26.01.2017

1

Gicht –

Alte und neue Möglichkeiten der Bildgebung

Prof. Dr. Gustav Andreisek, MD, MBA, SCMR

Chefarzt RadiologieKantonsspital MünsterlingenSpital ThurgauSchweiz

1

Ziel

Review typischer radiologischer Befunde

Case Based - Dual Energy CTs

2

Übersicht

Arthritis

Chondropathic

OsteoarthritisCPPD

Hemochromatosis Hemophiliac Neuropathic

Inflammatory

RAJRA

Psoriatic ArthritisReiter’s disease

Ank spondInfection

Depositional

GoutAmyloid

HyperlipidemiaSarcoidPVNS

Other

SLEScleroderma

HOAHyperparathyroidism

3

Gicht

4

26.01.2017

2

Gicht

5

Gicht

6

Gicht

7

Gicht

8

26.01.2017

3

GichtBursitis olecrani

9

Rheumatoide Arthritis - Frühform

10

Rheumatoide Arthritis - Spätform

11

Rheumatoid arthritis

12

26.01.2017

4

Das 5. MCP- Gelenk ist in Füssen zuerst betroffen.

13 14

Rheumatoid arthritis

15

Fingerpolyarthrose

16

26.01.2017

5

Erosive Fingerpolyarthrose

17 18

Kalzium Pyrophosphat Krankheit (CPPD)

19 20

26.01.2017

6

Psoriasis

21

Psoriasis Arthritis

22

Psoriasis Arthritis“pencil in cup”

23

Psoriasis ArthritisPeriostitis

24

26.01.2017

7

Case Based

25Courtesy TY Moon, MD

44MRight knee pain

26

T2 FS Gd T1 FS

27 T2 FS Gd T1 FS

Tophaceous gout involving the patellar tendon and prepatellarsoft tissue

28

26.01.2017

8

Fall 16 - 61 J. Frau

• Schwellung am linken Grosszehengrundgelenk

• Befund:– derber Schwellung – nicht verschieblich – Druckdolent, gerötet

29

Rx Füsse bds., Fragestellung: Gicht?

wie weiter ?

30

Dual-energy CT

31

CT Fuss links mit dual-energy

Beurteilung: keine Gichttophi

32

26.01.2017

9

33

• Gonarthritis rechts

• bekannte schwere Pangonarthrose (externe MRT)

• St.n. antibiotischer Behandlung mit Co-Amoxicillin von 02/13-05/13 bei DD infektiöser Arthritis (CRP max. 148 mg/L, jedoch kein bakterielles Wachstum im Punktat nachgewiesen)

• aktuell: wieder stärkere Schmerzen Knie rechts

• ad dual-energy CT: Frage nach Uratablagerungen?

Fall 18: 66 jähriger Mann

34

CT Knie rechts mit dual-energy

35 36

26.01.2017

10

37

1983… Die 3. Sackgasse in der CT-Entwicklung: Elektronenstrahl-CT (EBT)Der Elektonenstrahl wird elektromagnetisch gesteuert. Die Abtastung einer gesamten Schicht ist in 50 msec möglich, da die neue Technik keine rotierenden Geräteteile benötigt (die Detektoren sind stationär befestigt). Die Schichtdicke beträgt 0.8 mm. Durch die hervorragende Zeitauflösung sind sowohl morphologische als auch funktionelle Herzuntersuchungen möglich. Im Vergleich zur Spiral-CT oder der späteren MDCT ist der Unterhalt der Geräte um ein Vielfaches teurer.

CT-Entwicklung: 1983 – EBT

38

1989… Der Durchbruch zur Volumenbildgebung: „Von der Scheibe zur Kugel“Entwicklung und erste klinische Untersuchungen mit der Spiral-CT. Äußerlich sehen die Geräte nahezu unverändert aus. Da sich die Untersuchungsliege allerdings kontinuierlich bewegt, wird ein spiralförmiger Datensatz gewonnen, aus dem erst durch mathematische Umrechnungen die klassischen Schichtbilder erzeugt werden.

CT-Entwicklung: 1989 – Spiral-CT

39

1998… Einführung der Mehrzeilen-Technik (MDCT)Erstes kommerziell erhältliches 4-Zeilen Gerät. Mit der neuen Technik werden vier parallele „Spiralen“ gleichzeitig aufgenommen. Somit wird die Untersuchungszeit drastisch reduziert, was insbesondere in der Diagnostik bewegter Organe, z.B. dem Herzen deutliche Vorteile bringt. Die neue Technik kann aber auch zur Verbesserung der räumlichen Auflösung durch Anfertigung dünnerer Schichten verwendet werden.

CT-Entwicklung: 1998 – MDCT

40

26.01.2017

11

Fortsetzung der MDCT-Entwicklung: „Wieviele Zeilen braucht der Mensch?“

2001… 16-Zeilen CT

2004… 64-Zeilen CT

2007… 128- bis 320-Zeilen CT

CT-Entwicklung: 1998-… = Mehr Zeilen

41

Aug. 1999

Ein-Zeilen-CT

Aug. 2001 Aug. 2003 Okt. 2004

4-Zeilen-CT 16-Zeilen-CT 64-Zeilen-CT

Entwicklung der Bildqualität:

Follow-up bei Patientin mit Fibromuskulärer Dysplasie

CT-Entwicklung: 1998-… = Mehr Zeilen

42

2006… Multisource-CT: „Escaping the Hounsfield cage of CT density values…“

Die Verwendung von mehr als einer Röntgenröhre verbessert die Scangeschwindigkeit und die zeitliche Auflösung. Die Applikation von unterschiedlichen Röhrenspannungen an beiden Röhren erlaubt eine verbesserte Gewebedifferenzierung.

