Neuroradiolog y/Head and Neck Imaging • Original Research Salibi et al. Cancer From CT After Traumatic Brain Injury
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Neuroradiology/Head and Neck Imaging Original Research
Lifetime Attributable Risk of Cancer From CT Among Patients Surviving Severe Traumatic Brain Injury Patrick N. Salibi1 Vikas Agarwal2 David M. Panczykowski1 Ava M. Puccio1 Michael A. Sheetz 2 David O. Okonkwo1 Salibi PN, Agarwal V, Panczykowski DM, Puccio AM, Sheetz MA, Okonkwo DO
OBJECTIVE. The purpose of this study was to determine the lifetime attributable risk of cancer from CT among patients surviving severe traumatic brain injury. MATERIALS AND METHODS. A retrospective cross-sectional study was conducted with prospectively collected data on patients 16 years old and older admitted with a Glasgow coma scale score of 8 or less to a single level 1 trauma center from 2007 to 2010. The effective dose of each CT examination the patients underwent was predicted with literature-accepted effective dose values of standard helical CT protocols. The lifetime attributable risk of cancer and related mortality incurred as a result of CT were estimated with the cumulative effective dose incurred from the time of injury to a 1-year follow-up evaluation and with the approach established by the Biologic Effects of Ionizing Radiation VII report. RESULTS. The average patient was a 34-year-old man. The median number of CT examinations received during the first 12 months after injury was 20, and the average cumulative effective dose was 87 ± 45 mSv. This resulted in increases in the lifetime incidence of all cancer types from 45.5% to 46.3% and in the lifetime incidence of cancer-related mortality from 22.1% to 22.5%. CONCLUSION. Radiation exposure from the use of CT in the evaluation and management of severe traumatic brain injury causes negligible increases in lifetime attributable risk of cancer and cancer-related mortality. Treating physicians should not allow the concern for future risk of radiation-induced cancer to influence decisions regarding radiographic evaluation in the acute treatment of traumatic brain injury.
P
Keywords: cancer, CT, radiation, traumatic brain injury DOI:10.2214/AJR.12.10294 Received November 11, 2012; accepted after revision May 6, 2013. 1 Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA. 2 Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop St, Presby South Tower, 8N, Pittsburgh, PA 15213. Address correspondence to V. Agarwal ([email protected]).
AJR 2014; 202:397–400 0361–803X/14/2022–397 © American Roentgen Ray Society
atients receiving care for neurosurgical emergencies undergo myriad imaging studies with ionizing radiation throughout both the acute and chronic phases of treatment. This is especially true of patients with traumatic brain injury (TBI), who undergo extensive CT studies during their initial evaluation and hospitalization and adjunct assessments during clinical follow-up. The utility of CT studies in acute trauma management is indisputable; however, concern has mounted over the perceived increased risk of cancer and attendant mortality after exposure to ionizing radiation. Susceptibility to radiation is highest among younger patients and increases with greater cumulative radiation dose exposure [1, 2]. With regard to TBI, this may have major public health implications: More than 5 million TBI survivors currently reside in the United States, and patients with severe TBI who survive more than 6 months after injury have a life expectancy similar to that of their uninjured peers [3–5]. Concern over the po-
tential harm to patients from imaging procedures with ionizing radiation has led to efforts to reduce the frequency and radiation dose of CT studies [2, 6, 7]. Patients with TBI frequently receive large numbers of radiographic examinations during hospitalization; however, the degree to which this may augment their lifetime risk of cancer is unknown. Studies evaluating the risk of cancer attributable to radiographic examinations have focused mainly on general medical populations, and studies germane to neurotrauma have been limited to patients with neurovascular disease (e.g., stroke and aneurysmal subarachnoid hemorrhage) and general trauma [8–11]. To our knowledge, no study has yet specifically examined the cumulative radiation dose received during CT examinations for severe TBI, a disease process with a predilection for younger patients, who may be more susceptible to the risks of CT radiation. This study was conducted to determine the lifetime attributable risk of cancer from CT among patients surviving severe TBI.
AJR:202, February 2014 397
Salibi et al. Radiation exposure incurred from each CT study performed during the first 12 months after TBI was measured by effective dose. The effective dose quantifies the biologic effects of radiation absorbed by the anatomic region irradiated as determined by absorbed dose, radiation type, and energy, in this case during imaging studies performed with x-rays. Effective dose is one of the most frequently reported measurements of radiation exposure and allows comparison between CT and other imaging modalities. Literature-reported effective dose values were used so that results could be applicable to the broad spectrum of facilities treating these patients. Effective dose values for the CT techniques assessed in this study were obtained from three separate articles. Mettler et al. [12] were cited for effective dose values of the head (2 mSv), neck (3 mSv), chest (7 mSv), abdomen (8 mSv), pelvis (6 mSv), chest angiography (15 mSv), coronary angiography (16 mSv), and abdominal angiography (12 mSv). The following effective dose values were from Berrington de González et al. [13]: cervical spine (5 mSv), thoracic spine (7 mSv), lumbar spine (9 mSv), extremity (0.1 mSv), and pelvis angiography (9 mSv). Values from Mnyusiwalla et al. [14] were used for the following studies: head perfusion (4.9 mSv), head angiography (1.6 mSv), and neck angiography (3.8 mSv). The effective doses of individual studies were summed to obtain the cumulative effective dose each patient received. Lifetime attributable risk (LAR) of cancer is defined as the proportion of cancer in a population that can be ascribed to a specific risk factor, in this case radiation exposure during CT studies. LAR was calculated with the National Research Council Biologic Effects of Ionizing Radiation (BEIR) VII report formulas, which incorporate the magnitude of radiation exposure (measured in this study
5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
0
10
20
30 40 50 Age at Exposure (y)
60
70
80
Excess Cancer Mortality per 100,000 Patients Exposed to 0.1 Gy (no.)
