| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short communication |
Radiology Department, Northern General Hospital, Herries Road, Sheffield, South Yorkshire S5 7AU, UK
 |
Abstract |
|---|
 Top Abstract References |
|---|
Pulmonary embolism (PE) is the leading cause of maternal death [1]. The rate of PE in pregnancy is five times greater than that for non-pregnant women of the same age and is about 1 in 1500 deliveries; the risks are even higher in the puerperium. The clinical diagnosis of PE is difficult in the general population, but is further complicated in pregnancy as some of the clinical symptoms of PE can be normal/expected symptoms of pregnancy. Precise PE diagnosis in pregnancy is vital to prevent unnecessary treatment of PE as treatment is associated with side effects for both the mother and fetus. Accurate imaging is essential, but there are frequently anxieties relating to fetal radiation exposure during diagnostic procedures. Despite the complexities of the clinical scenario, the British Thoracic Society (BTS) guidelines on the management of PE do not address the issue of imaging for PE in the pregnant patient [2].
There are an array of clinical, biochemical and radiological tests available for the investigation of PE, some of which have not been validated in pregnancy. In general terms, the BTS guidelines recommend clinical assessment of the pre-test probability of PE for each patient with d-dimer assessment for patients with a low or intermediate pre-test probability followed by either a ventilation/perfusion scan or CT pulmonary angiogram (CTPA; depending on the availability of nuclear medicine scans locally and the presence of a chest X-ray abnormality). The Wells criteria are the most frequently used tool for assessing the clinical probability of PE. However, pregnant patients were excluded from the analysis group for validation of the criteria [3]. The d-dimer is known to increase in pregnancy. The d-dimer is usually normal in the first trimester of uncomplicated pregnancy, starts to rise during the second trimester and returns to baseline levels at 4–6 weeks post partum [4–6]. The d-dimer is not affected by bleeding, breast feeding or heparin prophylaxis, but is elevated in association with many pregnancy-related complications such as pre-eclampsia [5, 7]. Pregnant women with on-going thrombosis have been shown to develop a significant rise in d-dimer [8].
It is estimated that 70% of patients with a proven PE have proximal deep venous thrombosis (DVT). However, the proportion in pregnant women with PE is unknown. Therefore, Doppler ultrasound of the leg veins is recommended for the investigation of PE if the patient has symptoms and signs suggestive of DVT [2]. The diagnosis of DVT in pregnant women can be problematic. There is increased lower extremity vein diameter and decreased flow secondary to hormonal effects and the direct compressive effect of the enlarged uterus on pelvic veins, hence the legs are frequently swollen in the absence of DVT [9]. There is an increased risk of iliac vein thrombosis, which is not routinely assessed by leg Doppler studies. In addition, the accuracy of Doppler ultrasound, including iliac vein assessment, has not been validated in pregnant patients. However, Doppler ultrasound does not involve ionizing radiation and, as there are concerns about the radiation exposure to mother and fetus, bilateral leg Doppler has been proposed as the initial investigation of suspected PE in pregnancy [9]. This would conform to the general principle of maintaining doses "as low as reasonably achievable" whilst still offering a valuable test that may preclude the need for further investigations associated with a significant radiation dose [10].
Ventilation/perfusion (V/Q) imaging is well established for imaging PE and, in a survey relating to imaging practice for the investigation of PE in pregnancy in 1998, was the most frequently employed test for this sub group [11]. In pregnant women the radiation dose can be minimized by using a half-dose perfusion scan and only proceeding to ventilation imaging if a defect is identified on the perfusion scan [12]. However, for the general population 50–70% of V/Q scans are indeterminate. For pregnant women the proportion of patients falling into each of the reporting categories (high probability, normal and indeterminate) is different: fewer pregnant women have high probability scans (less than 5%) and many more have a normal scan (75%) [13–15]. Therefore, only 20% of this patient group have indeterminate scans. This change in distribution of patients within the probability groups is thought to reflect the younger average age and reduced presence of co-morbidities compared with the general population. PE can be confidently excluded with a normal V/Q scan, but this test throws up a relatively high proportion of indeterminate results and, in the high probability group, up to 20% of the patients may not have PE. A small study involving 113 de novo cases of potential PE showed that withholding anticoagulation in pregnant women with normal or indeterminate scans is probably safe [13]. However, larger studies are required to confirm this finding, especially as the incidence of PE in the proportion of pregnant women investigated is so low. It has been suggested that CTPA has a greater discriminatory power than V/Q scanning with a low pre-test probability, but that CTPA and V/Q scanning have a similar discriminatory power in those with a high pre-test probability [16].
