Radiation Doses from the ACR CT Accreditation Program: New Diagnostic Reference Values and Pass/Fail Limits Cynthia McCollough, Ph.D. Michael McNitt-Gray, Ph.D. J. Thomas Payne, Ph.D. Tom Ruckdeschel, M.S. Doug Pfieffer, M.S. Dianna Cody, Ph.D.

Robert Zeeman, M.D. Vince Herlihy, M.D. Theresa Branham, RT(R) (CT) (QM) Krista Bush, RT(R) (M) (CT) MBA Lavonne Robbins B.S., C.N.M.T. Mythreyi Bhargavan, Ph.D.

• Have been shown to lower average dose in other modalities and/or other countries • Represent the upper third or quartile of doses sampled from actual practice data • Do not represent ideal or suggested doses • Identify when dose is unusually high Gray JE, et al . Reference values for diagnostic radiology: application and impact. Radi ology 2005; 235: 354-358. Tsapaki V, et a l. Dose reduction in CT while mainta ining diagnostic confidence: Diagnostic reference levels at routine head, chest, and abdo minal CT--IAEA-coordinated resea rch project. Radiology 2006; 240:828 -834. Hart D, et al. Doses to patients from medical x -ray examinations in the UK —1995 review. In: Chilton: NRPB -R289, 1996. Shrimpton PC, et al. Doses from Computed Tomog raphy (CT) Examination s in the UK - 2003 Review. In: National Radiological Protection Board, Oxon: NRPB -W67, 2005.

Phantom size affects CTDI values

• ACR CT Reference Doses – Adult Head – Adult Abdomen – Pediatric (5 yr old) Abdomen

Reference Doses

Same kVp, kVp, collimation, pitch

60 mGy* 35 mGy* 25 mGy

Body (32 cm)

Head (16 cm)

280 mAs

116 mAs

÷ 2.4 =

20

• Currently no pass/fail dose criteria

16.6

– Justification or corrective action requested • New CTDI data and images • Low contrast resolution images • Statement that clinical image quality is acceptable

20

10

20

*From European Commission EUR 16262 (2000) European Guidelines on Quality Criteria for Computed Tomography

CTDIw = 16.6

20

16.6

16.6

16.6

16.6 CTDIw = 16.6

Materials & Methods • Use of smaller phantom and lower reference value implies that a reduction in tube output by a factor of to 3 - 4 is expected for a 5 y.o. abdomen exam • CTDIvol values displayed on the scanner console use large CTDI phantom – Need to address with appropriate standards, professional and manufacturer organizations, as well as clearly educate users

Site Dose Measurements

Excel® “Dose Calculator” spreadsheet Dose Calculator spreadsheet available for exposure or air kerma meters

• CTDIw (mGy) for – Routine head (cerebrum/brain) – Adult abdomen – Pediatric abdomen (5 y.o)

• CDTI phantom images filmed to verify correct technique

Center Measurement 1 (mR)

197

Measurement 2 (mR)

199

Measurement 3 (mR)

199

Average of above 3 measurements (mR)

2:9 198.3

Body CTDI at isocenter in phantom (mGy)

11.2

12 o'clock position Measurement 1 (mR)

401

Measurement 2 (mR)

422

Measurement 3 (mR)

401

Average of above 3 measurements (mR) Body CTDI at 12 o'clock position in phantom (mGy) CTDIw (mGy)

408.0 23.0 19.0

CTDIw = 2/3 CTDI100 (edge) + 1/3 CTDI100 (center)

Reviewer Validation • Adult Head – 16 cm CTDI phantom, in head holder CTDIw (mGy) 19.0 Clinical exam dose estimates (using measured CTDIw and site's Adult Abdomen Protocol from Table 1) CTDIvol (mGy)

=CTDIw*N*T/I

25.4

DLP (mGy-cm)

=CTDIvol*25

634.2

Eff Dose (mSv)

