The diagnosis of inflammatory bowel disease (IBD) typically occurs during childhood and adolescence. The initial workup as well as the management of complications related to IBD often involves imaging studies. These studies, particularly the abdominal computed tomography (CT) and small bowel series, involve an appreciable amount of radiation. Concern regarding the risks of radiation exposure in the pediatric population has heightened as the use of these imaging studies has become more widespread. Studies of atomic bomb survivors observed for nearly 60 years have demonstrated that the malignancy-inducing potential of ionizing radiation supports a linear dose relation with no threshold, with an excess of solid tumor development even at low doses. In addition, the relative risk of malignancy persists throughout life, is greater in girls than in boys, declines with age, and is the highest for those exposed during childhood (1). Irradiation with as little as 50 mSv from imaging-related radiation exposure (IRRE) has been implicated in the development of certain solid tumors, particularly of the large bowel and bladder (2). In a study involving 15 developed countries, it was estimated that between 0.6% and 1.8% of all malignancies occurred as a result of diagnostic medical radiation (3). A recent study in adults with IBD demonstrated a significant increase in the use of CT in patients with Crohn disease (CD) during the last 15 years (4). CT enteroclysis, a new technique that combines CT with fluoroscopy-guided infusion of contrast into the small bowel, has been heralded as a preferred method of depicting mucosal abnormalities, bowel thickening, fistulae, and extraintestinal manifestations of CD (5). This technique, however, comes at the cost of increased radiation exposure (6).
IBD itself has long been considered an established risk factor for certain cancers such as colorectal carcinoma. Furthermore, common medications involved in treating IBD, such as 6-mercaptopurine, azathioprine, and infliximab, are associated with an increased risk of lymphoma (7,8). Children, in comparison to adults, have more biologically active tissue with less intervening space between susceptible organs and thus are thought to be at increased risk for radiation-related malignancy (9,10). Moreover, the pediatric age group has a lifetime to develop radiation-induced complications. Children with IBD, therefore, can be considered a subgroup of patients at increased risk for the development of cancer.
At the present time, there is a paucity of information regarding the degree of exposure to radiation in children with IBD. The present retrospective study attempts to estimate the amount of radiation exposure in children with CD and ulcerative colitis (UC) seen at a single tertiary care pediatric center. In addition, we sought to determine whether identifiable clinical characteristics and laboratory findings could predict which children with IBD receive the highest amounts of imaging-related radiation.
PATIENTS AND METHODS
Data were derived from an institutional review board–approved retrospective chart review. Subjects were selected from the population of pediatric patients with IBD followed by the members of the Division of Pediatric Gastroenterology at Cohen Children's Medical Center in New Hyde Park, NY. To eliminate selection bias, the charts of consecutive children with IBD seen by members of the division between January 1, 2008 and May 30, 2008 were evaluated. Data recorded included pertinent demographic characteristics and the number and type of all of the diagnostic radiographic tests. The total number of tests involving exposure to radiation since diagnosis was assessed for each subject by review of each child's inpatient and outpatient chart and companion records maintained by the radiology department. All of the charts were carefully reviewed to identify instances in which radiographic procedures were performed outside the health system's facilities. Details regarding each patient's initial clinical course including symptoms, laboratory evaluation, disease severity, and therapy were recorded. In addition, charts were reviewed to determine duration of follow-up, nature and extent of disease, and history of intraabdominal IBD-related surgery since diagnosis. Subjects were included in the study if they carried the diagnosis of IBD for at least 1 year and were younger than 18 years old at the time of diagnosis. Subjects were excluded if their medical records were deficient in terms of documentation of radiology tests and/or clinical course. In addition, any subject who received a significant amount of radiation due to a diagnosis other than IBD was excluded.
The effective dose of radiation received from each x-ray, contrast study, and nuclear imaging test was estimated from typical effective doses (assessed in mSv) from the published literature (11–13) (Table 1). CT doses were based on our institution's present CT dose index values (CTDI). CTDI provides an indication of the magnitude of radiation dose delivered to a patient as a function of the particular CT scanner model and operation specifics. The CT scanners at our institution provide dose information including CTDI (units mGy) and dose-length product units (mGy, cm) for each prescribed scan series. Effective dose estimates were derived from the dose-length product using recently published coefficients specific to the pediatric population. These values vary according to age and body region and are the result of the work done by the European Commission Concerted Action on CT (14). CDTI values were recorded at our institution since May 2005. Radiation exposure from CT tests done at outside institutions or before the use of CTDI was estimated by using the most recent norms from the published literature (12). Cumulative effective dose (CED) was calculated for each subject by summing the effective doses of radiation from the time of presentation with IBD until the time of chart review.