This study was a retrospective cross-sectional study of prospectively collected clinical and radiographic data on all trauma patients admitted to a level 1 trauma center with severe TBI (Glasgow coma scale [GCS] score ≤ 8) between 2007 and 2010. The GCS, a standard neurologic assessment scale, is used to stratify TBI into mild, moderate, or severe on the basis of a score of 3–15, which takes into account verbal, eye, and motor capabilities. A GCS score of 3–8 is considered severe, moderate is 9–12, and 13–15 constitutes a mild brain injury. Data were collected with an institutional review board–approved protocol under the aegis of the brain trauma research center at our institution. All patients 16 years old and older who had a postresuscitation GCS score of 8 or less and had a signed informed consent form from a legal authorized representative to participate in the research center database were sampled for analysis. Patients were excluded from the study if they died or were lost to follow-up within 1 year of injury. All CT studies performed for each patient were tallied by complete review of the medical record. Body regions examined with dedicated CT studies included head, maxillofacial region, orbit, neck and cervical spine, chest, abdomen, pelvis, thoracic spine, lumbar spine, and extremities. CT angiographic studies of the head, neck, coronary arteries, chest, abdomen, and pelvis and cerebral perfusion studies were also included. Validation of data abstraction for 12 of the 67 patients was performed separately by two authors to ensure accuracy and consistency in radiologic scan counts. Imaging series created by postprocessing of chest and abdominal CT examinations were excluded to prevent double counting of the doses for these studies (specifically, thoracic and lumbar spinal series obtained as part of the initial trauma assessment).
Excess Cancer Incidence per 100,000 Patients Exposed to 0.1 Gy (no.)
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Materials and Methods
as effective dose), sex, and patient age at the time of the exposure [2]. Cancer types considered in calculation of LAR include lung, colon, stomach, liver, bladder, hematologic, and other malignancies. Sex-specific malignancies, such as breast, uterine, and ovarian cancer in women and prostate cancer in men, were also taken into consideration. The BEIR VII report draws on data collected from observational studies of Japanese atomic bomb survivors, populations near nuclear facilities during accidental releases of radioactive materials, workers with occupational exposures, and populations who receive exposure from diagnostic and therapeutic medical studies adjusted for U.S. population demographics. Tables 12D-1 and 12D-2 in the BEIR VII report were used to calculate cancer incidence and cancer-related mortality per 100,000 people exposed to 0.1 Gy (100 mSv) of radiation [2] (Fig. 1). The LAR for each patient was calculated with the cumulative effective dose received during the first year after injury, sex, and age at time of injury. These values were used to extrapolate exponential lines of best fit. In addition, the number of patients undergoing CT that would lead to the development of one case of radiation-induced cancer (number needed to harm) in a similar hypothetical cohort (also based on cumulative effective dose, age, and sex) was calculated. The BEIR VII report ascribes greater cancer risk to exposed women and inverse risk based on age at time of exposure. As such, LAR variance between the sexes and among those individuals above and below age 40 was assessed. Continuous demographic characteristics were reported as mean ± SD. Discrete (e.g., number of CT examinations) and multilevel ordinal measurements (e.g., GCS Score) were reported as median with interquartile range (IQR). All data were analyzed with Stata software (version 11, StataCorp).
2000 1800 1600 1400 1200 1000 800 600 400 200
A
0
0
10
20
30 40 50 Age at Exposure (y)
60
70
80
B
Fig. 1—Cancer statistics per 100,000 people exposed to 0.1 Gy (100 mSv) of radiation. A, Graph shows lifetime cancer incidence among female (dotted line) and male (solid line) patients. B, Graph shows cancer-related mortality among female (dotted line) and male (solid line) patients.
398
AJR:202, February 2014
Cancer From CT After Traumatic Brain Injury
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Other 19%
Other 22%
Head 29%
Pelvis 8% Head 51% Pelvis 12% Abdomen 9%
Cervical Spine 8%
Chest 7% Abdomen 17%
Cervical Spine 6%
A
Chest 12%
B
Fig. 2—Contribution of CT type. A, Chart shows contribution of CT type to total number of examinations performed. B, Chart shows contribution of CT type to cumulative effective dose received.
