Diagnostic and Interventional Imaging (2013) 94, 1165—1168

TECHNICAL NOTE / Genito-urinary imaging

Urinary stones: The contribution of dualenergy CT and material decomposition

C. Sanavia, C. Werquina, A. Fekira,b, C. Pinsona,H. Bugelb, J.-N. Dachera,c,∗

a Radiodiagnostics and Medical Imaging, Rouen University Hospital, Charles-Nicolle Hospital,1, rue de Germont, 76031 Rouen cedex, Franceb Urology, Rouen University Hospital, Charles-Nicolle Hospital, 1, rue de Germont, 76031Rouen cedex, Francec INSERM U1096, University of Rouen, 22, boulevard Gambetta, 76183 Rouen cedex, France

KEYWORDSUrinary lithiasis;Computedtomography;Dual energy

Renal lithiasis is a common pathology whose incidence is gradually increasing. Here, weexamine the contribution of dual energy CT by studying three cases.

Contribution of conventional CT

In the course of our urological imaging work, we are often asked to carry out assessmentsto map renal lithiasis.

In this pathology, computed tomography (CT) of the abdomen and pelvis without a con-trast agent is indicated. This technique provides information on the number, size, location,and density of the calculi, which is essential information in terms of management of theillness [1].

Knowing the biochemical composition of the calculi plays an important role in choosingtreatment: for example, a uric acid stone may benefit from specific hygiene and dietarytreatment [2].

This information, identified by measuring the spontaneous density of the urinary cal-culi, is too often imprecise. There are 16 minerals that may be part of the biochemicalcomposition of urinary stones [3]. On conventional CT, composition is estimated by mea-suring mean density in Hounsfield units (HU): on a spiral CT scan at 120 kV, it is knownthat the mean density of uric acid is around 500 ± 100 HU, while it is between 579 and

753 HUfor struvite, 483 and 939 HUfor cystine, 1183 and 1651 HU for calcium phosphate,and 1407 and 1883 HU for calcium oxalate [4,5]. This means that this kind of assessment isnot enough because calculi with a density of between 483 and 1407 HU could be composedof a number of different minerals.

∗ Corresponding author. Radiodiagnostics and Medical Imaging, Rouen University Hospital, Charles-Nicolle Hospital, 1, rue de Germont,76031 Rouen cedex, France.

E-mail address: jean-nicolas.dacher@univ-rouen.fr (J.-N. Dacher).

2211-5684/$ — see front matter © 2013 Éditions françaises de radiologie. Published by Elsevier Masson SAS. All rights reserved.http://dx.doi.org/10.1016/j.diii.2013.08.003

1

Ct

T

TdFctdfe

HoaZ

ttutlt

ica

Fon

cf

D

TcsvaibtatTcwbcatgve

Several applications are available in the dual energy post-treatment software.

166

We wished to illustrate the contribution that dual energyT can make in the biochemical study of urinary stones, ashis technique seems to us to be promising.

echnical description

he patients all underwent dual energy single-source 64-etector row CT (General Electric HealthCare 750 HD, Buc,rance) of the abdomen and pelvis without the use of aontrast agent. The following parameters were used: sec-ions of 0.625/0.625 mm, pitch 1.375, rotation time 0.7 s,etector width 40 mm, field of view 50 cm, matrix 5122, withast switching of the kilovolt setting for the acquisition (lownergy 80 kVp/high energy 140 kVp).

Dedicated post-treatment software (GSI Viewer®, GEealthCare) allowed a two-fold analysis to be carriedut: firstly a mapping of material decomposition (MD)nd secondly an assessment of effective atomic number.

The mapping compared all pairs two by two that includedhe following materials: calcium oxalate, uric acid, and cys-ine, minerals that form part of the composition of 95% ofrinary stones. The atomic number was determined afterracing an elliptical region of interest (ROI) measuring ateast 8 mm2, covering the largest surface area possible ofhe stone being studied and avoiding the edges.

We illustrate this technique using three cases treated

n our department (Figs. 1—3). Each time the biochemicalomposition of the calculi was confirmed by biochemicalnalysis after extraction.

igure 1. Uric acid urinary stone. a: material decomposition maps: thenly shows calculi on maps that do not remove the uric acid signal; b: stumbers within the pixels of the region of interest shows a peak (green)

o

C. Sanavi et al.

In this study, the mean DLP was evaluated at 508 mGym2, the diagnostic reference standard being 800 mGy cm2

or a standard abdominal pelvic CT scan.

