12/21/2016

1

A comprehensive CT radiation dose reduction and protocol standardization program in a complex tertiary hospital system

Prabhakar Rajiah, Jeffrey Guild, Travis Browning, Seth Toomay, Viswanathan Venkataraman, Orhan K. Oz, Anthony Whittemore, Lakshmi Ananthakrishnan, Avneesh Chhabra, Sanjeeva Kalva, Suhny Abbara

UT Southwestern Medical Center, Dallas, Texas, United States

Disclosures• Prabhakar Rajiah - Institutional Research Grant, Koninklijke Philips NV Speaker, Koninklijke

Philips NV

• Travis Browning - Advisor, McKesson Corporation

• Avneesh Chhabra - Consultant, ICON plc Author with royalties, Wolters Kluwer nv Author with royalties, Jaypee Brothers Medical Publishers Ltd

• Sanjeeva Kalva - Consultant, CeloNova Biosciences, Inc

• Suhny Abbara -Author, Reed Elsevier Editor, Reed Elsevier Institutional research agreement, Koninklijke Philips NV Institutional research agreement, Siemens AG

12/21/2016

2

Background• Radiation doses from CT- Leading cause of non-background radiation exposure

• Radiation dose should be maintained As Low As Reasonably Achievable

• Revised requirements of The Joint Commission

- Radiation dose of every CT exam should be recorded

- Investigation of cases where radiation dose exceeds reference levels

• Establishing a real-world radiation dose reduction program is challenging

Purpose• Establishing a CT radiation dose reduction program in a large complex

health system is not widely reported

• We describe a comprehensive radiation dose reduction and protocol homogenization program in a large complex system using

- Iterative process of lowest common denominator using phantom and clinical test cases

- Novel web-based information distribution system

12/21/2016

3

The problemReview of the CT protocols and radiation doses identified the following problems

Extensive heterogeneity of CT radiation doses

No established parameters

Lack of training and reliable dissemination

No robust radiation dose tracking process

Lack of uniform data storage

Extensive heterogeneity of CT protocols

Challenges

3 health systems

4 large hospitals

5 outpatient centers

Several remote locations

3 PACS systems

4 major manufacturers

21 CT systems +

9 hybrid CT

Brand new to > 10-year-old

equipment

6-slice hybrid CT to 320 slice

scanners

12/21/2016

4

Problems• Several, heterogeneous protocols

• Each location had local imaging protocols governed by local administrative body

• Different machines- multiple vendors; different software, hardware, weight limitations, etc

• Heterogeneous radiation doses for same protocols at different sites

• Variable maintenance of protocols at sites, paper or electronic

• Difficult to obtain protocols from another location without a phone call/email

CT Radiation Task Force

• A CT radiation task force was created

• Weekly CT operations meeting was established

• Coordination of stakeholders- Physicians, physicists, technologists and hospital administrators

12/21/2016

5

CT Radiation Task Force- Aims

• To optimize patient radiation dose across scanners

• To standardize and homogenize the scan protocols and their names

• To maintain or improve image quality

• To establish mechanisms to continually track the dose

• To establish a reliable training and dissemination processes

• To make protocols readily available

• To ensure adherence to imaging protocols

• All existing protocols were reviewed on a divisional basis

• 3 studies selected from the PACS for each protocol from each CT scanner

• Evaluated by a radiologist on a five point Likert scale for image quality (1-5)

• Physicist quantified the radiation dose (DLP) and image quality (CNR) using anthropomorphic phantoms

• If the image quality was maintained at the lowest dose, that protocol was programed in the scanner.

The Optimization Process

12/21/2016

6

• If the image quality was not maintained at the lowest dose, the protocol was optimized based on CNR metrics and previously optimized CT protocols

- Acquisition parameters

- Reconstruction kernels

- Iterative reconstruction levels

• The review process was repeated with the proposed protocol.