Results Patient Demographics During the study period, 195 patients admitted with severe TBI were enrolled into the research center study protocol. Thirty-seven patients were lost to follow-up before 12 months had elapsed, and another 91 patients died during the course of the study. These patients were excluded from analysis because of incomplete clinical and radiographic records. The BEIR VII LAR of cancer has a 2-year latency period for leukemia and a 5-year latency period for all solid tumors after radiation. All patients who died during the study period were within this latency period, negating radiationinduced cancer as a factor in their deaths. A total of 67 consecutively registered patients with severe TBI who survived 1 year with complete clinical and radiographic records were available for analysis. The mean age at injury was 34 ± 14 years, and 81% of the patients were male. Injury severity was assessed through GCS and injury severity score. The median GCS score was 7 (IQR, 1), and the median injury severity score was 43 (IQR 15). CT and Radiation Exposure The median number of examinations received by each patient during the first 12 months after injury was 20 (IQR, 8; range, 8–71). The mean cumulative effective dose in the sample was 87 ± 45 mSv (range, 34– 234 mSv). Unenhanced CT of the head was the most common procedure performed. Patients received a median of nine (IQR, 6) head CT examinations, resulting in 51% of each patient’s total examinations (range, 15–86%). Despite their frequency, head CT examina-
tions contributed a mean cumulative effective dose of only 25 ± 20 mSv, or 29% of the total cumulative effective dose (range, 5–65%). Of note, CT examinations of the chest, abdomen, and pelvis combined made up only 24% of all examinations performed (range, 6–63%) but contributed an average of 41% ± 14% to each patient’s cumulative effective dose (range, 13– 76%). Of the 18 CT examination types tallied, the five most common—head, cervical spine, chest, abdomen, pelvis—accounted for 81% of the CT effective dose and 78% of the cumulative effective dose patients received. Figure 2 shows the contribution of each type of CT examination to the total number of scans and the contribution to cumulative effective dose. Injury severity at admission was associated with neither number of CT studies performed nor cumulative effective dose (GCS score, p = 0.32 and 0.71; injury severity score, p = 0.73 and 0.20). Lifetime Attributable Risk of Cancer and Cancer-Related Mortality The LARs of cancer and cancer-related mortality were calculated with the patient’s age, sex, and cumulative effective dose received during CT studies in the first year after injury. The LAR was then extrapolated by use of known cancer and cancer-related mortality rates per 100,000 people exposed to 100 mSv of radiation within that same age-sex demographic. The baseline incidence of cancer for this age-sex demographic is 45.5% [2]. The mean LAR of all types of cancer due to CT exposure in the TBI patient population was 0.81% ± 0.54% (range, 0.12–2.49%). Thus radiation exposure from
CT use in the management of severe TBI contributes to a negligible increase in cancer incidence from 45.5% to 46.31% ± 0.54%. The mean incidence of cancer-related mortality for this age-sex demographic is 22.1% [2]. The mean LAR of cancer-related mortality due to CT exposure in the TBI patient population was 0.44% ± 0.27% (range, 0.09– 1.31%), leading to a negligible increase in the incidence of cancer-related mortality from 22.1% to 22.54% ± 0.27%. The attributable risk for each patient (based on cumulative effective dose, age at exposure, and sex) was used to calculate the number of patients with similar CT examination profiles that would lead to the development of one case of radiation-induced cancer (number needed to harm). On average, 183 ± 132 patients (range, 40–821 patients) with severe TBI would have to be exposed to similar cumulative effective doses before one patient contracted cancer attributable to CT examinations. Women had a slightly greater LAR of cancer and mortality than men did, though this difference was not statistically significant (0.94% vs 0.77%, p = 0.34; 0.48% vs 0.43%, p = 0.41). Patients younger than 40 years had significantly greater LAR of both cancer (0.99% vs 0.45%, p < 0.01) and (0.52% vs 0.29%, p < 0.01) death than did their older cohorts. Discussion For the 67 patients with severe TBI in the analysis, the average cumulative effective dose in the sample was 87 ± 45 mSv, leading to a negligible 0.81% increase in the LAR of cancer as a result of radiation exposure. For context, an individual is exposed to nearly 3
AJR:202, February 2014 399
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Salibi et al. mSv of background radiation per year, and exposure from a chest radiograph is approximately 0.1 mSv [2]. The most telling findings, though, is that the mean LAR of cancer-related mortality was only 0.44%, whereas the mortality of severe TBI itself is 20–39% [15, 16]. For every 285 patients with severe TBI undergoing the necessary CT studies, one patient may die decades later of cancer related to the scans obtained, whereas as many as 111 die of the TBI in the acute phase. Patients admitted to the hospital for severe TBI face life-threatening injuries as a result of the trauma. It is estimated that 20% of the 275,000 patients hospitalized for TBI every year have a severe head injury. The mortality among this subpopulation ranges from 20% to 39% [15, 16]. When comparing the mortality rate from TBI with the LAR of cancer mortality, one can appreciate that TBI poses a 60- to 115-fold increase in risk over radiation-induced cancer mortality. Thus the immediate risk of undiagnosed traumatic injuries far outweighs the risk of mortality from radiation-induced cancer in the future. Laack et al. [8] came to a similar conclusion. They determined that even among patients with less severe trauma (median injury severity score, 8), trauma mortality was 6 times as high as the estimated risk of radiation-induced cancer mortality. The small risk, however, does not negate the treating physician’s duty to thoughtfully consider the benefits and harm of all imaging procedures. In this study, we also evaluated whether multiple CT series should be used with caution in the evaluation of younger individuals, especially women. The reason for age discrepancy is twofold. First, the radiosensitivity of many organs has been found to decrease with age while at the same time there is a long latency period from radiation exposure to cancer development [1]. Thus younger survivors of severe TBI in the sample were both more sensitive and more likely to survive to cancer development and hence were assessed to have higher LAR. Across all ages, women have a higher LAR, primarily due to increased radiosensitivity in two tissues conferring the majority of cancer risk: lung and breast [2]. However, in this study sex did not confer a statistically significant difference in cancer risk, and there was less of a discrepancy in LAR of related mortality. The BEIR VII risk estimates were developed for use in the general U.S. population with typical life expectancy for age and sex.
