What do we know about imaging in pregnancy?

Introduction

Many conditions in pregnancy require noninvasive paraclinical examinations. These kinds of tests should never replace a good clinical examination, a thorough physical exam, and physicians should know what radiological test is best suited for the patient. Our lack of understanding of safety for a certain investigation can be dangerous and can lead to an undesirable result, and the worst-case scenario, when harm to the patient can have consequences, especially if there is a delay in making the right diagnosis. This review is aimed at presenting the benefits and risks of various imaging techniques in pregnancy and to highlight the appropriate techniques to image women presenting with common medical symptoms.

Ionizing radiation

This imaging techniques include X-rays and computed tomography. These imaging techniques that expose the fetus to ionizing radiation may be required in pregnancy to help with decision making and for establishing a diagnosis. Sometimes, a fetus may be exposed to radiation by accident, especially in the first trimester. During pregnancy, there are data to suggest that a fetus is exposed to a background radiation of 1 mGy⁽¹⁾.

Any physician should know that electromagnetic waves can affect the human tissue in  two ways: through stochastic and deterministic effects⁽²⁾. The first effect – occurrence of malignancies – can appear at any radiation dose, as a result of cellular damage following a germline mutation⁽³⁾. We do not know what would be the cut-off value for this effect not to take place. There is a strong correlation between the stochastic effect and the increased risk of childhood malignancy⁽⁴⁾. The second effect involves the loss of cellular function because of cell death, and it results from very high radiation doses. Aneuploidies can occur as a result of such a destructive injury at cellular level⁽³⁾. As a consequence, major risks include fetal structural defects (skeletal, ophthalmic and genital tract anomalies), placental injuries (small babies with growth restriction) and brain lesions (microcephaly, intellectual or developmental disability)⁽¹⁾. In Table 1, we have summarized some of the potential consequences of imaging in pregnancy at different gestations and the radiation doses thresholds. 

Fetal radiation doses for common radiological investigations are presented in Table 2. The exact fetal exposure dose vary with maternal characteristics like gestational age, maternal Body Mass Index and parameters of the machine, such as image acquisition indices⁽⁵⁾.

Radiography

X-rays are used to investigate various symptoms and causes of a disease. Given the minimal risks to the fetus and potential delayed diagnosis, there should be no delay in performing such an investigation in pregnancy, the same as for non-pregnant women⁽⁵⁾.

Computed tomography (CT)

In a review of 5270 examinations, performed in more than 3000 women in a 10-year period, Lazarus et al.⁽⁶⁾ noted an annual increase of 25% in the use of CT in the time period studied. The indications for the use of CT in pregnancy are numerous, including assessing injuries following trauma⁽⁷⁾, diagnosing pulmonary embolism^(8,9), investigating gastrointestinal complications (acute abdomen, bowel obstruction)^(10,11) and malignancy^(12,13).

The ALARA (“as low as reasonably achievable”) principle is well known by most physicians when it comes to the use of radiation⁽¹⁴⁾. The International Commission on Radiological Protection⁽¹⁵⁾ recommends that “imaging radiation must be applied at levels as low as reasonably achievable, while the degree of medical benefit must counterbalance the well-managed levels of risk”. Shielding techniques in our times have significantly reduced the dose of ionizing radiation; for example, the fetal radiation dose received during mammography is in the order of 0.001-0.01 mGy⁽⁵⁾. The use of lead shielding can further reduce this risk by an additional 50%⁽¹³⁾.

During computed tomography, the fetal exposure to radiations depends on a number of factors: the anatomical region of interest, machine set-up, X-ray tube voltage, tube current and number of image acquisitions⁽¹⁶⁾. As a general rule, the fetal radiation dose is at a maximum when the fetus is captured directly in the X-ray beam – for example, when using abdominal-pelvic CT. For head and chest CT, the fetus is exposed to scatter radiation, which confers a low-dose radiation exposure⁽¹⁶⁾. Radiologists should need to plan very well a scan before the actual procedure takes place, monitor the length of time for the procedure, and check the quality of images collected to minimize the exposure time⁽¹⁷⁾.