=DLP*0.015

9.5

• Pediatric abdomen (5 y.o.) – 16 cm CTDI phantom, on table

• Adult Body – 32 cm CTDI phantom, on table

Volume CTDI = CTDIw / pitch

• Axial scan mode • Correct detector configuration • Invalid data omitted from analysis

Data Analysis

Results I

• Average, standard deviation, and histogram determined

Mean ± standard deviation

– By exam (head, abdomen, pediatric abdomen) – By year (2002, 2003, 2004, and 2002-2004) – By CTDIw and CTDIvol

• Statistical significance of changes in average doses by year tested using a 2-tailed t-test • Percent of scanners above references dose determined – Current reference dose using CTDIw and CTDIvol – Proposed reference dose using CTDIvol

Shown for CTDIvol only

Adult Head

90

Adult Abdomen

30

80 25

66.7 59.1

57.8

60

CTDIvol (mGy)

CTDIvol (mGy)

70

54.6

50 40 30 20

P < 0.0001

10

18.4 17.0

15 10 NS

5

P < 0.0001

19.2

18.7

20

NS

P < 0.005

NS

0

0 2002 (n=117)

2003 (n=305)

2004 (n=208)

2002 (n=113)

2002-2004 (n=630)

2003 (n=290)

% > Current Reference (CTDIw)

Pediatric (5 y.o.) Abdomen

30

50

2004 (n=197)

2002-2004 (n=600)

% > Current Reference (CTDIvol)

Head

45

CTDIvol (mGy)

25 20

40 35

17.2

15.9

30

15.5 14.0

15

%

25 20

Pediatric Abdomen

10 15 P < 0.05

5

10

NS

P < 0.01

Abdomen

5

0 2002 (n=91)

2003 (n=224)

2004 (n=151)

2002-2004 (n=466)

0 2002

2003

2004

20022004

2002

2003

2004

20022004

2002

2003

2004

20022004

Conclusions I Results II

• Dose for three high-use exams have decreased significantly in the U.S. since 2002 – Adult head

• Percent above references dose • Establishing new reference doses

– Adult abdomen – Pediatric abdomen

– Maintained 5 mGy “step size”

• Sites are “dialing down” the dose for kids

• Mandatory dose limits

– About a factor of 3

• ACR CT Accreditation program has developed a valuable database to monitor dose trends and to establish new reference doses • ACR will switch to CTDIvol to include the effect of pitch

Pediatric Abdomen

28

Adult Abdomen

32 75%tile

26.6 26

30.6

75%tile

29.5

30

90%tile

25.6

29.5 90%tile

24

CTDIvol (mGy)

CTDIvol (mGy)

24.9 23.4

22 20.6

20.5

28

24

23.4 22.6

20.0 20

25.8

26

22.2

22

21.1

18.4 18

20 2002 (n=91)

2003 (n=224)

2004 (n=151)

2002-2004 (n=466)

2002 (n=113)

2003 (n=290)

2004 (n=197)

2002-2004 (n=600)

Adult Head

Adult Head

2002 Frequency 2002 Cumulative %

2003 Frequency 2003 Cumulative %

2004 Frequency 2004 Cumulative %

30

100% 90%

25

80% 70%

20

60% 15

50% 40%

10

Cumulative %

• 75%tile difficult to determine because initial reference values altered the practice distribution • Numerous sites felt the 60 mGy was not clinically acceptable • Multiple reports of sites increasing head dose after accreditation process completed

30% 20%

5

10% 0

0% 120

CTDIw (mGy)

