Premium
T 1‐weighted MR image contrast around a cryoablation iceball: A phantom study and initial comparison with in vivo findings
Author(s) -
Overduin Christiaan G.,
Bomers Joyce G. R.,
Jenniskens Sjoerd F. M.,
Hoes Maarten F.,
Haken Bennie,
Lange Frank,
Fütterer Jurgen J.,
Scheenen Tom W. J.
Publication year - 2014
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4896824
Subject(s) - imaging phantom , cryoablation , magnetic resonance imaging , nuclear medicine , materials science , in vivo , contrast (vision) , nuclear magnetic resonance , biomedical engineering , medicine , radiology , physics , optics , microbiology and biotechnology , ablation , biology
Purpose: To correlate T 1‐weighted ( T 1w) magnetic resonance (MR) image contrast around a cryoablation iceball with temperature in a phantom study and to compare this to its in vivo appearance during MR‐guided focal cryoablation of prostate cancer. Methods: A MR‐compatible cryoneedle was inserted into identical gel phantoms ( n = 3) on a 1.5 T MR system. Two fiber‐optic temperature sensors were placed parallel to the needle. A fast 3D T 1w gradient echo (GRE) sequence (TR/TE/FA = 4.81/1.98/6°) was used to monitor iceball progression. Normalized signal intensity (SI) was correlated with temperature. The same T 1w sequence was used during MR‐guided prostate cryoablation in ten consecutive patients at the authors’ institution. In vivo findings were quantitatively compared to the phantom data. Results: In the phantom study, the cryoablation iceball appeared in the T 1w MR images as a sharply delineated signal void. A 2.2 ± 0.2 mm wide hyperintense rim directly surrounded the iceball at cooled but nonfreezing temperatures (<20 °C) in the gel. Normalized SI was maximum at 8.4 ± 2.4 °C, showing a 35.6%–43.0% (mean 40.5%) increase with respect to baseline before cooling. In the clinical procedures, the same image contrast was observed in vivo in all patients. In vivo , width of the hyperintense rim was 1.6 ± 0.6 mm. Normalized SI increases with respect to nontreated prostate ranged 28.4%–55.6% (mean 36.8%). On quantitative analysis, normalized SI changes along a linear region of interest from surrounding tissue onto the iceball center were similar between the patients and phantom setting (root mean square difference 0.06). Conclusions: The hyperintense rim around the iceball in fast T 1w GRE images corresponded to cooled but nonfreezing temperatures (<20 °C) proximal to the frozen zone. The same image contrast was observed both in a phantom study as well as in vivo in the human prostate during cryotherapy. Potentially, monitoring of this rim could be useful in order to maintain a safe margin from at‐risk tissues during MR‐guided prostate cryoablation procedures.