Diagnostic radiation in pregnancy: perception versus true risks.

Abstract

Significant numbers of therapeutic abortions are performed for radiation-exposed pregnant women because of concerns about the teratogenic risk. However, available data suggest that current diagnostic radiation procedures are not teratogenic.

Full text

Page 1

DRUGS IN PREGNANCY
MOTHERISK ROUNDS
Diagnostic Radiation in Pregnancy:
Perception Versus True Risks
Raanan Cohen-Kerem, MD,1Irena Nulman, MD,1Maria Abramow-Newerly, BSc,1
Damien Medina, BSc,1Ronnen Maze1, BSc, Robert L. Brent, MD, PhD,2Gideon Koren, MD1
1Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics, the Hospital for Sick Children, Toronto ON
2Department of Pediatrics, Jefferson Medical College, Philadelphia PA, and DuPont Hospital for Children, Wilmington DE
Abstract
Significant numbers of therapeutic abortions are performed for
radiation-exposed pregnant women because of concerns about the
teratogenic risk. However, available data suggest that current
diagnostic radiation procedures are not teratogenic.
Résumé
Un nombre imposant d’avortements thérapeutiques sont pratiqués
chez des femmes enceintes soumises à une radiothérapie en
raison du risque de tératogénicité. Toutefois, les données
disponibles suggèrent que les procédures diagnostiques courantes
en matière de radiothérapie ne sont pas tératogéniques.
J Obstet Gynaecol Can 2005;28(1):43–48
CASE PRESENTATION
T
presentation.
he following is not a single case counselled by
Motherisk, but a composite that illustrates a typical
A 26-year-old woman, unaware of her pregnancy of seven
weeks’ gestation, had an upper gastrointestinal (GI) X-ray
series to investigateepigastricpain. She had beenon an oral
contraceptiveformorethantwoyears.Afterpregnancywas
diagnosed, she was told by theradiologist she should termi-
nate the pregnancy because of a high likelihood of malfor-
mations. The obstetrician who was asked to perform the
pregnancy termination indicated to the patient that this
diagnostic procedure does not appear to cause malforma-
tions. Unconvinced, she asked to go ahead with the termi-
nationbutagreedtobecounselledbyMotherisk.Motherisk
calculatedthattheaveragefetalradiationexposurecould be
up to 1.1 mGy.1The radiologist in this case estimated it to
be 0.9 mGy (1 Gy = 100 rads).
DISCUSSION
It has been estimated that approximately 30 000 fertile
women were exposed to an abdominal radiographic diag-
nostic procedure during the 1980s in the United States.2
The Health Physics Society’s website, Ask the Expert
(ATE)receives600 to700 consultationseachyearconcern-
ing radiation exposure during pregnancy, obviously only a
small sample of the number of occurrences. Newer radio-
logical techniques such as computerized tomography (CT)
and therapeutic uses of radiation and radionuclides can
expose the fetus to greater than 50 mGy. Handbook 54 of
the National Council on Radiation Protection (NCRP)
established the 50 mGy level in 1977 for two reasons: first,
that most diagnostic radiological procedures do not exceed
50 mGy, and second, that the threshold for birth defects
during the most sensitive period of embryonic develop-
ment is > 200 mGy.3,4Furthermore, protraction and frac-
tionation of radiation decreases the radiation’s teratogenic
potential,5,6sothatlengthyprocedures,multipleprocedures
separated by hours or days, or use of radionuclides have a
lower reproductive risk than an acute exposure.4
The Motherisk Study
A cohort of pregnant Canadian women who contacted the
Motheriskprogrambetween1999and2002inquiringabout
exposuretoaradiographicdiagnosticprocedureunderwent
prospective follow-up.7They were exposed to various
radiographic diagnostic procedures that involved ionizing
radiation.Excludedfromthestudywerewomenexposedto
radiation therapy for conditions such as cancer or for
JANUARY JOGC JANVIER 2006 ?
43
DRUGS IN PREGNANCY
Key Words: Pregnancy outcome, ionizing radiation, imaging,
induced abortion, abnormalities

