Pulmonary Embolism Evaluation in the Pregnant Patient: A Review of Current Imaging Approaches
Article Outline
- Abstract
- Pathophysiology
- Implications/Treatment
- Diagnosis
- CTPA and Lung Scintigraphy During Pregnancy
- Optimizing CTPA During Pregnancy
- Radiation Dose Considerations
- Methods to Reduce Dose
- What Is the Most Appropriate Modality?
- Conclusions
- References
- Copyright
Pregnancy is characterized by a higher incidence of pulmonary embolism (PE) than in age-matched nonpregnant women. However, the diagnosis of PE during pregnancy might prove to be more difficult than in the general population. Clinicians strongly rely on imaging studies to establish a prompt diagnosis. On reviewing this article, the reader will learn the pros and cons of the 2 main imaging studies used in the evaluation for PE, computed tomography of the pulmonary arteries and lung scintigraphy. Radiation dose and other important factors to consider during the evaluation for PE in pregnancy are highlighted so that clinicians and radiologists can choose the most appropriate imaging study for diagnosis.
The overall incidence of pulmonary embolism (PE) in the United States is approximately 600,000 cases annually.1 During pregnancy, the rate of PE is 4-fold to 6-fold greater than in nonpregnant women of similar age.2 Despite being relatively common, the diagnosis of PE often proves to be challenging in the setting of pregnancy. The consequences of a missed or delayed diagnosis might be fatal. PE remains the leading nonobstetric cause of death during pregnancy and the peripartum period in the developed world.3 PE might occur with almost equal frequency in all 3 trimesters of pregnancy,4 and the elevated risk does not return to baseline until the sixth week post partum.5 Clinicians must rely on imaging studies to make the diagnosis of PE; in turn, the radiologist must be familiar with the advantages and disadvantages of computed tomography of the pulmonary arteries (CTPA) versus lung scintigraphy, basic radiation dosimetry, and circumstances unique to pregnancy that must be considered in attempting to make this diagnosis during pregnancy.
Pathophysiology
First, it is important to review the factors that increase the likelihood of PE during pregnancy versus the general population. The hyperestrogenic state of pregnancy results in a relative hypercoagulability of blood to help prepare the mother for blood loss associated with delivery.3 Hemodynamic changes (ie, increased lower extremity venous stasis) and vascular endothelial injury also occur, completing the classically described Virchow triad.6 Other factors such as prolonged bed rest, obesity, multiparity, history of venous thromboembolism, or underlying thrombophilia further increase the risk of PE.
Implications/Treatment
A diagnosis of PE during pregnancy carries with it several important ramifications, particularly with regard to anticoagulation therapy. Management of PE during pregnancy is challenging, and empiric treatment with a questionable diagnosis is discouraged. Warfarin is contraindicated because it crosses the placenta and results in fetal anticoagulation, spontaneous abortion, and warfarin embryopathy including central nervous system abnormalities in 4%-5% of exposed fetuses.2 Conversely, heparin does not cross the placenta, so there is no risk of teratogenicity or fetal bleeding.5 The Pregnancy and Thrombosis Working Group consensus statement2 recommends that therapy should be initiated immediately on the diagnosis of PE with either low-molecular-weight heparin (LMWH) or unfractionated heparin for 5-10 weeks followed by LMWH. Deep vein thrombosis (DVT) prophylaxis must then be considered for all future pregnancies, because the incidence of recurrent PE during each subsequent pregnancy is 4%-15%.4 A history of PE might inhibit the future use of oral contraceptives or hormone replacement therapy. These implications should not be treated lightly and highlight the importance of making an accurate diagnosis.
Diagnosis
The clinical presentation of PE is variable. Although typical signs and symptoms might be present (sudden-onset dyspnea, pleuritic chest pain, tachycardia, tachypnea, syncope or near-syncope), these findings are nonspecific. Many patients will have atypical findings, and the diagnosis of PE might not be of primary consideration. The clinical picture can be even more clouded during pregnancy; signs and symptoms that would usually raise suspicion for PE or DVT might overlap with findings normally expected during the course of pregnancy.