iscussion

he multi-energy CT allowed us to access new informationoncerning the chemical composition of calculi [6,7]. Single-ource, dual energy CT relies on the acquisition of a targetolume using X-ray beams of two different energy levels (80nd 140 kVp) that are emitted alternately with fast switch-ng between the two. The attenuation curves depend onoth the nature of the material and on the energy fromhe monochromatic X-ray beam. A material therefore gives

unique response depending on the level of energy fromhe X-ray beam, resulting in a specific attenuation curve.hanks to chromatography techniques, these attenuationurves are known for quite a number of materials (iodine,ater, calcium, uric acid, etc.). In order for these data toe handled by computer, a dual energy index is then cal-ulated. This index corresponds to the slope of the specificttenuation curve; which is unique to each material. Post-reatment software can then use this dual energy index toenerate new images that depart entirely from the con-entional Hounsfield system: these are colour-coded dualnergy index maps.

study of the uric acid—calcium oxalate and uric acid—cystine pairsudy of the atomic number of calculi: the representation of atomic

corresponding to the atomic number of uric acid (red).

To study urinary stones we use paired colour-coded mapsf material decomposition (MD): within the selected voxels,

Urinary stones: The contribution of dual energy CT and material decomposition 1167

Figure 2. Calcium oxalate urinary stone in contact with a JJ stent. a: material decomposition maps: the study of the calcium oxalate—uricacid and calcium oxalate—cystine pairs only shows calculi on maps that do not remove calcium oxalate; b: study of the atomic number ofcalculi: the representation of atomic numbers within the pixels of the region of interest shows a peak (green) corresponding to the atomicnumber of calcium oxalate (purple).

Figure 3. Cystine urinary stone. a: material decomposition maps: the study of the cystine—struvite and cystine—uric acid pairs only showscalculi on maps that do not remove cystine; b: study of the atomic number of calculi: the representation of atomic numbers within the

to th

cbosc

mn

pixels of the region of interest shows a peak (green) corresponding

the dual energy indices of the selected materials are com-pared two by two. The maps that are currently used in renallithiasis compare uric acid, calcium oxalate, calcium phos-phate, cystine, and struvite, as they constitute over 95% ofstones.

It is also possible to extract the atomic numbers (Zn)of the voxels studied in the form of a pattern of distribu-tion of material. One region of interest is placed in thestructure to be analysed and a diagram is generated, repre-senting the concentrations of the different pixels according

to their atomic number. A concentration peak is visualisedwhen the region of interest predominantly contains one ele-ment with the same atomic number. For example, uric acidcalculi are made up of light elements, such as hydrogen,

C

Ue

e atomic number of cystine (orange).

arbon, nitrogen, and oxygen and present a low atomic num-er that only varies very slightly. In all of our cases, the studyf the atomic number of the chosen ROI within the calculustudied correlated with the biochemical analysis, showingoncentration peaks next to the expected atomic numbers.

Nonetheless, one limitation is that the assessment ofixed calculi remains difficult even when using this tech-

ique.

onclusion

rinary lithiasis is a common pathology in which non-nhanced CT imaging is at the forefront of diagnostic and

1

tad

D

JC

R

[

[

[

[

[

[

168

herapeutic management. Today, the use of dual energy CTllows us to address the composition of urinary stones, a keyeterminant in identifying suitable treatment.

isclosure of interest

ean-Nicolas Dacher works as a consultant for GE Health-are.

eferences

1] Saw KC, McAteer JA, Monga AG, Chua GT, Lingeman JE, WilliamsJr JC. Helical CT of urinary calculi: effect of stone composi-tion, stone size, and scan collimation. AJR Am J Roentgenol2000;175:329—32.

[

C. Sanavi et al.

2] Rassweiler JJ, Renner C, Chaussy C, Thüroff S. Treatmentof renal stones by extracorporeal shockwave lithotripsy: anupdate. Eur Urol 2001;64:237—40.

3] Mercks E, Beers MH, Fletcher AJ. Stones in the urinary tract. In:The Merck Manual of Medical Information Home Edition. 3rd ed;2008. Chapter 148.

4] Bellin MF, Renard-Penna R, Conort P, Bissery A, Meric JB, DaudonM, et al. Helical CT evaluation of the chemical composition ofurinary tract calculi with a discriminant analysis of CT attenua-tion values and density. Eur Radiol 2004;14(11):2134—40.

5] Richmond J. Radiological diagnosis of kidney stones. Nephrology2007;12:S34—6.

6] Eliahou R, Hidas G, Duvdevani M, Sosna J. Determination ofrenal composition with dual-energy computed tomography; an

emerging application. Semin Ultrasound CT MR 2010;31:315—20.

7] Hartman R, Kawashima A, Takahashi N, Silva A, Vrtiska T, LengS, et al. Application of dual energy ct in urologic imaging: anupdate. Radiol Clin N Am 2012;50:191—205.