CT-Entwicklung: 2006 – Dual Source CT

43

RöntgenröhreKollimator

Detektor

Aufbau des Computertomographen

44

26.01.2017

12

45

Viel Wissen stammt von Atombombenabwürfen und Berechnungen

46

Wir werfen in der Medizin aber nicht mit Atombomben,

sondern bewegen und in anderen Skalen

47 48

26.01.2017

13

Clinical Audits

(www.imagewisely.org) 49

Pauschale Urteile sind Blödsinn

50

Hintergrundbelastung

Wie hoch schätzen Sie die jährliche Belastung pro Person pro Jahr in der Schweiz?

– 0.04 mSv

– 0.4 mSv

– 4.0 mSv

– 40 mSv51

~ 4-5 mSv

Courtesy Prof. Verdun52

26.01.2017

14

53 54

3 mm Schichtdicke 10 mm Schichtdicke

Dickere Schichten haben weniger Rauschen und bessere

Kontrasterkennbarkeit für Weichteile, aber geringere Konturschärfe.

Welches Bild gefällt Ihnen besser ?

55

Konventionelles Röntgen

56

26.01.2017

15

Benefit versus Risiko~Thorax-Röntgen ~MDCT

57 58

Begriffe am Bsp Thorax

• CT «normal» 2-4 mSv• Low dose 1-2 mSV

• Ultra Low Dose 0.1-0.05 mSv= Röntgen Thorax

59

Zusammenfassung

Review typischer radiologischer Befunde Case – based

Literatur

60

COMPUTED TOMOGRAPHY

Gouty arthritis: the diagnostic and therapeutic impactof dual-energy CT

Tim Finkenstaedt1 & Andrei Manoliou1& Martin Toniolo2 & Kai Higashigaito1 &

Gustav Andreisek1& Roman Guggenberger1 & Beat Michel2 & Hatem Alkadhi1

Received: 16 July 2015 /Revised: 14 January 2016 /Accepted: 21 January 2016# European Society of Radiology 2016

AbstractObjectives To determine the diagnostic and therapeuticimpact of dual-energy computed tomography (DECT) ingout.Methods Forty-three patients with (n=20) and without a his-tory of gout (n=23) showing non-specific soft tissue depositsunderwent DECT after unrewarding arthrocentesis. Twoblinded, independent readers evaluated DECT for the pres-ence of urate crystals. Clinical diagnosis, clinically suspectedurate crystal locations, diagnostic thinking and therapeuticdecisions were noted before and after DECT. Clinical1-month follow-up was obtained.Results DECT showed urate in 26/43 patients (60 %).After DECT, clinical diagnosis of gout was withdrawnin 17/43 (40 %) and was maintained in 16/43 patients(37 %). In 10/43 patients (23 %) the diagnosis was main-tained, but DECT revealed urate in clinically unsuspectedlocations. In 23/43 patients (53 %), a treatment-changebased on DECT occurred. Changes in diagnostic thinkingoccurred more frequently in patients without a history ofgout (p< 0.001), changes in therapeutic decisions morefrequently in patients with a history of gout (p= 0.014).Clinical follow-up indicated beneficial effects of DECT-based diagnoses in 83 % of patients.

Conclusions In patients with or without a history of gout anda recent suspicion for gouty arthritis with an unrewardingarthrocentesis, DECT has a marked diagnostic and therapeuticimpact when hyperdense soft-tissue deposits are present.Key Points• This study evaluates the concept of evidence-basedradiology

• In patients with suspected gout, DECT can help cliniciansmake the diagnosis

• DECT has a marked impact on therapy• Clinical follow-up after 1 month indicated reliable results ofDECT

Keywords Gout . Gouty arthritis . Urate . Dual-energycomputed tomography . Evidence-based radiology

AbbreviationsAAD Acute anti-inflammatory drugsCPPD Calcium pyrophosphate dihydrate deposition

diseaseDECT Dual-energy computed tomographyEBR Evidence-based radiologySD Standard deviationUALD Uric-acid-lowering drugs

Introduction

Detection of urate crystals is crucial for the management ofgouty arthritis [1, 2]. The clinical reference standard test forthe diagnosis of urate crystals is microscopic analysis of jointfluid samples; however, arthrocentesis features several issuessuch as invasiveness [3] and poor sensitivity [4, 5]. The pre-cipitated urate crystal deposits in further stages of disease can

* Hatem Alkadhihatem.alkadhi@usz.ch

1 Institute of Diagnostic and Interventional Radiology, UniversityHospital Zurich and University of Zurich, Raemistrasse 100,8091 Zurich, Switzerland

2 Department of Rheumatology, University Hospital Zurich andUniversity of Zurich, Raemistrasse 100, Zurich 8091, Switzerland

Eur RadiolDOI 10.1007/s00330-016-4237-2

be appreciated on radiographs or conventional computed to-mography (CT) images as hyperdense soft-tissue deposits.However, the imaging appearance of such deposits is relative-ly non-specific, and soft tissue hyperdensity can be seen inother conditions such as calcium pyrophosphate dihydrate de-position disease (CPPD) as well. These alternative diagnosescan have similar clinical presentations, leading to ambiguity[6].

Single-energy CT in gout allows for depiction of bone ero-sions and hyperdense soft tissue deposits. While urate crystaldeposits occasionally are less dense than calcified deposits [7],single-energy CT usually does not allow for an accurate dis-tinction between the various soft-tissue deposits [8]. In con-trast to single-energy CT, dual-energy CT (DECT) allows thedistinction of two materials as calcium and urate crystals,based on the effective atomic numbers and electron densities[9–11].