Exposure to radiation was analyzed using 2 cutoffs. Subjects with CED in the upper quartile of the study population were compared with those with a CED in the lower 3 quartiles. In addition, subjects with a high CED, defined as ≥50 mSv (2), were compared with subjects with a CED <50 mSv. The analysis related to upper quartile exposure was carried out for all of the patients. Because only 1 subject with UC had a CED ≥50 mSv, the high CED analysis was limited to subjects with CD. For both endpoints, the analysis was carried out in a similar manner.
Univariate screening of factors associated with lifetime CED greater than or equal to either cutoff was performed using the χ2 test or Fisher exact test, as appropriate, for categorical factors, and the Mann-Whitney test for continuous factors. Factors that were significantly associated with a CED greater than or equal to either cutoff in the univariate analysis (P < 0.05) were included in a multivariate logistic regression model. Best subsets selection was used to select the final multivariate model and was carried out using SAS version 9.2 (SAS Institute Inc, Cary, NC). This analysis was performed by computing the score χ2 for each model and identifying the subset with k* variables, where the change in the score χ2 for the subset with k*+1 variables was <5%. The final model is the model with k* variables.
Complete records were available for 257 of the 372 initially screened subjects. Demographic and clinical characteristics of these 257 pediatric patients with IBD are presented in Table 2. The demographic characteristics of the children with CD and UC were similar in terms of age, sex, severity of disease at time of diagnosis, and length of follow-up. Compared with subjects with UC, those with CD were more likely to have a history of IBD-related surgery (P = 0.003) and a family history of IBD (P = 0.003).
Figure 1 illustrates the distribution of radiation exposure in our study population. The mean CED was 17.56 ± 15.91 mSv. The mean CED was greater in patients with CD than in those with UC (20.5 ± 17.5 vs 11.7 ± 9.9 mSv, respectively; P < 0.0001). Fifteen children (5.8%) had a CED ≥50 mSv, including 14 of 171 (8.2%) with CD and 1 of 86 (1.2%) with UC (P = 0.02).
Univariate analysis was used to examine the degree of radiation exposure from IRRE in terms of quartiles. Patients in the upper quartile had a CED ≥21.2 mSv. Several clinical characteristics and laboratory findings were associated with radiation exposure in this range (Table 3). Type of IBD, age at diagnosis, disease duration, body mass index at diagnosis, physician's global assessment at diagnosis, history of IBD surgery, hemoglobin at diagnosis, hematocrit at diagnosis, erythrocyte sedimentation rate at diagnosis, elevated platelet count at diagnosis, corticosteroids at diagnosis, and failure to thrive at diagnosis were included in a multivariate logistic regression model. IBD-related surgical history (P < 0.0001), longer disease duration (P = 0.0001), elevated erythrocyte sedimentation rate at diagnosis (P = 0.007), and use of corticosteroids at diagnosis (P = 0.009) were associated with a CED in the upper quartile (Table 4).
Fourteen of the 15 children with CED ≥50 mSv had CD. Analysis was therefore restricted to the subpopulation of study subjects with CD. Univariate analysis demonstrated that the following factors were associated with a CED ≥50 mSv: female sex (P = 0.04), stricturing and/or penetrating disease (P = 0.0001), IBD-related surgery (P = 0.0001), longer disease duration (P = 0.0005), and elevated platelet count (>440) at the time of diagnosis (P = 0.005) (Table 5). Sex, disease behavior, surgical history, elevated platelet count at diagnosis, and disease duration were included in a multivariate logistic regression model. On the basis of best subset selection, the final model included surgical history, elevated platelet count at diagnosis, and disease duration. Of these, only IBD-related surgery (odds ratio 42, 95% confidence interval 8–223, P < 0.0001) and elevated platelet count at diagnosis (odds ratio 16, 95% confidence interval 1.5–175, P = 0.02) remained significant when analyzed by the multivariate model.
Seventeen of the 257 children had no imaging at all. Of the 240 children who had imaging, those with high CED received 61% of their radiation exposure (mean 43 mSv) from CT scans, whereas those with CED <50 mSv received 75% of their radiation from upper gastrointestinal series (mean 5.8 mSv) and small bowel series (mean 5.2 mSv) (Fig. 2).
The significant increase in use of diagnostic radiography in the last 2 decades has led to concern about the malignant potential of ionizing radiation. Studies of survivors of atomic bombs dropped on Japan in 1945 and more recently of radiation workers in the nuclear industry have confirmed an association between radiation dose and mortality from cancer (1,15). It is therefore conceivable that ionizing radiation to the abdomen and pelvis may be contributing to the development of certain cancers in patients with IBD. Children may be especially at risk because they are inherently more radiosensitive and because they have more remaining years of life during which a radiation-induced cancer could develop (9). Concern regarding an apparent increased risk for lymphoma, including hepatosplenic T-cell lymphoma, a rare, lethal cancer reported in young patients with IBD, makes it critically important that clinicians investigate the various factors that may promote tumor development (16). It has been suggested that drugs such as 6-mercaptopurine and infliximab may play a role in the development of hepatosplenic T-cell lymphoma in patients with IBD. How such drugs act pathogenetically to induce lymphoma has not been elucidated, and it is possible that the treatments serve as surrogate markers for patients with complicated disease, the same patients likely to undergo repeated diagnostic studies. Support for such a hypothesis may be found in other disease populations, in which widespread use of anti-TNF therapy with concomitant immunomodulators does not appear to convey the same risk of malignancy as is seen in the IBD population (17). Because patients with IBD undergo more testing that exposes the abdomen and pelvis to ionizing radiation, IRRE should be considered as a possible factor in cancer development.
