Dose report from a non-contrast CT of the abdomen pelvis for suspected kidney stone. The two reported values are the CT dose index (CTDIvol) and the dose length product (DLP). The DLP is equal to the CTDIvol × the scan length. In this case, the scan length from top of the kidneys was 50 cm, so the DLP of 622.4 mGy-cm is equal to the CTDIvol of 12.42 mGy × 50 cm. In terms of effective dose, using a conversion factor of 17 microseiverts per mGy-cm in the abdomen-pelvis would correspond to an effective dose of approximately 10.6 mSv (slightly below the national average dose of 11.2 mSv for renal colic protocol CTs)
While DLP is the more frequently reported dose metric, it is important to understand that DLP is calculated based on a standardized phantom (i.e., simulated patient) and does not take patient size into account in terms of actual effective organ dose. To account for this, the American Association of Physicists in Medicine has recently proposed conversion factors for CTDIvol into a size specific dose estimate (SSDE), which utilizes conversion factors based on patient size.26 The key concept here is that for a given CTDIvol or DLP, a larger patient will have relatively lower actual dose delivered to the radiosensitive organs because of absorption by non-radiosensitive tissue.
Regarding actual radiation dose delivered to a patient and the potential for organ damage, several metrics are in use, and ongoing controversy about the best reporting methods can make comparison and estimation of risk at the individual and population levels confusing. Both rad and gray (gray is the SI unit) refer to the amount of radiation absorbed, while rem and sievert (seivert is the SI unit) refer to the effective organ dose that should be most directly related to risk of malignancy. For CT scanning, the most commonly used units will be milligray (mGy) and millisieverts (mSv). As mentioned earlier, a CT report will provide the total dose delivered as the DLP in mGy-cm. Converting to effective dose can be estimated depending on what body part is scanned.27 For example, the conversion factor for a CT of the abdomen and pelvis is ~15–17 microseiverts per milligray-centimeters, so a CT scan with a DLP of 622 mGy-cm (see Figure 12.1; Table 12.2) would be estimated to deliver an effective dose of approximately 10.6 mSv. To put this in perspective, the average yearly background radiation (without any medical radiation) in the United States is approximately 2–3 mSv (~6 mSv per year with medical radiation).11
| Imaging Modality | Approximate Delivered Dose (mSv) | Equivalent # of CXRs | Conversion Factor from DLP to Effective Dose in mSv per mGy-cm | Estimated Additional Lifetime Risk of Malignancy: 20 yo Female | Estimated additional lifetime risk of malignancy: 60 yo male |
|---|---|---|---|---|---|
| Chest X-ray, anteroposterior (AP) | 0.02 | 1 | |||
| Chest x-ray (CXR) – AP and lateral | 0.05–0.08 | 3 | |||
| Abdominal X-ray (”KUB”) | 1 | 50 | |||
| Yearly background radiation in U.S. without any medical imaging | 2–3* | 100–150 | |||
| Average yearly background radiation in U.S. including medical imaging | 6 | 300 | |||
| Head CT | 1–2 | 50–100 | 2.0–2.3 | 1 in 4,360 | 1 in 14,680 |
| Cervical spine CT | 4 | 200 | 5.1–5.9 | 1 in 2,390 | 1 in 8,030 |
| Chest CT pulmonary angiography | 7 | 350 | 17–19 | 1 in 330 | 1 in 1,770 |
| CT coronary angiogram | 12 | 600 | 17–19 | 1 in 150 | 1 in 790 |
| Chest CT “triple rule‐out” | 17 | 850 | 17–19 | 1 in 320 | 1 in 840 |
| Abdomen-pelvis CT | 10 | 500 | 15–18 | 1 in 500 | 1 in 1,330 |
| “Low dose” CT (for kidney stone) | 2 | 100 | 15–18 | 1 in 5,000 | 1 in 10,000 |
| “Pan scan” for trauma | 32 | 1,600 | variable | sum of risks of each scan | |
* Higher at higher elevations, for example ~10 mSv per year in Denver CO.