The concept of evidence-based radiology (EBR) pro-poses a six-level scale for the systematic evaluation ofthe impact of the results of imaging studies [12].Regarding DECT in patients with gout, studies so far ad-dressed the technical performance (level 1) [9, 13, 14] andthe diagnostic accuracy and reproducibility (level 2)[15–21]. However, it is also mandatory to know the im-pact of the test on patient management (levels 3 and 4).This includes knowledge about the percentage of patientsfor whom the test is judged useful for rendering the diag-nosis, or for whom the diagnosis is substantially modifiedafter the test. Furthermore, it is crucial to know the per-centage of patients for whom a test is judged useful fortreatment planning and for whom treatment planning ismodified on the basis of the information provided [12,13].

So far, no study—to the best of our knowledge—hasevaluated the role of DECT in gout disease at thesehigher levels of EBR. Thus, the purpose of this studywas to determine the diagnostic and therapeutic impactof DECT imaging findings in patients with or without ahistory of gout and a recent suspicion for gouty arthritiswith an unrewarding arthrocentesis of the currently af-fected joint.

Materials and methods

Patient population

Institutional review board approval was obtained. Written in-formed patient consent was waived by our local ethics com-mittee due to the retrospective nature of the study. All CTexaminations were clinically indicated and no CT was per-formed for the mere purpose of the study.

Ninety-three consecutive patients with a history of gout(n = 26) and without a history but with suspected gout(n = 67) who were referred for DECT of the extremities(n=90), spine (n=2) and hip (n=1) between February 2012and December 2013 were screened for study inclusion. DECTwas performed in all 93 patients after the diagnostic referencestandard approach (i.e. polarisation microscopy of joint fluidsamples) of the scanned joint was unrewarding (i.e.arthrocentesis could not be performed or was inconclusive).There were no other exclusion criteria; especially no selectionwas made whether or not patients had rheumatic disease.None of our patients had a known rheumatoid arthritis.

History of gout was defined as gout being known from thepatient's previous medical history anywhere in the body,which was based at that time on clinical symptoms, bloodanalysis and/or arthrocentesis.

No history was defined as gout being suspected in patientswith typical symptoms, like sudden intense joint pain accom-panied by inflammation and redness without a medical historyof gout. Differential diagnoses of gouty arthritis included de-generative osteoarthritis, rheumatoid arthritis, CPPD and se-ronegative inflammatory arthritis.

Forty-three of the 93 patients (9 women, mean age 66.1± 12.2 years [range 52-81 years]; 34 men, 59.9 ± 18.1 years[32-84 years]) had hyperdense soft tissue deposits on CTand thus were included into further analyses (Fig. 1 andTable 1). This was done because urate crystal deposits canbe detected with DECT only in the presence of hyperdensesoft-tissue deposits above a certain attenuation threshold[22]. The time period between onset of joint symptomsand DECT examinations ranged between 9 and 17 weeks(median 12 weeks).

Twenty out of 43 patients (47 %) with soft-tissue depositshad a history of gout, which was based at that time on clinicalsymptoms, blood analysis and arthrocentesis in 14 (70 %) andon clinical symptoms and blood analysis in 6 patients (30 %)of the patients. None of these 20 patients had an actual suc-cessful arthrocentesis of the currently affected joint(arthrocentesis was inconclusive in 7/20 patients [35 %] andcould not be performed in 13/20 patients [65 %]). The remain-ing 23 patients (53 %) with soft tissue deposits had no historyof gout. In 9/23 patients (39 %), clinical symptoms of thecurrently affected joint were ambiguous and arthrocentesiswas inconclusive, whereas in 14 patients (61 %) clinicalsymptoms were ambiguous and arthrocentesis could not beperformed. The time period between attempted arthrocentesisand DECT examinations ranged between 0 and 11 days (me-dian 7 days).

CT data acquisition and image reconstruction

All scans were performed using second-generationdual-source CT (SOMATOM Definition Flash; Siemens

Eur Radiol

Healthcare, Forchheim, Germany). Tube potentials were80 kVp and 140 kVp, the latter with tin filtration (Sn).Depending on the scanned body region, quality reference tubecurrents ranged between 130 to 170 mAs for tube A andbetween 200 to 250 mAs for tube B. Automated attenuation-based tube current modulation was used in all examinations.The tube currents of DECT were set to equal the radiationdoses of our corresponding single-energy CT protocols.The average radiation doses of DECT as taken from theelectronically logged patient protocols were as follows: vol-ume CT dose index (CTDIvol) spine, 11.3 mGy (estimated

effective dose, 4.4 mSv); CTDIvol hip, 13.7 mGy (3.6 mSv);CTDIvol foot/ankle, 10.3 mGy (0.03 mSv); CTDIvol knee,13.2 mGy (0.1 mSv); CTDIvol hand/wrist, 6.7 mGy(0.03 mSv).