400
Although the life expectancy among persons who have sustained severe TBI is not equal to that of uninjured cohorts, depending on functional status it may approach adjusted norms: a 16- to 28-year difference among severely disabled persons and a 5- to 10-year difference among moderately disabled persons, depending on age at injury [5]. Therefore, the results may be overestimates of the LAR of cancer and cancer-related mortality but are more or less applicable across the age spectrum. The limitations of this study center on the use of models for risk estimation. The LAR of cancer and related mortality are not based on actual epidemiologic data of malignancy rates after modern CT. This method of risk assessment would require a prohibitively long follow-up time, and given the relatively recent adoption of CT would not be available for some time. Until such an extensive study is completed, linear nonthreshold models such as those provided by BEIR VII are the best estimates of assessing risk in patient populations. The BEIR VII conclusions are based on populations with different radiation exposures from those analyzed in this study (atomic bomb, occupational, therapeutic) and are adjusted for U.S. demographics. Effective dose is used as a measure of generic risk for a given imaging procedure. It does not take into account the habitus, age, or sex of the patient. Effective dose should therefore not be used to measure an individual’s risk. Use of the BEIR VII model and effective dose requires inherent assumptions, which in turn limits the conclusions made. Even with these limitations, the widespread use of effective dose stems from its utility in comparisons of imaging techniques and body regions scanned. Thus this approach of using standardized models is the safest and most effective method of risk calculation currently available. Another limitation was that all of the patients selected for this study were treated at a single center, possibly restricting the applicability of the results because of both practitioner bias and scan protocols. To compensate, literatureaccepted effective dose values in risk assessment calculations were used to generalize the results to the national TBI population. References 1. Thompson DE, Mabuchi K, Ron E, et al. Cancer incidence in atomic bomb survivors. Part II. Solid tumors, 1958–1987. Radiat Res 1994; 137:S17–S67 2. Monson RR, Cleaver JE, Abrams HL, et al. Health risks from exposure to low levels of ioniz-
ing radiation: BEIR VII—phase 2. Washington, DC: National Academies Press, 2006 3. Brown AW, Leibson CL, Malec JF, Perkins PK, Diehl NN, Larson DR. Long-term survival after traumatic brain injury: a population-based analysis. NeuroRehabilitation 2004; 19:37–43 4. Faul M, Xu L, Wald MM, Coronado VG. Traumatic brain injury in the United States: emergency department visits, hospitalizations and deaths 2002–2006. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2010 5. Ratcliff G, Colantonio A, Escobar M, Chase S, Vernich L. Long-term survival following traumatic brain injury. Disabil Rehabil 2005; 27:305–314 6. Linton OW, Mettler FA. National conference on dose reduction in CT, with an emphasis on pediatric patients. AJR 2003; 181:321–329 7. Mettler FA, Thomadsen BR, Bhargavan M, et al. Medical radiation exposure in the U.S. in 2006: preliminary results. Health Phys 2008; 95:502–507 8. Laack TA, Thompson KM, Kofler JM, Bellolio MF, Sawyer MD, Laack NN. Comparison of trauma mortality and estimated cancer mortality from computed tomography during initial evaluation of intermediate-risk trauma patients. J Trauma 2011; 70:1362–1365 9. Einstein AJ, Henzlova MJ, Rajagopalan S. Estimating risk of cancer associated with radiation exposure from 64-slice computed tomography coronary angiography. JAMA 2007; 298:317–323 10. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med 2009; 169:2078–2086 11. Yamauchi-Kawara C, Fujii K, Aoyama T, Yamauchi M, Koyama S. Radiation dose evaluation in multidetector-row CT imaging for acute stroke with an anthropomorphic phantom. Br J Radiol 2010; 83:1029–1041 12. Mettler FA, Huda W, Yoshizumi TT, Mahesh M. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology 2008; 248:254–263 13. Berrington de González A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009; 169:2071–2077 14. Mnyusiwalla A, Aviv R, Symons S. Radiation dose from multidetector row CT imaging for acute stroke. Neuroradiology 2009; 51:635–640 15. Marmarou A, Lu J, Butcher I, et al. IMPACT database of traumatic brain injury: design and description. J Neurotrauma 2007; 24:239–250 16. Thurman D, Guerrero J. Trends in hospitalization associated with traumatic brain injury. JAMA 1999; 282:954–957
AJR:202, February 2014
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Neuroradiology/Head and Neck Imaging Original Research
Lifetime Attributable Risk of Cancer From CT Among Patients Surviving Severe Traumatic Brain Injury Patrick N. Salibi1 Vikas Agarwal2 David M. Panczykowski1 Ava M. Puccio1 Michael A. Sheetz 2 David O. Okonkwo1 Salibi PN, Agarwal V, Panczykowski DM, Puccio AM, Sheetz MA, Okonkwo DO