Page 2

thyroid ablation therapy, as well as women simultaneously
exposedtoknownteratogenics(e.g.,anticonvulsants,heavy
use of alcohol or tobacco, or retinoids).
The primary outcome measure of the study was the
frequency of major birth defects, defined as structural
abnormalities requiring surgical or cosmetic correction or
functional abnormalities with major impact on a child’s life
quality.8Details of radiation exposure included the type,
location, gestational age at the time of the procedure, indi-
cation, number of sessions, and the use of protective
shielding.
Data on radiation dose were taken from Health Canada
guidelines1where average values of fetal dose are presented
for different types of procedures (Table 1). Each value
assumes the actual amount of radiation, taking into account
an average number of exposures.
Data from a control group were retrieved from the
Motherisk database and compared with data from the
selected cohort. The women in the control group were not
exposed to a radiographic diagnostic procedure, cancer
therapy, thyroid ablation, or any other teratogen during
pregnancy and were queried by Motherisk on exposure to
non-teratogenic medications. This group of women was
matchedwiththeradiation-exposedgroupformaternalage,
gravidity, and parity.
Characteristics of pregnant women who were exposed to a
radiographic diagnostic procedure and of their controls are
shown in Table 2. We completed follow-up on 198 women
whowereexposedtoradiographicdiagnosticproceduresin
their pregnancy. Most (77.2%) of the diagnostic imaging
procedures were performed in the first trimester of
pregnancy.
Radiation data
The fetal dose (average ± standard deviation) following
exposure to a radiographic diagnostic procedure was 2.3 ±
0.59 mGy (range 0.01–68.0 mGy). The average gestational
age of exposure was 9.1 ± 0.7 weeks (range 1–38 weeks).
The average number of views per pregnancy was 2.7 ± 0.2
DRUGS IN PREGNANCY
44
? JANUARY JOGC JANVIER 2006
Table 1. Average fetal dose from X-ray procedures (from Health Canada)1
Diagnostic study type
Average dose (mGy) Diagnostic study type Average dose (mGy)
Dental
< 0.01* Barium meal (upper GI fluoroscopy) 1.1
Chest
< 0.01Barium enema (fluoroscopy) 6.8
Mammography
< 0.05* Head CT< 0.005
Pelvis
1.1Chest CT0.06
Abdomen
1.4Lumbar spine CT2.4
Lumbar spine
1.7 Abdominal CT8.0
Natural background radiation
(entire pregnancy)
0.5*Pelvis CT 25
*Estimates made by Health Canada.
CT: Computerized tomography.
Table 2. Characteristics of pregnant women who were exposed to a radiographic diagnostic
procedure and their controls
Variable
Radiation (n = 198)
Control (n = 198)
P
Mean maternal age (years) ± SD
32.1 ± 5.2 32.0 ± 4.80.81
Gravidity (mean ± SD)
2.57 ± 1.7 2.36 ± 1.80.72
Parity (mean ± SD)
1.97 ± 1.3 1.96 ± 0.80.35
Weight gain (mean ± SD)
14.71 ± 8.615.26 ± 9.8 0.64
Delivery method
Vaginal 125 (71.8%)134 (69.8%)0.75
Caesarean section33 (19%) 33 (17.2%)0.76
Vaginal w/assist16 (9.2%)25 (13%)0.32
SD: Standard deviation.

Page 3

(range 1–20 views), and 89.8% of the views were obtained
in one session. The radiation data are shown in Table 3.
No direct exposure to the embryo or fetus occurred in 149
of the patients out of the 198 who sought counselling.
However, even dental X-rays expose the embryo to very
low doses from scatter radiation.
Pregnancy outcome
The reported major anomalies in the radiation and control
groups are listed in Table 4. All newborns with major birth
defects were exposed to a dose of less than 0.1 mGy.
Elevenwomenintheradiation-exposedgroupchosetoter-
minate pregnancy compared with none in the control
group. Six of thosewomenstated thattheyterminatedtheir
pregnancy because of radiation (P = 0.04). These women
were exposed in the first trimester to a radiographic diag-
nosticprocedureinvolvingradiationlevelof0.01–5.6mGy.
None of them had an underlying condition that indicated a
termination was medically appropriate. The outcomes of
pregnanciesintheexposedandcontrolgroupsareshownin
Table 5. The rate of major anomalies among offspring of
women who were exposed to radiation in the first trimester
was 2.6% versus 3% in the control group.
A subgroup analysis of patients exposed to a relatively low
dose of radiation (? 1.0 mGy) versus a higher dose
(? 1.1 mGy) was performed (Table 6). The reason for this
cut-off point was apparent dose exposure distribution. The
high-dose group had significantly more miscarriages
(18.6%) than the low-dose group (2.6%). However, logistic
regression revealed that the only parameter that was signifi-
cantly associated with miscarriage rate was the mothers’
gestational age of contact with Motherisk. Mothers in this
subgroup who were exposed to a higher dose contacted
Motherisk earlier in pregnancy than the low-dose group,
suggesting that the high-dose group were observed over a
longer period of time with more potential for miscarriage.
No anomaly was observed in the high-dose group. The rate
oflivebirthswassignificantlyhigherinthelow-dosegroup,
probably because of the high rate of miscarriages.
Exposuretoaradiographicdiagnosticprocedureduringthe
first trimester of pregnancy led six out of 149 women (4%)
to choose to terminate an otherwise wanted pregnancy.
Diagnostic Radiation in Pregnancy: Perception Versus True Risks
JANUARY JOGC JANVIER 2006 ?
45
Table 3. Radiation data
Number (% within group)
Diagnostic study typeDental
100 (50.8)
Chest
31 (15.7)
Mammography
8 (4.0)
Upper GI fluoroscopy
6 (3.0)
Barium enema
2 (1.0)
Abdomen
7 (3.5)
Lumbar spine
15 (7.6)
IVP
1 (0.5)
Pelvis
1 (0.5)
Head and neck CT
10 (5.1)
Extremities CT
11 (5.6)
Abdominal CT
4 (2.0)
Lumbar spine CT
1 (0.5)
Non-specified
1 (0.5)
Indication (n = 196)Pain
98 (50.0)
Routine
86 (43.9)
Infection
8 (4.1)
Combination
4(2.0)
Multiple X-ray views Yes
120 (60.6)
No
78 (39.4)
Shielding (n = 188)Yes
165 (87.8)
No
23 (12.2)
CT: Computerized tomography.