With regard to a uniform diagnostic algorithm for PE in the setting of pregnancy, consensus is lacking. Clinical probability scores such as the Wells Model and Geneva Criteria have not been validated in pregnant patients.7 However, clinicians might follow one of several previously published guidelines. The Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) II committee recommends D-dimer testing with initial clinical assessment.8 D-dimer rapid enzyme-linked immunosorbent assay is a laboratory test with a high negative predictive value (NPV) and low positive predictive value (PPV) for PE. A low probability clinical assessment combined with a negative D-dimer test effectively excludes PE in the general population.9 This screening test loses value in pregnant patients because of a normal progressive increase in D-dimer values that occurs during pregnancy. Nevertheless, it is important to realize that approximately 50% of pregnant patients will still have a normal D-dimer assay, and this normal value has the same high NPV as in a nonpregnant patient.1 If D-dimer results are positive, lower extremity deep venous compression ultrasound (US) is recommended before proceeding to other imaging studies that would require ionizing radiation. If compression US is positive for DVT, then the work-up for PE is complete because treatment for either entity is the same in most instances. If compression US is negative and there remains clinical concern for PE, further testing is required. However, the role of compression US in the work-up for suspected PE during pregnancy has been debated. Only a small proportion of these studies are found to be positive, and as a result, most patients undergo further testing regardless.7
Classically described plain radiographic findings of PE, a peripheral wedge-shaped infiltrate (Hampton hump) or decreased pulmonary vascularity (Westermark sign), are rare findings.6 Chest radiography, however, plays an important role in excluding alternative diagnoses (such as pneumonia, pulmonary edema, or pneumothorax) at the cost of minimal radiation dose. Results of a chest radiograph (ie, normal or abnormal) might also sway one toward performing lung scintigraphy versus CTPA as the next step. Many published results comparing the effectiveness and appropriateness of these 2 modalities for the diagnosis of PE during pregnancy are contradictory and based on small sample sizes.10 Therefore, differing opinions exist throughout the literature with regard to what imaging modality should be the standard of care in evaluation for PE during pregnancy.
Lung scintigraphy, either perfusion scanning alone or in combination with ventilation imaging, was the main imaging test before the introduction of CTPA. It is associated with overall lower radiation exposure to the patient and fewer allergic reactions than CTPA. Most PIOPED II investigators (69%) recommended lung scintigraphy over CTPA.8 However, the use of CTPA has dramatically increased in the interim and is now accepted as the primary imaging modality to evaluate for PE at many institutions. Lung scintigraphy and CTPA each have their own strengths and weaknesses; the role of these different imaging modalities in the assessment for PE during pregnancy continues to be studied and debated.
CTPA and Lung Scintigraphy During Pregnancy
The main advantage of CTPA is that it is capable of providing a more rapid, definitive answer to the question of whether PE is present. Because PE is anatomically demonstrated as an intraluminal filling defect within the pulmonary arteries (Fig. 1), CTPA has a very high rate of interobserver agreement. CTPA has a proven high sensitivity (83%) and specificity (96%) for the diagnosis of PE.11 Multidetector CTPA can depict emboli in vessels as small as the sixth-order branches.3 Lung scintigraphy assigns a probability of PE, which leads the perception among some clinicians that this nuclear medicine study would be more appropriately termed “unclear medicine.” CTPA also has the advantage of at times identifying a significant abnormality or alternative diagnosis to PE not previously suspected on the initial chest radiograph.7, 10 However, CTPA is not without its own limitations, mainly higher cost, overall higher radiation dose, and reliance on iodinated contrast material (which carries its own risk of adverse reactions).12 Patients with a history of serious contrast allergy and many with impaired renal function are not candidates for CTPA.
Figure 1.
CTPA allows definitive diagnosis of PE by demonstrating intraluminal filling defects within the pulmonary arteries (arrows). CTPA provides the added benefit of demonstrating other anatomical information that might be pertinent to the patient's clinical outcome, such as the presence of right heart dilatation in this example (asterisks).
Of further debate is the added theoretical risk of iodinated contrast material inducing hypothyroidism in the fetus. Iodinated contrast is a U.S. Food and Drug Administration category B drug; animal reproduction studies have not demonstrated a fetal risk, but its safety has not been formally established because there have been no controlled studies in pregnant women. Thyroid hormone plays a critical role during the fetal and neonatal period, and congenital hypothyroidism remains the most common treatable cause of mental retardation in the United States.13 Therefore, it is believed that iodinated contrast should be used only after assessing the potential risk-to-benefit ratio for the patient. Despite this, there remains debate regarding the potential effect of even a single fetal exposure to iodine during pregnancy.
