutsson, Anina Seidemo, Anna Rydhög Scherman, Karin Markenroth Bloch, Rita R. Kalyani, Mads Andersen, Pia C. Sundgren, Ronnie Wirestam, Gunther Helms, Peter C Van Zijl, Xiang Xu

Research output: Contribution to journalArticle

Abstract

Dynamic glucose-enhanced (DGE) imaging uses chemical exchange saturation transfer magnetic resonance imaging to retrieve information about the microcirculation using infusion of a natural sugar (D-glucose). However, this new approach is not yet well understood with respect to the dynamic tissue response. DGE time curves for arteries, normal brain tissue, and cerebrospinal fluid (CSF) were analyzed in healthy volunteers and compared with the time dependence of sampled venous plasma blood glucose levels. The arterial response curves (arterial input function [AIF]) compared reasonably well in shape with the time curves of the sampled glucose levels but could also differ substantially. The brain tissue response curves showed mainly negative responses with a peak intensity that was of the order of 10 times smaller than the AIF peak and a shape that was susceptible to both noise and partial volume effects with CSF, attributed to the low contrast-to-noise ratio. The CSF response curves showed a rather large and steady increase of the glucose uptake during the scan, due to the rapid uptake of D-glucose in CSF. Importantly, and contrary to gadolinium studies, the curves differed substantially among volunteers, which was interpreted to be caused by variations in insulin response. In conclusion, while AIFs and tissue response curves can be measured in DGE experiments, partial volume effects, low concentration of D-glucose in tissue, and osmolality effects between tissue and blood may prohibit quantification of normal tissue perfusion parameters. However, separation of tumor responses from normal tissue responses would most likely be feasible.

Original languageEnglish (US)
Pages (from-to)164-171
Number of pages8
JournalTomography (Ann Arbor, Mich.)
Volume4
Issue number4
DOIs
StatePublished - Dec 1 2018

Fingerprint

Blood Glucose
Glucose
Cerebrospinal Fluid
Noise
Gadolinium
Brain
Microcirculation
Osmolar Concentration
Volunteers
Healthy Volunteers
Arteries
Perfusion
Magnetic Resonance Imaging
Insulin
Neoplasms

Keywords

  • AIF
  • CEST
  • glucoCEST
  • glucose
  • MRI
  • perfusion
  • sugar

Cite this

Arterial Input Functions and Tissue Response Curves in Dynamic Glucose-Enhanced (DGE) Imaging : Comparison Between glucoCEST and Blood Glucose Sampling in Humans. / Knutsson, Linda; Seidemo, Anina; Rydhög Scherman, Anna; Markenroth Bloch, Karin; Kalyani, Rita R.; Andersen, Mads; Sundgren, Pia C.; Wirestam, Ronnie; Helms, Gunther; Van Zijl, Peter C; Xu, Xiang.

In: Tomography (Ann Arbor, Mich.), Vol. 4, No. 4, 01.12.2018, p. 164-171.

Research output: Contribution to journalArticle

Knutsson, Linda ; Seidemo, Anina ; Rydhög Scherman, Anna ; Markenroth Bloch, Karin ; Kalyani, Rita R. ; Andersen, Mads ; Sundgren, Pia C. ; Wirestam, Ronnie ; Helms, Gunther ; Van Zijl, Peter C ; Xu, Xiang. / Arterial Input Functions and Tissue Response Curves in Dynamic Glucose-Enhanced (DGE) Imaging : Comparison Between glucoCEST and Blood Glucose Sampling in Humans. In: Tomography (Ann Arbor, Mich.). 2018 ; Vol. 4, No. 4. pp. 164-171.
@article{fb31cab9f9f5463794254df82a7f3c98,
title = "Arterial Input Functions and Tissue Response Curves in Dynamic Glucose-Enhanced (DGE) Imaging: Comparison Between glucoCEST and Blood Glucose Sampling in Humans",
abstract = "Dynamic glucose-enhanced (DGE) imaging uses chemical exchange saturation transfer magnetic resonance imaging to retrieve information about the microcirculation using infusion of a natural sugar (D-glucose). However, this new approach is not yet well understood with respect to the dynamic tissue response. DGE time curves for arteries, normal brain tissue, and cerebrospinal fluid (CSF) were analyzed in healthy volunteers and compared with the time dependence of sampled venous plasma blood glucose levels. The arterial response curves (arterial input function [AIF]) compared reasonably well in shape with the time curves of the sampled glucose levels but could also differ substantially. The brain tissue response curves showed mainly negative responses with a peak intensity that was of the order of 10 times smaller than the AIF peak and a shape that was susceptible to both noise and partial volume effects with CSF, attributed to the low contrast-to-noise ratio. The CSF response curves showed a rather large and steady increase of the glucose uptake during the scan, due to the rapid uptake of D-glucose in CSF. Importantly, and contrary to gadolinium studies, the curves differed substantially among volunteers, which was interpreted to be caused by variations in insulin response. In conclusion, while AIFs and tissue response curves can be measured in DGE experiments, partial volume effects, low concentration of D-glucose in tissue, and osmolality effects between tissue and blood may prohibit quantification of normal tissue perfusion parameters. However, separation of tumor responses from normal tissue responses would most likely be feasible.",
keywords = "AIF, CEST, glucoCEST, glucose, MRI, perfusion, sugar",
author = "Linda Knutsson and Anina Seidemo and {Rydh{\"o}g Scherman}, Anna and {Markenroth Bloch}, Karin and Kalyani, {Rita R.} and Mads Andersen and Sundgren, {Pia C.} and Ronnie Wirestam and Gunther Helms and {Van Zijl}, {Peter C} and Xiang Xu",
year = "2018",
month = "12",
day = "1",
doi = "10.18383/j.tom.2018.00025",
language = "English (US)",
volume = "4",
pages = "164--171",
journal = "Tomography (Ann Arbor, Mich.)",
issn = "2379-1381",
publisher = "Grapho Publications LLC",
number = "4",