Page 4

Those women were reassured by Motherisk counsellors
during positively framed advice that no harm was expected
to their fetus following such a low dose of radiation. More-
over, in this cohort, the rate of major anomalies was within
the anticipated range of 1% to 3% and similar to the rate of
major anomalies in the matched control cohort. Other out-
come measures of this study were also similar between the
two groups. Subgroup analysis of the radiation-exposed
group did not show any adverse effect following a higher
radiation dose.
To date, guidelines regarding radiation exposure in preg-
nancy rely mostly on animal studies and extrapolation of
dataretrievedfromyearsofresearchonsurvivorsofatomic
bombdetonations.Becauseof thefearof radiation,women
tend to misinterpret the risk of teratogenicity and overesti-
mate an effect on the fetus even when reassuring
evidence-based data are provided for them.9,10Animal
studies dealing with the reproductive effects of ionizing
radiation are significantly different from teratology studies
in animals using drugs and chemicals.3,4,11–16Drugs and
chemicals have to be absorbed, possibly metabolized, and
transported via the placenta to the implantation site,
embryo, or fetus. There can be species differences in mam-
mals for each of these functions. Ionizing radiation, on the
other hand, has its effect directly on the fetus. There are no
concerns about differences in metabolism, absorption, or
placentaltransferbetweenananimalmodel and thehuman.
Many studies indicate that deleterious effects including
birth defects occur only if the fetus is irradiated.11,17–22
The United States National Council on Radiation Protec-
tion (NCRP) states that a radiation exposure of less than
50mGyhasnomeasurableincreasedreproductiveeffecton
the fetus.4,11,23However, clinical studies examining the
actual effect of low-dose radiation associated with
DRUGS IN PREGNANCY
46
? JANUARY JOGC JANVIER 2006
Table 4. Major congenital anomalies in the radiation-exposed and control groups
GroupType of anomalyMagnitude and timing of exposureProcedure
Radiation*Trisomy 18 0.04 mGy week 1Extremities X-ray
Radiation Extra toe 0.01 mGy week 3 Head CT
RadiationPyloric stenosis 0.06 mGy week 4Extremities X-ray
Radiation Down syndrome 0.06 mGy week 14Dental X-ray
Control Cleft palate––
Control Club foot––
ControlInguinal hernia surgically corrected––
Controlventricular septal defect––
ControlUndescended testes surgically corrected––
ControlPyloric stenosis––
*Major anomalies that were observed in first trimester exposure to radiation (n = 149).
CT: Computerized tomography.
Table 5. Pregnancy outcomes of women exposed to radiation versus controls
Variable
Radiation
Controls
P
Live births
173/198192/1980.001
Miscarriages
12/1986/1980.23
Elective termination
11/1980/1980.001
Elective termination due to radiation6/1980/1980.04
Stillbirth/Fetal death
2/1980/198 0.48
Major malformations*
4/1496/1980.79
Birth weight (g) ± SD
3498 ± 5663515 ± 6050.79
Prematurity
12/173 24/190 0.1
*Only those who were exposed in the first trimester.
SD: Standard deviation.