The Risk of Radiation from CT
A dramatic paper published in 2009 estimated that there could be as many as 29,000 future malignancies resulting from CTs performed in the United States in the year 2007 alone.28 Published population-based risks of developing a malignancy from CT scans depend on age, gender, and type of scan performed but have been estimated to range from approximately 1 in 150 for a 20-year-old female undergoing a CT angiogram to about 1 in 15,000 for a 60-year-old male undergoing a routine head CT (see Table 2).29
The estimates of risk were derived from the rates of malignancy in survivors of the atomic bomb attacks on Japan during World War II. The current generally accepted model is called the “no threshold linear model,” meaning each increased radiation dose is directly related to the increased risk of malignancy, and there is no single threshold amount that becomes more dangerous.30 However, the number of people in these cohorts developing cancer after receiving smaller doses of radiation was small, and a recent study has called this model into question.25 Larger epidemiologic studies are being conducted now on data from patients receiving CT scans that should yield more accurate estimates; some early studies show increased risks, although definitive estimates may take decades.11,31
Dose Variability from CT Scanning
Given the potential harms of ionizing radiation, radiological societies have endorsed the concept of using a dose that is “as low as reasonably achievable” (ALARA). While this is good in theory, in practice dosing may vary considerably for similar protocols between centers. In certain states (California, Texas, and Connecticut, to date), estimated radiation dose is a required element of the CT report. Additionally, the National Quality Forum (NQF) has endorsed “participation in a systemic dose index registry” as a quality measure, and the American College of Radiology founded the national Dose Imaging Registry (DIR, part of the National Registry of Diagnostic Radiology) in 2011.32 However, there is no actual requirement to lower dosing for a given protocol, and although these protocols are typically set by manufacturers, they may be modified by radiologists and CT technicians. A recent report from the DIR found that dosing for CTs looking for kidney stones could vary by a factor of 10, with less than 2% done using the low-dose protocols (less than ~3 mSv) that have been shown to be effective in diagnosing kidney stones while lowering risk from radiation.33,34 The range of doses used for a given study, therefore, can vary substantially from institution to institution.
Gender Considerations in Specific Clinical Situations
Imaging in Chest Pain and/or Dyspnea
Chest pain is the most common presenting ED chief complaint in adult males, and the second leading complaint in adult females, accounting for more than 5 million annual visits to U.S. EDs.1 The ED evaluation of chest pain and dyspnea involving advanced medical imaging has risen from 3.4% in 1999 to 15.9% in 2008.35 It is estimated that 40% of all medical radiation is related to cardiovascular imaging and intervention.36
Almost all chest pain evaluated in the ED will receive a plain chest X-ray. While the utility of this as a routine test has been questioned for more than 30 years,37 it is a common and inexpensive test without much radiation and typically functions as a screen for pneumonia, heart failure, pneumothorax, and aortic pathology. Whether imaging beyond a chest X-ray is obtained will depend on the suspicion for serious pathology, with the three most serious pathologies being acute coronary syndrome (ACS), pulmonary embolism (PE), and thoracic aortic dissection (TAD).
Echocardiography offers an attractive modality with no ionizing radiation that can be helpful in evaluation of the ED patient with chest pain and/or dyspnea. While access to formal echocardiography is often restricted in the ED setting,38 point-of-care ultrasound of the heart and lungs involves no radiation, is easily performed at the bedside, and can help provide diagnostic guidance in chest pain and dyspnea.39,40 Bedside echocardiography can exclude pericardial effusion, provide information about left ventricular function (affected by coronary disease), right ventricular strain (seen in large PEs), and aortic root size (associated with TAD), although echocardiography in particular represents a user-dependent diagnostic technology and the accuracy for these conditions will vary based on the experience of the clinician performing it.