CTPA is now a well-validated investigation with a sensitivity and specificity between 94% and 100% [17, 18]. The negative predictive value of a normal CTPA is over 99% [19]. The clinical validity of a CT scan to rule out PE is similar to that reported for conventional pulmonary angiography [19]. Anticoagulants can be safely withheld if the CTPA is negative for PE. CTPA is advantageous as the emboli are directly visualized (unlike for V/Q scanning) and alternative causes for the patient's symptoms may be diagnosed. However, there are concerns regarding the radiation dose received from CT scanning, particularly to the fetus. Recent studies have shown that the fetal radiation exposure for CTPA varies from 3.3 µGy to 130.0 µGy; the dose increasing during each trimester as the fetus enlarges and approaches the imaged area in the thorax [20]. However, the estimated fetal radiation dose for V/Q scanning is estimated as 100–370 µGy, i.e. the dose may be more than 3 times greater than for CTPA. In addition, CTPA has a superior sensitivity and specificity for PE [21].
All radiation to the fetus carries a potential risk. This risk must be balanced against the risk to the mother/fetus if PE is not diagnosed or treated and against the risk of treatment of non-confirmed PE. Everyone is exposed to radiation all the time from the atmosphere, ground and from ingested food and drink. The average "background radiation" for an individual in the UK is 2.7 mSv per year, which equates to about 1000 µGy for a fetus in utero for 9 months [22]. The worst estimated absorbed dose for the fetus in the third trimester undergoing CT pulmonary angiography is 130 µGy, i.e. approximately 7 times less than the natural background radiation. Central nervous system malformations can be associated with excess radiation [23]. A threshold fetal dose greater than 100 000–200 000 µGy is required to cause such a problem. A dose of 100 µGy to the fetus is associated with an excess death from cancer up to the age of 15 years of 1 in 300 000. This can be put into perspective by comparison with everyday activities and their relative annual risk of death in the UK, e.g. smoking 10 cigarettes each day has a risk of 1 in 200 and relative risk of death from uncomplicated pregnancy for the mother is 1 in 170 000 [24]. The annual risk of death in the UK for all cancer is 1 in 400 and for death from all causes at the age of 40 years 1 in 700 [24]. Therapeutic termination would not be considered for a fetal dose below 100 000 µGy. However, it is recommended that the patient is given more detailed information regarding the risks of radiation for procedures where the fetal dose is expected to be greater than 1000 µGy [23]. Therefore, although the radiation risk to the fetus cannot be ignored, the risk is very low. The risk of fetal death is much greater if the mother has untreated PE.
In addition to the radiation risk to the fetus, the breast radiation dose from CTPA must be considered. The female breast is extremely radiosensitive, and it has been shown that a sufficiently large radiation dose can cause breast cancer [25]. The exposure of the immature breast during early development and around the time of menarche carries a higher risk than at other times of a woman's life. There is little evidence that radiation exposure after 45 years of age increases the incidence of breast cancer. The effect of radiation on the breast in pregnancy is unclear. However, there has been shown to be a relatively linear relationship between radiation dose and subsequent breast cancer, although this relationship does not extend into the highest radiation dose exposures used for radiotherapy. The vast majority of people exposed to radiation do not develop a cancer related to that exposure. The 25 000 female atomic bomb survivors in Japan have been followed for over 50 years, but only 173 breast cancer deaths have occurred of which 41 were attributed to the radiation received in 1945.
A radiation dose of 100 cGy is associated with an increased risk of breast cancer of 40% in young Western women [25]. This is the same risk a woman experiences secondary to several common conditions/lifestyle choices, e.g. never being pregnant, menarche before the age of 11 years or a late menopause. Epidemiological studies have not detected a significantly increased risk of breast cancer below a dose of 20 cGy. Parker et al have specifically investigated female breast radiation exposure during CTPA and calculated an effective minimum dose of 20 mGy (2 cGy) [26]. This dose concurs with other data estimating a dose between 20 mGy and 50 mGy (2–5 cGy) for a standard chest CT scan [27]. These estimates are significantly below the level of 20 cGy, below which no effect on the breast can be demonstrated, but significantly higher than the estimated breast radiation dose of 0.28 mGy associated with ventilation/perfusion scanning [28]. Although this radiation exposure is associated with an immeasurable low malignancy risk, this exposure should not be ignored and the development of breast shields that may reduce this dose by up to 73% may be considered in the future [29].