Bourjeily et al13 recently published a retrospective study on the effect of in utero exposure to a single dose of water-soluble iodinated intravenous contrast medium (isohexol) on thyroid function at birth. The subjects were all pregnant women who underwent CTPA for suspected PE. The developing thyroid gland can concentrate iodine and synthesize thyroxine (T4) after the 11th week of gestation.13 The mean gestational age at the time of contrast administration was 27.8 weeks. All had normal T4 levels at birth, and none had a lasting effect on thyroid-stimulating hormone. The authors concluded that a single high-dose exposure to a low osmolar iodinated contrast agent is unlikely to have any clinically important effect on thyroid function at birth.
Lung perfusion scintigraphy is performed after the intravenous injection of technetium (Tc)-99m labeled macroaggregated albumin (MAA) and is safe to perform in patients with poor renal function or iodinated contrast allergies. Different radiopharmaceuticals might be used for ventilation imaging, such as xenon-133 inhaled gas and Tc-99m–diethylenetriaminepentaacetic acid aerosol. Guidelines for interpretation include the (revised) PIOPED criteria, Hull criteria, and so-called Gestalt interpretation (which combines existing guidelines with the radiologist's own personal experience). Hagen et al14 found all 3 criteria to have good accuracy and interobserver variability among 2 experienced readers. No matter what criterion is used, lung scintigraphy can usually provide very useful information to the referring clinician. A very low probability or normal ventilation-perfusion (V/Q) scan (Fig. 2) is very reassuring because it has a specificity of 99.7%.12 It is also accepted that a high probability V/Q scan (Fig. 3) has very high specificity for PE. Recent studies have found that sensitivity of a high probability scan is also good (77.4%).12 Hagen et al report that a high probability V/Q scan has a PPV of 90% and justifies treatment with anticoagulation.
Figure 2.
A 29-year-old pregnant woman who presented with chest pain. Note the sharp, normal contours of the lungs in all 8 standard planar images of this normal perfusion scan. A normal or very low probability scan is very reassuring, with specificity approaching 100%. LT, left; RT, right; RPO, right posterior oblique; RAO, right anterior oblique; LPO, left posterior oblique; LAO, left anterior oblique.
Figure 3.
Multiple large wedge-shaped perfusion defects in right lung (arrows) and decreased perfusion to large portion of left lung (asterisks), with sparing of the apicoposterior segment of the left upper lobe. Ventilation portion of this study and recent chest radiograph (not pictured) were both unremarkable. Findings are consistent with high probability for PE.
There have been several recent investigations comparing the efficacy of CTPA versus lung scintigraphy in the detection of PE during pregnancy. On the basis of their retrospective study, Shahir et al10 concluded that the 2 imaging techniques are equally effective, with 99% NPV for CTPA and 100% NPV for lung perfusion scintigraphy. Diagnostic quality was also equal, because there were similar percentages of indeterminate studies for CTPA (3.7%) and lung perfusion scintigraphy (2%). Revel et al7 also reported equal numbers of indeterminate studies for CTPA (8/43, 19%) and lung scintigraphy (17/91, 19%). In this particular study, the authors classified V/Q scan results as “indeterminate” if they did not fall into either the “normal” or “high probability” categories. If low probability studies were also deemed negative, the number of indeterminate V/Q scans would have been only 7 of 91 (8%). Ridge et al15 reported a greater number of indeterminate results with CTPA (10/28, 36%) than V/Q scan (1/25, 4%) in their small sample population. The authors acknowledged that selection bias played a role in their data because patients with abnormal chest radiographs underwent CTPA instead of V/Q scan. Some have suggested that indeterminate V/Q scan results are less common during evaluation for PE in pregnancy because this population is generally younger with less comorbidity such as cardiac or respiratory disease.7 On the other hand, the studies just mentioned are concordant with previously published data that show indeterminate CTPA results are more common in pregnancy than in the general population.
Optimizing CTPA During Pregnancy
Common factors that limit assessment of the pulmonary arteries on CTPA in any patient include respiratory motion and poor contrast opacification. Hemodynamic changes occurring during the course of pregnancy might further increase the risk of a nondiagnostic study owing to greater dilution of contrast material in the blood pool.16 Blood volume increases up to 50% greater than in nonpregnant women, with a corresponding increase in cardiac output. This hyperdynamic circulation shortens the arrival time of intravenous contrast with the pulmonary arteries, and poor peak arterial enhancement might result if using standard triggered scan delays.15 Another phenomenon that results in greater contrast dilution during pregnancy has received recent attention, transient interruption of contrast by an influx of unopacified blood from the inferior vena cava (Fig. 4). Ridge et al15 identified this process in the majority of nondiagnostic CTPA studies in pregnant patients (80%).