We are aware of only 1 study that addresses the issue of IRRE in children with IBD. Palmer et al (18) used an insurance claims database to approximate the number of radiologic examinations obtained for children with IBD. They found that 23% of children with UC and 34% of children with CD were exposed to a moderate amount of IRRE. A moderate dose was defined as having had at least 1 CT or 3 fluoroscopic procedures during a 2-year period. In our study we quantified the amount of radiation exposure in millisieverts from IRRE from the time of diagnosis with IBD. Duration of follow-up was similar in our subjects with CD and UC (5.3 ± 3.5 and 5.4 ± 3.4 years, respectively), yet the mean dose of radiation was significantly higher in the subjects with CD. These results are similar to a recent study in adults with IBD. Levi et al (19) found that patients with CD received an average of 21.1 mSv, whereas patients with UC were exposed to 15 mSv during an average follow-up period of >5 years. Although our study demonstrated similar amounts of IRRE, it is concerning that this degree of exposure occurred in the pediatric age group. The average child with CD in our study population had already accumulated nearly half of the total CED associated with an increased risk for malignancy while still studied in a pediatric practice. Nearly 10% had already exceeded this amount of exposure to ionizing radiation. In view of the chronic nature of IBD, we find this amount of exposure to be alarming. Furthermore, although the threshold for the high-CED group in our study was 50 mSv, exposure less than this range should not be considered benign. The risks of low-dose ionizing radiation are not completely understood and are likely most harmful in the youngest subjects. The US National Research Council estimates that for every 1000 patients undergoing a 10-mSv CT examination of the abdomen, 1 patient will develop a radiation-induced neoplasm in his or her lifetime (20).
In a similar fashion to studies done in adults, we sought to examine factors associated with increased exposure to IRRE. Our study had in common with the previous studies the finding that IBD-related surgery was independently associated with being in a high-CED subgroup. This is likely due to repeated assessment before surgical management. For example, up-to-date imaging is common in the context of abscess formation, in which evaluating the success of intravenous antibiotic therapy and/or drainage by interventional radiology is imperative.
Our study has several limitations. The CED for each subject was an approximation. Some tests done at outside institutions may not have been recorded in the patient's chart. Thus, the CED may have been underestimated in some subjects. Moreover, radiation exposure from tests performed at an outside center was estimated based on norms from the established literature. Depending on the techniques used in those centers, the estimated radiation dose may have been greater than or less than the true exposure. Second, the present study represents the experience of 1 center and may be in part a reflection of 1 group's practice patterns. As a tertiary care center with an active IBD clinical and research practice, the subjects in our study population may represent a more severely ill population than seen in many pediatric practices, and as a consequence may require more imaging studies. Alternatively, the practice is based in a children's hospital with a dedicated pediatric radiology faculty that is familiar with strategies aimed at reducing the amount of radiation incurred during radiologic procedures. This is in line with the Image Gently campaign, which encourages radiological protocols that minimize radiation exposure to the lowest extent possible (21). Institutions without a pediatric radiologist on staff may be less likely to be familiar with such guidelines. IRRE may thus be higher in pediatric patients with IBD studied at other institutions.
This study's radiation exposure data were strictly cumulative; year-by-year exposure was not assessed. It is plausible that the bulk of radiation exposure in some patients occurs during certain peak periods such as at the time of diagnosis or disease complications. Potential fluctuations in exposure over time were thus not captured in the present study. However, epidemiological data suggest that the risk of malignancy from ionizing radiation is based solely on one's lifetime cumulative exposure, irrespective of timing (1).
Although children with CED ≥50 mSv incurred a greater level of radiation exposure from all forms of imaging considered in the present study compared with children with CED ≤50 mSv, the greatest discrepancy came from CT scans, which accounted for 61% of the IRRE in this subgroup. In light of this, it is important for physicians caring for children with IBD to consider comparable tests with little or no IRRE whenever possible. Fortunately, a number of imaging modalities such as magnetic resonance imaging, ultrasound, and capsule endoscopy are becoming viable options in many centers. Magnetic resonance enterography is particularly useful in providing information regarding disease activity and can aid in distinguishing inflammation from fibrosis in areas of bowel wall thickening (22). Unfortunately, most children younger than 7 years require general anesthesia for prolonged and multiple sequences, thus limiting its use. In the hands of an experienced radiologist, ultrasound can be accurate for the detection of IBD and can identify disease complications such as small bowel obstruction and abscess formation (23). Capsule endoscopy has the benefit of providing direct images of small bowel mucosa, and the recent development of the patency capsule has lessened concerns of capsule retention in a strictured segment (24).