• The lowest dose which did not compromise image quality was selected

• Redundant protocols were also eliminated, merging protocols which could give similar results- Eg- Bony pelvis from CT abdomen, pelvis; Lumbar spine from CT abdomen

The Optimization Process

The Optimization Process

Tech/adminCollects 3 studies from

PACS from each scanner for each

protocol

• Radiologist- Image quality scored on Leikert scale (1-5)

• Physicist- DLP & CNR measured on anthropomorphometricphantoms

Uncompromised image quality

(3/4) at lowest dose

YES

NOTech/physics/radiologistAdjust protocol on select scannersAdjust dose based on CNRMatch other low dose scanners

Tech/physics/radiologistObtain 3-5 new studiesNew patient scanned under supervisionRecon from other CT of the same anatomy

Physics/tech programs the protocol on

the CT scanner

12/21/2016

7

Radiation dose storage• Radiation doses were tracked consistently and stored in a database

• Radimetrics was utilized to store, retrieve and analyze the radiation dose programs

Protocol homogenization

• Imaging protocol was defined across subspecialty radiologist teams

• Each protocol was reviewed by subspecialty radiologist groups

• Overlapping protocols combined

• Duplicated/outdated protocols eliminated

• New protocols developed if there was a clinical requirement

• Imaging protocols reflect specific modalities and available equipment

• After consensus, pdf document of protocol created

12/21/2016

8

• Database created linking clinical imaging protocols to scanner/machine specific acquisition parameters

• Implemented in Microsoft Access

- Contrast administration

- Imaging phases

- Radiation dose

- Electronic orders

Protocol homogenizationCLINICAL IMAGING PROTOCOL

LINK BETWEEN CLINICAL AND MACHINE PROTOCOLS

• Protocol library was made available to everybody regardless of location

• Sharepoint site or “Radpoint’

• “Source of truth” for all protocols

• Protocol pdf documents stored

Protocol storage

12/21/2016

9

Protocol change process• A “Protocol Czar” was tasked to manage protocol change process

• Standard process established to manage protocol change process

Protocol change request

Send to CT Ops Chair and Division

Chief/delegate

LIMITED

DENIED

NEW/MAJOR CHANGE

Rapid minor protocol change

workflow

New/major protocol change

workflow

Returned to requestor with

reasons

For further details, please visit QS131-ED-THA2

Results- Protocol homogenization

• Project start- May 2014

• Protocols reviewed- >2000 individual scanner protocols

• Optimization proceeded from division to division

• First division (Cardiothoracic) was completed in two months

• Entire optimization process completed in 9 months

• Total number of types of CT protocols decreased from 222 to 136

12/21/2016

10

Results- Radiation dose

• Significant improvements in radiation doses

• Improvements ranging from 23-58 % dose reduction

Results- Radiation dose

ProtocolGeometric Mean

DLP, mGy-cmImprove

ment (%)Before After

Routine abdomen,pelvis 900 690 23Renal colic 900 660 27

Routine chest 710 400 44Pulmonary embolism 960 600 38

Mesenteric CTA 1845 975 47Bony Pelvis 1170 490 58

L-Spine 1850 970 48

Boxplots showing the radiation doses of average sized patient in 7 protocols, before and after optimization

Evaluation of radiation doses for each optimized protocol over a period of one year before and one year afteimplementation was done using geometric mean to measure differences in dose

12/21/2016

11

ResultsAbdomen and Pelvis, portal venous phase

0

100

200

300

400

500

600

700

800

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Nu

mbe

r of

Exa

min

atio

ns

DLP, mGy-cm

Portal Venous Abd/Pelvis

Before

After 0

10

20

30

40

50

60

70

80

90

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mm

Portal Venous Abd/Pelvis Before

OPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0

10

20

30

40

50

60

70

200 250 300 350 400 450Av

erag

e CT

DIVo

l, m

GyPatient Size, WED, mm

Portal Venous Abd/Pelvis AfterOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

Scatter plot distribution of radiation dose, before (green) and after (blue)

Renal colic

0

50

100

150

200

250

300

350

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Nu

mbe

r of

Exa

min

atio

ns

DLP, mGy-cm

Renal Colic

Before

After

0102030405060708090

100

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mm

Renal Colic BeforeOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0

10

20

30

40

50

60

70

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mm

Renal Colic AfterOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

Scatter plot distribution of radiation dose, before (green) and after (blue)

12/21/2016

12

Routine Chest

0

100

200

300

400

500

600

700

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Nu

mbe

r of

Exa

min

atio

ns

DLP, mGy-cm

Routine Chest w/o

Before

After 0

10

20

30

40

50

60

70

80

200 250 300 350 400 450

AVer

age

CTDI

Vol,

mGy

Patient Size, WED, mm

Routine Chest w/o BeforeOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0

10

20

30

40

50

60

70

200 250 300 350 400 450Av

erag

e CT

DIVo

l, m

GyPatient Size, WED, mm

Routine Chest w/o AfterOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

Scatter plot distribution of radiation dose, before (green) and after (blue)