OBJECTIVE. The purpose of this study was to determine the lifetime attributable risk of cancer from CT among patients surviving severe traumatic brain injury. MATERIALS AND METHODS. A retrospective cross-sectional study was conducted with prospectively collected data on patients 16 years old and older admitted with a Glasgow coma scale score of 8 or less to a single level 1 trauma center from 2007 to 2010. The effective dose of each CT examination the patients underwent was predicted with literature-accepted effective dose values of standard helical CT protocols. The lifetime attributable risk of cancer and related mortality incurred as a result of CT were estimated with the cumulative effective dose incurred from the time of injury to a 1-year follow-up evaluation and with the approach established by the Biologic Effects of Ionizing Radiation VII report. RESULTS. The average patient was a 34-year-old man. The median number of CT examinations received during the first 12 months after injury was 20, and the average cumulative effective dose was 87 ± 45 mSv. This resulted in increases in the lifetime incidence of all cancer types from 45.5% to 46.3% and in the lifetime incidence of cancer-related mortality from 22.1% to 22.5%. CONCLUSION. Radiation exposure from the use of CT in the evaluation and management of severe traumatic brain injury causes negligible increases in lifetime attributable risk of cancer and cancer-related mortality. Treating physicians should not allow the concern for future risk of radiation-induced cancer to influence decisions regarding radiographic evaluation in the acute treatment of traumatic brain injury.
P
Keywords: cancer, CT, radiation, traumatic brain injury DOI:10.2214/AJR.12.10294 Received November 11, 2012; accepted after revision May 6, 2013. 1 Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA. 2 Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop St, Presby South Tower, 8N, Pittsburgh, PA 15213. Address correspondence to V. Agarwal ([email protected]).
AJR 2014; 202:397–400 0361–803X/14/2022–397 © American Roentgen Ray Society
atients receiving care for neurosurgical emergencies undergo myriad imaging studies with ionizing radiation throughout both the acute and chronic phases of treatment. This is especially true of patients with traumatic brain injury (TBI), who undergo extensive CT studies during their initial evaluation and hospitalization and adjunct assessments during clinical follow-up. The utility of CT studies in acute trauma management is indisputable; however, concern has mounted over the perceived increased risk of cancer and attendant mortality after exposure to ionizing radiation. Susceptibility to radiation is highest among younger patients and increases with greater cumulative radiation dose exposure [1, 2]. With regard to TBI, this may have major public health implications: More than 5 million TBI survivors currently reside in the United States, and patients with severe TBI who survive more than 6 months after injury have a life expectancy similar to that of their uninjured peers [3–5]. Concern over the po-
tential harm to patients from imaging procedures with ionizing radiation has led to efforts to reduce the frequency and radiation dose of CT studies [2, 6, 7]. Patients with TBI frequently receive large numbers of radiographic examinations during hospitalization; however, the degree to which this may augment their lifetime risk of cancer is unknown. Studies evaluating the risk of cancer attributable to radiographic examinations have focused mainly on general medical populations, and studies germane to neurotrauma have been limited to patients with neurovascular disease (e.g., stroke and aneurysmal subarachnoid hemorrhage) and general trauma [8–11]. To our knowledge, no study has yet specifically examined the cumulative radiation dose received during CT examinations for severe TBI, a disease process with a predilection for younger patients, who may be more susceptible to the risks of CT radiation. This study was conducted to determine the lifetime attributable risk of cancer from CT among patients surviving severe TBI.
AJR:202, February 2014 397
Salibi et al. Radiation exposure incurred from each CT study performed during the first 12 months after TBI was measured by effective dose. The effective dose quantifies the biologic effects of radiation absorbed by the anatomic region irradiated as determined by absorbed dose, radiation type, and energy, in this case during imaging studies performed with x-rays. Effective dose is one of the most frequently reported measurements of radiation exposure and allows comparison between CT and other imaging modalities. Literature-reported effective dose values were used so that results could be applicable to the broad spectrum of facilities treating these patients. Effective dose values for the CT techniques assessed in this study were obtained from three separate articles. Mettler et al. [12] were cited for effective dose values of the head (2 mSv), neck (3 mSv), chest (7 mSv), abdomen (8 mSv), pelvis (6 mSv), chest angiography (15 mSv), coronary angiography (16 mSv), and abdominal angiography (12 mSv). The following effective dose values were from Berrington de González et al. [13]: cervical spine (5 mSv), thoracic spine (7 mSv), lumbar spine (9 mSv), extremity (0.1 mSv), and pelvis angiography (9 mSv). Values from Mnyusiwalla et al. [14] were used for the following studies: head perfusion (4.9 mSv), head angiography (1.6 mSv), and neck angiography (3.8 mSv). The effective doses of individual studies were summed to obtain the cumulative effective dose each patient received. Lifetime attributable risk (LAR) of cancer is defined as the proportion of cancer in a population that can be ascribed to a specific risk factor, in this case radiation exposure during CT studies. LAR was calculated with the National Research Council Biologic Effects of Ionizing Radiation (BEIR) VII report formulas, which incorporate the magnitude of radiation exposure (measured in this study
5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
0
10
20
30 40 50 Age at Exposure (y)
60
70
80
Excess Cancer Mortality per 100,000 Patients Exposed to 0.1 Gy (no.)
