Radiation safety and risk management Estimating Cancer Risk Attributable to Computed Tomography Coronary Angiography Koos Geleijns Radiology department Leiden University Medical Center Leiden, The Netherlands Estimating Cancer Risk Attributable to Computed Tomography Coronary Angiography Spectacular technical developments in computed tomography (CT), like 64-slice CT, dual-source CT and 320-slice volume CT, led to fast growing application of CT. At the same time, concerns are expressed about radiation exposure and associated radiation risks of CT examinations. Estimating Cancer Risk Attributable to Computed Tomography Coronary Angiography What do we need to know? • Output of the scanner (CTDI, DLP) • Organ dose (effective dose) • Radiation (late) risk as a function of age and gender (dependent on organ dose, age of exposure, gender) • Competing risks (procedure and disease related acute and late risks, false positives, false negatives) • Decision model based on Disability Adjusted Life Expectancy (DALE) Dosimetry Radiation Risk Assessment Comprehensive Risk Assessment Justification: Balancing Risks and Benefits Estimating Cancer Risk Attributable to Computed Tomography Coronary Angiography
12
Embed
Estimating Cancer Risk Attributable to Computed ... - AAPM · PDF file1 x 5 4 x 1 16 x 0.5 64 x 0.5 320 x 0.5 1998 2001 2004 2008 Calculation of organ dose for the Toshiba Aquilion
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
qx: Conditionalprobablitythatanindividual who hassurvivedto start of theageintervalwill die in theageinterval.
How to assessradiation risk, e.g. yearslife lost
Life tables that represent age and gender related functionspertaining to mortality.
ax Fraction of the age interval lived by those in the cohort populationwho die in the interval.
Mx Age-specific death rate.qx Conditional probability that an individual who has survived to start
of the age interval will die in the age interval.px Conditional probability that an individual entering the age interval
will survive the age interval.lx Life table cohort population.dx Number of life table deaths in the age intervalLx Number of years lived during the age interval.Tx Cumulative number of years lived by the cohort population in the
age interval and all subsequent age intervals.ex Life expectancy at the beginning of the age interval.
Survival, exposureto 15 mSvat age40
0
20000
40000
60000
80000
100000
0 20 40 60 80 100 120
Age, years
Su
rviv
al,c
oh
ort
of
100
000
Males, no radiationFemales, no radiation
Males, with radiation (15mSv)
Females, with radiation (15 mSv)
In a cohort of 100000males61 radiationinduceddeaths,in a cohort of100 000females147radiationinduceddeaths.
Radiationinducedmortality is a late effect: reductionof life expectancy3 days (males)and9 days(females).
Probability of CAD in % asfunction of age,gender and type of chestpain
Hamm CW, Goldmann BU, Heeschen C, Kreymann G, Berger J, Meinertz T. Emergency room triage of patients with acute chest pain bymeans of rapid testing for cardiac troponin T or troponin I. N Engl J Med 1997; 337:1648-53.
Bayesiannetwork for probability of CAD basedon the mostimportant test resultsprior to imaging or intervention
Decision analysisis basedon thepremisethathumansare reasonablycapableoffr aming a decisionproblem, listing possibledecisionoptions,determiningrelevantfactors, and quantifyinguncertaintyandpreferences,but are ratherweak in combining thisinfor mation into arational decision.
A Bayesiannetwork, or beliefnetwork, showsconditionalprobability andcausalityrelationships betweenvariables.
Probability of CAD,male45 yearsold
35NegativePositiveNon anginal
32EquivocalNegativeTypical
16PositiveNegativeAtypical
15NegativeNegativeTypical
5EquivocalNegativeAtypical
4PositiveNegativeNon anginal
2NegativeNegativeAtypical
Probability CAD(%)
Stress-ECGTroponinChest pain
Probability of CAD, male45 yearsold
96PositivePositiveAtypical angina
98EquivocalPositiveTypical angina
99PositivePositiveTypical angina
96NegativePositiveTypical angina
88EquivocalPositiveAtypical angina
83PositivePositiveNon angina
74NegativePositiveAtypical angina
61PositiveNegativeTypical angina
58EquivocalPositiveNon anginal
Probability CAD (%)Stress-ECGTroponinChest pain
Disabili ty adjustedlife expectancies (LE) usedin themodel
0.160.611319.8Female65
0.130.511116.3Male65
0.301.172437.8Female45
0.271.042233.5Male45
Reduction ofLE due to
false positive(years)
Reduction ofDALE due tomissed CAD
(years)
LE fordiagnosed
CAD(years)
NormalLE
(years)
GenderAge
Disability-adjustedlife expectancies and yearsof life lost(YLL) basedonrandomisedtrials.MukherjeeD et al., Am HeartJ 2002;The PRISM-PLUS StudyInvestigators.N Engl J Med 1998;AndersonHV et al. J Am Coll Cardiol1995
Excessmortality from imaging
0.00450.0232Female65
0.00840.0192Male65
0.01080.0452Female45
0.00690.0387Male45
YLL due toCCTA
YLL due to ICAGenderAge
Metaanalysis: ROC64-sliceCT for detection of CAD
Influencediagramto computeoptimalpolicy
YLL = yearsof life lostDALE = disability-adjustedlife expectancy
End point DALE isoptimisedasa functionof- age- genderand- probability of CAD afterclinical evaluation
Yielding- theoptimal imagingpolicy- way of handlinguncertainscans and- diagnosisfor eachof thesepatient groups
Rangeof clinical probability of CAD whereCTCAis optimal as afunction of gender andage
5062Female65
5562Male65
5062Female45
5062Male45
Upperbound (%)
Uncertainnegative
below (%)
Lowerbound (%)
GenderAge
Conclusion1
• A preliminary study on the efficacy of CCTA for themost important clinical CTCA indications has been
performed for males and females for the ages of 45and 65.
• The model is based on meta-analysis results of
CCTA, average radiation dose and radiation risks.
• Outcomes in years of life lost (YLL) and disability-adjusted life expectancy (DALE) were estimated.
• The study integrates cancer risk attributable to
CCTA in a medical decision making model
Conclusion2
• The current practice of the use of 40- and 64-slice CT
coronary angiography in suspected CAD seems to be justified
in patients with a low to intermediate probability (i.e. 2-50%
probability for CAD) after clinical evaluation.
• Age and gender have little effect on the range of clinical
probability where CTCA is optimal *)
• For very low probabilities (i.e. below 6% probability for CAD),
it seems most effective to assume non-diagnostic scans as
negative results.
* ) Note:ageandgenderhavea major impact on clinical probability
Thank you for your attention …
Ying Li e O
Alex Meijer
Job Kievit
JaapSont
Albert de Roos
Lucia Krof tSafety and Efficacy in Computed Tomography: a broad perspectiveEC-EURATOM 6 Framework Programme call 2003 Project no.FP6/002388.2005– 2007.