Axial images with a soft-tissue convolution kernel(I50s, sinogram-affirmed iterative reconstruction, strengthlevel 3) with single and weighted energies (DE composi-tion factor 0.1) and a slice thickness of 0.75 mm (incre-ment 0.5 mm) were reconstructed. In addition, regulargreyscale reformations in all three orientations with a slicethickness of 1 mm (increment 1 mm) were reconstructed

Fig. 1 Flow chart of the study

Table 1 Patient demographics

All patients witharthropathy referredfor DECT (n= 93)

Patients withhyperdense soft tissuedeposits (n= 43)

Patients with DECTpositive for urate crystaldeposits (n= 26)

Patients with DECTnegative for uratecrystal deposits(n = 17)

p valuea

Age, mean (SD), years 60.1 (16.4) 63 (18.6) 64.7 (19.1) 59 (17.9) 0.339

Gender 0.445Male 63 (68) 34 (79) 21 (81) 12 (71)

Female 30 (32) 9 (21) 4 (15) 5 (29)

Patients with Soft tissue deposits 43 (46) 43 (100) 26 (100) 17 (100) 1.000

Body study area:Cervical spine 3 1 - 1

Hand/wrist 37 21 10 8

Knee 7 7 3 2

Foot/ankle 47 25 13 6

Hip 1 1 1 -

History of gout as per medical records 26 (28) 20 (47) 18 (69) 2 (12) <0.001

Except where stated otherwise, values in parentheses are the percentage (%) of patients

SD standard deviation, DECT dual-energy computed tomographya The p values refer to comparison of patients with DECT results positive (n= 26) and negative (n = 17) for urate crystal deposits

Eur Radiol

from the weighted-DE images (50 % 80 kVp and 50 %140 kVpSn) using a bone tissue convolution kernel(Q30s). The following parameters were used: minimum(HU), 150; iodine ratio, 1.4; air distance, 5; bone distance,10; resolution, 4; material definition ratio, 1.25. Post-processing of DECT was performed by radiologicaltechnologists on a workstation equipped with dedicatedsoftware (syngo.via VA31, syngo Dual Energy Gout;Siemens), which colour-codes urate crystal deposits asgreen, and allows for fusing coloured images ontomulti-planar reformations and volume-rendered 3D im-ages (Fig. 2). Using this software, colour-coded multi-planar images were reconstructed using a soft tissue ker-nel (I50s) with a slice thickness of 2 mm (increment2 mm). Colour-coded volume-rendered three-dimensional(3D) images were generated, highlighting urate crystaldeposits in green. Post-processing for each case takes ap-proximately 2 min (Figs. 3 and 4).

CT data analysis

Two radiologists (with 3 and 4 years of experience in ra-diology) blinded to the patients’ medical records indepen-dently evaluated the images in random order in a dichoto-mous manner (yes/no) for presence of soft-tissue depositson greyscale CT reformations and for urate crystal depositson colour-coded images (multi-planar reformations andvolume-rendered 3D images). Locations of urate crystaldeposits were noted. In addition, the total volume of uratecrystal deposits (gout burden) was quantified (in cm3) au-tomatically by the software. In case of disagreement be-tween readers, consensus was achieved by a third blindedreader (10 years of experience in radiology), who was notinvolved in primary reading. All readers were fully aware

of the known artefacts of DECT for urate deposit detectionand of the potential underestimation of urate deposits withDECT [14, 18].

Clinical evaluation

The rheumatology department of our hospital follows currentinternational guidelines [23, 24]. Electronic medical recordsof all patients were reviewed by a physician (with 8 years ofexperience in rheumatology), who was blinded to the resultsfrom CT. The median time interval between medical recordentries was 8 days before and 15 days after DECT. The fol-lowing information was retrieved:

Change in diagnosis

A change in diagnosis was defined as any difference in thediagnosis made by the clinicians before and after DECT asfollows:

& Suspected diagnosis before DECT and after DECT.& Clinically suspected location of urate crystal deposits be-

fore and after DECT.& Other relevant clinical assessment of the scanned area be-

fore and after DECT.

Change in therapy

Change in therapy was defined as a difference in the therapeu-tic strategy by the clinicians before and after DECTas follows:

Fig. 2 DECT of the right hand in a 61-year-old female patient with nohistory of gout and suspected gouty arthritis at all joints of Dig. II-IV. aAxial regular greyscale CT reformations show non-specific hyperdensesoft tissue deposits (arrows) in the extensor tendon sheaths and inperiarticular structures. b Sagittal reformations of colour-coded images

and c volume rendered 3D images indicate that the soft-tissue depositscoded in green contain urate crystals dorsally to the MCP II-IV and inaddition at the proximal and distal interphalangeal joint of Dig. II-IVandinterphalangeal joint of Dig. I. The total urate crystal burdenwas 1.62 cm3

Eur Radiol

& Medication and dosages before and after DECT.Medications were grouped according to the followingmain classes:

– Acute anti-inflammatory drugs (AAD)—non-steroi-dal anti-inflammatory drugs, colchicine and steroids

– Uric-acid-lowering drugs (UALD)—xanthine oxi-dase inhibitor (allopurinol and febuxostat), uricosu-rics (probenecid and sulfinpyrazone) and uric acidmetaboliser (pegloticase)

& Invasive procedures (infiltration with local anaesthetic/steroids) after DECT.

& Other relevant changes in treatment: start of antibiotics,initiation of weight-relief, prostheses replacement andconduction of targeted physiotherapy.

Clinical follow-up

The electronic medical records were reviewed regarding clin-ical follow-up focussing on those body parts examined byDECT. The following was noted:

& Subjective relief of the symptoms in the affected body part.