The use of intravenous contrast agents may improve the diagnostic accuracy of CT by enhancing soft tissue and vascular structures⁽¹⁸⁾. The most commonly used intravenous contrast media contain iodine, and this carries a small risk of maternal side effects, like nausea, vomiting, flushing and anaphylactoid reactions⁽¹⁹⁾. Iodinated contrast medium can easily cross the placenta and enter the placental circulation or pass directly into the amniotic fluid from the fetus; however, animal studies have not revealed any teratogenic effects from its use⁽²⁰⁾.

Ultrasound

Ultrasound quickly became a relied upon and widely used imaging technique in pregnancy, being first introduced by Professor Ian Donald in Glasgow⁽²¹⁾. Initially, ultrasound imaging was assumed to be safe in pregnancy. The early ultrasound machines used relatively low output settings. Color flow, power Doppler or three-dimensional or four-dimensional imaging of the fetus were introduced at a much later stage⁽²²⁾. Thermal effects on tissue temperature and mechanical effects resulting in tissue cavitation were regarded as potential harmful to the mother or fetus. Thermal injury can potentially affect the fetal central nervous system; animal studies have demonstrated associations with neural tube defects, arthrogryposis, disorders of muscle tone, miscarriage and fetal growth restriction⁽²³⁾. The risk of temperature elevation is lowest in B-mode imaging and is higher with color Doppler and spectral Doppler imaging⁽²⁴⁾. Cavitation refers to the development of gas bubbles in tissues exposed to ultrasonic vibration⁽²⁵⁾. These bubbles can cause inertial (transient) or non-inertial (stable) cavitation effects. Inertial cavitation effects have been shown to cause genetic damage in vitro⁽²⁵⁾, while in vivo studies of non-inertial effects have remained inconclusive⁽²⁶⁾. There is a general consensus that no significant effects of ultrasound occur, unless fetal exposure is prolonged for more than 60 minutes⁽²³⁾.

Magnetic resonance imaging (MRI)

Magnetic resonance imaging does not use ionizing radiation and permits the visualization of deep soft tissue structures⁽²⁴⁾. MRI is useful for assessing a variety of medical conditions – for example, encephalopathy syndrome, thrombosis of venous plexuses⁽²⁷⁾, acute abdomen⁽¹⁰⁾, inflammatory bowel disease⁽²⁸⁾, and suspected placenta accreta spectrum disorders⁽²⁹⁾. Antenatal MRI is increasingly used to further evaluate structural fetal anomalies, including cranial lesions (agenesis of the corpus callosum, ventriculomegaly, posterior fossa anomalies, gyral or sulcation pattern)⁽³⁰⁾, neural tube defects, pulmonary sequestrations, congenital diaphragmatic hernia and heart anomalies (tumors or large vessels anomalies)⁽²⁹⁾.

In obstetrics, many times, performing an MRI can be challenging due to fetal movements and lack of trained personnel. However, MRI has several advantages over antenatal ultrasound. This imaging modality has improved image quality, and imaging of the fetal head allows the direct visualization of the entire brain. Additional limitations of sonography, resulting from reduced amniotic fluid, breech or transverse presentation and lack of visualization from the ossifying head bones, can be overcome using this type of investigation⁽²⁹⁾. Other maternal complications include phobia and back or abdominal pain, particularly at term, when it is difficult to lie in a certain position for a long period of time^(31,32).

MRI does expose the fetus to a magnetic field far greater than that of Earth (50 lT), but there are no side effects from ionizing radiation because it doesn’t use this⁽⁵⁾. The possible virtual risks are: firstly, a teratogenic risk as a result of fetal exposure to the static magnetic field and cell destruction due to quick movement, proliferation and differentiation; secondly,  tissue heating due to exposure to intermittent radiofrequency fields; thirdly, hearing damage, given the fetal exposure to high-gradient electromagnetic fields, needed in order to acquire good quality images⁽³³⁾. The American College of Radiology stipulates clearly that MRI is safe to use in pregnancy, but we should always balance risks versus benefits⁽³⁴⁾. The  consequences of first-trimester MRI on fetal growth, pregnancy loss and optic nerve damage have not been proven in studies performed in humans⁽³²⁾. Strizek et al.⁽³⁵⁾ evaluated the effects of exposure to MRI in pregnancy (1.5 T) on fetal growth and neonatal hearing function in a group of newborns at low risk for congenital hearing impairment or congenital deafness (n=751). The median gestational age at first MRI exposure was 37 weeks of gestation (range: 16-41+6 weeks). None of the exposed neonates had hearing loss or impairment. In terms of growth, both cohorts had similar mean percentile (50.6%) and controls (48.4%; p=0.22)⁽³⁵⁾.