This study was a retrospective cross-sectional study of prospectively collected clinical and radiographic data on all trauma patients admitted to a level 1 trauma center with severe TBI (Glasgow coma scale [GCS] score ≤ 8) between 2007 and 2010. The GCS, a standard neurologic assessment scale, is used to stratify TBI into mild, moderate, or severe on the basis of a score of 3–15, which takes into account verbal, eye, and motor capabilities. A GCS score of 3–8 is considered severe, moderate is 9–12, and 13–15 constitutes a mild brain injury. Data were collected with an institutional review board–approved protocol under the aegis of the brain trauma research center at our institution. All patients 16 years old and older who had a postresuscitation GCS score of 8 or less and had a signed informed consent form from a legal authorized representative to participate in the research center database were sampled for analysis. Patients were excluded from the study if they died or were lost to follow-up within 1 year of injury. All CT studies performed for each patient were tallied by complete review of the medical record. Body regions examined with dedicated CT studies included head, maxillofacial region, orbit, neck and cervical spine, chest, abdomen, pelvis, thoracic spine, lumbar spine, and extremities. CT angiographic studies of the head, neck, coronary arteries, chest, abdomen, and pelvis and cerebral perfusion studies were also included. Validation of data abstraction for 12 of the 67 patients was performed separately by two authors to ensure accuracy and consistency in radiologic scan counts. Imaging series created by postprocessing of chest and abdominal CT examinations were excluded to prevent double counting of the doses for these studies (specifically, thoracic and lumbar spinal series obtained as part of the initial trauma assessment).
Excess Cancer Incidence per 100,000 Patients Exposed to 0.1 Gy (no.)
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Materials and Methods
as effective dose), sex, and patient age at the time of the exposure [2]. Cancer types considered in calculation of LAR include lung, colon, stomach, liver, bladder, hematologic, and other malignancies. Sex-specific malignancies, such as breast, uterine, and ovarian cancer in women and prostate cancer in men, were also taken into consideration. The BEIR VII report draws on data collected from observational studies of Japanese atomic bomb survivors, populations near nuclear facilities during accidental releases of radioactive materials, workers with occupational exposures, and populations who receive exposure from diagnostic and therapeutic medical studies adjusted for U.S. population demographics. Tables 12D-1 and 12D-2 in the BEIR VII report were used to calculate cancer incidence and cancer-related mortality per 100,000 people exposed to 0.1 Gy (100 mSv) of radiation [2] (Fig. 1). The LAR for each patient was calculated with the cumulative effective dose received during the first year after injury, sex, and age at time of injury. These values were used to extrapolate exponential lines of best fit. In addition, the number of patients undergoing CT that would lead to the development of one case of radiation-induced cancer (number needed to harm) in a similar hypothetical cohort (also based on cumulative effective dose, age, and sex) was calculated. The BEIR VII report ascribes greater cancer risk to exposed women and inverse risk based on age at time of exposure. As such, LAR variance between the sexes and among those individuals above and below age 40 was assessed. Continuous demographic characteristics were reported as mean ± SD. Discrete (e.g., number of CT examinations) and multilevel ordinal measurements (e.g., GCS Score) were reported as median with interquartile range (IQR). All data were analyzed with Stata software (version 11, StataCorp).
2000 1800 1600 1400 1200 1000 800 600 400 200
A
0
0
10
20
30 40 50 Age at Exposure (y)
60
70
80
B
Fig. 1—Cancer statistics per 100,000 people exposed to 0.1 Gy (100 mSv) of radiation. A, Graph shows lifetime cancer incidence among female (dotted line) and male (solid line) patients. B, Graph shows cancer-related mortality among female (dotted line) and male (solid line) patients.
398
AJR:202, February 2014
Cancer From CT After Traumatic Brain Injury
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Other 19%
Other 22%
Head 29%
Pelvis 8% Head 51% Pelvis 12% Abdomen 9%
Cervical Spine 8%
Chest 7% Abdomen 17%
Cervical Spine 6%
A
Chest 12%
B
Fig. 2—Contribution of CT type. A, Chart shows contribution of CT type to total number of examinations performed. B, Chart shows contribution of CT type to cumulative effective dose received.