Fig. 3 DECT of the right knee in a 52-year-old male patient with nohistory of gout and suspected gouty arthritis in the medial and lateralfemorotibial compartment. a Coronal regular greyscale CT reformationsshow non-specific hyperdense soft tissue deposits adjacent to the femoralinsertion of the lateral collateral ligament (arrow). On colour-coded (b)

and volume rendered 3D images (c) no deposits were encoded in green,which indicates that none of the soft-tissue deposits contain urate crystals.This favours the diagnosis of calcium pyrophosphate dihydrate deposi-tion disease (CPPD)

Fig. 4 DECTof the left ankle in a 79-year-old male patient with a historyof gout. Referring physicians suspected gouty arthritis at the upper anklejoint. a Coronal regular greyscale CT reformations show non-specifichyperdense soft tissue deposits (arrows) in the deltoid ligament and ad-jacent to the medial malleolus as well as in the sinus tarsi and anteriortalofibular ligament. b Coronal reformations of colour-coded and c

volume rendered 3D images demonstrate gout deposits encoded in greenat the clinically expected location and additionally in the tarsal sinus.Three-dimensional images furthermore show urate crystal deposits(arrow) adjacent to the fifth dorsal tarsometatarsal ligament. The totalurate crystal burden was 2.47 cm3

Eur Radiol

& Objective relief of arthropathy symptoms based on clinicalevaluation.

Statistical analysis

Descriptive statistics were used to evaluate the potentialimpact of imaging on diagnostic thinking and therapeuticdecision-making, as previously shown [25]. Regarding di-agnostic thinking, diagnoses obtained before and afterDECT were compared: (1) DECT results negative forurate crystal deposits; (2) DECT results confirmed clinicaldiagnosis and suspected location of urate crystal deposits;(3) DECT results demonstrated urate crystal deposits at adifferent, clinically not suspected location; (4) DECT re-sults demonstrated urate at the clinically suspected and inan additional location.

Regarding therapeutic decision making, the followingtherapeutic consequences were assessed: (1) start of treat-ment with AAD; (2) start of treatment with UALD; (3)intensification of the current treatment; (4) discontinua-tion of the current treatment; (5) performance of invasiveprocedures.

Descriptive parameters were provided as frequenciesand percentages. Inter-reader agreement for qualitativemeasures was analysed with Cohen’s kappa (κ) coeffi-cient: 0.21-0.40 indicated fair, 0.41-0.60 moderate,0.61-0.80 good and >0.81 indicated excellent agreement.To test for differences between groups (patients with andwithout a history of gout and with DECT-positive orDECT-negative for urate crystal deposits, respectively)Student's t-tests and chi-squared tests were used.Statistical analyses were performed using commerciallyavailable software (GraphPad Prism version 5.04 forWindows, GraphPad Software, La Jolla, CA, USA; IBMSPSS Statistics for Windows, Version 22.0.0.1, IBMCorporation, Armonk, NY, USA). A p< 0.05 was consid-ered statistically significant.

Results

Inter-reader agreement

Inter-reader agreement between readers for detection ofsoft-tissue deposits and for detection of urate crystal was ex-cellent (κ=1.0 and κ=0.94, respectively). From the 43 pa-tients, colour-coded images were positive for urate in 25 asdetermined by both readers. In one patient, a reader consid-ered a small soft-tissue deposit colour-coded in green to be anartefact, whereas the other reader classified it as containingurate. Here, consensus was achieved by consulting the thirdreader, and the patient ultimately was rated as being positivefor urate crystals.

Diagnostic impact of DECT

In 17/43 patients (40 %) with hyperdense soft-tissue deposits,no urate crystals were detected by DECT and the cliniciansdismissed their initial clinical suspicion of gouty arthritis afterDECT (Fig. 3 and Table 2). In these patients, the aetiology ofsoft tissue deposits was assumed to be due to degenerativeosteoarthritis, rheumatoid arthritis or CPPD. Two of these 17patients (12 %) had a history of gout.

Following DECT, gouty arthritis was confirmed but uratedeposits were not found at the clinically suspected location butat a different location in 4/43 patients (9 %), and additionalurate crystal deposits beyond the clinically suspected locationswere found in 6/43 patients (14 %) (Fig. 4). In 16/43 patients(37 %), DECT results confirmed the clinically suspected lo-cation (Fig. 2). After DECT no other diagnostic procedures(e.g. ultrasound) were performed.

Changes in diagnostic thinking occurred more often in pa-tients without than in those with a history of gout (p<0.001)(Table 3). In patients with a history of gout, urate crystal de-posits were found more frequently, and thus the diagnosis ofgout was more often confirmed than changed.

Therapeutic impact of DECT

In 23/43 patients (53 %) the treatment plan was changed afterDECT (Table 4). AAD was initiated in 17 % and UALD in29 %, pre-existing medication was intensified in 31 % andinvasive procedures (infiltration with local anaesthetic/ste-roids) were conducted in 23% of the total number of treatmentchanges. Sixteen of these 23 patients (69 %) had a history ofgout.

In 13/16 patients (81 %) in whom the location of uratecrystals was confirmed by DECT, the following treatmentchanges occurred: AAD were started in four and UALD werestarted in six patients, pre-existing medication was intensifiedin six and invasive procedures were conducted in three pa-tients. In the remaining three patients, medical therapy wasmaintained and no change in treatment plan occurred.

In the four patients whose urate crystal deposits were de-tected byDECTat a different location other than that clinicallysuspected, the change in treatment plan was as follows: AADwere started in two patients and UALD were started in onepatient, while the pre-existing medication was intensified intwo patients and invasive procedures were conducted in threepatients.

In the six patients whose urate crystal deposits were foundby DECT in addition to the location that was suspected byclinicians, the change in treatment plan was as follows: UALDwere started in three patients, while pre-existing medicationwas intensified in three patients and invasive procedures wereconducted in two patients.