Children with IBD represent a population at increased risk for malignancy, one that may be enhanced by exposure to high doses of imaging-related radiation. Factors associated with significant amounts of radiation exposure from imaging tests include surgery and CD. It is important for health care providers to be aware of these risks and use strategies that minimize radiation exposure whenever possible.
1. Preston DL, Shimizu Y, Pierce DA, et al
. Studies of atomic bomb survivors. Report 13: solid cancer and noncancer disease mortality: 1950–1997. Rad Res 2003; 160:381–407.
2. Brenner DJ, Doll R, Goodhead DT, et al
. Cancer risks attributable to low doses of ionizing radiation: assessing what we really know. Proc Natl Acad Sci USA 2003; 100:13761–13766.
3. Berrington de Gonzalez A, Darby S. Risk of cancer from diagnostic x-rays: estimates for the UK and 14 other countries. Lancet 2004; 363:345–351.
4. Desmond AN, O’Regan K, et al
. Crohn's disease: factors associated with exposure to high levels of diagnostic radiation. Gut 2008; 57:1524–1529.
5. Sailer J, Peloschek P, Schober E, et al
. Diagnostic value of CT enteroclysis compared with conventional enteroclysis in patients with Crohn's disease. Am J Roentgenol 2005; 185:1575–1581.
6. Maglinte D, Bender GN, Heitkamp DE, et al
. Multidetector-row helical CT enteroclysis. Radiol Clin North Am 2003; 41:249–262.
7. Kandiel A, Fraser AG, et al
. Increased risk of lymphoma among inflammatory bowel disease patients treated with azathioprine and 6-mercaptopurine. Gut 2005; 54:1121–1125.
8. Infliximab [package insert]. Malvern, PA: Centocor; 2006.
9. Brenner D, Hall E. Computed tomography — an increasing source of radiation exposure. N Engl J Med 2007; 357:2277–2284.
10. Brenner DJ. Estimating cancer risks from pediatric CT: going from qualitative to quantitative. Pediatr Radiol 2002; 32:223–228, discussion 242–4.
12. Mettler FA Jr, Huda W, Yoshizumi TT, et al
. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology 2008; 248:254–263.
13. Buls N, Pagés J, de Mey J, et al
. Evaluation of patient and staff doses during various CT fluoroscopy guided interventions. Health Phys 2003; 85:165–173.
14. Thomas K, Wang B. Age specific effective doses for pediatric MSCT examinations at a large children's hospital using DLP conversion coefficients: a simple estimation method. Pediatr Radiol 2008; 38:645–656.
15. Cardis E, Vrijheid M, Gilbert E, et al
. The 15-country collaborative study of cancer risk among radiation workers in the nuclear industry: estimates of radiation-related cancer risks. Rad Res 2007; 167:396–416.
16. Mackey AC, Green L, et al
. Hepatosplenic T cell lymphoma associated with infliximab use in young patients treated for inflammatory bowel disease. J Pediatr Gastroenterol Nutr 2007; 44:265–267.
17. Lipsky PE, van der Heijde DM, St Clair EW, et al
. Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Engl J Med 2000; 343:1594–1602.
18. Palmer L, Herfarth H, Porter CQ, et al
. Diagnostic ionizing radiation exposure in a population-based sample of children with inflammatory bowel diseases. Am J Gastroenterol 2009; 104:2816–2823.
19. Levi Z, Fraser A, Kongrad R, et al
. Factors associated with radiation exposure in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2009; 30:1128–1136.
20. National Academies Press. Health risks from exposure to low levels of ionizing radiation: BEIR VII-Phase 2. 2005.
21. Goske MJ, Applegate KE, Boylan J, et al
. The ‘Image Gently’ campaign: increasing CT radiation dose awareness through a national education and awareness program. Pediatr Radiol 2008; 38:265–269.
22. Darbari A, Sena L, Argani P, et al. Gadolinium-enhanced magnetic resonance imaging: a useful radiological tool in diagnosing pediatric IBD. Inflamm Bowel Dis
23. Gasche C, Moser G, Turetschek K, et al
. Transabdominal bowel sonography for the detection of intestinal complications in Crohn's disease. Gut 1999; 44:112–117.
24. Signorelli C, Rondonotti E, Villa F, et al
. Use of the Given Patency System for the screening of patients at high risk for capsule retention. Dig Liver Dis 2006; 38:326–330.