Results Patient Demographics During the study period, 195 patients admitted with severe TBI were enrolled into the research center study protocol. Thirty-seven patients were lost to follow-up before 12 months had elapsed, and another 91 patients died during the course of the study. These patients were excluded from analysis because of incomplete clinical and radiographic records. The BEIR VII LAR of cancer has a 2-year latency period for leukemia and a 5-year latency period for all solid tumors after radiation. All patients who died during the study period were within this latency period, negating radiationinduced cancer as a factor in their deaths. A total of 67 consecutively registered patients with severe TBI who survived 1 year with complete clinical and radiographic records were available for analysis. The mean age at injury was 34 ± 14 years, and 81% of the patients were male. Injury severity was assessed through GCS and injury severity score. The median GCS score was 7 (IQR, 1), and the median injury severity score was 43 (IQR 15). CT and Radiation Exposure The median number of examinations received by each patient during the first 12 months after injury was 20 (IQR, 8; range, 8–71). The mean cumulative effective dose in the sample was 87 ± 45 mSv (range, 34– 234 mSv). Unenhanced CT of the head was the most common procedure performed. Patients received a median of nine (IQR, 6) head CT examinations, resulting in 51% of each patient’s total examinations (range, 15–86%). Despite their frequency, head CT examina-
tions contributed a mean cumulative effective dose of only 25 ± 20 mSv, or 29% of the total cumulative effective dose (range, 5–65%). Of note, CT examinations of the chest, abdomen, and pelvis combined made up only 24% of all examinations performed (range, 6–63%) but contributed an average of 41% ± 14% to each patient’s cumulative effective dose (range, 13– 76%). Of the 18 CT examination types tallied, the five most common—head, cervical spine, chest, abdomen, pelvis—accounted for 81% of the CT effective dose and 78% of the cumulative effective dose patients received. Figure 2 shows the contribution of each type of CT examination to the total number of scans and the contribution to cumulative effective dose. Injury severity at admission was associated with neither number of CT studies performed nor cumulative effective dose (GCS score, p = 0.32 and 0.71; injury severity score, p = 0.73 and 0.20). Lifetime Attributable Risk of Cancer and Cancer-Related Mortality The LARs of cancer and cancer-related mortality were calculated with the patient’s age, sex, and cumulative effective dose received during CT studies in the first year after injury. The LAR was then extrapolated by use of known cancer and cancer-related mortality rates per 100,000 people exposed to 100 mSv of radiation within that same age-sex demographic. The baseline incidence of cancer for this age-sex demographic is 45.5% [2]. The mean LAR of all types of cancer due to CT exposure in the TBI patient population was 0.81% ± 0.54% (range, 0.12–2.49%). Thus radiation exposure from
CT use in the management of severe TBI contributes to a negligible increase in cancer incidence from 45.5% to 46.31% ± 0.54%. The mean incidence of cancer-related mortality for this age-sex demographic is 22.1% [2]. The mean LAR of cancer-related mortality due to CT exposure in the TBI patient population was 0.44% ± 0.27% (range, 0.09– 1.31%), leading to a negligible increase in the incidence of cancer-related mortality from 22.1% to 22.54% ± 0.27%. The attributable risk for each patient (based on cumulative effective dose, age at exposure, and sex) was used to calculate the number of patients with similar CT examination profiles that would lead to the development of one case of radiation-induced cancer (number needed to harm). On average, 183 ± 132 patients (range, 40–821 patients) with severe TBI would have to be exposed to similar cumulative effective doses before one patient contracted cancer attributable to CT examinations. Women had a slightly greater LAR of cancer and mortality than men did, though this difference was not statistically significant (0.94% vs 0.77%, p = 0.34; 0.48% vs 0.43%, p = 0.41). Patients younger than 40 years had significantly greater LAR of both cancer (0.99% vs 0.45%, p < 0.01) and (0.52% vs 0.29%, p < 0.01) death than did their older cohorts. Discussion For the 67 patients with severe TBI in the analysis, the average cumulative effective dose in the sample was 87 ± 45 mSv, leading to a negligible 0.81% increase in the LAR of cancer as a result of radiation exposure. For context, an individual is exposed to nearly 3
AJR:202, February 2014 399
Downloaded from www.ajronline.org by University of Utah on 07/04/14 from IP address 155.97.178.73. Copyright ARRS. For personal use only; all rights reserved
Salibi et al. mSv of background radiation per year, and exposure from a chest radiograph is approximately 0.1 mSv [2]. The most telling findings, though, is that the mean LAR of cancer-related mortality was only 0.44%, whereas the mortality of severe TBI itself is 20–39% [15, 16]. For every 285 patients with severe TBI undergoing the necessary CT studies, one patient may die decades later of cancer related to the scans obtained, whereas as many as 111 die of the TBI in the acute phase. Patients admitted to the hospital for severe TBI face life-threatening injuries as a result of the trauma. It is estimated that 20% of the 275,000 patients hospitalized for TBI every year have a severe head injury. The mortality among this subpopulation ranges from 20% to 39% [15, 16]. When comparing the mortality rate from TBI with the LAR of cancer mortality, one can appreciate that TBI poses a 60- to 115-fold increase in risk over radiation-induced cancer mortality. Thus the immediate risk of undiagnosed traumatic injuries far outweighs the risk of mortality from radiation-induced cancer in the future. Laack et al. [8] came to a similar conclusion. They determined that even among patients with less severe trauma (median injury severity score, 8), trauma mortality was 6 times as high as the estimated risk of radiation-induced cancer mortality. The small risk, however, does not negate the treating physician’s duty to thoughtfully consider the benefits and harm of all imaging procedures. In this study, we also evaluated whether multiple CT series should be used with caution in the evaluation of younger individuals, especially women. The reason for age discrepancy is twofold. First, the radiosensitivity of many organs has been found to decrease with age while at the same time there is a long latency period from radiation exposure to cancer development [1]. Thus younger survivors of severe TBI in the sample were both more sensitive and more likely to survive to cancer development and hence were assessed to have higher LAR. Across all ages, women have a higher LAR, primarily due to increased radiosensitivity in two tissues conferring the majority of cancer risk: lung and breast [2]. However, in this study sex did not confer a statistically significant difference in cancer risk, and there was less of a discrepancy in LAR of related mortality. The BEIR VII risk estimates were developed for use in the general U.S. population with typical life expectancy for age and sex.