Figure 4.
Coronal reconstructed image from CTPA in a pregnant woman demonstrates mixing of large influx of unopacified blood from IVC (arrow) and injected contrast bolus (asterisk). Contrast dilution from this phenomenon, owing to the hemodynamic effects of pregnancy, might lead to poor opacification of the pulmonary arteries and a greater number of indeterminate studies if preventative steps are not taken.
Kuzo et al17 investigated this phenomenon by using velocity-encoded magnetic resonance imaging to measure blood flow in the inferior vena cava (IVC) and superior vena cava (SVC) during different respiratory maneuvers. They proved that there was a significant increase in caval flow during deep inspiration, with greater effect in the IVC than SVC (2.4:1 ratio), which explains why a large influx of IVC blood would dilute contrast material and lead to poor opacification of the pulmonary arteries. So why is this finding prevalent on CTPA done in pregnancy? Images are usually acquired after a deep inspiration (“take a deep breath and hold”) shortly after contrast injection. Venous return to the right heart increases during this deep inspiration, mostly from the IVC because of decreased intrathoracic pressure and the resultant negative pressure gradient.15 Resting pressure in the IVC is already high in a supine patient because of compression by the gravid uterus, especially during late-stage pregnancy.
Several authors have offered suggestions on how to optimize CTPA to overcome the hemodynamic effects of pregnancy. The initial deep inspiration can be avoided in favor of shallow breaths or none at all during the time of scanning.17 This serves to reduce or eliminate the large influx of unopacified contrast from the IVC during deep inspiration that would otherwise dilute the contrast bolus. Careful instructions should be given to the patient before scanning. Contrast bolus triggering can still be used, but with shorter start delays to adapt to the hyperdynamic state of pregnancy.16 Higher flow rates of contrast injection, higher contrast medium concentration, and low kilovoltage (kVp) techniques have also been suggested.15, 16
Radiation Dose Considerations
There is evidence to support an increased childhood cancer risk as a result of irradiation of the fetus in utero. Despite this, the use of diagnostic imaging in pregnancy has dramatically increased during the past decade. Lazarus et al18 found 107% increase in the number of radiologic imaging studies performed at one institution in pregnant patients from 1997-2006. This trend was greatest in the number of CT studies (mostly CTPA), which increased by an average of 25% more performed per year. This dramatic rise in CTPA did not result in any substantial decrease in the number of V/Q scans performed.
The radiologist must understand basic radiation dosimetry as it applies to the use of CTPA and lung scintigraphy during pregnancy, especially in light of growing concerns among clinicians and the general population about the very real and potential risks of ionizing radiation. However, recent surveys such as the one by Groves et al19 demonstrated a general lack of knowledge among health care professionals, including radiologists, about radiation considerations in imaging of pregnant women suspected of having PE.
During CTPA, the fetus is only exposed to scatter radiation because it lies outside of the scanned volume. Therefore, the dose to the fetus is relatively small. During lung scintigraphy, the fetus is exposed to radiation from renal excretion of Tc 99m that collects within the bladder. Exact doses depend on the particular imaging protocol used at various institutions. Winer-Muram et al4 demonstrated that helical chest CT (ie, CTPA) delivers a smaller dose to the fetus than V/Q scan during all 3 trimesters. In this study, the mean fetal dose in 23 subjects (0.003-0.131 mGy) was calculated by using Monte Carlo geometrical model design. Hurwitz et al20 used a different technique to establish fetal radiation dose during CTPA by directly measuring dose with an anthropomorphic phantom. They concluded that the mean fetal dose (0.24-0.660 mGy) with CTPA was no smaller than that of V/Q scan with 2 mCi of Tc-99m–labeled MAA and 10 mCi of xenon-133 gas (0.32-0.36 mGy).