Eur Radiol

Tab

le2

Impactof

DECTon

diagnosis(n=43)

DECTresults

Num

berof

DECT

exam

inations

Changein

diagnosisafterDECT

Diagnosis

changed

Diagnosis

andlocatio

nmaintained

Diagnosismaintained

butu

ratedepositsata

different

locatio

n

Diagnosismaintained

butu

ratedepositsat

anadditio

nallocation

CTprovided

furtherinform

ation(not

based

onthetwo-materiald

ecom

positio

nability

ofDECT)with

anim

pacton

diagnostic

thinking

DECTnegativeforuratedeposits

1717

00

07 →

degree

oferosions

anddegenerativ

echanges(n=3),presenceof

rheumatoid

nodules(n=1),presenceof

CPP

D(n=3)

DECTpositiveforurate

depositssolely

atclinically

suspectedlocatio

n

160

160

04 →

presence

ofCPP

D(n=1),degreeof

gout

burden

(n=1),loosening

ofthe

prosthesis(n=1),presenceof

phlegm

on/

abscess(n=1)

DECTdemonstratedurate

depositsatadifferent

locatio

n4

00

40

1 →presence

ofafracture

(n=1)

DECTdemonstratedurate

depositsatan

additio

nal

locatio

n

60

00

61 →

degree

oferosions

anddegenerativ

echanges(n=3)

Overall

43(100)

17(40)

16(37)

4(9)

6(14)

13/43(30)

Valuesin

parenthesesarethepercentage

(%)of

patients

DECTdual-energycomputedtomography,CPPDcalcium

pyrophosphatedehydratedisease

Eur Radiol

Changes in therapy were more frequent in patients with ahistory of gout as compared to those without (p= 0.014)(Table 3), with the main change of therapy being the startand intensification of medication as well as the performanceof invasive treatments (Table 4).

Clinical follow-up

The time period between DECTand clinical follow-up rangedbetween 10 and 42 days (median 32 days). Follow-up wasavailable in 32/43 patients (74 %). It was not available in threepatients with positive DECT and in eight patients with nega-tive DECT.

In 19/23 patients (83 %) with positive DECT, consequentgout-related changes in therapy and available clinicalfollow-up, treatment response was good showing both subjec-tive and objective relief of symptoms. In the remaining fourpatients (17 %), follow-up showed no change in subjectiveand objective symptoms.

In 5/9 patients (56 %) with negative DECT and conse-quently no gout-related changes in therapy and available clin-ical follow-up, treatment response was good showing bothsubjective and objective relief of symptoms. In the remainingfour patients (44 %) that were treated for an alternative diag-nosis like degenerative osteoarthritis, symptoms did not im-prove either subjectively or objectively.

Discussion

Evaluation of EBR levels 3 and 4 for DECT in gout diseasehas not yet been performed. Our study indicates that DECThas a considerable impact on diagnosis and therapy, becausein 40% of patients the clinical suspicion of gouty arthritis wasdismissed and the treatment plan was changed in 53 %.

For diagnosis of gouty arthritis, the clinical reference stan-dard test is microscopic analysis of joint fluid samples.However, percutaneous puncture carries the risk of infection.In addition, arthrocentesis can be problematic due to potentialrisk of infection, in joints that are anatomically challenging toaccess, when there is not sufficient joint effusion, in patientswith blood clotting levels prohibiting punctures, and if urate

crystal deposits are located in periarticular soft tissue but notin the synovial fluid. Besides these challenges, the diagnosticperformance of polarising microscopy for detecting uratecrystals is poor, showing a sensitivity of only approximately63-78% [4, 5]. These difficulties were encountered also in ourstudy, in which arthrocentesis of the recently affected jointwas unrewarding in all patients.

Several studies have tested the EBR level 1 for DECT ingout [9, 13, 14]. Johnson et al. [9] showed the feasibility ofDECT with an acceptable noise and image quality without aradiation dose penalty compared to single-energy CT. This isaccordance with our study in which the radiation dose ofDECT was kept dose-neutral to corresponding institutionalsingle-energy CT protocols. Several studies also evaluatedthe EBR level 2 [15–21]. Bongartz et al. [15] found a sensi-tivity and specificity of DECT for diagnosing gout of 0.90 and0.83, respectively, which is in line with other studies [17].Although it was not the scope of our study focusing on EBRlevels 3 and 4, it is important to mention that DECT also haslimitations. For example, Bongartz et al. [15] showed thatsensitivity of urate detection by DECT decreases in patientswith recent-onset gout. Also, Melzer et al. [22] showed in acase report that the density of the urate-containing tissue mustexceed an attenuation threshold of 150 HU (corresponding toapproximately 15-20 vol.% urate in a tophus) to be detectablewith DECT. Thus, detection of urate cyrstals occurring only injoint fluid might be detectable with ultrasonography but notwith DECT [26]. Thus, soft-tissue deposits with a densityabove this threshold are a prerequisite and consequently clini-cians should judge a DECT examination without hyperdensesoft-tissue deposits as not suitable for urate detection, ratherthan as a negative result. These issues should be taken intoconsideration for the correct interpretation of negative DECTresults.

As expected, changes in diagnosis based on DECT weremore frequent in patients with no history of gout than in thosewith a history of gout. In the latter group, urate crystal depositswere more frequent, and thus the diagnosis was more fre-quently maintained instead of changed. In patients with nohistory of gout, in distinction, DECT confirmed or excludeddeposits, since the reference standard for confirmation of agouty arthritis (i.e. arthrocentesis) was unrewarding.