Adult Head

2002 Frequency 2002 Cumulative %

2003 Frequency 2003 Cumulative %

Adult Head

2004 Frequency 2004 Cumulative %

2002 Frequency 2002 Cumulative %

2003 Frequency 2003 Cumulative %

2004 Frequency 2004 Cumulative %

30

100%

30

100% 90%

90% 25

25

80%

80%

74.60%

15

50% 40% 75

10

30% 20%

5

Cumulative %

60%

70%

20

60% 15

50% 40%

10

30% 20%

5

10%

10% 0

0% 120

0

0% 120

Cumulative %

70%

20

2002 Frequency 2002 Cumulative %

2003 Frequency 2003 Cumulative %

Adult Head

2004 Frequency 2004 Cumulative % 100%

2002 Frequency 2002 Cumulative %

2003 Frequency 2003 Cumulative %

2004 Frequency 2004 Cumulative % 100%

30

90%

90% 25

25

80%

77.57%

70% 60% 50%

15 65

40%

10

70% 60% 50%

15

40% 10

30% 20%

5

80%

20 Cumulative %

20

30% 20%

5

10%

10% 0

0

0% 120

CTDIw (mGy)

CTDIw (mGy)

2003 Frequency 2003 Cumulative %

2004 Frequency 2004 Cumulative %

30%

Multiple Scan Average Dose (comparable to CTDIw for contiguous sca

100%

30

2000-01 (n = 203)

60

25%

90% 25

1990

80% 76.17%

50% 40%

10

30%

Cumulative %

60% 15

(n = 249)

20%

70%

20

15%

28% > 60 15% > 70

10% 5%

20%

0% 120

0%

MSAD (mGy)

Data courtesy of Stanley Stern, Ph.D. - U.S. FDA 2001

>1 00

10% 0

010 10 -2 0 20 -3 0 30 -4 0 40 -5 0 50 -6 0 60 -7 0 70 -8 0 80 -9 0 90 -1 00

5

Cumulative %

Adult Head 30

ACR vs. NEXT data (Head) New UK Diagnostic Reference Levels

• NEXT 1990: 45.9 mGy ± 18.1 (n=249) Solid state detectors become standard, spiral CT and higher power tubes introduced, slice width begins to decrease

• NEXT 2000: 50.3 mGy ± 19.4 (n = 203) MDCT introduced in 1999, SDCT techniques used on MDCT, slice width continues to decrease

Brain Posterior Fossa

SSCT 65 55

MSCT 100 65

• ACR 2002: 66.8 mGy ± 23.2 (n = 127) • ACR 2003: 58.1 mGy ± 17.4 (n = 321) Shrimpton et al. Doses from CT Examinations in the UK – 2003 Review. NRPB-W67 (NRPB, Chilton) 2005.

• ACR 2004: 55.5 mGy ± 15.5 (n = 214)

Adult Head

110 99.0

CTDIvol (mGy)

100

75%tile

• New ACR CT Reference Doses

90%tile

– Adult Head – Adult Abdomen – Pediatric (5 yr old) Abdomen

90 82.2 80

81.3

76.8

70

74.0 64.3

63.9 60.0

60 50 2002 (n=117)

2003 (n=305)

2004 (n=208)

2002-2004 (n=630)

60 → 75 mGy 35 → 25 mGy 25 → 20 mGy

% > Current Reference (CTDIw)

% > Proposed Reference (CTDIvol)

Conclusions II

50

Head 45

• Have sufficient data for new U.S. reference doses

40

• Based on CTDIvol to include the effect of pitch

35

Pediatric Abdomen

30

%

25

Abdomen

20

• Reference doses (site given educational information) – Adult Head

60 → 75 mGy

– Adult Abdomen

35 → 25 mGy

– Pediatric (5 yr old) Abdomen

25 → 20 mGy

• Maximum allowable doses (site fails if these are exceeded)

15 10

– Adult Head

80 mGy

5

– Adult Abdomen

30 mGy

0

– Pediatric (5 yr old) Abdomen

25 mGy

2002

2003

2004

20022004

2002

2003

2004

20022004

2002

2003

2004

20022004

Additional Program Refinements • Simplified Film Page 1 – Elimination of need to convert spiral into axial – CT number of non-water rods only at 120 kVp – Fewer slice thickness scans

• New results database • WIP – – – –

New performance limits Accommodation of non-traditional CTDIvol measurements Quality control manual Electronic submissions

• Effective January 1, 2008

Thank you