A 2003 survey of members of the Society of Thoracic Radiology found that 53% of responding radiologists would use CTPA as a first line investigation for excluding PE in pregnant patients, but only 60% of radiologists obtained informed consent from any pregnant patient undergoing CTPA, only 16% of departments had a written policy for this group and only 40% modified the imaging protocol in an attempt to reduce radiation dose [30].
The risks of iodinated contrast media on the fetus have not been fully investigated. Many pregnant patients have received intravenous contrast for the investigation of other problems such as renal colic. However, there are no reports in the literature of any ill effects, despite the theoretical risk of contrast induced hypothyroidism.
In summary, d-dimer concentration is known to rise in normal pregnancy, but a normal value is clinically helpful. Doppler ultrasound of both legs has a low pick up rate, but does not involve ionizing radiation. If DVT is found, the patient should be treated as for PE. This test requires further validation in pregnant women. Ventilation perfusion imaging is associated with a higher radiation dose to the fetus, but lower radiation dose to the mother/breast than CTPA. The majority of patients have a low or indeterminate probability of PE from the V/Q scan and it is probably safe to withhold anticoagulants in these groups, but this has not been validated in a large controlled study. V/Q scanning has a lower discriminatory threshold for PE than CTPA in patients with a low or intermediate pre-test clinical probability. The mother receives a higher radiation dose from CTPA, especially to the breast, but the fetal dose is lower than V/Q imaging. However, all quoted radiation doses to breast or fetus are below the thresholds estimated to be associated with any significant risk. CTPA has a greater sensitivity and specificity than V/Q scanning and is able to diagnose alternative causes for the patients' symptoms in cases where PE is absent. However, the risk of iodinated contrast to the fetus is not known.
Multiple clinicians may interpret these same facts and derive differing imaging protocols based on their perceived significance for each factor discussed. The following imaging protocol is proposed to optimize the accuracy of PE diagnosis whilst minimizing radiation dose to the fetus: All patients should be clinically assessed by a senior clinician and a pre-test probability assigned. Patients in the first two trimesters of pregnancy should have the d-dimer measured, provided there is no specific contraindication. A normal d-dimer and low pre-test probability of PE can be used to exclude PE in the pregnant patient. Exclusion of patients from further imaging on the basis of intermediate pre-test probability and normal d-dimer needs further investigation. All pregnant patients requiring further investigation for PE should proceed to bilateral leg Doppler ultrasound. If this test is positive for venous thromboembolism, the patient should be treated for PE. If this test is negative, all patients should proceed to CTPA. Informed consent should be sought from the pregnant patient prior to CTPA. The patient should be given simple information explaining the risks of fetal and maternal radiation and risks to the mother and child of failing to accurately diagnose PE. The CTPA protocol should be modified to minimize the radiation dose; in particular the length of the thorax along the z-axis should be reduced. It is recommended that the baby has thyroid function testing within the week of birth due to the theoretical risk of contrast induced hypothyroidism [31].
Further research is advised to validate the above protocol and to clarify the in utero affects of iodinated contrast on the neonate.
Received for publication August 1, 2005. Revision received November 17, 2005. Accepted for publication January 19, 2006.
|
References |
|---|
|
TopAbstractReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Remy-Jardin, M. Pistolesi, L. R. Goodman, W. B. Gefter, A. Gottschalk, J. R. Mayo, and H. D. Sostman Management of Suspected Acute Pulmonary Embolism in the Era of CT Angiography: A Statement from the Fleischner Society Radiology, November 1, 2007; 245(2): 315 - 329. [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. McCollough, B. A. Schueler, T. D. Atwell, N. N. Braun, D. M. Regner, D. L. Brown, and A. J. LeRoy Radiation Exposure and Pregnancy: When Should We Be Concerned? RadioGraphics, July 1, 2007; 27(4): 909 - 917. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. D. Stein, P. K. Woodard, J. G. Weg, T. W. Wakefield, V. F. Tapson, H. D. Sostman, T. A. Sos, D. A. Quinn, K. V. Leeper Jr, R. D. Hull, et al. Diagnostic Pathways in Acute Pulmonary Embolism: Recommendations of the PIOPED II Investigators Radiology, January 1, 2007; 242(1): 15 - 21. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