While chest X-ray and ultrasound can be helpful, definitive diagnosis in coronary disease, PE, or TAD is typically obtained using imaging modalities with ionizing radiation. For PE and TAD, the test of choice is a contrast-enhanced CT scan of the chest. For coronary disease, cardiac catheterization with coronary angiography is the reference standard. However, in both PE and ACS evidenced-based approaches allow effective exclusion of disease in a substantial proportion of patients without using ionizing radiation. Data suggest that testing occurs more than is warranted by the evidence, and thus ionizing radiation is used more than is necessary. When ionizing radiation of the chest is over utilized, women are disproportionally at risk due to the exposure of breast tissue to the potentially carcinogenic radiation.41
Chest Pain concerning for Coronary Artery Disease
Heart disease is the leading cause of death in the United States for both men and women, and efforts to minimize missed diagnoses have led to extensive diagnostic testing. While cardiac catheterization and coronary angiography are the gold standard for diagnosis of coronary disease, these are usually not immediately performed on patients with chest pain unless the history is convincing or there is a documented myocardial infarction or positive stress test. The majority of chest pain seen in the ED can be considered “low-risk chest pain,” with a normal or nonspecific EKG and negative cardiac biomarkers.42 About 2% to 4% of people presenting with low-risk chest pain will ultimately be found to have significant coronary disease.43
If the history is concerning, it is often standard care in the United States to evaluate the coronary arteries either with a CT angiogram or a stress test.42 Options for a stress test include stress EKG, stress echo, and nuclear perfusion studies. More than10 million nuclear stress tests are currently performed annually in the United States, accounting for about 20% of the medical radiation received by the population, equivalent to half the radiation from all non-cardiac CT scanning put together.44 While nuclear stress tests are an effective tool for risk stratification, other options such as exercise echocardiography may be as effective and do not involve radiation. A recent survey of practitioners using nuclear stress tests found that only 7% were using a stress-first approach (i.e., only moving to a nuclear study if stress EKG or stress echocardiography were not conclusive) that could decrease radiation by as much as 75%.45
Increasingly, coronary CT angiogram (CCTA) is an option for detecting coronary disease.46 However, the radiation involved is higher than that typically delivered for a CT of the abdomen and pelvis and includes irradiation of the breasts in female patients.47 However, CCTA may allow clinicians to make an immediate disposition decision on a patient, avoiding lengthy observation admissions or stress testing, and can reduce negative cardiac catheterizations.46 There is also the potential for using CT to do a “triple rule-out” of all serious causes of chest pain (coronary disease, PE, and TAD). A triple rule-out CT protocol typically adds nearly 5 mSv to the radiation dose, but a recent study showed minimal added diagnostic benefit.48
Our recommendation is that in younger, lower-risk patients (including younger females) if provocative testing for coronary artery disease is desired, stress EKG, stress echo, or CT/nuclear strategies that minimize radiation should be considered first, with nuclear studies and CCTA reserved for cases for which these tests will not provide adequate risk stratification.
Chest Pain and/or Dyspnea concerning for Pulmonary Embolism
PE is estimated to occur in as many as 600,000 patients annually in the United States. CT pulmonary angiogram (CTPA) is now considered the diagnostic gold standard; since its introduction, the diagnosis of PE has nearly doubled in the United States. However, with increased diagnosis, mortality from PE has only been minimally affected, while complications from anticoagulation have risen, suggesting that we are treating more and potentially less clinically significant PEs.49 The authors of a study on breast irradiation from CT scanning for PE concluded that we should “encourage the judicious use of CT pulmonary angiography and lower doses and nonionizing radiation alternatives when appropriate.”41
Several clinical decision rules can stratify patients with symptoms of PE below the testing threshold for CT scan, either using clinical presentation alone or with d-dimer testing.50 The PE rule-out criteria (PERC) can stratify patients into a group that is low enough that a d-dimer or other imaging is likely to do more harm than good.50,51 The Wells score indicates whether patients have a low enough pretest probability that if a d-dimer is negative further imaging is not needed.52 Women are more likely to have a positive d-dimer (which is more likely to be falsely elevated), and this may contribute to the higher rate of imaging even if decision rules are followed.53 However, it is estimated that as many as a third of CTPAs (in both genders) could be avoided by utilizing these evidence-based rules.54 While women and men have an approximately equivalent incidence of PE, women undergo two-thirds of the CT scans for PE, and thus the diagnostic yield in women overall is substantially lower.55
Imaging in Abdominal Pain
Abdominal pain is the leading reason that adult females and the second leading reason that adult males visit the ED.1 Abdominal pain is a symptom that arises from a myriad of pathologies that can range from benign to serious both inside and outside of the abdominal cavity. A substantial number of patients with abdominal pain will not have a definitive diagnosis established during their ED visit. Like chest pain, evaluation of abdominal pain in the Emergency Department setting has seen a marked rise in advanced medical imaging. There are approximately 7 million annual ED visits for abdominal pain, with use of advanced medical imaging in its evaluation increasing from 19.9% to 44.3% between 1999 and 2008.35 Specifically, the use of CT for the evaluation of abdominal pain has seen the most dramatic rise, from 1.4% of visits in 1996 to 31.7% of visits in 2007.56 Despite this rise in CT imaging, there has been no change in the diagnosis of appendicitis, diverticulitis, or gallbladder disease, and admission rates for patients with abdominal pain have not changed.17 This suggests that utilization of CT for the evaluation of abdominal pain may not be adding value to patient-centered outcomes.