Pulmonary embolism

0

20

40

60

80

100

120

140

160

180

200

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Nu

mbe

r of

Exa

min

atio

ns

DLP, mGy-cm

Pulmonary Embolism

Before

After

0

10

20

30

40

50

60

70

80

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mGy

Pulmonary Embolism BeforeOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0

10

20

30

40

50

60

70

200 250 300 350 400 450

Av

era

ge

CT

DIV

ol,

mG

y

Patient Size, WED, mm

Pulmonary Embolism AfterOPC CT1 OPC CT2 OPC CT3

Scatter plot distribution of radiation dose before (green) and after (blue)

12/21/2016

13

CTA mesenteric

0

1

2

3

4

5

6

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Nu

mbe

r of

Exa

min

atio

ns

DLP, mGy-cm

CTA Mesenteric

Before

After0

10

20

30

40

50

60

70

80

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mm

CTA Mesenteric BeforeOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0

10

20

30

40

50

60

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mm

CTA Mesenteric AfterOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

Scatter plot distribution of radiation dose before (green) and after (blue)

Bony pelvis

0

5

10

15

20

25

30

35

40

45

50

0 500 1000 1500 2000 2500 3000 3500 4000

Nu

mbe

r of

Exa

min

atio

ns

DLP, mGy-cm

Bony Pelvis

before

after

0

20

40

60

80

100

200 250 300 350 400 450Av

era

ge

CT

DIV

ol,

mG

y

Patient Size, WED, mm

CT Bony Pelvis Before

OPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0

10

20

30

40

50

60

70

200 250 300 350 400 450

Av

era

ge

CT

DIV

ol,

mG

y

Patient Size, WED, mm

CT Bony Pelvis After

OPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

Scatter plot distribution of radiation dose before (green) and after (blue)

12/21/2016

14

Lumbar spine

0

2

4

6

8

10

12

14

16

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

Num

ber o

f Exa

min

atio

ns

DLP, mGy-cm

L-Spine

Before

After

0

20

40

60

80

100

120

140

160

180

200 250 300 350 400 450

Aver

age

CTDI

Vol,

mGy

Patient Size, WED, mGy

L-Spine BeforeOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

0102030405060708090

100

200 250 300 350 400 450Av

erag

e Pa

tient

CTD

IVol

, mGy

Patient Size, WED, mm

L-Spine AfterOPC CT1 OPC CT2 OPC CT3 OPC CT4 PMH RM5

Scatter plot distribution of radiation dose, before (blue) and after (orange)

• Protocol access through Radpoint

• Number of CT protocol page visits from July 2015 till date - 21,037

• Average/month - 1315

Results- Protocol Usage

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Jul-15 Aug-15 Sep-15 Oct-15 Nov-15 Dec-15 Jan-16 Feb-16 Mar-16 Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16 Oct-16

Radpoint Usage statistics for CT protocols

12/21/2016

15

Discussion

• It is possible to establish a robust radiation dose reduction and protocol homogenization program in a complex health system

• Requires participation of all stakeholders, including radiologists, technologists, physicists and hospital administrators

• Each protocol can be optimized by an iterative process that uses both clinical and phantom data

Discussion• Training of technologists is an important component of the program and

we achieved this by incorporating this as a part of protocol change process

• Imaging protocol homogenization requires subspecialty operating committees and specific individuals to manage the process

• Dissemination of protocols was made easy by a novel web-based information distribution system

• Periodic protocol and dose review ensures consistent maintenance of quality

12/21/2016

16

ProblemExtensive heterogeneity of CT radiation doses

No established parameters

Lack of training and reliable dissemination

No robust radiation dose tracking process

Lack of uniform data storage

Extensive heterogeneity of CT protocols

Radiation doses optimized

CT protocols & names homogenized

Parameters established

Training pathway established as part of protocol change process

Radimetrics used for radiation dose tracking

Radiation doses- RadimetricsCT protocols- Protocol library, RADPoint

Solution

Conclusion• We successfully managed the complex process of homogenizing CT

protocols and optimizing radiation doses without compromising image quality

• Key elements are-

- Establishment of CT Dose task force and CT operations committee

- Iterative process of protocol optimization using phantom and clinical tests

- Novel web-based information distribution system for protocols

- Establishment of a protocol change process

- Establishment of radiation dose tracking process