Table 3 Impact of DECT ondiagnosis and therapy dependingon whether gout was known fromthe previous medical history ornot

History of gout(n= 20)

No history of gout(n= 23)

Diagnostic impact Change in diagnosis after DECT (n = 27) 7 (35 %)a 20 (87 %)a

No change in diagnosis after DECT (n= 16) 13 (65 %)a 3 (13 %)a

Therapeuticimpact

Change in treatment plan after DECT (n = 23) 15 (75 %)b 8 (35 %)b

No change in treatment plan after DECT (n = 20) 5 (25 %)b 15 (65 %)b

a Comparison between subgroups revealed a significant difference (p = 0.001)b Comparison between subgroups revealed a significant difference (p = 0.014)

Eur Radiol

Our study showed that after DECT the treatment planwas changed in 53 % of patients. Therapy for gouty arthri-tis differs from that for other forms of inflammatory arthri-tis. Thus, correct identification of urate crystal deposits isimportant for avoiding unnecessary and non-effectivetreatment. In our study, DECT led to the initiation of ther-apy with AAD in 17 %, UALD in 29 %, intensification ofpre-existing medication in 31 % and performance ofinvasive procedures in 23 %. Interestingly, changes in

therapeutic decision-making were more frequent in pa-tients with a history of gout, as opposed to those with nohistory of gout. This is most probably due to the higherfraction of patients with (by DECT) confirmed urate de-posits, resulting in starting and intensifying treatment, aswell as in performing invasive procedures. When uratedeposits could not be confirmed with DECT, differentialdiagnoses such as degenerative osteoarthritis were associ-ated with less treatment change.

Table 4 Impact of DECT ontherapy DECT imaging results (n= 43) Change in treatment plan

(n = 23 patients with 35treatment changes in total)

No change intreatment plan(n = 20)

DECT negative for urate deposits (n = 17) 0

→ Start of AAD - 0

→ Start of UALD - 0

→ Drug intensification - 0

→ Drug discontinuation - 0

→ Invasive procedures - 0

17

DECT confirmed clinical diagnosis (n= 16) 13

→ Start of AAD - 4

→ Start of UALD - 6

→ Drug intensification - 6

→ Drug discontinuation - 0

→ Invasive procedures - 3

3

DECT confirmed clinical diagnosis, butdemonstrated urate deposits at a differentlocation (n= 4)

4

→ Start of AAD - 2

→ Start of UALD - 1

→ Drug Intensification - 2

→ Drug Discontinuation - 0

→ Invasive procedures - 3

0

DECT confirmed clinical diagnosis, butdemonstrated urate deposits at anadditional location (n= 6)

6

→ Start of AAD - 0

→ Start of UALD - 3

→ Drug intensification - 3

→ Drug discontinuation - 0

→ Invasive procedures - 2

0

Total of 35 treatment plan changes:

→ Start of AAD (6)

→ Start of UALD (10)

→ Drug intensification (11)

→ Drug discontinuation (0)

→ Invasive procedures (8)CT provided further information (n= 13) 5

→ Start of antibiotics - 1

→ Initiation of weight relief - 1

→ Prostheses replacement - 1

→ Targeted physiotherapy - 2

8

One or more change(s) in treatment plan might result following DECT imaging for each patient

DECT dual-energy computed tomography, AAD acute anti-inflammatory drugs, UALD uric-acid-lowering drugs.

Eur Radiol

We included a clinical 1-month follow-up evaluation fortesting whether therapeutic decisions based on DECT influ-enced patient outcome. Here we could see that in 83 % ofpatients the response to treatment, which was changed basedon positive DECT findings, was good, leading to subjectiveand objective relief of symptoms. In 56 % of patients withnegative DECT findings and no gout-related change in thera-py treatment, response was also good with relief of symptoms.The differences in frequencies are most probably related to theresulting diagnoses: in patients without urate crystal depositsaccording to DECT, the differential diagnoses degenerativeosteoarthritis, rheumatoid arthritis and CPPD are known torespond less well to therapy as opposed to the relatively fastresponse to anti-inflammatory and uric-acid-lowering medica-tion in gouty arthritis.

We have to acknowledge the following study limitations.First, this is a retrospective study and thus diagnostic andtherapeutic impact as well as follow-up data were registeredthrough electronic medical records. Second, the time periodbetween DECT and the next electronically recorded consulta-tion varied so that the impact of DECT might have been bi-ased. However, time periods were in the range of 8-15 daysonly. Third, only 43 patients were included in this study andthe clinical follow-up evaluation after an average of 30 days israther short-term. A larger population in a prospective studywith a questionnaire for clinicians to evaluate diagnostic andtherapeutic thinking before and after DECT would be desir-able. Fourth, as discussed above, there might be false-positiveor false-negative DECT findings and there was an inherentlack of a Bgold standard^ reference test or additional diagnos-tic parameter to compare them with. Fifth, only patients withhyperdense soft tissue deposits were included into furtheranalysis, which is explained by the above-mentioned short-coming of DECT being feasible only above a certain attenu-ation threshold of deposits. Finally, our study represents theexperience of a single centre, and the diagnostic and therapeu-tic approachmay vary in other institutions. However, manage-ment and therapy of patients with gout at our institution wereconducted in accordance with current guidelines [23, 24].

In conclusion, this study shows that—in patients with orwithout a history of gout and an unrewarding arthrocentesis ofthe currently affected joint being suspected for gouty arthritis—DECT has a marked impact on diagnosis and therapy whenhyperdense soft-tissue deposits are present. Preliminary clinicalfollow-up data indicate that DECT-based treatment decisionsare beneficial to the patients.

Acknowledgements The scientific guarantor of this publication isHatem Alkadhi, MD, MPH, EBCR. The authors of this manuscript de-clare no relationships with any companies, whose products or servicesmay be related to the subject matter of the article. The authors state thatthis work has not received any funding.

One of the authors has significant statistical expertise. InstitutionalReview Board approval was obtained. Written informed consent was

waived by the Institutional Review Board. Methodology: Retrospective,case-control study, performed at one institution.