Contrast agents

In a large retrospective cohort study conducted by Ray et al.⁽³⁶⁾, the long-term safety of MRI was evaluated in more than a million matched maternal-child pairs. MRI was carried out either in the first trimester of pregnancy or with gadolinium at any time during pregnancy. The MRI exposure was 3.97 per 1000 pregnancies. Comparing first-trimester MRI (n=1737) with no MRI (n=1,418,451), there was no significant difference in morbidity and mortality rate between the exposed and unaffected groups (19 versus 9844; adjusted relative risk [RR] 1.68; 95% CI; 0.97-2.90). When MRI was performed in the first trimester, there was no increase in the risk of structural defects, malignancies, affected hearing or vision. Inflammatory skin diseases and other autoimmune diseases were more likely in the group exposed to MRI with gadolinium contrast (123 versus 384,180 births; adjusted HR 1.36; 95% CI; 1.09-1.69). In regards to other pregnancy outcomes, the stillbirth and neonatal death rate was higher in the gadolinium-exposed group (adjusted RR 3.70; 95% CI; 1.55-8.85). Given the available evidence, it is therefore recommended that gadolinium contrast be avoided in pregnancy, unless the benefits clearly outweigh the possible risks to the fetus.

There is no requirement to cease breastfeeding following the administration of gadolinium to a lactating woman, as it has low lipid solubility, with less than 0.04% measurable in breast milk⁽³⁷⁾.

Nuclear medicine imaging

The following investigations are included: pulmonary ventilation/perfusion (V/Q scan), thyroid, bone and renal scans⁽¹⁸⁾. Technetium-99m is a commonly used isotope in V/Q scanning to diagnose pulmonary embolism (PE) in pregnancy. It has a half-life of approximately 6 hours. In general, V/Q scans result in a fetal radiation exposure below 5 mGy⁽¹⁹⁾. A normal chest radiograph can be used as a surrogate marker of ventilation and half-dose perfusion scans could be considered to reduce the exposure time. In contrast, radioactive iodine (iodine-131), for the assessment of thyroid pathology, readily crosses the placenta, has a half-life of eight days and may cause fetal hypothyroidism, especially if used after 10-12 weeks of gestation. When considering imaging of the thyroid gland, technetium-99m is the isotope to use, but it shouldn’t be used routinely^(18,19).

Imaging of malignancy during pregnancy may warrant investigation with a positron emission tomography (PET) scan⁽³⁸⁾. In pregnant women with cancer, the use of PET imaging has been debated, because it uses radioactive-labelled tracers, therefore increasing the risk of exposing the fetus to radiation. The most commonly used radiotracer is 2-deoxy-2-[fluorine-18] fluoro-D-glucose (18F-FDG). The highest exposure risk is in the first trimester, but the total absorbed dose of radiation is well below the threshold for non-cancer health effects throughout pregnancy⁽³⁹⁾. Studies in the second and third trimesters of pregnancy also indicate that the fetal radiation dose from 18F-FDG administration is low. When medically indicated in pregnant women, 18F-FDG PET scanning should not be withheld for fear of excessive radiation exposure to the fetus⁽⁴⁰⁾.

Common investigations

Calf pain/swelling

First-line imaging for suspected deep venous thrombosis (DVT) is with compression duplex ultrasound, and women should remain on therapeutic anticoagulation until the diagnosis is ruled out. If the ultrasound is negative and a high level of clinical suspicion remains, anticoagulant treatment should be stopped, but the ultrasound should be repeated on days 3 and 7⁽⁸⁾.