For some of the most common diagnosable causes of abdominal pain, use of ultrasound is a safe and effective alternative to CT, and the benefit: risk ratio may be higher for women. Biliary disease represents a disorder that is more common in women and is most effectively diagnosed with ultrasound. More than 20 million people in the United States suffer from biliary disease, and 70% of these are women.57 Biliary disease accounts for as many as 9% of admissions for abdominal pain, and 700,000 cholecystectomies are performed annually in the United States.58 While classically there is pain and tenderness in the right upper quadrant, midline epigastric pain is also consistent with biliary colic. Ultrasound is the initial test of choice if biliary disease is suspected. Point-of-care ultrasound has also been shown to be accurate for detection of biliary pathology and may be performed at the bedside by experienced practitioners.59,60
Appendicitis occurs in about 7% of the population in their lifetime. While appendicitis occurs more commonly in males (male-to-female ratio is 1.4:1), the diagnosis may be more challenging in women due to other etiologies of pain in the right lower quadrant.61 CT scanning is often considered the test of choice for appendicitis and is highly accurate. However, ultrasound is typically the first-line approach for pediatric and pregnant patients. A recent analysis of an ultrasound-first approach for appendicitis estimated that adopting such an approach could potentially avoid about 180 excess cancer deaths per year along with substantial cost savings.62 Such an approach would again be most effective in younger female patients.
The elderly with abdominal pain represent a special population for both genders, as they are more likely to harbor a serious diagnosis, particularly a vascular etiology.63 Abdominal aortic aneurysm (AAA) causes about 15,000 deaths annually in the United States with twice as many deaths occurring in men.64 While AAA is more common in men, women face an increased risk of rupture at smaller sizes.65 Ultrasound screening for AAA is currently recommended by the U.S. Preventive Services Task Force for asymptomatic men between the ages of 65 and 75 who have ever smoked, but routine screening is not recommended for women.66 Identification of AAA may be accomplished with bedside ultrasound in the ED.67 While ultrasound may help discern the etiology of abdominal pain in elderly men or women, a non-diagnostic ultrasound should generally be followed by a CT scan due to the prevalence of serious causes of pain and the lower lifetime risk of radiation in this population.
Sex-specific causes of lower abdominal pain must also be considered. In men with lower abdominal pain, a testicular etiology should be considered and a testicular examination should be performed. If there are concerning testicular findings on physical examination, scrotal ultrasound is typically the test of choice. For women, pelvic etiologies should be considered in most cases of abdominal pain, and ultrasound remains the test of choice for examination of the ovaries and uterus.
As noted earlier, CT scanning is frequently performed in the evaluation of abdominal pain in the emergency setting and is an excellent test, particularly for ruling out serious bowel and vascular pathologies. A recent review in radiology concluded that “CT can therefore be considered the primary technique for the diagnosis of acute abdominal pain, except in patients clinically suspected of having acute cholecystitis,” but it went on to note that “When costs and ionizing radiation are primary concerns, a possible strategy is to perform US as the initial technique in all patients with acute abdominal pain, with CT performed in all cases of nondiagnostic US.”68 Ultrasound as a primary or initial modality is used more frequently outside of the United States.
Imaging in Flank or Back Pain concerning for Renal Colic
Kidney stones are estimated to occur in about 10.6% of men and 7.1% of women during their lifetime.69 Despite CT having a lower diagnostic yield and a higher risk of later malignancy in women, a recent large study in the United States showed that women were undergoing 53.9% of CTs for renal colic and that the average age of CT scanning for renal colic was 45.33
Two reasons are cited to obtain a CT in a patient with suspected renal colic: (a) to determine the size and location of a stone, helping with prognosis and need for intervention and (b) to ensure no other important diagnoses are mimicking renal colic. However, approximately 80% of kidney stones will pass spontaneously, and those that are unlikely to pass will be clinically evident over time. While there are several mimics of renal colic, acutely important alternate causes of symptoms occur uncommonly (<3%) in patients without typical symptoms or evidence of infection or typical symptoms.70 A clinical prediction rule for ureteral stone (the “STONE score”) has been derived and validated that includes gender as an element.71 While women are still less likely to have a kidney stone, the incidence in women has been increasing, and women with a stone are more likely to be admitted.72–74
A recent position paper from the American College of Emergency Physicians (ACEP) asserts that “North American practice is to perform a noncontrast CT for new-onset flank pain/renal colic.”75 While this appears to be true in terms of practice patterns, the only evidence to support this practice is apparently a single study of 121 patients more than 10 years ago.76 Larger epidemiologic studies have shown no difference in rate of diagnosis, intervention, or admission despite the large increase in CT scanning for this condition.18,19 While low-dose CT protocols are recommended by both the ACR and the ACEP for evaluation of kidney stones, these are currently being used less than 2% of the time in the United States. The average effective dose in the United States for renal colic protocol CTs is approximately 11.2 mSv with a large variation in median institutional dose ranging from 3.5 to 19.8 mSv.33,75,77 For CT scans in younger persons, a 10 mSv effective dose may lead to an additional malignancy for approximately every 1,000 scans, potentially causing malignancy while diagnosing a usually self-limited condition.