Even if studies were to consistently demonstrate a statistically significant difference in fetal dose delivered by CTPA and lung scintigraphy, the effects of these small fetal doses have so far not been found to be clinically significant.10 Each modality delivers a fetal dose less than that of the natural background radiation (1.2-2.6 mGy) experienced during pregnancy from cosmic rays, radon, etc.4 A fetal radiation exposure of at least 100 mGy is necessary before pregnancy termination is considered.4 Although the potential risks to the fetus from either diagnostic study should not be taken lightly, the risks do seem small in comparison with the normal risks of pregnancy: 3% risk of spontaneous birth defects, 15% risk of spontaneous abortion, 4% risk of prematurity and growth retardation, 1% risk of mental retardation.21 In any case, it is important to inform patients undergoing CTPA or lung scintigraphy about these radiation risks so that they too can be involved in the decision-making process. The potential consequences of delaying or refusing imaging should also be clearly explained to the patient, because the risk of death from undiagnosed PE outweighs the small risk of radiation-induced cancer.1
Methods to Reduce Dose
To help minimize the potential risks of CTPA or lung scintigraphy, certain steps might be taken to minimize the radiation dose delivered to the fetus. For instance, strongly encouraging the patient to drink plenty of liquids and empty her bladder frequently after lung scintigraphy is a very simple method to minimize fetal radiation. Also, many centers routinely omit the ventilation portion of a V/Q scan during pregnancy because the efficacy of performing a lung perfusion scan alone has been established.10 A perfusion scan combined with a chest radiograph can be as accurate as a V/Q scan.22 The perfusion scan can also be done with “low dose” technique by using one-half of the radiopharmaceutical dose that a nonpregnant woman would receive. Shahir et al10 reported similar fetal radiation doses for a low-dose perfusion only scan (0.1-0.37 mGy) and CTPA (0.1-0.66 mGy). The scan time might need to be doubled to achieve adequate counts with this low-dose technique. With regard to CTPA, newer CT protocols use lower tube voltage and tube current and limit scanned volume to only necessary anatomical coverage, and these modifications can significantly lower radiation dose.10 Fetal irradiation during CTPA can also be reduced substantially when oral barium shielding is used, a method Yousefzadeh et al3 postulate could also be expanded to include pregnant patients undergoing lung scintigraphy.
Radiation dose to the fetus has not been the only subject of debate when it comes to deciding which modality to choose when evaluating for PE during pregnancy. As the use of CTPA has dramatically risen, so have concerns about radiation dose delivered to the maternal breast and the potential increased risk of breast carcinoma (Fig. 5). CTPA delivers an estimated dose of 20 mGy per breast in an average-size woman.10 To put this into perspective, the American College of Radiology recommended dose per breast for a standard 2-view screening mammogram is about 6.7 times less (3 mGy) than the dose incurred by CTPA. In contrast, the radiation dose delivered to the breast during lung scintigraphy is considerably smaller (0.11-0.31 mGy).10 Bismuth shields have been shown to reduce the amount of radiation exposure to the breast during CTPA, although this breast shielding is not routinely used in adult women in the United States.5
Figure 5.
Axial CTPA image in 19-year-old pregnant woman demonstrates the patient's breast tissue directly within the scanned volume. CTPA delivers much greater radiation dose to the maternal breast than does lung scintigraphy. The potential increased risk for radiation-induced breast carcinoma should not be disregarded when deciding on method of diagnostic evaluation.
What Is the Most Appropriate Modality?
Because several investigators have shown that CTPA and lung scintigraphy have similar diagnostic accuracy in the evaluation for PE during pregnancy, an emphasis should be placed on other factors when deciding which modality is most appropriate for a particular patient. Those with a history of a serious contrast allergy or impaired renal function are better off undergoing lung scintigraphy. As described earlier, recent studies have negated the idea that fetal radiation dose is significantly lower with CTPA than with lung scintigraphy. However, it is clear that maternal breast dose is much lower with latter modality. Therefore, lung scintigraphy should be the imaging modality of choice in the setting of a normal chest radiograph and no clinical suspicion for an alternative diagnosis that would only be apparent by CT.10 A low-dose perfusion only scan might allow for the least radiation exposure.
Factors such as the availability of each imaging modality at a particular institution might limit options. Multidetector CT is available even at smaller community hospitals on a 24 hours/day, 7 days/week basis. In contrast, many institutions do not have nuclear medicine technologist support during evenings or on weekends. At the authors' institution, 24-hour technologist coverage is available with the nuclear medicine technologist on beeper call for emergencies occurring after normal work hours. Despite this, some clinicians might see any added time from having a technologist come in from home and prepare the radionuclide beforehand as a deterrent to ordering lung scintigraphy, because CTPA can be obtained so readily and theoretically might allow for quicker triage. Given today's heightened concern for radiation exposure and with all other factors being fairly equal, the much lower maternal breast dose from lung scintigraphy versus CTPA must be given serious consideration with respect to the patient. This is especially true when one considers that it is not unusual for these same patients to undergo multiple repeat examinations to exclude PE in the future. Kline et al23 reported that at least one-third of all emergency department patients who undergo CTPA for suspected PE will undergo a second study within the next 5 years; young women comprise a significant percentage of this group.