}

TY - JOUR

T1 - Arterial Input Functions and Tissue Response Curves in Dynamic Glucose-Enhanced (DGE) Imaging

T2 - Comparison Between glucoCEST and Blood Glucose Sampling in Humans

AU - Knutsson, Linda

AU - Seidemo, Anina

AU - Rydhög Scherman, Anna

AU - Markenroth Bloch, Karin

AU - Kalyani, Rita R.

AU - Andersen, Mads

AU - Sundgren, Pia C.

AU - Wirestam, Ronnie

AU - Helms, Gunther

AU - Van Zijl, Peter C

AU - Xu, Xiang

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Dynamic glucose-enhanced (DGE) imaging uses chemical exchange saturation transfer magnetic resonance imaging to retrieve information about the microcirculation using infusion of a natural sugar (D-glucose). However, this new approach is not yet well understood with respect to the dynamic tissue response. DGE time curves for arteries, normal brain tissue, and cerebrospinal fluid (CSF) were analyzed in healthy volunteers and compared with the time dependence of sampled venous plasma blood glucose levels. The arterial response curves (arterial input function [AIF]) compared reasonably well in shape with the time curves of the sampled glucose levels but could also differ substantially. The brain tissue response curves showed mainly negative responses with a peak intensity that was of the order of 10 times smaller than the AIF peak and a shape that was susceptible to both noise and partial volume effects with CSF, attributed to the low contrast-to-noise ratio. The CSF response curves showed a rather large and steady increase of the glucose uptake during the scan, due to the rapid uptake of D-glucose in CSF. Importantly, and contrary to gadolinium studies, the curves differed substantially among volunteers, which was interpreted to be caused by variations in insulin response. In conclusion, while AIFs and tissue response curves can be measured in DGE experiments, partial volume effects, low concentration of D-glucose in tissue, and osmolality effects between tissue and blood may prohibit quantification of normal tissue perfusion parameters. However, separation of tumor responses from normal tissue responses would most likely be feasible.

AB - Dynamic glucose-enhanced (DGE) imaging uses chemical exchange saturation transfer magnetic resonance imaging to retrieve information about the microcirculation using infusion of a natural sugar (D-glucose). However, this new approach is not yet well understood with respect to the dynamic tissue response. DGE time curves for arteries, normal brain tissue, and cerebrospinal fluid (CSF) were analyzed in healthy volunteers and compared with the time dependence of sampled venous plasma blood glucose levels. The arterial response curves (arterial input function [AIF]) compared reasonably well in shape with the time curves of the sampled glucose levels but could also differ substantially. The brain tissue response curves showed mainly negative responses with a peak intensity that was of the order of 10 times smaller than the AIF peak and a shape that was susceptible to both noise and partial volume effects with CSF, attributed to the low contrast-to-noise ratio. The CSF response curves showed a rather large and steady increase of the glucose uptake during the scan, due to the rapid uptake of D-glucose in CSF. Importantly, and contrary to gadolinium studies, the curves differed substantially among volunteers, which was interpreted to be caused by variations in insulin response. In conclusion, while AIFs and tissue response curves can be measured in DGE experiments, partial volume effects, low concentration of D-glucose in tissue, and osmolality effects between tissue and blood may prohibit quantification of normal tissue perfusion parameters. However, separation of tumor responses from normal tissue responses would most likely be feasible.

KW - AIF

KW - CEST

KW - glucoCEST

KW - glucose

KW - MRI

KW - perfusion

KW - sugar

UR - http://www.scopus.com/inward/record.url?scp=85061961835&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061961835&partnerID=8YFLogxK

U2 - 10.18383/j.tom.2018.00025

DO - 10.18383/j.tom.2018.00025

M3 - Article

C2 - 30588502

AN - SCOPUS:85061961835

VL - 4

SP - 164

EP - 171

JO - Tomography (Ann Arbor, Mich.)

JF - Tomography (Ann Arbor, Mich.)

SN - 2379-1381

IS - 4

ER -