Page 5

radiological diagnostic procedures were rarely performed.
Kinlen and Acheson who studied this issue did not find an
increased rate of malformations or spontaneous abor-
tions.24In contrast, Jacobsen and Mellemgaard reported
some eye anomalies in children exposed in utero to
low-dose radiation.25Bohnen and colleagues showed an
association of decreased head circumference with low-dose
radiation (< 3 mGy) in the second and third trimesters.26
There are significant differences in radiation dose depend-
ing on techniques and equipment. For example, fluoros-
copy time (usually recorded by the technician) may vary, as
may the use of digital radiography equipment. These vari-
ables were not available to us.
The tendency to misinterpret the risk of drug teratogenicity
and overestimatea drug effect on the fetus even when reas-
suring data are available is also shown to be the case with
respect to very mild alcohol drinking10and cocaine use.27
The misperception of teratogenic risk was investigated by
Motherisk with respect to diagnostic radiation in 1991.28
Regardless of reassuring information and a position state-
ment from an authoritative agency in the US,23the fear of
harm from low-dose radiation exposure in pregnancy still
exists.
A large number of animal studies indicate that the repro-
ductive effects of teratogenesis, growth retardation, embry-
onicdeath,andstillbirtharedeterministiceffects(threshold
effects)andthatthethresholdfortheseeffectsare200mGy
at the most sensitive period of mammalian embryonic
development (22nd–40th day postconception in the
human). The threshold for these effects is even higher
beyond the period of early organogenesis. This means that
the 50 mGy limit set by the NCRP is very conservative and
that exposures between 50 mGy and 100 mGy, for exam-
ple, are also without measurable reproductive risks.3,29
Otherconfoundersmightcomeintoplayandaffecttherate
ofmajormalformationsandthemother’sdecisiontotermi-
nate pregnancy. Other possible effects of radiation, includ-
ingminormalformations
development, were not assessed in this study.
and effectson neuro-
Lack of information among the public regarding the safety
of low-dose radiation exposure may lead to misperception
of the risk and to termination of an otherwise wanted preg-
nancy. Motherisk patients were reassured during counsel-
ling. It is conceivable that the rates of pregnancy termina-
tion are higher among pregnantwomen who do not receive
appropriate counselling. Therefore, an attempt should be
made to increase awareness of the safety associated with
low-dose radiation, and thus to prevent termination of
otherwise wanted pregnancies.
REFERENCES
1. Health Canada. Diagnostic X-rays and pregnancy. Available at:
www.hc-sc.gc.ca/iyh-vsv /med/xray-radiographie_e.html. Accessed
November 21, 2005.
2. Mossman KL. Medical radiodiagnosis and pregnancy: evaluation of options
when pregnancy status is uncertain. Health Phys 1985;48:297–301.
3. Brent RL, Moseley RD Jr, Russell LB, Wilson J, Casarett GW, Taylor LS.
Medical radiation exposure of pregnant and potentially pregnant women.
NCRP Report No. 54, Ad Hoc Committee on Exposure of Potentially
Pregnant Women in Diagnostic Radiology of the National Council on
Radiation Protection and Measurements. Washington, DC: 1977.
4. Brent RL. Utilization of developmental basic science principles in the
evaluation of reproductive risks from pre- and postconception
environmental radiation exposures. Teratology 1999;59:182–204.
5. Brent RL. The response of the 9 and one-half-day-old-rat embryo to
variations in exposure rate of 150 R x-irradiation. Radiat Res
1971;45:127–36.
6. Brizzee KR, Brannon RB. Cell recovery in fetal brain after ionizing
radiation. Int J Radiat Biol Relat Stud Phys Chem Med 1972;21:375–88.
7. Cohen Kerem R, Nulman I, Brent R, Koren G. A prospective controlled
study of diagnostic radiation in pregnancy. Birth Defects Part I: 2003;67.
8. Holmes LB. Need for inclusion and exclusion criteria for the structural
abnormalities recorded in children born from exposed pregnancies.
Teratology 1999;59:1–2.
Diagnostic Radiation in Pregnancy: Perception Versus True Risks
JANUARY JOGC JANVIER 2006 ?
47
Table 6. Subgroup analysis of radiation-exposed women with low-dose versus high-dose
diagnostic radiation exposure
Variable
? 1.0mGy (n = 151)
? 1.1mGy (n = 43)
P
Live births
137 (90.7%) 32 (74.4%)0.01
Elective termination
5 (3.3%)00.51
Elective termination due to radiation
4 (2.6%)2 (4.6%)0.96
Miscarriages
4 (2.6%)8 (18.6%)
? 0.001*
Stillbirths
1 (0.7%)1 (2.3%)0.92
Major anomalies†
4 (2.6%)0 0.64
* Gestational age at first maternal contact was significantly earlier in higher exposed group.
† Major anomalies that were observed in first trimester exposure to radiation.