Conclusions
Pregnancy is associated with an elevated risk of PE, although making this diagnosis might prove more difficult than in the general population. Clinicians depend on imaging to establish the diagnosis. At the time of initial work-up, chest radiography might provide an alternate diagnosis to PE and at the very least plays an important role in deciding the next most appropriate study to order. Recent studies have shown that similar overall diagnostic accuracy for PE can be achieved with both CTPA and lung scintigraphy, with little to no significant difference between the 2 modalities in regard to fetal dose. Radiation dose to the maternal breast is clearly greater with CTPA, an important factor that should not be dismissed. Hence, in the setting of a normal chest radiograph, lung scintigraphy is the imaging modality of choice for evaluation of PE during pregnancy. If CTPA is performed, standard protocols might need to be modified to compensate for normal hemodynamic changes of pregnancy that might otherwise hinder diagnostic quality. Additional methods might be applied with either modality to help minimize radiation dose.
References
- . Challenges, controversies, and hot topics in pulmonary embolism imaging . AJR Am J Roentgenol . 2011;196:497–515
- Antithrombotic therapy and pregnancy: Consensus report and recommendations for prevention and treatment of venous thromboembolism and adverse pregnancy outcomes . Am J Obstet Gynecol . 2007;197:457.e1–457.e21
- . Internal barium shielding to minimize fetal irradiation in spiral chest CT: A phantom simulation experiment . Radiology . 2006;239:751–758
- Pulmonary embolism in pregnant patients: Fetal radiation dose with helical CT . Radiology . 2002;224:487–492
- . Peripartum pulmonary embolism . Clin Chest Med . 2011;32:147–164
- . Thromboembolism in pregnancy . Obstet Gynecol Clin North Am . 2007;34:481–500
- Pulmonary embolism during pregnancy: Diagnosis with lung scintigraphy or CT angiography? . Radiology . 2011;258:590–598
- Diagnostic pathways in acute pulmonary embolism: Recommendations of the PIOPED II Investigators . Radiology . 2007;242:15–21
- . Don't bury the V/Q scan: It's as good as multidetector CT angiograms with a lot less radiation exposure . J Nucl Med . 2008;49:5–8
- Pulmonary embolism in pregnancy: CT pulmonary angiography versus perfusion scanning . AJR Am J Roentgenol . 2010;195:214–222
- Multidetector computed tomography for acute pulmonary embolism . N Engl J Med . 2006;354:2317–2327
- Acute pulmonary embolism: Sensitivity and specificity of ventilation-perfusion scintigraphy in PIOPED II study . Radiology . 2008;246:941–946
- Neonatal thyroid function: Effect of a single exposure to iodinated contrast medium in utero . Radiology . 2010;256:744–750
- Comparison of observer variability and accuracy of different criteria for lung scan interpretation . J Nucl Med . 2003;44:739–744
- Pulmonary embolism in pregnancy: Comparison of pulmonary CT angiography and lung scintigraphy . AJR Am J Roentgenol . 2009;193:1223–1227
- . CTPA for the diagnosis of acute pulmonary embolism during pregnancy . Eur Radiol . 2008;18:2705–2708
- Measurement of caval blood flow with MRI during respiratory maneuvers: Implications for vascular contrast opacification on pulmonary CT angiographic studies . AJR Am J Roentgenol . 2007;188:839–842
- Utilization of imaging in pregnant patients: 10-Year review of 5270 examinations in 3285 patients—1997-2006 . Radiology . 2009;251:517–524
- CT pulmonary angiography versus ventilation-perfusion scintigraphy in pregnancy: Implications from a UK survey of doctors' knowledge of radiation exposure . Radiology . 2006;240:765–770
- Radiation dose to the fetus from body MDCT during early gestation . AJR Am J Roentgenol . 2006;186:871–876
- Imaging the pregnant patient for nonobstetric conditions: Algorithms and radiation dose considerations . Radiographics . 2007;27:1705–1722
- Comparison of lung scintigraphy with multi-slice spiral computed tomography in the diagnosis of pulmonary embolism . Clin Nucl Med . 2009;34:424–427
- Incidence and predictors of repeated computed tomographic pulmonary angiography in emergency department patients . Ann Emerg Med . 2009;54:41–48
PII: S0887-2171(11)00117-X
doi:10.1053/j.sult.2011.09.001
© 2012 Elsevier Inc. All rights reserved.