400
Although the life expectancy among persons who have sustained severe TBI is not equal to that of uninjured cohorts, depending on functional status it may approach adjusted norms: a 16- to 28-year difference among severely disabled persons and a 5- to 10-year difference among moderately disabled persons, depending on age at injury [5]. Therefore, the results may be overestimates of the LAR of cancer and cancer-related mortality but are more or less applicable across the age spectrum. The limitations of this study center on the use of models for risk estimation. The LAR of cancer and related mortality are not based on actual epidemiologic data of malignancy rates after modern CT. This method of risk assessment would require a prohibitively long follow-up time, and given the relatively recent adoption of CT would not be available for some time. Until such an extensive study is completed, linear nonthreshold models such as those provided by BEIR VII are the best estimates of assessing risk in patient populations. The BEIR VII conclusions are based on populations with different radiation exposures from those analyzed in this study (atomic bomb, occupational, therapeutic) and are adjusted for U.S. demographics. Effective dose is used as a measure of generic risk for a given imaging procedure. It does not take into account the habitus, age, or sex of the patient. Effective dose should therefore not be used to measure an individual’s risk. Use of the BEIR VII model and effective dose requires inherent assumptions, which in turn limits the conclusions made. Even with these limitations, the widespread use of effective dose stems from its utility in comparisons of imaging techniques and body regions scanned. Thus this approach of using standardized models is the safest and most effective method of risk calculation currently available. Another limitation was that all of the patients selected for this study were treated at a single center, possibly restricting the applicability of the results because of both practitioner bias and scan protocols. To compensate, literatureaccepted effective dose values in risk assessment calculations were used to generalize the results to the national TBI population. References 1. Thompson DE, Mabuchi K, Ron E, et al. Cancer incidence in atomic bomb survivors. Part II. Solid tumors, 1958–1987. Radiat Res 1994; 137:S17–S67 2. Monson RR, Cleaver JE, Abrams HL, et al. Health risks from exposure to low levels of ioniz-
ing radiation: BEIR VII—phase 2. Washington, DC: National Academies Press, 2006 3. Brown AW, Leibson CL, Malec JF, Perkins PK, Diehl NN, Larson DR. Long-term survival after traumatic brain injury: a population-based analysis. NeuroRehabilitation 2004; 19:37–43 4. Faul M, Xu L, Wald MM, Coronado VG. Traumatic brain injury in the United States: emergency department visits, hospitalizations and deaths 2002–2006. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2010 5. Ratcliff G, Colantonio A, Escobar M, Chase S, Vernich L. Long-term survival following traumatic brain injury. Disabil Rehabil 2005; 27:305–314 6. Linton OW, Mettler FA. National conference on dose reduction in CT, with an emphasis on pediatric patients. AJR 2003; 181:321–329 7. Mettler FA, Thomadsen BR, Bhargavan M, et al. Medical radiation exposure in the U.S. in 2006: preliminary results. Health Phys 2008; 95:502–507 8. Laack TA, Thompson KM, Kofler JM, Bellolio MF, Sawyer MD, Laack NN. Comparison of trauma mortality and estimated cancer mortality from computed tomography during initial evaluation of intermediate-risk trauma patients. J Trauma 2011; 70:1362–1365 9. Einstein AJ, Henzlova MJ, Rajagopalan S. Estimating risk of cancer associated with radiation exposure from 64-slice computed tomography coronary angiography. JAMA 2007; 298:317–323 10. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med 2009; 169:2078–2086 11. Yamauchi-Kawara C, Fujii K, Aoyama T, Yamauchi M, Koyama S. Radiation dose evaluation in multidetector-row CT imaging for acute stroke with an anthropomorphic phantom. Br J Radiol 2010; 83:1029–1041 12. Mettler FA, Huda W, Yoshizumi TT, Mahesh M. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology 2008; 248:254–263 13. Berrington de González A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009; 169:2071–2077 14. Mnyusiwalla A, Aviv R, Symons S. Radiation dose from multidetector row CT imaging for acute stroke. Neuroradiology 2009; 51:635–640 15. Marmarou A, Lu J, Butcher I, et al. IMPACT database of traumatic brain injury: design and description. J Neurotrauma 2007; 24:239–250 16. Thurman D, Guerrero J. Trends in hospitalization associated with traumatic brain injury. JAMA 1999; 282:954–957
AJR:202, February 2014