Ultrasound offers an effective method for diagnoses in younger women with flank pain.78 While ultrasound may not always visualize a ureteral stone, it is effective in showing the presence of hydronephrosis, which is nearly always present with a ureteral stone that may require intervention.79,80 For women in particular, several mimics of renal colic are readily diagnosable using ultrasound. Adnexal torsion in particular is an emergent condition more likely to occur in younger women that is poorly diagnosed by CT scan, and for which ultrasound is the test of choice.81 Ovarian cysts without torsion or other pelvic pathologies represent possible mimics of renal colic as well. Appendicitis may be positively identified with ultrasound, although if the appendix is not visualized and the diagnosis remains on the differential, CT should be considered.
While more definitive evidence-based trials are warranted, it is our recommendation that younger patients, particularly younger women (younger than ages 40 to 50) should always undergo an ultrasound of the retroperitoneum, abdomen, and pelvis prior to consideration of CT scan in suspected renal colic (even new onset renal colic). Point-of-care ultrasound may allow this to be done at the bedside if appropriately trained clinicians are available. If ultrasound is non-diagnostic and symptoms are persistent or severe, CT may then be warranted.
Imaging in Trauma
Each year about 1 in 10 persons in the United States will visit an ED for a traumatic injury, with the majority male and about 70% younger than age 45.82 From 1998 to 2007, the use of advanced imaging (CT or MRI) increased from 6% to 15%; however, the trends in prevalence of serious injuries did not match this increase, and admission rates for trauma did not change.83 A recent study from Germany that looked retrospectively at whole body CT (sometimes called a “pan-scan”) in “polytrauma,” concluded that incorporation of this into routine care could increase survival.84 In high-mechanism trauma, where the likelihood of injury is relatively high, the benefit of CT no doubt outweighs the risk.85 However, because pan-scanning delivers a substantial dose of radiation (~32 mSv), often to younger patients who are more vulnerable to its carcinogenic effects, more selective use of pan-scanning has been advocated.86,87
An ACEP clinical policy published in 2011 on blunt abdominal trauma concluded that guidance for patients who may not require an abdominal CT included (level of evidence “C”):88
Patients with isolated abdominal trauma, for whom occult abdominal injury is being considered, are at low risk for adverse outcome and may not need abdominal CT scanning if the following are absent: abdominal tenderness, hypotension, altered mental status (Glasgow Coma Scale score <14), costal margin tenderness, abnormal chest radiograph, hematocrit <30% and hematuria.
The focused assessment with sonography in trauma (FAST) exam represents a rapid method to evaluate for serious injury to the thorax or abdomen in blunt or penetrating trauma without any ionizing radiation. FAST is recommended as the initial imaging modality if the patient is hemodynamically unstable.88 If the patient is hemodynamically stable, a positive FAST should be followed with a CT scan and may expedite appropriate evaluation and therapy. In a stable patient with a negative FAST exam, further evaluation depends on suspicion for significant injury. A negative initial FAST exam does not rule out all significant traumatic injuries. However, if there is a low suspicion for significant injury, a period of observation and a repeated negative FAST 6 hours later may be a reasonable approach.89
While we are not aware of studies specifically targeted at imaging based on gender in trauma, many of the same principles apply as in other areas of imaging. If there is a true suspicion for serious injury to the head, chest, abdomen, pelvis, or spine, CT is likely indicated regardless of age or gender. However if suspicion for injury is low, a careful physical examination and period of observation may safely replace CT scanning, with long-term benefits to patients and society. Young males, who most commonly incur a traumatic injury, are the group in which radiation risk could be most reduced by the appropriate use of CT.
Related posts:
Sex and Gender; Pharmacology, Efficacy, Toxicity, and Toxicology
Overcoming Resistance: Importance of Sex and Gender in Acute Infectious Illnesses
A Sex and Gender Based Perspective on Traumatic Injury
Digesting Sex and Gender: Gastroenterology
Sex and Gender in Medical Education: The Next Chapter
Special Populations
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