Shortness of breath

Women presenting with signs or symptoms of an acute pulmonary embolism (PE) should be investigated as a matter of urgency. Initial investigations should include an electrocardiogram and a chest radiograph. In those presenting with a suspected PE, but without symptoms and signs of deep vein thrombosis, a V/Q scan or a CT pulmonary angiogram (CTPA) should be performed. A Cochrane review looked at the diagnostic accuracy of CTPA and V/Q scanning for the diagnosis of pulmonary embolism and has shown that the median frequency of inconclusive results was 5.9% for CTPA and 4% for V/Q scanning⁽⁹⁾.

CTPA is many times preferred to V/Q imaging, due to disponibility, relatively low fetal radiation exposure and superior identification of other pathologies, including chest infection (5-7%) and pulmonary oedema (2-6%). A downside of CTPA is the delivery of up to 20 mGy of radiation to maternal breast tissue, so that the risk of breast cancer is slightly increased. Delivery of 10 mGy radiation to a woman’s breast before the age of 35 years old is expected to increase her lifetime risk of developing breast cancer by 13.6% above that of the general population⁽⁴¹⁾.

V/Q scanning may withhold an increased childhood malignancy risk when compared with CT owing to a slightly higher fetal radiation dose. The International Commission on Radiological Protection⁽¹⁵⁾ estimates an increased risk of fatal childhood cancer up to the age of 15 years old following in utero radiation exposure of 0.006% per mGy, which equates to a risk of 1 in 17,000 per mGy.

Abdominal pain

The most common causes of non-obstetric pain in pregnancy are appendicitis and cholecystitis⁽¹⁰⁾. Ultrasound and MRI (without contrast) are the primary imaging modalities recommended for the evaluation of abdominal pain in pregnancy; however, abdominal radiography can be used⁽¹⁰⁾. Ultrasound is a useful first-line investigation to image the appendix, bowel, hepatobiliary tree, renal tract and adnexa. In the setting of active infection or inflammation, particularly inflammatory bowel disease, MRI can also help identify bowel obstruction, fistulas or abscess formation⁽¹⁰⁾.

Headache

Headaches remain the most frequent reason for referral to an outpatient neurology clinic⁽²⁷⁾. Most headaches are benign, but they should be urgently investigated. In many cases, headaches are a primary disorder: migraine, tension-type headache and cluster headache. However, clinicians must be alert to causes of secondary headache in pregnancy, including preeclampsia, posterior reversible encephalopathy syndrome, reversible cerebral vasoconstriction syndrome and acute arterial hypertension. Other possible causes for secondary headache in pregnancy include cerebral venous thrombosis, intracranial hemorrhage, subarachnoid hemorrhage, ischemic stroke, pituitary adenoma and malignancy⁽²⁷⁾.

Women with focal neurological deficits or signs of raised intracranial hypertension should be referred for urgent intracranial imaging after a thorough history taking and clinical evaluation⁽⁴²⁾. Table 3 summarizes the intracranial imaging techniques that can be used to investigate the underlying causes of headache in pregnancy⁽²⁷⁾.

Breast mass

Women of young age still have a very high risk of dying before they turn 55 years old because of breast cancer (a lifetime risk of 1 in 9)⁽¹²⁾. When there is a suspicion of malignancy, the suspicious looking breast should be scanned and a tissue biopsy must be taken. Samples should be sent for histology, and if malignancy is identified, mammography is advised to evaluate the spread of the disease, visualize small calcifications and assess the contralateral breast. The sensitivity of mammography during pregnancy is thought to be between 78% and 90% in women with clinical abnormalities. Both breasts should always be evaluated⁽³⁸⁾. Mammography exposes the tissue to a radiation dose in the order of 0.001-0.01 mGy.

Conclusions

It is important to know which investigations can be performed in pregnancy with minimal effects on the fetus. A review of the literature has shown that the effects of a radiation below 50 mGy are likely to be negligible in pregnancy, with minimal or no consequences in regards to the fetus. MRI is the investigation of choice when it comes to pregnancy. The informed consent should always be sought and, when possible, the written connect, as well. 

Conflict of interest: none declared  
Financial support: none declared
This work is permanently accessible online free of charge and published under the CC-BY.