References

1. Hak AE, Choi HK (2008) Lifestyle and gout. Curr Opin Rheumatol20:179–186

2. Lawrence RC, Felson DT, Helmick CG et al (2008) Estimates of theprevalence of arthritis and other rheumatic conditions in the UnitedStates. Part II. Arthritis Rheum 58:26–35

3. Cheng J, Abdi S (2007) Complications of joint, tendon, and muscleinjections. Tech Reg Anesth Pain Manag 11:141–147

4. Segal JB, Albert D (1999) Diagnosis of crystal-induced arthritis bysynovial fluid examination for crystals: lessons from an imperfecttest. Arthritis Care Res 12:376–380

5. Swan A, Amer H, Dieppe P (2002) The value of synovial fluidassays in the diagnosis of joint disease: a literature survey. AnnRheum Dis 61:493–498

6. Ogdie A, Taylor WJ, Weatherall M et al (2014) Imaging modalitiesfor the classification of gout: systematic literature review and meta-analysis. Ann Rheum Dis. doi:10.1136/annrheumdis-2014-205431

7. Paparo F, Fabbro E, Ferrero G et al (2011) Imaging studies ofcrystalline arthritides. Reumatismo 63:263–275

8. Nicolaou S, Yong-Hing CJ, Galea-Soler S, Hou DJ, Louis L, MunkP (2010) Dual-energy CT as a potential new diagnostic tool in themanagement of gout in the acute setting. AJR Am J Roentgenol194:1072–1078

9. Johnson TR, Krauss B, Sedlmair M et al (2007) Material differen-tiation by dual energy CT: initial experience. Eur Radiol 17:1510–1517

10. Botsikas D, Hansen C, Stefanelli S, Becker CD, Montet X (2014)Urinary stone detection and characterisation with dual-energy CTurography after furosemide intravenous injection: preliminary re-sults. Eur Radiol 24:709–714

11. Nattenmuller J, Hosch W, Nguyen TTet al (2015) Hypodense liverlesions in patients with hepatic steatosis: do we profit from dual-energy computed tomography? Eur Radiol. doi:10.1007/s00330-015-3772-6

12. Sardanelli F, Hunink MG, Gilbert FJ, Di Leo G, Krestin GP (2010)Evidence-based radiology: why and how? Eur Radiol 20:1–15

13. Mallinson PI, Coupal T, Reisinger C et al (2014) Artifacts in dual-energy CT gout protocol: a review of 50 suspected cases with anartifact identification guide. AJR Am J Roentgenol 203:W103–W109

14. Diekhoff T, Kiefer T, Stroux A et al (2015) Detection and charac-terization of crystal suspensions using single-source dual-energycomputed tomography: a phantom model of crystal arthropathies.Invest Radiol 50:255–260

15. Bongartz T, Glazebrook KN, Kavros SJ et al (2015) Dual-energyCT for the diagnosis of gout: an accuracy and diagnostic yieldstudy. Ann Rheum Dis 74:1072–1077

16. Glazebrook KN, Guimaraes LS, Murthy NS et al (2011)Identification of intraarticular and periarticular uric acid crystalswith dual-energy CT: initial evaluation. Radiology 261:516–524

17. Choi HK, Burns LC, Shojania K et al (2012) Dual energy CT in gout:a prospective validation study. Ann Rheum Dis 71:1466–1471

18. Huppertz A, Hermann KG, Diekhoff T, Wagner M, Hamm B,SchmidtWA (2014) Systemic staging for urate crystal deposits withdual-energy CT and ultrasound in patients with suspected gout.Rheumatol Int 34:763–771

19. Bacani AK, McCollough CH, Glazebrook KN et al (2012) Dualenergy computed tomography for quantification of tissue urate de-posits in tophaceous gout: help from modern physics in the man-agement of an ancient disease. Rheumatol Int 32:235–239

Eur Radiol

20. Glazebrook KN, Kakar S, Ida CM, Laurini JA, Moder KG, Leng S(2012) False-negative dual-energy computed tomography in a pa-tient with acute gout. J Clin Rheumatol 18:138–141

21. Dalbeth N, House ME, Aati O et al (2015) Urate crystal depositionin asymptomatic hyperuricaemia and symptomatic gout: a dual en-ergy CT study. Ann Rheum Dis 74:908–911

22. Melzer R, Pauli C, Treumann T, Krauss B (2013) Gout tophusdetection-A comparison of dual-energy CT (DECT) and histology.Semin Arthritis Rheum. doi:10.1016/j.semarthrit.2013.11.002

23. Khanna D, Fitzgerald JD, Khanna PP et al (2012) 2012American College of Rheumatology guidelines for manage-ment of gout. Part 1: systematic nonpharmacologic and phar-macologic therapeutic approaches to hyperuricemia. ArthritisCare Res (Hoboken) 64:1431–1446

24. Khanna D, Khanna PP, Fitzgerald JD et al (2012) 2012American College of Rheumatology guidelines for manage-ment of gout. Part 2: therapy and antiinflammatory prophylax-is of acute gouty arthritis. Arthritis Care Res (Hoboken) 64:1447–1461

25. Muller MA, Mayer D, Seifert B, Marincek B, Willmann JK(2008) Recurrent lower-limb varicose veins: effect of directcontrast-enhanced three-dimensional MR venographic findingson diagnostic thinking and therapeutic decisions. Radiology247:887–895

26. Pineda C, Amezcua-Guerra LM, Solano C et al (2011) Joint andtendon subclinical involvement suggestive of gouty arthritis inasymptomatic hyperuricemia: an ultrasound controlled study.Arthritis Res Ther 13:R4

Eur Radiol